ML19327C153
| ML19327C153 | |
| Person / Time | |
|---|---|
| Site: | University of Virginia |
| Issue date: | 11/30/1989 |
| From: | VIRGINIA, UNIV. OF, CHARLOTTESVILLE, VA |
| To: | |
| Shared Package | |
| ML19327C150 | List: |
| References | |
| NUDOCS 8911210025 | |
| Download: ML19327C153 (233) | |
Text
4 , ,
)
i r
li !
- ['
l,
. ,3 ,
I l
l l
)
h f,
I i
i SAFETY ANALYSIS 1EPORT ;
r-
?
FOR THE i
I I4W ENRICHED URANIUM FUELED t
i UNIVERSITY OF VIRGINIA REACTOR I
t l
'l
(" LEU-SAR") >
}
i h
t i
1 l
I
, NOVEMBER 1989 i l
i i.
l i
t^
i 7 l-
[
/~N '
[
I
- j. -
i
.(( k o '
g.
I2 ,
P 1
r~ ~N l
i
'\s -)
L i.
SAFETf ANALYSIS REPORT FOR THE LOW ENRICHED URANIUM FUELED UNIVERSITY OF VIRGINIA REACTOR I
l 4
(" LEU SAR*)
Contributors, past and present:
n
() J.R. Ball M.G. Bickel T.G. Foster V.R Johnson J.S. Brenizer R.U. Mulder M. Fehr S. Vasserman J.A. Dahlheimer J.L. Kelly R.A. Rydin P. Benneche R.D. Derry A.B. Reynolds D.W. Freeman J..P. Farrar J.H. Rust 'B. Hosticka This report is issued in support of a license amendment to License R 66 for the conversion of the University of Virginia's 2 MW reactor (UVAR) from high enriched uranium (HEU) to low enriched uranium (LEU) fuel. The report supersedes the original SAR for HEU cores and all its revisions and amendments as well as the so called "UVAR Design and Analysis Handbook", which was the heretofore updated SAR. The present SAR may be referred to as the " LEU SAR" to distinguish it from the previous SAR.
V .
E p.
I !
i i
/ x l
(j Table of contents i i I
k .
ZA&R i 1.0 Introduction 11 1.1 Summary of Previous Documentation 12 !
2.0 cenerel Descrintion of racility 21 l 2.1 Reactor Site 27 L 2.2 Reactor Building 27 l 2.3 New Construe. tion 27 !
2.4 Wind Direction and Velocity 2 13 l 2.5 Hydrology 2 13 2.6 Seismology 2 15 l
3.0 Reactor comnonents and control 31 [
3.1 Reactor Assembly 31 ;
3.2 Fuel Elements 34 l 3.3 Fuel Plates 39 3.4 Control Rods and Drives 39 i 3.5 Reactor Reflectors 3-12 l 3.6 Core Loadings 3 12 !
3.7 Fuel Storage Facilities 3 14 3.8 Reactor Data 3 16 :
p 3.9 Reactor Kinetics 3 20 ;
i") 3.10 Fission Product Inventory 3 21 3.11 Nuclear Instrumentation 3 21 (
3.11.1 General Description 3 21 j 3.11.2 Source Range Circuit 3 24 '
3.11.3 Intermediate Range Circuit 3 26 :
3.11.4 Power Range Drawer 3 29 l 3.11.5 Scram Logic Drawer 3 31 ;
3.12 Scrams, Interlocks and Alarms 3 38 :
3.12.1 Scrams 3 38 !
3.12.2 Interlocks 3 39 3.12.3 Alarms 3 39 ,
3.13 Automatic Control for Maintainitig Constant Power 3 40 ;
4.0 Reactor Systems 41 4.1 Pool 41 4.2 Filling and Draining the Pool 41 ,
4.3 Primary Cooling System 43 {
4.4 Measurement of Temperature DJfferential 46 !
4.5 Secondary Cooling System 46 '
4.6 Design Specifications 47 i 4.7 Water Purification 47
,. 4.8 Liquid Waste Disposal System 47 4
4.9 Building Ventilation System and Airborne Effluents 4 12 i
4.10 Core Spray Systen 4 26 O
v i l
l
~
y ~
}
j Table of Contents (cont.)
EA&R l 5.0 Exoeriment racilities 51 '
5.1 Baamports 51 i 5.2 Large Access Facilities 56 5.3 Rabbit Facility 57 i 5.4 Fueled Experiments 59 '
6.0 Rabbit Hazards 61 5 6.1 Confinement 61 6.2 Shiciding 66 6.3 Hazards During Normal Operations 6 13 ;
7.0 Health Physics 71 7.1 General Information 71 7.2 Education in Health Physics 71 i 7.3 Personnel Monitoring and Protection 72 .
7.4 Permanent Monitoring and Surveys 72 7.5 Prohibitions and Sanctions 73 7.6 Waste Disposal 74 7.7 Shipping and Transport 74 8.0 Administration 81 7s
() 8.1 General Organization 8.2 Radiation Safety Committee 81 81 -
8.3 Reactor Safety Comnittee 81 8.4 Procedures 83 9.0 Safety Analysis P.1 9.0 Safety Analysis 91 i 9.1 Thermal Hydraulic Analys'.s of the UVAR 91 i 9.2 Forced Convection Heat ".ransfer 94 9.3 Prediction of Incipien'; Boiling 95 -
9.4 Burnout Heat Flux 96 ;
9.5 Flow Instability 9 12 9.6 Burnout Ratio 9 18 !
9.7 Nomenclature Used in Thermal Hydraulic Analysis 9 19 9.8 Hot Channel and Minimum Core Loading 9 21 9.9 Allowance for Error in the Burnout Determination 9 24 9.10 Safety Limit 9 24 9.11 Limiting Safety System Settings and Measurement Errors 9 31 a) Coolant Inlet T6mperature 9 31 b) Flow Rate 9 31 c) Reactor Power 9 32 ;
9.12 Short Period Transient 9 32 9.13 Loss of Flow Transient and Natural Convection 9 35 [
9.14 HEU Analysis for Loss of Flow Transient 9 36 9.15 HEU Analysis for Natural Convection 9 50 ;
() 9.16 Maximum LEU 22 Puel Temperatures Following LOCA 9 51 a) Introduction 9 51 ,
b) Calculation of Peak Fuel Temperature Following LOCA 9 54 '
f b
?~ . . , ,
.t t
>'8 l t
'1 Table of contenta feant.)
g b 9.17 Emergency Core Spray Systen Analysis 9 61' 9.18 Time to Uncover Core Following a LOCA 9 65 n a) Flow Rate Without Frictional Losres 9 66-b4 b) Flow Rate With Frictional Lesses 9 69 c) Time to Uncover Core With Double Ended Pipe Break. 9471 l d) Time to Uncover Core With Crack in Pool Wall 9 72 b References for Chapter 9 9 73 i,
f i
i h'
=
0 4
l I
i iv list of Tirures Firure Eggg 21 Aerial View of Reactor Site and Immediate Vicinity ,
(1967) 22 22 Aerial View of Reactor Site and Immediate Vicinity +
(1964) 23 23 Contour Map of UVAR Site with Exclusion Fence 25 j 24 Map of Charlottesville and Vicinity 26 25 U.Va Research and Training Reactor Facility 2-8
[-
26 First Floor Plan of UVAR Section of Building 29 1
27 Mezzanine Level of UVAR Section of Building 2 10 28 Cround Floor Plan of UVAR Section of Building 2 11 ;
31 Cross Section View of Reactor foul and Reactor Room 32 32 8 x 8 UVAR Cridplate 33 33 Top View of UVAR Standard and Partial LEU Fuel Element 35 34 Side View of UVAR Standard and Partiel LEU Fuel '
Element 36 35 Top View of UVAR Control Rod LEU Element 37 36 Side View of UVAR Control Rod LEU Element 38 37 Nuclear Instrumentation System 3 22 i fs 38 Source Range Drawer 3 25 f 39 Intermediate Range Drawer
\' ') 3 28 '
3 10 Power Range Drawer 3 30 3 11 Scram Logic Drawer 3 32 3 12 High Power Trip 3 36 i 41 Cooling System Flow Diagram 44 42 Exhaust System to Stack 4 13 43 Spray Header Mock up 4 28 44 Core Spray System Elevation View 4 29 45 Core Spray System Plan View 4 30 51 Beam Hole Detail 52 5 la Top View of North Beamport Shielding and Access ,
control Walls 53 '
5 lb North Beamport Drain Fill System 54 52 Large Access Facility 5-5 53 UVAR Optimum Configuration 58 ,
61 Personnel Door 62 62 Exit Manhole 63 -
63 Pressure Tight Air Duct 6-5 64 UVAR Confinement Room Count Rate Data 67 65 Thermal Neutron Fluxes 68 66 Fast Neutron Dose Rates 69 67 Camma Ray Dose Ra es 6 10 i 68 Dose Rates Near Surface of Pool 6 12 8-1 Organizational Structure of U,Va. Reactor Faci 2ity 82
r i
i ,
I !
- i y i
'x ) List of Firures (cont.) ,
! t k
Firure EA&E [
t 91 Heat Flux Distributions 93 .
92 Calc. vs. Exp.; Incipient Boiling in the ORR 97 93 Calculated and Experimental Burnout Heat Fluxes 9 11 94 Core Loading Configuration (showing peak flux location) 9 22 !
95 Vertical Flux Traverse 9 23 96 LEU Core Power versue System (Core) Flow 9 30 9 6a HEU Core Pewer versus System (Core) Flow 9 37 '
97 Flow snd Power Coastdown (Header Up) 9 38 '
98 Flow Header Jammed in Cocked Position 9 42 -
99 Flow Reversal After LOF Transient from 3.45 MW 9 43 '
9 10 Comparison Betweon OVR and UVAR Corrilations 9 03 9 11 Maximum Fuel Temperature After LOCA, LEU 22, 4x4 {
9 60 ;
9 12 Power of Hottest Element 6 ECSS Cooling After Shutdown 9 64 9 13 Primary Piping 9 67 7
9 14 Ceometry For Calculating V2 9 68 i
4 F
l i
<-'s .I
~J i
1
(- _
L i
I L l t < s vi !
l !.
) List of Tables Page !
i Iahin t 2.1 Hourly Wind Speeds 2 14 2.2 Hourly Wind Direction {
2 14 l
' t 3.1 LEU.22 Reactor Data '
I. Typical 4x4 and 4x5 Core Parameters 3 17 i II. Fuel Element Parameters 3 17 l
[ III ruel Plate Parameters 3 18 IV. Side Plate Parameters 3 18 ;
V. Cuide Plate Parameters 3 18 i VI. Control Rod Parameters 3 18 I
- 1. Safety (Shim) Rods 3 18
- 2. Regulating Rods 3 19 [
VII. Feedback Coefficients 3 19- :
VIII. Kinetic Parameters 3 19 !
3.2 Prompt Nautron Lifetimes for LEU and HED t Fueled UVAR Cores
- 4.1 Heat Exchanger Specifications 48 4.2 Cooling tower Specifications 49 Secondary Pump Specifications 4.W t. 10
,_ 4.4 Primary Pump Specifications 4 11
( ') 4.5 Ar.41 Releases 4 18
'xd 4.6 [
Radioactive Cases 4 24 4.7 Iodine concentrations 4 25 i i
5.4.1 Exclusion Radius Fractional Exposures 5 14 f 5.4.2a Reactor Room.5 min, exposure 5 15 !
5.4.2b Experimental.5 min, exposure 5 16 i
9.1 LEU.22 Data And Parameters 9 27 l 9.2 CR Time to Drop from Predetermined Position 9 33 '
9.3 Maximum Peak Puel Temperature Following LOCA 9 59 ,
9.4 Chrracteristics of Core Spray System 9 63 .
9.3 Time to Uncover Core for various Leakage Mechanisms 9 65 i i
l '
I r i
e r%
\~- l 1
P i-l- t i
1.0 Introduction (v') {
The University of Virginia Reactor (WAR) first went into
(
operation in 1960 at a maximum licensed power level of one megawatt I under facility license No. R 66. The reactor core consisted of HTR plate type high enriched uranium (HEU) fuel elements. In 1971 ,
the authorized power limit of the WAR was increased to two megawatts. The operating license for the WAR was extended for 20 :
years in September of 1982. '
In 1989, high enriched to low enriched uranium fuel (LEU) :
conversion studies were concluded by the University of Virginia reactor staff. The NRC mandated conversion resulted in adoption of a higher fuel loading, with the number of plates per standard element increasing from 18 to 22, to maintain the operating ;
Q<m characteristics of the reactor. The present safety analysis report is an updattd version of the original HEU Safety Analysis Report, '
and for differentiation is called the Low Enriched Uranium (LEU)
Safety Analysis Report (SAR). This report was reviewed and f t
approved by the Reactor Safety Ceimittee at the end of the summer '
1 I
of 1989 ;
The WAR is operated by the Department of Nuclear Engineering .
and Engineering Physics, which is part of the School of Engineering l- and Applied Science of the University of Virginia. The reactor is i l
primarily utilized as a research and training facility of nuclear engineering students and for the generation of radioisotopes, neutron activation analysis, neutron radiography, radiation damage f 11
l f
g studies and other research. The reactor and experimental facilities l
<i
) .- are made available to the entire uni"ersity as well as to outside [
! i agencies under suitable contract arrangements. The facility is l
, also made available to students from other colleges and ;
universities in the state under a reactor sharing program sponsore.1 ;
by the Department of Energy.
1.1 Summary of Previous Documentation Docur.entation relevant to the WAR and Facility License No. R. ,
66 is summarized chronologically below. Most of these docner.ts ,
are on file under Docket No. 50 62, Lieunse No. R.66 in the 1)ivision of Reactor Licensing, Nuclear Regulatory Commission,
- 1) March 14, 1957, application to AEC for Class 104 license and construction permit by Colgate W. Darden, Jr. , President, !
f University of Virginia, and Lawcence R. Quarles, Dean, School of Engineering.
I
- 2) March 14, 1957, enclosed with the above application was ;
UVAR 3, "A Hazards Su'. mary of the Proposed Research and Training Reactor," by Lawrence R. Quarlos and Walter P. L 1ker. {
- 3) June 7,1957, Amendment to Hazards Summary submitted by ;
Quarles and Walker.
l
- 4) Fall of 1957 Construction Permit No. CPPR.15 issued, I
l l signed by H.L. Price.
- 5) September 23, 1958, Appilcation to AEC to convert Construction Permit CPPR.15 to a class 104 license, signed by Darden and Quarles.
12 i O
I c ,
l l
i I
f
(' !
l
- 6) September 23, 1958, enclosed with the above application was WAR 8, "The University of Virgit.ia Reactor, Description and Operation,
- by J .L. Meem. i
- 7) May 25, 1959, Amendment No.1 to WAR 8 submitted by Meem and Quarles.
- 8) December 4,1959. Amendment No. 2 to WAR 8 submitted by Mnem and Quarles. .j
- 9) February 5,1960, Amendment No. 3 to WAR.8 by Meem and Quarles.
- 10) June 24, 1960, Facility License No. R.66 issued and signed ;
i by R.L. Kirk.
i
- 11) January 27, 1961, Amendment requested to License R 66 for j the use of boron stainless steel control rods signed by Muem and !
Quarles.
- 12) April 13, 1961, Amendment requested to License R 66 to '
I permit irradiation of rare earths and uranium isotopes, signed by {
Meem and Quarles. l
- 13) May 17, 1961, Submission of surPl ementary information on l the use of boron stainless steel control rods, signed by Meem and P
Quarles.
- 14) September 1, 1961. Am.endment No. 1 to Facility License R. [
i 66 granted, authorizing the use of boron stainless steel control rods and authorizing the Irradiation of rare earths and uranium isotopes, signed by Edson G. Case.
O 13
_ ~ _ _ _. _
g _ _ . -
j l
l
- 15) Noveaber 21, 1961 Amendment requested tn License R 66 to l
utilize a concentration reduction factor of 500 for argon and noble 1
gases, signed by Quarles and Meta. I l
- 16) February 22, 1962, (UVAR 14), Amendment requested to License R 66 for the conduct of a broad irradiation program, signed l J
by Merin and Quarles. !
i i
- 17) April 26, 1962, Amendment No. 2 to Facility License R 66 I granted, authorizing the use of a concentration factor of 500 for l argon and noble gases, signed by Robert H. Byran. l l
- 18) August 23, 1962, Amendment No. 3 to Facility License R 66
- granted, authorizing the conduct of a oroad irradiation program, ;
signed by Robert H. Bryan, i
- 19) December 17, 1962, Amendment requested to License R 65 to '
i !
reduce the frequency of inspection of the control rods, signed by j Meem and Kuhlthau. [
- 20) January 15, 1963 Amendment requested to License R 66, for i the use of 250 grams of U 235 in a fission plato, signed by J.L. t r
Meem. ,
- 21) February 18, 1963, Supplementary information provided for January 15, amendment; for use of a fission place, signed by J.L.
Meem.
- 22) March 19, 1963, Amendment No. 4 to License R-66 granted to i reduce the frequency of inspection of the control rods, signed by Isbert H. Bryan.
14
.v,,,-y-- , - - . = - - - - - - - . - - - . , - - . - ,e
h 2
() 23) March 19, 1963 Amendment No. 5 of Facility License R 66
]
granted, authorizing use of 275 grams of V 235 in a fission plate, I
signed by Robert H. Bryan.
- 24) February 27, 1964 Amendment requested to License R.66 for i
the use of 135 grams of U 235 in a second fission plate, signed by ?
Villiamson and Meem.
- 25) May 18, 1964 Amendment No. 6 to License R.66 granted authorizing the use of 135 grams of U 235 in a second fission plate, signed by Roger S. Boyd. ;
- 26) Submitted August 22, 1967, UVAR 17, Safety Analysis in l sopport of amendmert of Licet6se R.66 for two megawatt operation. ,
- 27) November 18, 1968, Change No. I to License R 66 granted 1 authorizing round grooves vs. square grooves in the safety shim i
rods. l
- 28) January 29, 1969, Change No. 2 to License R 66 granted I authorizing the replacement of the weather recording instruments t with a wind vane and an anemometer, signed by Donald J. Skovholt.
- 29) June 4,1969 Amendment No. 7 to License R 66 which authorized the increase of U 235 inventory limit from 6.9 kilograms to 12.0 kilograss and increase in allocation of special nuclear material from 6.9 kilograms to 12.0 kilograms U.235, signed by ,
1 Donald J. Skovholt.
- 30) August 4, 1971, Amendment No. 8 to License R-66 to allow storage of 70,000 curies of Cobsit 60 in reactor pool, aigned by l Donald J. Skovholt.
O V 15
, ,v g -
I t
i l
[ 31) November 4, 1971, Amendment No. 9 to License R 66 to allow operation at 2 MW and incorporate Technical Specifications into the I i
lieer.se and to receive, possess, and use up to 7.9 kilograms of !
contatried uranium 235 for use in connection with the operation of l the reactor and receive, possess, and store up to 6.1 kilograms e,f j cor.tained uranium 235 not for use in connection with operatina of l r
the reactot, signed by Donald J. Skovholt. j
- 32) February 6, 1975, Amendment No. 10 to License R 66 to receive, p.ssess and use up :.o 14.0 kilograms of contained uranium- l 235 and 16 grams of plutonium in a Pu Be source for use in ,
connection with operation of the reactor. Also change No. I to the Technical Specification 5.1 describing the fuel elements used in 1'
the reactor, signed by Karl R. Go11er.
- 33) May 17, 1976, Amendment No. 11 to License R-66, Chango in Technical Specification 3.7 to clarify the uce of fueled experiments in the reactor facility, signed by George lear.
- 34) December 19, 1978, Amendment No. 12 to License R 66 to 1
change the requirement of visual inspection of control rods as stated in Technical Specification 4.1.C, signed by Morton B. )
j Fairtile for Robert W. Reid.
l
- 35) December 22, 1978. Amendment No. 13 to License A 66 to allow the receipt, possession, and use of 1.0 grams of Neptunium- !
i 237 in connection with the operation of the reactor. Also a change l in Technical Specification 3.5 regarding the exit manhole hatch cover, signed by Morton B. Fairtile for Robert W. Reid.
16 j i l 4
. , , - r 9 - - - -
--e -- -
- - - - - * - - - - - - - - - - - - . - - - - - - - - -~ -
~_
l
)
l 1
y) 36) August 25 M 81, Amendment No. 14 to License R 66 adding the Physical Security Plan to the License, signed by James P. l l
Hiller. l
- 37) September 30, 1982, Amendment No. 15 to License R.66 !
l renewing the operating license for 20 yeata, signed by Cecil 0.
Thomas.
- 38) February 25, 1986 Final Rule on Conversion to LEU fuel j given in 10CTR50.64, effective March 27, l')S6. ;
- 39) April 25, 1988 Amendment No.16 to License R 66 to clarify '
the possession, storage and use in the WAR pool of up to 70,000 f I
Curies of Cobalt-60 in the form of doubly encapsuled rod sources.
- signed by Lester S. Rubenstein. [
I
- 40) December 16, 1988, Amendment No.17 to License R 66 to change the Technical Specifications for mi'itmua shutdown margin so ,
that it is provided by shim rods only instead of control rods, sir,ned by Charles L. Miller.
i
- 41) July 20,1989, Amendment No.18 to License R 66
- i change ;
the Technical Specifications changing the organizational structure to allow the Health Physicist to report to the Department of Nuclear Engineering.
17 ,
O ,
- l. ;
i L j l 2.0 CENERAL DESCRIPTION OF THE FACILITY 2.1 Etapter site )
i The UVA reactor (WAR) is located approximately 2000 feet west of i the of Charlottesville, in Albemarle County, Virginia, llatitude 380 )
2'30" N, longitude 780, 31' W) and at an elevation of 700 feet. The .:
i Reactor Facility housing the reactor is next to an abandoned reservoir (
(pond), 200 feet up the ridge that runs between Mt. Jefferson and 14wis l Mountain. The pond has a watershed area of 10 5 square feet. The reactor building is approximately 50 teet above the water level of the pond. l North, east, and south of the site, no closer than 2000 feet, are ;
city residential districts, and 3/4 mile west over the ridge are suburban j i
developments. The downtown business district of Charlottesville is two I l !
t miles e ay.
Figures 2 1 (1967) and 2 2 (1964) are serial photographs of the j reactor site. Additional construction near the Reactor Facility accurred
't between 1964 to 1967, as seen in Fig. 2 1, taken in March 1967. I comparison should be made with Fig. 2 2 taken in 1964, on which newer construction is marked with an asterisk (*). The nearby buildings are: ;
Radio Astronomy Laboratory (RAL*), a group of dormitories (Dorm #), and ,
t the 5.5 MeV Van de Graaf (VdCe). The Auildings and Grounds divisions j complex (B C) (now called Physical Plant), the City Water Filtration
[
Plant (Filt), the Obssrvatory (Obs), and the buildings of the former R+ search Laboratory for Engineering Sciences (RLES), now Aerospace ,
Research Laboratory (ARL),wera all in existcnce prior to constructien of i-the reactor. At. .ddition i;o the reactor building was completed in 1970.
21 I
n O
l
,s ,
.- .,,, y, .. ,,
- ~
.-/ '; @ * " #*
.< . . .N . I -
9, ' , ,,,, -
l 3 f. - p .. ..
W, ,,
.',f l$
A
. :/ ; .
.g . c . .. -r, s.
- 3. .
Qf h3[L. &((jff ['R:.
pk,g I",, 3 .- a..
- ~( ,.
.- sy ,'
3.,
fe"" ( ,n , .
.x. , .? .. ,
m ..
/ -+.
. .. .j i N ^
- f
[ * - W-. , t. .
,,[ / h . .,fh. ,. ~; v ,,
s ... a
- e
- 5. .
s
.g fo . ss
@e+4 ' .u
' y 4 BO e
- ~ "
,.')'( .' ' ' /, d ['. ','% ' '. . ? " '3 * '-
i' W ,
i
- b '
- ' . g g ,, .
- t f '
6.
~
.A . ... L. > .. , .
.. . 'h .
x ,7 m&q l+
[(J ;" W Y. i p[Qo-
,!3 y\A syeyp 2% ,
.. y. ., . . . . -
U'
'.,-l.,.',lJ';:f.f g.y . rs j%Q_)Yfh .g ) *.,%. aj' $ .. ., \ ', , , ~ . -
Q* ^ a l' . i >
- ". -'~'
4 / N.; A / ' ;, , l .i. . p -4 g **,' ~ ' 3"
'ga.l,,i'(.gg.lIijg5'Jgd(':'f ,
i y'f' ',j. en -- f.
-y '-
.) .,.
(' h' #N' ' " ,[ l, , 8
, ARI. ,
"'/,g <
7 ., . ,. j; , s
*^
, j -
4 -. ,
, y#;', k, ., sf f*
"ej') *
~
*f'. )
p6- . ., a i .
.' # r s ..
,g .,-
, n , , ,+
FIGURE 2 -1. AERIAL VIEW OF REACTOR SITE AND IMMED1%TE VICll41TY (1967) 4 2-2
+
;. - = - -
=.. !
l i L = M W= ';
. .s ;
l', d ' f g -- . l
.-+ s . l
( _ _ .600 g erc ,,)'[. j [ } "" i
- Zv 3.4y f)
L
.. T ,
~
i -- l . l ' t 0 l i " } a m Eh _ _ "uc::= _ fleac t ,r
- a -
a~: ; t i
, -hy __
N'
- 7.,[ .
d L g 1
- d'!
g , r' ,- Tr e- .
- E
- - -- - TE.T.E. f'. 2... M4 ',d 'd J FIGURE 2-2. AERIAL VIEW OF REACTOR. SITE AND
;i
[- J I i L ! h l
,,-~ ,
The Van de Graaf building is approximately 125 meters to the f r L southeast of the reactor facility and the Radio Astronomy laboratory is l t about 250 meters to the northeast of the reactor, nose buildings are j occupied by technical people able to understand and respond to a reactor ' emergency. Farther away to the east, the student dormitories are at a nearest distance of about 325 meters. A copy of the Reactor Emergency , t Actions List is posted at the nearby Office of Environmental Health and f Safety (EH&S). The University police force is prepared to evacuate any ; of these areas if necessary. By virtue of its position in the draw, the reactor has a natural 4 terrain shield for approximately 270 degrees of its circumference, with ! t the elevation of the h m ily wooded slopes ranging from 215 acters at the ; i lowest points to 265 meters at the point of highest elevation. ; () d herefore, the UVAR is exposed for only a 90 degree sector from the northeast clockwise to the southeast, as rhown in Fig. 2-3. In the j easterly direction, the elevation drops rapidly, so that approximately ; 1200 meters from the reactor, the elevation if 150 meters. As shown in Fig. 2 4, residential areas are found to the north, east ; and south, with major business districts to the northeast and east at I approximately 2500 meters and 3500 meters respectively. The population f of the city of Charlottesville is estimated at about 50,000. ; 24 i t i i O
\
)
)
i O / /, i t l
/ ,
j s'60 ' non ,#', i'
,/ /, / l l / / i
/a s
/ ]
l
.' l
-- 750' ",,,. ..
,/ , . . < , [ / ( , M, m/ <
'/ ( .
- \ \
l
/ '
f 30 nr i , 650)/
/
p i , I i fy'-{ f'?sf/'h 610 ,[ An/z.:.c700j. ,n' , , , , , ,,3 e, g,, . V,.- 3 '
- c
,...,,(
"' " ' ** ' ' ~ '
l f
.'?h, '
l,'h W'N O ,,lg; . Q"
. ,/- .
.- c .
\
......-. . - /
a ... , ss f s s . 660-
/ ,r' f . s ,, s i s.
g , . <
/,,, N '. : ., *<
y-j
= . ;
.. . s . ,, ,
., . ,.-., .....---. / ;.~. - .
x s s s,
,'N :
s, ,
~ 660 ' - - ~ ~ ~ ~ ' .
K\ 's' ! 670 - .,
.. _ h ,- ' N [
. - ~ . # j
.. J f ~
T,N A !
.- 3 s l N ;
. - 690 - T.q .' ~
s, j l T- .\ l
,700 - -
,s.
# '* " h
.-710 - Scale 1: 1200
N 0 50 meters : _.,m - - - - N's s,
*- Contour Interval ..
N, i
---I N # 10 Feet ..., 'N ..
'N ,
m , s .__...._. ._.-.--. - s N ,
's ,.
750-- N. tw-8
- i. Figure 2 3 contour Map of UVUt Site With Exclusion Fence l -770 - .
' ' ~ ~ ~ ~ N., ,
~.
. I f
l 2-5
1 l O
~ / b y , .
. W P . ~ . ~) ,
j s Y^ a
. q.- -
g ' g- 3
~ ..
,7 mQ,
.r. / w (
r s,
- e. .
.- b 'c f . .
pp
~>~! ...
.e\ .;.:. .
m 3 ' 8 g -~ g,
.7 .,
V. .
\_l s s. .
.I
. .:n..
.. i O'. '
. . . , ,, u
, ._ v. . . ;
\
~) Y '.: ..f' '
^
.l..
k.,. * '- '
~
i
... 4 .. ... . . . . .
; w. ....: ; ... r.. . !
_u
, p, < .
l' .,hY d :"
' .' '. '- ll\ .', .,l, '
mp . .
.. :2 . ,, . . ,
,M. .
. - ... . . ~ ugw L. a.
\ ._ t 4 /
^'
- r:
~'
b N- l[h41['WDT
.i -
. .,m _6@ y p : fi_ ' . ' ' ;
/4, 6/ \ '. men snoiar c =....... ~
- i>
FIGURE 2 -4 POPULATION DENSITY DISTRIBUTJON (1968) (Each Dot = 10 Persons) O l l 2-6 1
- , , . - . . . . ~ . - . . . . - - - - - - , , - - - , - -
'the Reactor Facility has an exclusion fence as shown in Fig. 2 3.
This fence is approximately 70 meters from the reactor building in the ; l terrain ' unshielded" direction. Within a 600 meter radius of the WAR l l there are very few buildings and all of these, with the exception of ! I approximately a dozen privately owned homes just inside this radius, are j operated either by the University, City of Charlottesville or State of Virginia. , 2.2 Reactor Buildine The Reactor Facility buildinr,, shown in Fig. 2 5, consists of the - main reactor room, radiation laboratory, supporting laboratories, hot ! l cell and office space. Figures 2 6, 2 7, and 2 8 show floor plans for ! the three levels. The construction is of conventional masonry, with the
,O exception of the main reactor reon. This portion of the building is cylindrical in shape to increase its ability to withstand internal pressne. The walls are of reinforced masonry, plastered on the inside ;
for gas tightness, while the roof is a concrete slab. This portion of i { the structure is windowless and the doors are gasketed. The reactor bay i ventilation is described in Section 4.9. ; 2.3 New Construction k i l Construction of an addition to the Reactor Facility was completed in l 1970. The new addition provides more office space, classroom, machine shop, electronics shop, low background counting room, health physics and i student laboratories. , 2., , 9 3
, ,, , , , - - , - , - , . - , s~ . - -- -- -, ,, - - - , - - -- - ,
j i' O ~ a- . I a: __1s.8 , n
- ertrus .
Wa . ~W.R EBBBBB i sunfMF EkREJrE t Bg
- l l
L
%j% ,C> g=nBE.RE = , l l o t
?
rigure 2-5 University of Virginia Research and Training Reactor Facility. . L i; P l l ! i l O . l 1 2-8 i
..-.,y _ _ . . - . _ _ _ - ._
l l l
, .. q We me M.W e 9 *.8s .@. f. ,
l l
. ! e y
=1 I
g . .
.} ,r- - --
l
}4 '
I
.. . ,i r. . .), , )
s z i
. t f. g? <
t . ,1, J ! j i s i
) .
a ! [ / -
.c [
8 i ! f[.d.k s) i l ' -- s.,,
. .J n!
% - ~ '
iy a !
, , g. x \; . '*
t 0 ;
.Q ___ fW p..4..** l f T.. * '
'-~ -}-- -r --
Y
--- ~~-
m ! )y 0 l tI i A 1.! _ s f J ; g j d, , ; b,g.. 'n - I i ,
.n.+ -.m-,.-7 p .....
3 g e .: F. 1 i b k. 0- ~ .j j W l a W
- f. $; '
* ~
l
. 1 y..v 3.,.s." ,y . - . s. - >,.
k) 9n . i ! s
- i. .
.w!l
. . ei -
l s.*
, .4 1% C 't . A A
- w. N
,n
./ (t $" #
k s l;a an ,- p . _,.__.It 3g. w s,.3,.,L . q y~ 3 L (.).' 2 d a -- sh F ll i > ,J,j W) l 'NJ.\* ,' .. - Nd {. G,, .M
+E
- e,'
p--o. ::v .- - 1 .. _ g
% ?,,e - ,.,.,
- 1 11 1 1 I. p y
*** '. .s * " - -%" ~".-~*'
i m ..-= * * * **- e O f 2-9 t
-....---,,--...--..._,--_...--..--......-.._...__.---,-.__..,-.-_-.-_,_____._..,_m--.,_-.m.,-y.,_,.m .,_.-,----.-,,a
O ==
.o es.
'm ..eu. Lpe
.eist . .ais .* a .a w - .ns e . ,T l- -
,gm
, ii ii ,, *t
-pr
- I
,, e 1* , :
3 i c 7.- ~ w.8 =* .- w ,. .
. ..s -
4 7l ' :- . - : -
.?
@I .. f g .
a 1
=*
i P s
% 9 i T ]
, .., .,.s.. -- --
.; .; ., g
- - sa. - -
.l ,
J]'u=r.u. )
,. , .3 n 1
, [ 4 s 3 s..,;
1 . . . . s. 1,
- . N. <
r i ' 1r _O
.,., .. .~..
r 7.,.
, ,t .
c,
-g t-
- .7' a
.v . l .
g5 '
. . . . .:. ,. .t . 46
. . g.
- p. .
.; . g
., 8
/ , -
,,1.
t
.-~ - -- ,. .N
\ .- ,
R r *1 's
,.s . r,- .. y
- s
, ,7 g .a *;-
3 a _4_ 3 ~ e -
,7 4 .:
6 e
!!c w s1pL. .,..~+-
./ Ui
[ h,gA
, 41 d=j 4..}'
i i c
%. .).
'u g , ,y _ }im~ -= *L - @
I. , 1,2 - j..i c y ,, i: 3
. 4 ,. . ..
4.,. ,. g. .. u,... . . . p. . ( ! l
.i O
I-2 - 1 ') L
r a '
?'
- i. .
..$oo i
39.i. j ... e
)'.
Q i (T l
. .. .. . .s
.t. . )
, t )
'A j ** a :
'l't .t
- o. l 7
n a . ,3, i 4-a
)
am . . . . . 4 / ei.r.,__. :!.Ls . .p .,. .'- 5
.si 1 g 1 mis-m.. . :: ..q t ., .g .t
. - s, l e
.v . e 5 r:. . - J,
/ e :
A h $N w, g :' Gc , I, O : 17 . s' v .i .
\> *
, 'e - .9.s -
I. ..gr ! u i *t ,! , it -4, '
..... . '; . - .,4,w x,-,- }, '4.?L !
.?, , . '., I . i;"
d' r --a i 0 g..s. " g,-- ., rs i., w- 'p ... c - .e . M
. .?
i 1.
- O-l
~
.s nt
- J,. .., - s _i. , '4 .'.e .l' l .
.- e. .<x M D' s. _ .A ,
'r' / U- ,
!' ? 3.
8/ i '9.
- .,,, n .
't .m - --
..._{~.,/'\. <
- ..... .. g a ;i 4 0
.pl.g_
! y, v
, p.s. r..- .
._., . . *: a o 2,
i
..:. . s .?'- -
.? ! ,
o W- .
. . . tj'. n J.
s - M.
*; li .p. t
%.. . .e - r d;
- l a. .e e :
... . i
, - .o. y .. .
g
.t }, ,
I -
..-- .el 9,. ;, n, g - . - . . . . . . , ; ;
. , y- ,
m., .e a ,; i
- ,, i. ,
g 9 : i;4- 1
- e W..
o a '9 e r
/;- ' "
'!. . .", . N' 'e C :
.- ?
't (*
. . . i
- i e
[ t *' . .1 as i
..,, o .. .
l ,. s.. s. . .:\
. . ,e
. ...:,, %. 3
-i >
. .. .- : ..s.
..s.,
I i e . hy B . 1 . p .* . ,<, t .
? , '4' 1 . 8 :.
O . fr-~
- 2. . .8 . . .
g a am. ,.i g g ,q 8 .. ' 0 5 j '
. 4
_ ge' e , gs. . ' '
.e .e 2-11
l i l O ! t lu, ~ 1 (( _ _. ___ ._
% i a =:_J= r r .
g., --k.i #ee 13 d 'ur. T l l
$o4 %
- , , T~a, ,.jVld -
i !
+ *"""
.. I 1 h g '
i : [L I h
','.'{.9'
' }A)W,h, MT- n :"J-Y
f
,,.s..-
, t
-i m . ,,)rl a us'
. 1
=&
se
\
3i Oj a e w %,,_m_ .w.-. _. .__c .
~
FIRST TLOCR PLAN i {. _Il . yA.a:,; w ' y- o - r cTIra > (77 I pg7 . A i I
.7]"y-4 y y w w -
EICdi~'..1._.Td. r. Nu Eda B
._ . '.J_A s
,[, >. .
s
~ -c_ t i.> 3.. 4
, s_\
m
\,. 9qw p u=.--- -
.f
**N .1 .//
I< - __ - -. ' r ; o i, .r.c j
- :. ! J,' , ru3.-
- M "-
PcH F' -. - t-" C'1. 1 1YJj T,9 PMC
~
, . , .g.
i- :: : m.- _ . - - . MEZ.ZAWINE LEVEL
] l m.u snnn .
l!i..l jii f."',_D
~
- i F,. ,
? -
L. , t, ,a ' m . n j - mx t ,'
-r --
\ u+*
- :.lic::,,
s T a ! g Ili 1 3 wa IL
.T. 4Wm j. .A t-- n
- ) \; et.-up., m s, ,' i 7 ,
1 I' g
.I
. L y '. ~
Ll.. = = l . i *
- e
.I. t . e
-]
\
}*bMe-mas *, ,
;. j i s i q , a _
f
, (Austue( . I
'CL .-
Er d
, _ =,_ . _ _ _ _ _ y2. '; _ _ _
C UN 11 s .J .4 l',.d.J CROUND TLOOR PLAN FIGURE 2 -9 PLANS OF THE MELEAR REACTOR FACILITY 2-12 l
o - ~ v 2.4 Wind Direction and Velocity During the first few years of University of Virginia Reactor operation, records of w'nd velocity and direction at the reactor site were maintained. Tables 2.1 and 2.2 represent the equivalent of one year's data, from 21 June 1961 to 21 June 1962. For that year, winds from F the Northeast to Northwest quadrant are v daminent with a strong contribution from the West. The summer seaso.- n e a high percentage of calms with principal vinds from the Southeast to the Southwest quadrants. Therefore, gaseous effluent discharged or fission products released into the atmosphere would, with greatest probability, be transported in the direction of highest population density, except in the summer season where the reverse would be true. 2.5 Hydro 1ori As mentioned elsewhere (Sectico 2.1 and 4.8) of this report, liquid I effluents from the reactor building may discharged, upon dilution with Eu water from the adjacent pond to concentrations below NPO, to Meadowbrook Creek which flows into the Rivanna River. Releases are made in 2 accordance with the restrictions set forth in 10 CFR Part 20. Due to its location on the side of a draw between Mt. Jefferson and L 7,_ Lewis Mountain, the reactor is not subject to flood conditions. !!= 2-13 M 4 e
b.. l I i 1 O l i V TABLE 2.1 l l RELATIVE FREQUENCY OF HOURLY WIND-SPEEDS IN PER CENT BY SEASON l
, 1 Wind Speed j Type - aph ' Summer Fall Winter Borina Year Cala 0-1 k3 5 9.0 35 20 7 18.7 .
Light 1-3 15 3 2.0 1.h 8.1 6.4 ( Gentle k-10 28.0 10.7 1k.3 31.7 20.0 Moderate 11-21 11.6 33.3 61.8 37 9 36.3 Strong 21-up 1.6 45 0 19 0 1.6 18.6 Total 100.0 100.0 100.0 100.0 100.0 Avg. Speed (mph) k.26 15.0 13.1 7 71 10.36" The average vind speed of 10.36 mph is equivalent to h.63 m/sec. 1 er 1 ! TAELE 2.2 . RELATIVE FREQUENCY OF HOURLY WIND DIRECTIONS L IN PER CD*T BY SEASON Direction Summer Fall Winter Spring Year l- North 5.6 27.4 30.3 8.4 1f.9 ! l Northeast 7.4 10.2 23.3 24.5 16 9 East 57 0.0 2.6 0.0 2.6 Southeast 7.8 12.h 8.0 2.6 8.1 - South 10.6 3.8 3.0 10.2 7.8 Southvaat 8.5 10.3 4.6 50 77 West 3.8 15 9 12 5 92 10.0 Northwes. 71 10.9 75 10.4 11.h calms k3.5 91 35 20 7 18.7 Total 100.0 100.0 1.00.0 100.0 100.0 L 2-14 ,
- j l
l
, t ; . (~'\ .
( ,) 2.6 Seimmalorv Tha' Central Appalachian region is characterized by a moderate amount : of low level earthquake activity. Because of the low seismic energy release, this region has received very little attention from earthquake seismologists. A study by Dr. G.A. Bollinger of Virginia Polytechnic Institute covering the period of 1758 through 1968, indicates a history of 9 earth tremors in the city of Charlottesville and Albemarle County during that period. The tremors felt in this area and their intensity on the modified Mercalli Scale, when of sufficient magnitude for assignment, are listed below: . Intensity DAlt at Eoicenter August 27, 1933 VI 4 April 29, 1852 VI , N-- September 1, 1886 V VI December 26, 1929 VI April, 1936 Not Available February 2, 1937 III-IV May 24, 1946 Not Available March 26, 1948 Not Available September 10, 1952 IV May 31, 1966 Not Available November 19, 1969 Not Determined Considering the low level of earthquake activity and intensity in this area and the reinforced construction of the pool, earthquake activity is not considered to present a danger to the facility. 2-15
- O
,O I
,, _.. _ , . _ - - . - . . , . ~ , - . - - .
. ~. - - - . - - -. . . - ; 1 t
L 3.0 REACTOR COMPONENTS AND CONTROL 3.1 Beactor Annambiv The reactor assembly is comprised of fuel elements, control rod ; fuel elements, control rods and graphite reflector elements; all of which sit in the reactor gridplate near the bottom of the reactor pool. The i reactor gridplate is supported by an aluminum framework from a movable bridge. The bridge can be rolled back and forth across the pool in the r i north south direction and is designed such that the minimum distance between the core and the pool wall is about 4 feet. This distance is sufficient to prevent significant activation af the concrete walls. The bottom of the core is about 4.5 feet above the pool floor and the top of L l the active core is about 2 feet below the pool surface (see Figure 3 1). The heat capacity of the pool is sufficient for steady state operation at O 200 kW with natural convection cooling; but for higher power operation, a , forced convection system is required. The forced convection cooling system, described in Section 4.3, uses downflow in order to minimize nitrogen-16 activity in the reactor room. The reactor gridplate, shown in Figure 3 2, is made of aluminum and I
- i. contains an eight by eight array of holes used for positioning reactor i components. Each positioning hole is approximately 21/2 inch .'n diameter. The center-to-center spacings are 3.2 inch in the East West l direction, and 0.0 inch in the North-South direction. Small holes (not
- l. shown in Figure 3-2) are interspaced between the positioning holes to l
! provide cooling flow between reactor components. ! Reactor components include: 1) fuel elements, 2) graphite elements, 1 o h 3) gridplate plugs and 4) in-core experiments. L D d 31 l; i. l n . . .. . --
. . . . - . . . _ _ ~_ _. . . - - _ -__.
p , l l l D V-l
== 1
)
sc'.
. 1, I .
5 TON UNDERiluMG CRMt-e < i E'XHAUST AIR , REACTOR ROOM S6' 6( ( 5 4'- 0"- ) FRl5h ;
^* d -8 RIDGE RNL. BRIDGE
( l , gcyeggo WATER WLE.T 2 AIR a L-WATER l L EVE L.3 g , , FL (^ s ~~ N' --
~
h, ,
/ ................,
yg 6 m.t 1 , y . 'T f-CONCRETE a + l
. REACTOR 'l - SMIELb i e, foot. . ', ,
! p a .8 , 4, l s' N 19'* 9 == * # .*
- I j MOVABL E a
' d
- ME2 RANINE
'~~~*-~~
~',.-
l r, G AT E --1 **' 4 *{yC-
$ea l- 8, REACTOR ,a foop:
l ,, g,/ %E % ..= e6 8 wile OUTLET gggggq U: CROUptD FLOOR o _~
.i I. e a.4, I:fl
*2 # 1 e . Ili e, . n. . . / ~. a n
- A
- u. d i e p8 .
of 6. = r ' P4AINS j n :! 11 FROM Coouve PvMP TD Co0 LING PUV,? O V Figure 3-1. Cross-Section View of Reactor Pool and Reactor Room 3-2
, , , , , , _ _ , . , , . _ . , r. . . ,,
. _ _ _ . + _- _ _ _ _ _ , , , ., _ . , ..-,--,,,.,.-,-,wm,. ,,
r.N , 4 . _ . . . L s ?', - , x. t P P i II 25.6" >l t p. O!O!OlO!O!O!OlO
-+---+---.t--_+___+___+___,___,
O!O!O!O!O!O!O!O
---+---Y---+---+ _..;___4 _4 _,,
o; OlOlOlOlO!OlOlO
---+---+---+---+-__+___4___4.__, ,
N 010101010lO!O!O
---+-- r---+---+-__+___4___4.__,
O!OlO!OlO!OlOlO
---+---+---+_--+---+---+---+----
O!Ol0lO!OlO!OlO
-~-+-~~+-~~+---+-_-4___4___4___,
O!O!OlO!O!OlO!O
---+---+---+---+__.+___4___4___,
t O l O ! O l O l O !. O j O ! O 3Te
- 3. 2 "A m
~
Figure 3-2. UVAR Gridplate 3-3
- y. _ . _ _ . _ _ _ ..
l 4 r
- () 3.2 Estl Elements Thrse types of fuel elements s.re used in the WAR: 1) standard fuel -
elements, 2) partial fuel elements and 3) control rod fuel eiereents. Each type of fuel element uses generic fuel. plates (described in Section 3.3) and has similar outer cross sectional dimensitis of about 3 inch by
.l, 3 inch. The standard and partial elements are about 34 inches in height '
and the control red element is about 38 inches in height. The bottom of each element consists of a cylindrical tapered nozzle which fits snugly into the gridplate positioning holes. The standard low enriched uranium fuel element, shown in Figures 3 3 and .3-4, contains 22 fuel plates. The initial loading of a standard foal L element is 275 grams of uranium 235. The water gap between fuel plates l L is 0.092 inch. The metal to water ratio of the active standard fuel
- 1 I 'O element is 0.76.
Partial fuel elements have the same dimensions as standard fuel elements but contain only half the amount of fuu. Partial elements are (c. . l' loaded into the core when relatively small changes in reactivity are desired. The' partial element has 21 plates, of which 11 are fuel plates and 11 are aluminum " dummy" plates, with fuel and dummy plates alternating. Dummy plates have the same outer dimensions as fuel plates. , l' I The partial fuel element initially contains 138 grams of ursniuc 235. 1 Control rod elements, shown in Figures 3-5 and 3-6, have dimensions similar to standard fuel elements and use Lhe same water gap between fuel i plates. Tb' control rod element contains 11 fuel plates (138 grams of ur - . ~
~
and has an open center (referred to as the tater hole) where
. a u! . 7. travels. Two guide plates on either side of the water 3-4 l :
I + I* I . . . . - . _ _ _ . - . . . , _ - . . . , . _ . . . - . . -
- g. _
i,
;(b 1
(
/
-l
'(v) . '
y I 1 r, a t: i e f 2.9980* : e . tu e l. N G . en G t T ! G. - G .
< m l
1 l- - p , E _ _ m -
.:V <
in , G. 8 2.6210*
=
l , 0.1875* l l.' i. 1 0.0770* '
- i. i I-l I
V Figure 3-3. Top View Of Standard and Partial LEU Fuel Element .
. 3-5 o
f. t c' (' , , . -..m.- , ,, .---r,.. . -- -m. -a. - - - - - ..- - . . . . , ,v.. m
, c ,
4a 8 I
\.;
.5
)
t'
- Fii,ure 3 4. Side View of Standard and Partiti LEU Fuel Element i i
h a u a m l < . si . . . . . . . . gr i N fl f* t j,i ng i R
- ?
a
- 4 i' i e l [!!s. e i yr..
. 3 , _f .' c._: t.i
~ : 3
- e , ,$ .
'[# g .:.u1 _,
i1 r) O 4 'i b M [sl,
, .g
- I fo
, N i,i, ,,o r r i in,,,,,s II b.
8 U [i f
}.
e D. !
"I t ? h ! I l
'a !'" om
'h. [hlglf l ; I LL.LN l
d - ,,,.,e . . . . , r ; I e 7:: ! ! ! ! !.! il l! - i i i3
-g.
3, 1 gs
,; . , , - - . . n !,.i,pr !j.....-
r
.x 1 4 l
, s. .
t-
.i.
- s af lggllgp 4 Ei
' T* m e
e *i;. i('.g~, l j l. 44 a ( 5*
- a
' El Q o
.s l \g,t'
- l
'ls 44
( -- s . !.
.; I....f , ,l, o zg
> y. m u t -
#E gI k$ .
6.s'
, J
=0 ?
- 1 i
u, c ~ , e li bd g, J <y g2 t 4 ss s&, l .g:._ s n a
/ oc i
N Bs . b-" > '
% 42 Nil l,1 , % T$
e . . n ,
- e as i ; a ! fl Q g I *
- t <@
h l l 1 c a rt-1 ; s 9 x[ s [\ i i/ , 7sk ' fg , i Y
. . i -
v
/g\ Je l s. _,,_
- /g s -
4 ef I - != ~ .d w s
. Al ,
t
\' - s
's LinW ,
b,=l1 o Q c N ( . >
- ti ; 3 0, R - W
\ o .
3-6 _. __ .. .,. ._ __ . - . _ . . . ~
<= % m m. . , $.vuew .>c w 4
' <.),.
' l r s
- h. . +
i y s g ..7 g!L '\
.v:) J i
v: i e : l s I b
! w. ,
sn
,.,. y .;-e a W g 1
1 l l m E @ 2.9960* ; j .. . J , f, -- .
' s ,
l.
- m +
a u . v
s l N , '
- - N p l'a * ,-
i
, s ,
N
- l. en
.- m ; w g
2.6210* ' O.1875* B.0770' t'
.g -
.iu .
+ ' figure 3-5. Top View Of Control Rod LEU Element .
3-7
\
- , + -,,,---.m . . * , . _ ~ , - - - . , . - , , _ . .m. ._.,.. --v. , , - - -, r,-..4 ,
p- , , p , \: g, , Tigur3 3-6. -Sida Vi;w of C:ntral Rod LEU Elcme3t ! .- o I o a m i <
.- . - - a a - It t L., _
h,- < '{' P,g - k F' 1 Dq'l y ti! $ . I [j 0 9 p o , f. oge .a . , { j *[gj i "E i r, ,r 5 .l;p'il?i_
! i
-gl9
_! . ..s _4 , r , : . i $; o e, - r e r, . !f f 8 g i s l 6 l n y L_ t I ! f I 5 8 l E , l ' LLL N
- t. _I .t' k ex ,
r I LL , 1- "3 p /l= 3
; ;[ !
~
hh '-.! m!! m! $ $ $ .n .It ,j '(d 3
, e$ r l- hi, zN 4 i
' ~ * - - - - '
l l,,de.n.. l illillt n '>[ ' i , n r , v i m(lj <j
. I
- .s s'/ sh
~.N I i -
4 s s s~N w , 1
- 5 N, ii + 1 n;j
(_ i l
./
se l e
!.!,Ei a -
g _ .I ' :::: i >. ~ l i g uh ; I o 5
~
i
.i e: n
' t ; A Ek Nf k^I h 13 T~
syl / m 3- l i ' - m [ !"' 6 *I e [. e I;j , 4 -
!n -
?2 N
ej - 4[g ' h
- g. ig s ~~ bn
- sar l-- a l
. e w i SgJ s I
~
N f,h !"!, t > 4 'd d _ E!I *- gs !. r- '!;
!! ii!;!!I js (I-!8 e-jiII:fIf!
y- !
,s a
,i, 7 )
g I fj ;j r gg -4 c:a
. _ i l5 1isri.I -
l
@ @ 5 (v,'a i . sr
[ i; h g-
- w !!re l
r $ ../. g j
" I '
v4 :!!!!N li! 4 +- i
'\:
.'5 o. I u t m I <
3-8 i - 1
.,c -
j
r
*/
l D !
~
,-~
p g,;t hole assure that the control rod will not contact or damage' the fusi . 4 . plates. The metal-to water ratio of the control rod element is 0.52. 'l C' Detailed' dimensions of UVAR LEU fuel elements are available in References [1] through [4) . , t -- 3.3 Fucl Plates The fuel plate are flat and each initially contains 12.5 grams of r uranium 235. The plates contain low enriched uranium (LEU) fuel meat clad in aluminum. The fuel plate is 0.05 inch thick, 2.80 inch wide, and 24.6 inch long. Detailed fuel plate dimensions are provided in Reference
'[5).
L The fuel meat is composed of uranium silicide (U3 Si2 ) dispersed in
- an aluminum matrix. The uranium is 19.75 percent enriched in uranium.
235. The fuel meat is 0.02 inch thick, 2.40 inch wide, and 23.3 inch long.
- 1. -
The aluminum clad is 0.015 inch thick and surrounds the fuel meat. 3.4 Control Rods and Drives r The rea: tor has four control rods. Three ot these, designated as > Shim rods (or Safety rods) are designed ' n +; ass control and . safety. Shim rods are magnatically coupled to thoir drive mechanisms and drop-l into the core by gtavity on a scram signal . The fourth rod is a l regulating rod which is fixed to its drive mechanism and is therefore non-scramable. The regulating rod is primarily used to compensate for i small changes in reactfvity associated with normal operations. LEU cores l uce the same centrol rods and drive uschanisms used previously in the HEU l l 39 l
./N
\.ed l
1 l ' 1
L H t
' (O ' cores. Detailed centrol rod and drive. specifications are presented in References [6] through [8).
The reactivity worth of the shim rods in HEU cores have typice!ly varied between about $3 and $5, depending on factors such am , core / reflector configuration, rod position in the core, core burn up. . etc. Analyses presented in Reference [9] shov that the shim rod worths V for the LEU core are not significantly different from shim rod worths in HEU core. Therefore, the. shim rods should provide adequate control for safe operacion. The reactivity worth of the regulating rod in HEU cores has typically varied between $0.3 and $0.5. The analysis providej in l Reference [9] eows that the reactivity worth of the r*Sulating rod in the LEU cores is similar to its wort.a in HEU cores. Therefore, the l f regulating rod is expected to ac'equately perform its function of compensating for small reactivity chtages in LEU cores. All of the rods are of the bayonet type, fitting into the control , rod fuel elencnt water hole. The control rod, rod drive, c.nd extension assembly is bolted to the top of the control rod fuel element, thus creating a sit,gle rod unit. A rod unit may be located in any core position by locating the control element nozzle into the desired gridplate position.
, Ths absorbing section of the shim rods is boron-srainless steel, clad in aluminum. The stainless stael is alloyed with about 1.5% boron by volume. E ch absorbing section is 24-13/16 inches long ar.d has an oval cross section of 2-1/4 x 7/8-inchac with semi-circular ends. Four 3-10
o I i
+ ;
Q . groves are cut in each side of an absorbing section to increase the 1 surface area. .) l The shim rods are suspended magnetically frem the drive mechanism. ! The drive is provided by a 115 volt, 60 cycle, split phase synchronous motor. The motor, lead screw drive, and position indicating equipment ] are contained in a cylindrical tube extending from the top'of the core to above the vater level where it is supported from the bridge. A scrca l signal vill de-energize the magnet holding the absorber section, allowing it to drop freely until it is hydraulically damped and stopped, fully in the core. The supporting asgr.et must be driven down to contact the bottom absorber section before the control rod can be raised after a scram. The shim tods are driven at about 3.7 inches per minute in both
- directions.
! A U The regulating rod has the same overall dimensions sa the safety rod, except there are no grooves in the regulating rod. The regulating rod is made of stainless steel ar.d is clad in aluminum. The regulating rod is permanently fixed to itc drive mechaniss' and doos not dtop on a scram signal. It is driven by a 115 volt, 60 cycle two phase c ontrol ! motor which, along with the lead screw and position indicating equipment, is contained in a cylindrical tube similar ta that employed for the l [ safety shir rods. The regula*:ing rod is connected to the automatic cortrol system of the :eactor described in Section 3.12. The regulating rod travels at a speed of appro.imately 24 in/ min in both directions. Each shim rod is removed from the reactor and visually inspected on an annual basis. The inspection includes checking for cracks and swelling, s. 3-11 1 e -- -
- . , . . ~ . . , . . . _ _ , . . . _ . , _ . _ . . . . . _ - . , , . _ . _ . _ . . _ , _ , , _ _ _ _ _ , , , . , , . _
~. _ , - _ _ _ . . . . _ . -
m27
- l. ?!
ll ' l ', r l L l' '(- Rod drop times are measured semi annual'ly and whenever a safety rod is' maintained er repositioned in the core. The maximum allowable time
, 'from scram initiation to full insertion is less than one second.
3.5 Reactor ReflectJgg The primary reflectors used in the UVAR are graphite elements and poo1 water. Experiments located near the core may also behave as ! reflectors, r Graphite elements have che same approximate outer dimensions as the fuel elements and consist of a solid graphite core surrounded by , aluminum. Grsphite elements are significantly better reflectors than pool , water. Gridplugs are used to prevent water flow through empty gridplate locations. A gridplug is a short metal cylinder, approximately 3 inches
- 1. diameter, mounted on a tepered nozzle. When inserted in the gridplace, the plug extends no more than a. few ine'nes above the gridplate effectively providing pool water reflector at that gridplate location.
Other types 3f reflectorr include items such as experiments or experimental facilities, located in cicse proximity to the core. ; 3.6 Core Leadines A wide variety of critical loadings are possible with the UVAR reactor. Core loadings are limited by Technical Specification restrictions on shutdown margin (0.50$) and excess reactivity (7.00$). The minimum critical loading is a graphite reflected four-by-four array of elements, including 12 standard fuci elements and 4 control rod elements. This loading has a mass of 3850 grams of uranium-235. 3-12
~
- c., . - a
F ,
]
1 l i n ; 1 1 I Excess reactivity and shutdown margin are functions of items such as 1 the amount of fuwl in the core, fuel and reflector configuration, control !
- n. 1 rod locations, fuel burn up, etc. The core configuration of the UVAR is l frequently changed to enhance :haracteristics of experimental facilities, i perform new experiments, or compensate for fuel bur" up. Shutdown targin and excess reactivity at. experimentally determined after each core .
configuration change. Additionally, shutdown margin and excess! reactivity are periodically reevaluated to account reactivity changes t essociated with fuel bura up. Initial loading of a new core is performed carefully. Sub critical multiplication data is collected with the addition of each fuel element. ! Analysis of this data allows a fairly accurate prediction of wh6n initial f-- criticality will be schieved. Final fuul additions are made in half-element increments until a desired core is achieved. Operation of the new core is limited to 1 kilowatt for the purpose of obtaining ! experimental data to calibrate the control rods. The shutdown margin and ; excess reactivity are determined from the rod calibration data and a j- determination o# acceptability of these parameters must be made prior to , i operating the new core in excess of 1 kilowatt. Analyses presented in Reference [9] show that the reactivity of unburned LEU and HEU fuelea cores are similar. The loading of LEU cores is expected to result in core reactivity parameters similar to those verified through past experience with the HEU cores. It should be noted that L"M 22 plate / element fuel is expected to have a nomewhat longer ccre life than 18 plate / element HEU fuel. This
.' 3-13 L , -; ,, -
t 1 i 3 l l
/ \
V should result in fewer core modifications to accomodate fuel burn up, j 1 which is beneficial with regards to experiments and fuel costs. 1 l
, .3.7 Fuel Storama Facilit:ita )
WAR Technical Specifications require all fuel elements not in the reactor core to be stored in a geometric array such that the 'k-effective 4 is less than 0.9. Existing fuel storage facilities at the WAR Reactor Facility have been evaluated for compliance with the Technical J
, Specification reactivity requirement, with regard to LEU 22 plate / element fuel.
The current fuel storage facilities are described below:
- 1) Fuel Storage Room - The Fuel Storage Room is located in the l CAVALIER Room and consists of an aluminum rack with a rectangular array of square holes in which fuel elements may be stored. The center-to center distance of the holes is 12 inches in both directions. The Fuel Storage Room is dry and is 1
located above ground. The Fuel Storage Room is primarily used for the storage of unburned fuel elements. Elements in the Fuel Storage Room are secured with chains bolted across the u rack openings. l L
- 2) Auriliary Fuel Storage tack - The Auxiliary Fuel Storage I Rack (AFSR) is normally stored at the pool bottom, The AFSR
.contains two linear arrays of fuel storage positions. The two j
linear arrays are separated by a distance of two feet. Each linear array consist of twelve fuel storage locations, i E separated by a center-to-center distance of 6 inches. Elements c O 3-14 t t
-r --,-...%- -
.-g- - . ,, -, p - rv c- *
, _ . _ . ..- - . - - .. ~
3- < - 3
, r fc5 j ;
' ji'i ,:( 9 ' : in the AFSR are mechanically held in place by the rack lids'
^
which may be bolted shut. (
- 3) Four Element Racks"Lumsr Landers" - There are currently 3
- Lunar Landers. Lunar Landers loaded with burned fuel are stored 7 at the pool bottom. The Lunar Landers each hold up to 4 i elenents in a square configuration withithe elements being .
loceted at the corners of the square. The center to center
'discance between elements in the x-y direction varies from 7 to 8 inches. Elements are secured by bolting an aluminum-top onto the lander.
- 4) Wall Rack - The Wall Rack is used for temporary underwater 9torage of fuel elements. It consists of a linear array of 12 fuel stotage locations. Each fuel s:orage location is
.i separated by a center-to-center distance of 3.5 inches.
An analvsis applicable to the Fuel Storage Room, Auxiliary Fuel j' l Storage Rack, and Lunar Landers has been performed using two dimensional , diffusion theory computer modeling. A unit LEU fuel cell surrcunded by ' water was modeled in x y geometry. Flat flux boundary conditions were used to simulate an infinite array. The k-effective of a water moderated a-infinite array of fuel was determined to be 0.8 for a center-to-center , distance of 5.5 inches. This analysis is conservative in that 1) it i neglects leakage from the boundaries of a storage facility and 2) the value of k-effective - 0.8 associated with the 5.5 inch spacing is significantly lower than the recuired value of 0.9. Based on this information, LEU-22 fuel storsge facilities with an x-y storage array with center-to-center spacings in excess of 5.5 inches meet the Technical
' \,/
Specification reactivity requirement. 3-15 I r*, e' *b + 4- -,
~
[1 ' i
!r '.. '
q l i l l'~)% (, An analysis applicable to the Wall Rack (and to the Auxiliary Fuel j Stort.ge Rack) was performed for an infinite linsar array of water moderated (and reflected) LEU fuel elements positioned side by side, a J (i.e. no separation between elements). The k effective for this array is ]
\
0.74, which also meets the Technical Specification criteria. ]
!' In summary, modeling results show that all of the e.xisting fuel i
storage facilities meet the Technical Specification reactivity criteria when loaded with. LEU-22 plate / element fuel. 3.4 Reactor Data For ease in reference, pertinent data for the UVAR reactor are v
- presented in Table 3.1. Information detailing the UVAR LEU 22 fuel
~
elements, fuel plates, and fuel meat is provided in Reference [10). E
%)
i. 1 l l 4 i r k 3-16
~
, , . , . , - - - +.--,-.n... , - . , ,.. - - . - , .n,, , , , - - , , - . - . , - _ . . ,- , - - , , ,-
l \O:
,r j
/~N .
9x_) Table 3.1 Reactor Data for LEU Cores j I. Typical 6 by 4 and 4-ty-5 Core Parameters i 1 4-by.4 Core 4-by 5 Core Parameter Confinuration confinuration ;
- 1. Core Dimensions i 23.3 inch 23.3 inch j a) Active height b) Width 12.6 fuch 12.6 inch l c) Length 1?.0 loch 15.0 inch l d) Volume 3,520 inch 3 4.400 inch3 '
2 Numbe. of Elements a) Standard 12 16 b) Cone.rol 4 4
- 3. Mass ?J 235 3.85 kg 4.95 kg :
- 4. Metal-to Water Ratio l -(by volume) 0.70 0.74 II. Fuel Element Parameters
,_s Ear.a meter Value :
l:
'-- 1. Outer Dimensions l
a) Height , 34.4 inch b) Width 3.14 c: Length - 3.00 ! l 2. Water Cap 0.092 inch
- 3. Standard Element
- Number of Fuel Plates 2
- Number of Side Plates 2
- U 235 Content 12.5 gram
- Metal-co Vater Ratio 0.76 ;
(by volume)
- 4. Control Element ?
- Number of Fuel Plates 11 L
- Number of Sice Plates 2
- Number of Guide Plates 2
- U-235 Content 6.25 gram L - Metal-to Water Ratio 0.52 3
- 5. Partial Element
- Number of Fuel Places 11 i
. .iumber of Side Plates 2 i
- Number of " Dummy" Plates 11 l
's
- U-235 Content 6.25 gram !
- Metal-to Water Ratio 0.76 R'
l 3-17 l l L 1 l- \ 1 .'
, . - ~ . ),b. ;+ <
1 i I III. Fuel Plate Parametera
~ ' ' Overall Dimensions j 1.
a) Length 24.6 inch , b) Vidth 2.80 inch. ) c)' Thickness 0.050 inch j i
- 2. Fuel Heat a) Type U3 Si2 in Al matrix b) Uranium Enrichment 19.8% U 235 )
c) U 235 Content 12.5 gram 2 d) Length 23.3 inch :. e). Width 2.40 inch ~ f)' Thickness 0.020 inch
- 3. Clad Thickness 0.015 inch IV. Side Plate Parameters l Length a) Standard 28.5 inch b) Control 30.9 inch Width 3.14 inch '
' Thickness 0.188 inch Composition Aluminum ,
V. Guide Plate Parameters l ,- i Length 27.6 inch
.j ' Width 2.80 inch
' Thickness 0.125 inch Composition Aluminum VI. Control Rod Parameters l
' 1. Safety (Shim) Rods Absorber Material Beton-Stainless Steel 1.5% Boron f L' Aluminum
, Clad a) Dimensions, Overall Width (Approx.) 1 inch
. Depth (Approx.) 2.38 inch L Length (Approx.) 27.5 inch Travel (Approx.) 24 inch Weight (Dropping Section) 5.5 inch b) Drive - Electric motor, 115 V 60 cycle, split phase. t 3600 rpm at 60 cps. From top by lead screw. c) Drive Speed ).74 in/ min d) Release - Magnetic; after release, mechanism must be driven down to re-engage absorber e) Typical reactivity, fully inserted $ 3 to $ 5 < f) Typical reactivity per inch $ 0.1 to $ 0.3 () g) Typichl rate reactivity increase in up travel (per second) h) Excess reactivity controllable
$ 0.01 to $ 0.02 '
with all rods, typically $ 9 to $15 3-18 E
.'9,.
*> ' 4 w eg - - 9- y e+yw ww v--ew w * -w---v,---+-*--,,m-m--w w-ew-----en- -ew.-e- v w-- -*=v= v i www m.-,weewww- - * - = *r-w'imr--<
^
{ '
,b '
l i i L i a
.,..e ".,<- j
- i. ,j 2. Regulating Rod ;
a) Abscrber Material Stainless Steel i b) Dimensions Same as Shia Safety (No grooves) I fi c' Travel- 24 in. l c Release None - does not drop on scram e) Typical Reactivity t , 7ully inserted $ 0.3 to $ 0.5 3 i f)- Typical Reactivity per in. $ 0.02 L g) Typical rate of Reactivity increase in up travel (per $ 0,01 i I VII. Feedback Coefficients (Reference (9)) t- 1. Doppler Coeff ($/ 0C) 1.5 x 10*3 l
- 2. Void Coeff ($/t void) l a) Uniform void (1 to 10% void) 0.3 <
1 b) Local void (1 to 14% void) 0.8 ,
- 3. Moderator coeff ($/00) 0.02 VIII. Kinetics Parameters (Reference (9)) .
l l ! _s 1. Prompt Neutron Lifetime 53 to 67 alcrosecs L ,
- 2. Effective Delayed Neutron 0.0074 !
Fractions 1 1-L 1. 9 1 L l '. l. 1 l' 3-19 l. p 1 o i
> ,. . . . , . , . . , - - . . , = , , . . . . - . , - - - . . . . , - . . . . . , . . . . - - , , , , - - . - - , . . , . . - - . - . . - . - . - - . . , , . - - - --
PE : , . ; i I e i- / 1 1 () 3.9 Ranctor Einetics ) An antlysis determining important kinetics parameters (i.e. prompt i
. neutron lifetimes and delayed neutron fract ons) i for the LEU fueled UVAR core is presented in Reference [9). This analysis provided a comparison i
of kinetics parameters associated with LEU and HEU fueled UVAR cores. Results of that analysis are discussed below. L Prompt neutron lifetimes determined for both LEU and HEU fueled ; UVAR cores are presented in Table 3.2. Both graphite and water reflected 1 i L corcs were esalaated. 1 J Table 3.2. Prompt Neutron Lifetimes For LEU and HEU Pueled UVAR Cores (Reference (9)) Prompt Neutron Lifetime (psecs) Core Descriotion LED HEU-18 l, L
) 1. Graphite Reflector 67 79
- 2. Water Reflector 53 64 i
l l Information presented in Table 3.2 shows that the prompt neutron
- 1 L lifetime of the ',EU core is about 15 t lower thnn that of the HEU core.
This is as expected because of the significantly higher uranium 238 > L ' loading associated with the LEU fuel. Because the prompt effect is , relatively small in the normal operations of the UVAR reactor, this difference is not expected to produce a noticeable change in the UVAR response to reactivity changes. The effective dslayed neutron fractions determined for both the LEU and HEU fueled cores were found to be essentially the same, at a value of L , about 0.0074. Ls
- ~
3-20
. c lD;;;
a rN
- 1) Based on the information provided above, UVAR LEU cores one expected ,
to behave' essentially the same as past HEU cores, with regard to reactor kinetics, 3.10 Fission Product Inventory Reference [11) states that the total inventory of fission products for LEU cores will not be significantly different from that associated with HEU cores. Additionally, because there are 22 plates in each LEU eleoent as opposed tc 18 plates in the HEU elements, the inventory per plate will be less in the LEU cores for the same operating histories. 3.11 Eggigar Instrumentation K. 3.11,1 Ceneral Descr1 9112n The UVAR nuclear instrumentation consists of those components ' necessary to monitor and display the operating parameters over all ranges , t of operation, from start-up to full power, and to automatically terminate I. N ! L L operation before any limiting safety system aceting is reached. The 1 overall system is shown in Figure 3-7. . L The Source Range contains the circuitry necessary to monitor reactor 1 power level and period from shutdown through six decades of power level L, increase. The circuit utilizes a fission chamber as a neutron detecting device. The fission chamber is movable by use of a switch on the 1 1- console. It is moved out of the core after start up in order to minimize the burn up of U-235 and buildup of fission products in the chamber while operating at high power levels. Both power level and period measurement l L are displayed on reactor console meters and the power level is repeated i on a chart. The Source Range instrumentation also prevents rod
;- withdrawal unle=s minimum source counta are present.
l' N~ 3 21
*h- 9 9 -'w- y .-w ,my., .e_. _
\ $&t E lk/o IMAGE EVALUATION //jg/'j [>4, y TEST TARGET (MT-3)
'4 4)'$,/;:
s' + '%,j[M,4 l.0 !!En EM [lj RM l-l E5: L: HM 1.25 ' l.4 1 1.6 l l_
< 150mm >
< 6" >
46 hyffyr
\zf /f4%
+$ . .
<pf4%
%eu v
m .__
%fs>th>i'*
f IMAGE EVALUATION pkg, g/ y >;,
,,///// , <> +
1.0 lf m na pllj I!!!E i.i [i= IlllE I.8 l.25 1.4 1.6 4 150mm > 4 6" >
- 1;fA,m,,
< $4
~
4);3:!;g+ 77777 D/ L '4 u ;
.@ $la l
+ * \\A IMAGE EVALUATION 7//j M TEST TARGET (MT-3)
*,[f <k, '
gg)77litp'/g/// p,f 'b,4;,y$*4 ///
+4 %
l.0 J' 2 Eaa 5lNE j I,l [t: Ele ) Ja l 1.25 1.4 1.6 ! 4 150mm > 4- - yo -> GL %,, e s 4% 4;gy%>77777-s
<Qf4%
L u%:w_ s 1 v __j
$ $* 4k* l Ylho O. i? IMAGE EVALUATlON [gf\ 4 yg+ TEST TARGET (MT-3) <e
#4:
4)R>,/ %NY , l.0 lf 2 m l y ,* D E l I.I tu = Illla i.25 n i.4 i.6 Il 4 150mm > 4 6" >
*%s /A
+>[ff>//,a;s<g<y#
$7777s .
o
.w n -- - > w
<s
-o ,
NUCLEAR INSTRU!EfRAT10N SYSTEM
' . r#. FIGURE 3-7 Source Huncti t
LUEL SOURCE PDIOD Mf h PRE-AMP 7 1 - RANCE j RECORDER' l l . 1 ff.Y. PWR. SUDPLY
! ! -# l[
f-
. Linear Power M
LYNEAR REO ROD Colf 3 MODE w y, pg, _
~
10N RECORDER CONTROLLER HONR c cvav m l o i t'PP;1 LEVEL SET M l SCRAM 14GIC SCRAM ~ j Power Rance = ' r DRAVER sm KAC!lE'lli ) H.V. FWR SUPPLY l/l -am l
,o l CitANNEL 1 PROCESS SCRM4S
- 7P M ,
GER p CMANNEL 2 1,
' ION 4 yp p p B.S.
I CitM!BER TRIP t i N.Y. PW. SUPPLY l/l l/l ] B.S. Intermedinte Ronce gIL . n INT. l COff. ION !.EVEL PERIOD RANCE CllAMBER MIP Mf RECORDER l a 1 l -1 I I l Colf & d.\ l ,A l l / l l/l / I
'WR. SUPPLY l
Core Cn-an N-16 1 ION - up MM. IO:: - up MM. CKM2ER CHM 3ER t I t I ---- P$ .SUFPLY ! P$ SUTTLY 1 I
\ ,
1 l l L 3-22
y : 1 p l 3
,\_f) , The Intermediate Range instrumentation receives its input from a .
v compensated ion chamber and provides indication of power level and period I over seven decades. Both power level and period measurement are displayed on the reactor console and power level indication is repeated on a chart. This instrument provides protection against a too rapid ( Figure 31 7. Nuclear-Instrumentation period by scramming the reactor if l.
" the period is too short.
The Power Range instrumentation contains two completely independent power range channels, each of which indicate reactor power over a range of 0 to.150 percent. Each channel is supplied with an input signal from an independent uncompensated ion chamber. The output from these channels are displayed by separate meters on the reactor console. Each of these g channels provides independent scram protection from high reactor power.
\'
There is a range switch to select full power indication of 2MW when in l forced cooling or 200 kV when in natural convection. , 1 Indication of reactor power is also provided by a linear power instrument over 9 decades of reactor power through a range selector switch. This instrument receives its input from a compensated ion , chamber and its output is displayed by a meter and a chart on the reactor console. The linear power channel recorder indication also serves as the l' sensing element for the automatic control system (see Section 3.12) which o l operates the regulating rod to maintain a set power level. l. In addition to the above instrumentation, the following indications l .. 'are provided for comparative observations of reactor power. An ionization chamber, located on the ground floor in the heat exchanger room (adjacent to the primary piping), is used to detect gamma l' (o,) 1 3-23 1 1
+ -
, - - - ,,,--,. . _ . ..-c...,-. - - - - - .. . . . . - . , ---- ,.--,,. ,---..- - , . - . , , ,
]-jn ,
/ ac
( ,f radiation from the decay of nitrogen-16. The signal from this detector is a function ~of reactor power and is displayed on a meter and a chart at [. the reactor console. An ionization chamber, referred to as the core gamma monitor, is suspended at a fixed position typically about 7.0 feet ab'ove the reactor ,
- ' core. The signal from this detector is displayed on a picoammeter on the secondary conrole and is a measure of core gamma flux. The core gamma monitor is provided for comparative observation of power and has a floating point alarm that sounds on the common alarm panel. ,
3.11.2 Source Ranne Circuit The Source Range Circuit (see Figure 3-8) contains circuitry required to monitor reactor power level and period from shutdown through six decades of power level increase. The circuit utilizes c fission chamber to detect neutrons. . A combination high voltage filter and pre amp, mounted on the , detector support pole, provides final high voltage filtering and a two atage pre amplifier in one module. On entering the drawer, the input ' si 5nal passes throu6h a pulse amplifier, a discriminator, and a scale of two counter. The discriminator has a fixed discrimination level of 2.2 l volts. Actual discrimination can be changed by varying the gain of the preamplifier or the pulse amplifier that precedes the discriminator. The scale of two counter divides the input frequency by two so the output is a square wave whose frequency is one-half of the input signal frequency. Log integrators A and B provide a DC voltage proportional to the log of pa the input frequency. O 3-24 t 1
*w ,,.r_ ~w., y . . . . . , . , - ~ , - .
m.-w-,, ,,me.v., . -- , - , %- e- . , . , - .-y - - 9 -,-- ,-
o o i O- r FIGURE 3-8 S0 W CE RANGE DRAHER i t P.S. P.S. OSCIL-P025 NO25 LATOR i 200 200 10 KC } S. C R 800 Y 1r f
- 25 VDC 25 VDC 10 KC 8 6Y 4 PWR SUPP 4
i LEVEL 1 METER
*
- RECORDER i
- w I i e i y - -
FILTER
- j. PULSE DISCRIN SCALE LOG LOG AMPLI BI-
---* AMP INA70R . OF
, INTEGRA . [NTEGRA FIER 4
3 w- PRE-AMF PfABM : l TWO TOR A [DR B -S- WIP IN COURT RATE INTM* 2 CPS i i t j NElfrRON DIFFER ' M 4- AMPLI- DDDD. j DETECIDR INTEG. TUR
+
FIER Pb p = * ) B 1 4 FISSION I l CHAMBER i 4 l i i
- METER i
i _. - .- , - , , , _ ~ . _ _ _ _ _ _ _ _ __ _ _ - _ _ . _ _ _ _ _ _ _ . _ _ _ _ _ _ _ _ _ _ _ _ . _ _ _ _ . . _ _ _ _ _ _ _ . _ . _
(.s ,
, ._,u___.
l.
/
') Amplifier S is a DC Amplifier whose output provides source range log Rj level, indication,' level bistable trip logic, input to the source range period second, and remote recorder output.
The' period.section-of the Source Range Drawer consists of a differentiator integrator (D-I) which converts the level input from Amplifier S to signal proportional to the rate of change of reactor l' power level. Modulator P' combines the DC input from the D-I with a 10
, KC input from the Or.cillator module. The resultant, a 10 KC signal l 1
proportional.in amplitude to the DC input signal, is applied to amplifier l B. Amplifier B is an AC amplifier which amplifies the signal from Figure 3 8. Source Range Drewer Moderator P and feeds it to demodulator
'P4 where the AC signal is converted back to a DC. signal proportional to I (m) the exponential rate of change of reactor power. The Demodulator P4 ,
l l feeds a front panel period meter. 1 1 The bistable module installed in the Source Range Drawer is a solid state multivibrator circuit which prevents rod withdrawal in the absence of a sufficient count rate. The output is a logic signal of 0 volts (tripped) if the count rate is less than 2 counts per second, or 10 volts i (untripped) if the count rate is greater than 2 counts per second. The Source Range Drawer contains a test module that performs alignment and operational checks of both the level and period sections without the use of additional test equipment. 3.11.3 Intermediate Ranne Circuit The Intermediate Range circuit (see Figure 3 9) receives its input from a compensated ion chamber (CIC) and provides log level indication
.r"% J
'd 3 26 i
( .; , t
. , over several decades, period. indication and scram and control logic L_)
- outputs,
.The power supply provides 200 800 volts DC high voltage to the CIC with regulation as'close at 1 0.1%. A variable DC power supply provides l variable compensating voltage to the CIC.
The DC currunt proportional to neutron level from the CIC is fed 'I first to Modulator L where it is dropped across a string of forward biased diodes to a voltage proportional to the Log of the current and is combined with a 10 KC sine wave. The resultant is a 10 KC signal proportional in amplitude to the Log of the DC input current. The Demodulator, as the name implies, converts the AC output from Amplifier B back to a DC signal which is proportional to the Logarithm of reactor
.r m%
%)
l l. l-l l l- 3-27
~
I ( ')
- s. ,
-. . . - . , , ~ . - - - ~ - . - ,
w.:
)
L)9 : .- 4 t-FIGLRE 3-9 INTERPEDIATE RANGE IMt. P.S. P.S. P.S. I.R. i XNPENSAT NO250TS 10 KC - P025200 TEST. 3SCILLAT-OR
! If l' -T
-25 VDC +25 VDC
! 10 EC 8 6Y
- SG
- ny PERIOD P800001 METER 1 / h f u '
I nv , PERIOD h 7 METER [ METER { MODU- AMPLI- DEMDD- DIFFER MODU- AMPLI- DDf0D- BI- INT. RANGE LATOR , FIER JLATOR , INTEG. . LATOR
, FIER . ULATOR STABM
( L B LP P B P- TRIP PERIOD TRIP. 11 12 I
? I RECORDER B/S Integrated Power Trip
{ II N
; c.I.C. NEuTRot DETECTOR s
I i l d
- r- - ' -- - --
v^ <=a ' ~ - ~ . - y r.v - r , _ _ ____2_ _ _ _ __________________a____.ma
a ,
- n
.(
, y 2
.3
~
-q- f
+ j
, flux level. The log level output of the demodulator feeds five devices:
- 1) Local 1og level meter l2) Remote log level meter 3).' Remote recorder D
- 4) Differentiator-Integrator for the period circuit i
- 5) A bistable that activates the power integrator p The Differentiator-Integrator accepts an input from the Demodulator and converts it to a signal proportional to the exponential, rate of )
A change of the neutron flux. Modulation of this DC period takes place in Modulator LP where the signal is combined with a 10 KC sine wave and the output feed to Amplifier B No. 2 for amplification. Demodulator P receives an amplified AC signal from Amplifier B No. 2 and converts it to f a DC signal which is proportional to the exponential rate of change of ;
~q i ! neutron flux (period). .The outputs of the Demodulator P are:
, x,
- 1. A front panel period meter
- 2. A remote period meter
- 3. A bistable period trip set at a less than 3 seconds The bistable modules of the intermediate range circuit are identical L '
~ to those in the source range circuit The Intermediate Range drawer contains a test module that performs ' alignment and operational checks of both the level and period sections without the use of additional test equipment. 3.11.4 Power Renne Drawer The Power Range drawer (see Fig. 3-10) contains two completely independent power range channels which will indicate reactor power over the range of 0 to 150 percent. Each power range channel has G JJ' its own detector power supply, and its own i 25 volt power supply. The 3-29 3
('y m r O U,. :u - FIGURE 3-10 POWER RANGE DRAMER U. I. P.S. P.S. P.S. 7 P800001 C. P025075 NO25075 i 1 y p
' %PWR +25 VDC -25 VDC CHANNEL 1 Integrated
- (when in natural convection) ,
1 AMPLIFIER BISTABLE TO SCRAM IDGIC
- ; P y TRIP 3 (SEE FIG. III-1k)
, HIGH POWER TRIP 4 w
' PR E p3 U pjoh001 MH M 33 u^sTEa y .
- -a I
j > AMPLIFIER > BISTABLE (SEE FIG. III-1k) I i P TRIP HIGH POWER TRIP i CHAIGEL 2 ! V U. - ** PWR ' I* MTR P.S. P.S. P.R. TEST C. P0250T5 NO25075 CIRCUIT
- 2 I V [
+25 VDC -25 ,VDC 4
t f- 'W r -- -- -- e -- - e v +- e * - L--- s-"--*- - - - - - - - - - ' - = - - - - - -m
rn .. . U . L ,4.[ l p [ ,
\ the power range drawer. Each power range channel is supplied an input K_ I 1
from an uncompensated ionization chamber (UIC). Outputs from each channel include a ecmmon local power level meter, separate remote power level meter, and a high power bistable trip logic signal. In eddition, Power Range #1 supplies ~a signal to the power integrator when in natural convection. A high voltage power supply provides 200 800 volts DC to the detector. The output of the UIC is a DC current proportional to reactor power. The signal is fed to Amplifier P, which is a DC amplifier, where the signal is amplified to a level suitable for use. When in the "200 kW" range the full signal is amplified. When in the "2MW" range only one tenth of the signal is amplified. The outputs of Amplifier P are: Local common percent power meter l 1. () 2. Separate remote power meter
- 3. High Power B1 stables
- 4. Power signal to the power integrator when in natural convection It should be noted that the amplifier is designed so that a short or open circuit, in any of the outputs, will not cause the other outputs to
. vary more than a fraction of a percent.
The bistable modules of the power range circuits are identical to those described previously for the Source Range Circuit. The Power Range drawer contains a test module that performs alignment and operational checks of each power range channel without the use of additional test equipment. 3.11.5 Scram Lonic Drawer The Scram Logic drawer (see Figure 3 11) contains the logic
'() circuitry necessary to process the scram function inputs and to shutdown 3-31 ,
-- -- ,.n.. ,- - -
~
. f \ - _
#1 Both MD's (j. .MD-#2lonly- V'~ ' ~
Power 1 Period Pump on/ Header down Power 2 . Pool level 1 Range switch Manual Reactor door Pool level 2 Pump off ' Pump on. Manual Ground floor Low flow . _ Face rad. Escape hatcli.- Bridge rad. Truck door key switch Fire / Evacuation . Pool temp. - Air to header w , j dh Scram MD #1 Annunciators MD #2 3 L 2 cps f.w hl Test Rese t- K2a -2 2 - - - - - - - - , SR Switch
, - - - - - -- - _7.1Kl a String y *
,' NA 45 0 ; ;
Solid State l Solid State Relay W
- - K2 M
Kl : Relay + r ACR i 1 5 i 1 Console ---- - - . - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - ---- Console a Scram Rod Scram pW mag l q ma9 qs- Permissive r3 'Vf - Relays 4 mag w _t 3. -
.r-
- Scram Relay Figure 3-11 Scram Logic Drawer i
4 - e e -- , . . - . . +a-ge _ v ____ _ .____m-_ m____ _w
, j Y . ,
l the reactor automatically should conditions warrant. The drawer also m (em') ) contains the interlock circuitry to prohibit safety rod withdrawal if ) certain minimum conditions are not met. To understand the operation of the Scram Logic circuit, it is only necessary to understand the operation of four basic modules. Negative logic is used throughout the system, in that the normal . safe, input , signal 110 volts and the abnormal, unsafe, signal is 0 volts. All available inputs to these modules may or may not be used.
- 1. Transistor Cate NA 45 consists of five separate, four input, y negative logic AND gates, hence the designation NA-45.
For each separate AND gate, all four inputs must fall to zero volts to obtain a 0 volt output signal. Any input at
+10 volts will hold the output of that gate at +10 volts. ,
The inputs (up to 20) are logic 0 or +10 volt signals and the outputs (up to 5) are also logic 0 to +10 volt signals. [ c As explained later, the rod withdrawal interlock is L derived from two logic inputs to NA 45. 7
- 2. The Auxiliary Control Module takes logic inputs in the form of 0 or +10 volt signals and provides a relay output capable of handling 110 VAC and 2 amps. The input stage consists of a four input OR gate where all inputs must be at +10 volts to have the relay energized. The relay is de-energized if any input is at zero volts.
1 b 3. There are two Mixer Driver B's. A Mixer Driver B is essentially a 28 input OR gate. Any one of the 28 logic inputs falling to 0 volta is sufficient to cause the output to fall to l. O 3-33 1 l
, , , - . - ~ . - - .-- . . - . . . , ,- . - - - . , - ~ . . - _ , , . -_ . . - , , . . --- ,, - - - , - -
r- - -
f . 1 J b.
. 1 1
N s p I h
).d '
y i ) 0. volts.. Only when all inputs are at +10 volts is the output [ at +10 volts. The input is 28 logic signals, while the outputis ) i one logic signal. l
- 4. There are two Solid State' Relay modules. A Solid State Relay j Module takes one logic signal (0 to +10 volts) as its input and provides up to 5 amps DC of output current to the scram ,
magnets if a safe input condition. exists. The Solid State L Relays are subject to one type of failure that could' render
, the model inoperable as a device for cutting off magnet current. This type of malfunction of the entire module ;
would necessitate the simultaneous failure of two components within the module, in the form of a short circuit in two series silicon controlled rectifiers (SCR). Built in' t* l circuitry within the drawer has been provided that will l l annunciate a short circuit in one SCR as a warning light. If a short circuit occurs subsequently in the other SCR, the solid state relay will continue to provide current to the scram l magnets even if a unsafe input is applied. As will be explained in the overall operation of the Scram Drawer, even i the existence of the above mentioned conditions will not render the entire system inoperable as a parallel network exists which could de-energize at least one safety rod and thereby shut the reactor down. i The Scram Logic Drawer, as a safety cystem, can be divided into two sections. Namely, a scram logic process section and an activation 1 section. The process section takes logic signals from the various t
~
3-34
~
fit . . j
- g. n p.
I 1 L Ib :
- l. J( bistables, power supplies and relays; processes these signals with it N) respect to preconditions and emits output signals that exercise either-i scram control or rod interlock control. The modules included in this j
section are the NA-45 gate circuits and the Auxiliary Control Relay. The actuation section, consisting of the Mixer-Drivers and the Solid State Relays, take only safe or unsafe logic inputs, and, depending on the y nature of the inputs, control current to the safety rod scram magnets. The rod withdrawal interlock is derived from two logic inputs to NA-45 (1). A +10 volt signal is emitted from the source range level bistable if the source count rate exceeds 2 counts per second. A second
+10 volt signal comes from a series circuit that is closed when source range, intermediate range, pool temperature, and power range instruments are not in test. These two signals are fed to one gate. If these inputs f3 are at +10 volts the appropriate output is at +10 volts on an Auxiliary 1
u ). Control Relay (ACR) which energizes its output relay permitting safety rod withdrawal. If either input to the ACR drops to zero, this zero volt input signal will cause the output relay to de energize, preventing the withdrawal of any shim rod. The reactor can be operated at 2 megawatts with forced convection cooling or at 200 kilowatts with natural convection cooling. The High Power Trips are initiated from the power range drawer (see Figure 3-10)
- and provide overpower protection in both the natural convection cooling mode and the forced convection cooling mode of operation as shown in l;
Figure 3-12. A range switch is used to determine the scram point for each mode. 1 In the natural convection mode (see Figure 3 12), the flow header is
\
u) down and header position relay CE is de-energized. A safe +10 volt logic 3-35 V i l l l
. ,, s .
L') t FIG'J RE ' 3-12 HIG4 POER TRIP as 3.CGI D?33 l I
}
-M MP P l lBI-UIC-1
-(SE FIG III- l
( N,lI ! r3 "'7pg 3IP l CI'. ACEL 1 r , HIGN PO'4R TRIPS : j 13) ll l TO MIXE3 DRIVERS - i I M Y I BI- , CHABEL 2 d l' h ll1 2 (s G III-13) y { _ . _ __j '81 I i _ a 4 L
- l. . 2w -
~ Q i ! Range s m en, s1
[
;I y
L' 2com (SHOW IN
/g ! w 2W POSITION) w -
1
* , ;o
- I, _m
/ DRIVERS d
t
+ 10 YOL? ., NOTE: hEEN SWITG l' . FG. SLTFLY LEVER IS UP, SWITG l ! ---- ---- WIPER IS DOWN.
' i I
._f t_ i
; <ca a - ,
l -.. --- l HEtZa ,
- sussG S e ca k'
110 YAC
~
- .. - ,.c. _ - - _ _ - - _ - _ _ - _ - - _ _ . _ _ .
(; - _ ,
- i. co
'l 1
'~^ signal is passed through the range switch only if the switch is in the j
200 kW position. .Under.this condition, the power range channels are sensitive over a range of 0 300 kW for an amplifier output of 0 150% of normal range,.i.e., 100% of output from the amplifier corresponds to 200 kW. If the reactor power level reaches 250 kW, the power range channel will indicate a power level of 125% and will send an unsafe (0 volt) . Figure 3 12 High Power Trip signal to the mixer drivers scramming the reactor. When the reactor is operating in the forced convection mode, the
' flow header is in the UP position. Therefore, relay CE is energized, which allows the range switch to be placed in the 2MW position without losing the 10 volt safe signal. In the 2MW position, the range switch also reduces the output sensitivity of the two power range amplifiers by "
t'h s-j a factor of ten. Each channel indicates over a range of 0-150%, which l corresponds to a maximum reactor power of 3MW. The scram trip point is l l also increased by a factor of ten by placing the range switch in the 2MW position and will scram the reactor when reactor power reaches 2.5 MW. When operating in the natural convection mode, any attempt to change the l position of the range switch to the 2MW position will immediately l l initiate the scram, through loss of the 10 volt signal through the L switch. Referring again to Figure 3 11, the intermediate range period trip 1 function originates at the period trip bistable, (see Figure 3-9). This p logic signal is sent directly to the Mixer Drivers. 1 i The scram actuating portion of the Scram Logic Drawer consists of two independent channels cross connected in such a way as to afford ( ,j 3-37 l l
~ '
[ , . - 1 ( t 4 p I.1 p b
,Y' ' maximum scram protection against component malfunction. The two Mixer
~
,t Drivers receive several logic inputs in parallel. If one of these inputs is an unsafe signal (0 volt), the output of both MD's will be at zero '
p r volts. The input to each Solid State Relay going to zero causes the i, outputs to stop conducting, cutting off current flow to the scram magnets. To' nullify the effects of a complete failure of one MD, the two ;
. scram channels are cross connected through relays R1 and R2.. If a g failure occurred in one of the MD's, say MDB1, which would prevent it
. from tripping off when an unsafe input signal is received, MDB2 would be tripped off by the same unsafe input signal, SSR2 would stop conducting, dropping magnet 3 and de-energizing relay R2. The contacts on R2 which p feed an input to SSR1 would be opened, causing SSR1 to also stop
- conducting the dropping magnets 1 and 2.
()
,m Both mixer drivers receive separate inputs from redundant channels l
l (i.e. high power, low pool level, and low flow). An unsafe (0 volt) l signal from one of these inputs causes the affected mixer driver to give - 1 L a 0 volt output causing a reactor scram. The simultaneous failure of both MD's in the safe mode or a failure of four diodes in two separate modules, the SSR's, could result in the j scram logic system being rendered as "can't scram." 3.12 Scrams. Interlocks. and Alarms 3.12.1 Scrams l . The following scrams are required by the UVAR Technical l l Specifications: 3-38 L g-w w ,- - y---o-- y e- -
- e. - --
L 1 /~~T SAIRE Ocaratino Mode Raouired N ,) . l
- 1. 2 Safety Channels for high power scram
]
a) range switch in high power position Forced convection b)' range switch in low power position Natural convection I
- 2. Bridge radiation monitor scram All modes -
- 3. Pool water high temperature scram All modes
- 4. Loss of power to primary pump scram Forced convection- :
- 5. Application of power to the primary - i pump scram Natural convection
- 6. Low primary coole-nt flow rate scram Forced convection
- 7. Manual scram button All modes ,
- 8. Fast reactor period scram All modes
- 9. . Air pressure to header scram All modes
- 10. Low pool level scram Forced convection
(]h N. ' The above listed scrams must be operational as required during reactor operation. Other scrams, not listed, may be added as appropriate
- to provide for added protection of personnel, experiments and equipment.
3.12.2 Interlocks The UVAR Technical Specifications requires an interlock on the Source range instrument which prevents shim rod withdrawal unless a minimum of 2 cps are detected. > 3.12.3 Alarras Both audible and visual alarms are used at the reactor console to alert the operator of alarm conditions. A continuous tone audible alarm and a red light visual alarm accompanies a reactor scran signal. This audible alarm may be silenced by either resetting the scram logic drawer
; 3 39 N.
t
cr..
' }' or pressing the " silence" button on the common alarm panel. The visual alarm is cleared only by resetting the scram.
An intermittent tone sounds and a visual indicator is actuated for each of th'e following conditions: a) Regulating rod control shifting from automatic to manual b) High radiation on any. area monitor or on either argon monitor c) High radiation on core gamma monitor I d) High radiation on criticality monitor e) High radiation on constant air monitor - f) Entry'into the demineralizer room g) Entry into the heat exchanger room h) High differential temperature across reactor core , i) High demineralizer conductivity
^
j 1 ~ j) Secondary pump de energized l -(}~ The audible alarm will automatically reset after about two minutes or may be reset manually by pressing the " silence" button. Local alarm bells are supplied at the heat exchanger room and demineralizer room when the key switch is on which warns personnel entering the area of a possible high radiation area. Other audible and visual alarms may be used, as necessary, to t provide extra personnel safety or equipment protection.
'3.13 Automatic Control For Maintainine Constant Power r A voltage signal proportional to reactor power is developed by a ,
slide wire potentiometer in the linear power recorder. This signal is l compared to the voltage developed by the " Power Set" potentiometer on the I control console. Any difference in these signals is displayed on a ' L .(
}
l 3-40 i {-
r~) ,; , .. l ,
; _.} deviation meter and supplied to the controller as an error signal.. The / . ;
I , controller converts this small error signal into 60 cycle power either in phase or 180 0 out of phase with line voltage. This power supplied to the servo motor for the regulating rod will drive the rod in if the linear f signal is higher than the power set voltage or out if the linear signal
, .is below the power set voltage.
Several con'ditions will automatically cause control-to shift into the manual mode and sound an alarm to alert the operator that pover is no l longer being controlled automatically. These are:
! 1) Any attempt to move the regulating rod with the normal control switch; this insures that manual control is always instantly , available to the operator.
- 2) The regulating rod either at its top limit or bottom limit; this
]n insures that regulating rod has free movement to control reactor power.
- 3) The error signal, as displayed'on the deviation meter, exceeds 7.5% (arbitrary units); this insures control is shifted if the regulating rod is unable to control power for any reason, such as 1
the reg rod being stuck. ;
- 4) The linear power recorder is turned off. This ensures the L feedback loop is complete.
- 5) A switch is provided that allows the operator to select either the manual or automatic mode of operation. l; P
3-41 l, i fn r . . . . - - - .. - .'
~ - -
o 1 k 1- . , i l
,o fl J
O References For Chapter 3 L > (f
- 1. EC&G Idaho, Inc.," University of Virginia Test Research And Training Reactor 2 Standard And Partial Fuel Element Assembly" Drawing Number 428500, '
Revised August, 1989. F I' i
- 2. EG6G Idaho, Inc.. " University of Virginia Test Research And Training Reactor .
!. '2, Control Fuel Element Assembly", Drawing Number 428501, revised August, 1 1989. , 4. EG&G Idaho, Inc.." University of Virginia Test Research And Training Reactor -
2, sideplates", Drawing Number 428502, Revised August, 1989, i
- 3. EG6G Idaho, Inc.," University of Virginia Test RSeearch And Training Reactor 2 -Shock Absorber Seat Adapter, & Bail", Drawing Number 428503, Revised;
[ August 1989.
- 4. EG6G Idaho, Inc. ," University of Virginia Test Research And Training Reactor
- 2, Fuel & Guide Plates", Drawing Number 428504, Revised August, 1989.
, 5. EG6G Idaho, Inc. '," Test Research And Training Reactor 1.EU Fuel Plate", ,
Drawing Number 422264, Revised August 1989. 6, Diamond Power Speciality Corporation, " Shim Safety Rod Assembly Top Magnet control", Drawing #EX 7935 0. April 21, 1958. .; O' O 7. Diamond Power Speciality Corporation, "27 Feet Regulating Rod Assembly", L Drawing #EX-7769-0. April 21, 1958,
- 8. Diamond. Power Speciality Corporation, " Element-Shim Safety", Drawing
#7006181122, October 26,1960. .
- 9. Freeman, D., Neutronic Analysis For The UVAR Reactor HEU To LEU Conversion
- Proiect, Master's Thesis For The Nuclear Engineering and Engineering Physics >
Department, University of Virginia, July,1989. I
- 10. EG6G Idaho, Inc.,"TRTR-2 For University of Virginia Standard, Partial And Control Fuel Element Assemblies For The University of Virginia Reactor",
Revision 1, June,1989. I l
- 11. U.S. Nuclear Regulatory Commission, Safety Evaluation By The Office Of l Nuclear Reactor Renulati,,on Suncortine Conversion Order To Convert From Hinh L Enriched To Low Enriched Uranium Fuel. Facility Ooeratine License NO. R-61.
Worchester Polytechnic Institute, Docket No. 50 134, September 12, 1988. l 3-42 , l L L, s (
t
' t
=
y
/^^g I l j ,) 4.0 REACTOR SYSTEMS 4.1 Reactor Pool ;
L The pool'in which the reactor operates is 32 feet long, 12 feet wide, 26 feet 4 inches deep, and holds about 75,000 gallons of water. An i
, aluminum gate is provided approximately mid way of the pool so that ]
'1 either side may be drained independently of the other while the reactor [
itself is adequately shielded by the water in the undrained side. With the. reactor in position at a far end of the pool, the radiation level in the emptied section will be low enough to permit personnel to alter or , set up experimental equipment and perform maintenance. l The south end of the pool is above ground level and faces the main t radiation laboratory experimental area. The shielding provided by the water and the pool wall is augmented by the addition of sufficient 'i concrete to reduce the radiation well below tolerance levels at maximum operating power. The south wall of the pool is penetrated by s [ experimental facilities which will be further explained in Section 6.0. l The level of the water is about 20 feet above the active core when , l the pool is full. An alarm and scram is actuated by two pool level devices if the pool level drops 19 feet 2 inches above the top of the active core. Figure 3-1 shows a vertical section of the pool and related building l- spaces. b l 4.2 Filline and Draininn the Pool l Two systems may be used for filling the UVAR pool; 1) a 4-1 l 1- I
+ , . _ . . , . _ , , . . . _ _ _ . . . , , - _ . , . , . ,, ,, , .., , ...,.,.,,, ._....,,._,,,.m- ... . - _ _ . .,_. . . _ _ , , , , , _ - . , ~ _ . -
~. -. _
i i i' ~~g
/
( ,); dedicated fresh water make-up system, and 2) the main domineralizer system. L The dedicated fresh water make up system is located in the UVAR room L and consists of a small filter and H OH mixed bed ion exchangers to which city water is supplied. The system discharges deionized water into the , UVAR pool through a stand pipe to preclude the possibility of pool water-backflow into the city water system. The system has a flow rate of about-5 gallons per minute. The pool can also be filled using the main domineralizer system as P shown in Figure 4 1. City water enters a catch tank through a hand valve' and a float valve. The water entering the pool through the domineralizer system is drawn from this tank. The float valve discharge is higher than the tank overflow, thus precluding the possibility of any backflow of
'( 'j
,s radioactive water into the city water system. The pool can be filled at ,
a rate up to about 20 gallons per minute through a mixed bed demineralizer. With the pool gate in place, either side of the pool may be filled independently. , The pool is drained throu6h two manually operated valves located in a sump pit in the heat exchanger room. These valves n'ce normally locked shut. Either side of the pool may be drained when the gate is in place. , The discharge from the drainage system empties into the waste pond and is described in Section 4.8. 1 e l
.O 4-2 l
l
. .~ . - - . . . . .
, . i r
i u je)' 4,3 Pri mary Coolina Synta==
~
With the rSactor in position at the south end of the pool, an air operated header may be raised into place. This header is actuated by I the emission of compressed air at about 50 psi into a header skirt which floats the header up to the reactor grid plate. Once positioned the 2 header is held in place by a differential pressure.of 0.30 psi created by-the downward flow in excess of 1000 gallons per minute through'the ' l Y reactor to the primary piping system. After the pump has started and t
sufficient flow is established to h31d the header in place, the air pressure is vented to allow the. header to drop automatically for natural convection cooling in case of pump failure. Such pump' failure initiates L
a scram as does the existence of 2 psi air pressure in the header. This ; l. precludes the possibility of raising the header while the reactor is ; operating in the natural convection mode with the pump secured. The reactor is in a scram condition if the pump is operating with the header down. l F lh
~
4-3 r l. 1 1
*d'*
- M - 4 ,a_
}
5 Ii, 4 h g [' T I sw
- l a
.s~ . , -
g - 9 1
- y. n : 1
" 5 i 5 -
f l 5
" LE' g ,,....
e i. 1 9 Ee 9 [ .' s '3 l N 9 lg } lP _)- . w ,
. I k,. . "'
hm w , , 7
- OO g< r q
o t
' gg-* ,
g y )n e()e , t 46 -}-1*,, }& -
} V f; ,, ?"" ;
aL nn j*
.' i a y
,,[ ', [
,,,..,..h h ' a 2,*,a j
w..:;.::.-: 9 : ::;, ; t r v 1, , Q,, a
- u. . .
..v 4 . . . . . . .
, .. m 4 th0
- e. C
- 'y t
.. . , e ... ;:
1' d 4 . o A h k ,, E s' - [ p , a 4 Y g Y G H i g .- k p,, , g ., 4::t .- w a
~S f7"" " . a e 4 1 l ', 4'4
- e a
a e e W' l e 4 d
- l. d k
! 4 l 9
&=***.*.",***se',* [w*4*6* 2 * * **
\
O 1.l1 es 1-l 1- 4-4 l l
- . . , . . . . - , . , . _ . _ . . . . _ _ _ . . - . . . _ . . . . _ - . _ , . _ - _ . . . . . . . _ _ _ _ _ . . _ , . . . - . , . - , _ . . _ . _ ~ . - - - ~ . . . - _ . . . - - - - . . . . . . . . -
b~ ~
]
1 j i l I t
/~N , I f ) ,
The water flows from the header through a six inch line to a i strainer and the primary pump. The primary pump is of the centrifugal i type with a stainless steel casing, impeller, and shaft. A self- . adjusting mechanical seal is used to seal the shaft. The pump is. driven by a 30 horsepower 1700 RPM induction motor and'is rated at ~ 1100 ; gallons per minute flow against a 50 foot head of water. The actual flow rate is "1055 gal / min. The flow rate is measured by the differential pressure across an orifice in the primary piping and displayed on the secondary console. The water proceeds through the six inch line, through an orifice, used to measure the flow rate, into the shell side of an aluminum tube and shell heat exchanger. The primary water makes one pass through the l l heat exchanger and is returned to the pool. The water enters the pool-through a flow diverter which directs it toward the wall at the greatest distance from the reactor, thus gaining the greatest mixing and minimizing nitrogen 16 gamma activity at the pool surface. Water temperature and pressure are measured on the inlet and outlet sides of the heat exchanger. The temperatures are displayed on the secondary console in the reactor room. A separate continuous monitor is ; located above the core which will initiate a scram if the pool water temperature exceeds 1050F. i 4-5
'O
" ~
i .
. , 4 k
1 [ I a ' A 4.4 Measurement of Tannarature Differential r: . , ~j )' , A system has been installed to continuously monitor the temperature i' differential across the reactor core when it is operated with forced t convection cooling. The AT value. measured directly by this system, can , be related to the reactor thermal power level. Separate coolant inlet and outlet temperatures are read periodically and subtracted manually to l obtain this information in addition to the constant monitor. , The sensing elements used in the system are platinum resistance . bulbs. One is located in the reactor pool, about three feet above and to j the side of the , top of the core, to indicate core inlet temperature. Core outlet temperature is sensed in the primary coolant line, just upstream of the primery coolant pump. The sensing elements are placed in
. a resistance bridge network where the differantial temperature values are measured, then amplified and displayed on a digital meter. An alarm with an adjustable setpoint is provided to give alarm signals for AT values in excess 'of pre set levels. The range of the instrument is 0 200F, while the normal AT across the reactor (at two megawatts) is about 130F.
L 4.5 Secondarv Cooline system Water on the secondary side of the heat exchanger is pumped at about 1200 gallons per minute from the basin of a conventional cooling tower rated at a cooling capacity of 2 MW with a 950F entering water l l r , 1 l 1 46 l ()) > I l 0 l 1
l ll g , G l V temperature and 730F wet bulb atmospheric temperature. The system
],
utilizes a conventional centrifugal pump driven by an induction motor. l l l.
.Water for evaporative cooling in the cooling tower comes directly from -
.the city water supply and overflows to the waste pond. Water temperature
.is measured on the inlet and outlet of the heat exchanger and is p
displayed on the secondary console in the reactor room. ' L 4' 6 Desian Snecifications , The design specifications for the Heat Exchanger, Cooling Tower, Secondary Pump and Primary Pump are given in Tables 4.1 through 4.4. , 4.7 Water Purification i The pool water purity is maintained by circulating it at a rate of l 20 gallons per minute through a carbon filter and a mired bed ion , exchange domineralizer. The water is normally maintained at a pH of 6.0 to 7.0 with a conductivity of less than 5 micro mohs. 4.8 Liould Waste Discosal System The reactor facility can collect radioactive liquid waste in two j underground retention tanks of 5000 gallons each located outside of the Reactor Facility building, but within the site area. The waste is recirculated and filteted, as well as given decay time before it is either discharged into the pond or discharged along with the pond as normal procedure. Other storage tanks within the Reactor Facility may t also be used to temporarily store liquid waste. The option for sanitary sewer release has not been exercised but is not precluded. All radioactive releases are made in conformance with applicable regulations. Two additional tanks of 250 gallons receive all waste from the Hot cell. These tanks were installed as underground retention tanks in the 4-7 _ . _ . _ _ . ~ _ . . _ _ _ _ . _ . _
r , 4 l j, ! i < i i i LO !
!. l i
TABLE 4.1 Best Exchanger Specifications ( r i Meat Transferred - 6.83x106 Stu/h
' Materials - Aluaisua. All asterials must be compatible with alusiana. I f; For this reason no copper containing alloys can be used, j Maximm length - 18 feet i Shell Side Tube Side Fluid Circulated Hirh purity water Cooling tower va.or Liguid Flow 1100 gallons / minute 1200 gallons / minute I T inlet 110.2'F 82*F T outlet p$'T 93. ls'F r Pressure Drop Allowed 8 psi 8 psi l Design Pressure 50 psi 50 psi !
Test Pressure 75 poi 75 psi Design Temperature 150'F 150'F Inlet and Outlet Pipe 8 inch 8 inch '! t l To be fabricated in accordance with ASME Code, Section VIII, Division 4
- 1. Shall be inspected, certified, and stamped with the code U - Symbol.
1 1 l t t t t t O l 4-8 l l 1 l
. . . , _ . . _ - . _ _ _ _. . . . . _ . . _ _ _ _ . _ . . . , _ _ . . _ _ . . _ . _ _ _ ~ _ . . _ _ _ _ . _ _ . . _ . _ . , _ _ . . _ _ . . . . _ _ . . _ _ _ _ _ . _ . . _ . .
c
$ 1 I
i l ):
'm ,
l i l i TABLE 4.2 Cooling Tower Specifications l Water Flow Rate 1200 aalices/ minute ! t Heat Transferred 6.03x10' st u/h l Wet Bulb Temperature 73'T i Water On 93.b'F f I Water Ott St*F i r l Fill and construction materials - Redwood ! All : materials must be compatible with aluminum. For this reason no ; copper-containinE alloys can be used. , t a,e s I I l ! L l l l i l l . O - 4-9 L ,,. . . . . - - _ . . - - . - . . - . - _
l 1 ( , l 1 l 1
'l TABLE 4.3 Secondary Pump Specification 1
i Centrifugal end suction pump : 1 To pump 855 chremated vater froni coolint tower , 1 Flow Rate: 1200 gallons / minute Dynamic Head 70 feet All vettable parts to be stainless steel . To include Tefien packint } Flooded suction ! O V Hounted on base plate with coupling t i l Motor: 1750 rps, 3-phase, 60-cycle, 220-volts l
?
l
- j. Open drip proof enclosure ,
i l l i
~
O F t 4-10
- i. .
n--. . . . . _ . . , . , . . . , , - - - , _ . _ ,._,,,,_m.,. ,,.--..,...__,,_m.--,_,,._,,..-..,-,,.,..,.,,.-,..-w,,
b du ,a e-t O I i TABLE 4.4 Primary Pumn Snacification Centrifugal end suction pump To pump 700 1000F Demineralizer water from pool through heat exchanger Flow Rate: -1100 gallons / minute Dynamic Head: 50 feet All wettable parts to be stainless stee1\ Size of Impeller: 10. Inches
\d Mounted on Base Plate with Coupling Motor: 30 H.P., 3 phase, 60 cycle, 220 volts b
I s [ t
)
4 11 {'
. . , . . , _ _ . _ . _ _ . _ , . , . . . , - . . , _ . _ . . _ _ . - . , . . . _ _ , . . , . , ,, . _ , . . . . . . , _ , . , , . _ . __-_._...~ _ , . _ . _ , . . .
l l event of waste discharge from the hot cell. These tanks empty into the l pond. ! l The water released from the pond is sampled prior to, at the ; beginning, during and at the end of each release. The results of these ] sample analyses are maintained by the reactor health physicist, No waste ! is released off-site with an active concentration in excess of 1x10*7 l 1 microcuries per milliliter, annual everage. This limit is based on 10 l l C W Part 20 limits for facilities producing waste where no iodine 129 or j l radium are present, j l 4.9 Buildine Ventilation System and Airborne Effluents The reactor building ventilation system consists of the following
- major items.
i
- 1. An axial vane exhaust fan on the reactor room roof, the primary function of which is to provide circulation for the reactor room.
- 2. A short, duct. work 31Agl on the reactor room roof.
- 3. A centrifunal blower on the mezzanine which exhausts the ground f floor experimental facilities, hot cell, and fume hoods, through duct. I work, to the suction of the roof top fan.
A schematic diagram of the system is shown in Figure 4 2 showing the l r fan, blower, nominal flow rates, and the general layout of the various t connecting duct work. Ducts leading from the reactor experimental ! i facilities (beasports, large access and thermal column acr.ess), the hot cell, and fume hoods are all provided with in line filters and either ; i have, or have provision for, in line radiation monitoring. As a result of normal reactor operations, and in the event of . certain types of experimental failures that may be expected to occur, 4 12 1
1 6600 Cygg Axial Vene Exhaust Fee d Stock _ 6&(Cm l }4---7080CtM h reseter room ventilation is elooed 7----_----__-----_-----.----. e i a 7000 CM : l t n
- I 8 l
- I _ gg. g,,g
, l t 1600 Cat
- F i - ,,0 0
, , I b I (Flow Values are Nominal) e e I 8 I I M I i d l 1 N l '
w e 8 Q ' ' 3
=
= . C.
. - - _ _8eet _ _ _ - _ - __ _ _ _Ger _rvi+1*=.
Sc=' ;
- 2. es ;
g . i i 3 Absolute Filter l I h o Centrifugal Blower ; ] M i _ _ 1 : 600 cres
- Abeolute T 1 I
- Filter f
~
I - -- 600 CM ""***I*' ' i
)
[ 6* duet G 4 7 e , 7 - - - - - - -- _3
# 10" duet $
2}" pipe ' ' istlos e E "g-knitors l CEL 8 t o wi.io.au , e[p hi Qj , 60 * - ." duct g L% ..,eac C
. .J "hgagl. ..: L- ------ L 250 cm j
- Ports Phc111ty Column l
i 1
, , ...--.-~,e - - - . , . , . - - ~,---+=-~--4 ,,..-,-e-- . .--v~.--~r .-..---*,w,* , - . - - - - - - - - - * = . - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
p ^\. V there may be some release of gaseous, or airborne particulate, radioactive material from the facility. In general, such effluents would be relear,ed from the stack. 'Ihe purpose of this section is to identify r normally occurring and potential sources of such effluents, and to evaluate the consequences of both. l 4.9.1 sources The gaseous radioactive isotope produced during normal operations which is of potential concern is Argon 41, produced by thermal neutron capture in the stable Ar.40 isotope present in air to about 1.3% weight. A small amount of Ar.41 is produced and released, as a result of air dissolved in the reactor pool water, being irradiated in passing through I the reactor, and then coming out of solution into the reactor room i l atmosphere. This isotope is subsequently released as the reactor room is O normally ventilated. While Ar.41 is continuously produced while the reactor is operating, its release is somewhat variable and to a large ! extent uncontrollable. However, the Ar.41 releases from the WAR are small and of negligible consequence, as shown below. l Ar.41 may also be produced if air is irradiated in the various { l l facilities on the ground floor experimental area. This source can be ! , s I controlled by limiting the volume of air exposed and is thus very small. t Brief releases of Ar.41 would be a normal consequence of reflooding a , beamport, in which air had been irradiated. However, when in operation l the beamports are usually filled with a low activatable gas, such as [. l helium or carbon dioxide. . t i Accidental gaseous releases may occur as a result of experiment failure, either in an experimental neutron irradiation facility, in a 5 1 l 4 14 1 e l l
_~ { . _ . . i ; i l [ fume hood or hot cell. A severe case of such a failure, the complete melting of a 1 watt U 235 fueled experiment, is analysed in Section 5.4 ' of this document. ! 4.9.2 conaeauences of casmous Erfluent Relamme - cenaral ; Figure 2 1 is an serial photograph which indicates the position of ! r buildings nearest to the reactor, including those ennstructed since the photograph shown in Figure 2 2 was taken. There follows an analysis of - l the atmospheric dilution of the reactor stack effluent at 2 points on the j i site boundary (fence) and at the seven buildings nearest to the reactor, I identified in Figure 2-1. This analysis is performed on a yearly average [ or long term basis and is thus pertinent to the calculation of the I consequence of materials continuously released during operations. The following assumptions are made in the calculations: ,
- 1. Wind frequency and velocity data taken at the site Table 11 1 and II-2, are representative of average conditions.
- 2. Pasquill D type of dispersion conditions exist on the average ;
over a year's tis.e, with a wind speed of 2/3 of the annual average measured (Reference 1 at end of this Chapter), f-
- 3. No elevation differences are accounted for (i.e., "0" stack i
height), s 4 Effluent materials are uniformly distributed over each 450 l sector (Reference 2 at end of this Chapter).
- 5. No credit is taken for the rise or mixing of materials leaving !
the stack, although the stack effluent velocity is ' 15 ft/sec. 7 O 4 15 _. _~ . - .._._. _ ,_-- _ _- . _ . _ __ _ _ _ .
~ _ _ . _ - .. __ .. _
(i l I l d The dispersion equation recommended in Reference 2 for j i releases is: X(xg ) fg
- (2,)1/2 o(x)E(2*x/N) g g g where X(xt) = the concentration of material at xg, Ci/M 3 {
Q - the average release rate of materials at the stack, Ci/sec.
'- ft - the fraction of time during a year that the wind blows into compass l section i.
xg a distance to site, in sector 1 M. . og(xt) - standard deviation of plume in vertical direction at i l distance xg, for Type D conditions M. l p - average wind velocity, M/sec (i - 4.63x2/3 - N/sec) [ N = number of compass sectors for which there is data (N-8) The results of the application of this equation to the 9 sites are I i presented in Table 4.5. Column 8 shows the percent of the 10 CFR Part 20 l t limit of Ar.41 that would result from a continuous release of 10 5 f Ci/sec. (10 p Ci/sec), i For long term releases, dilution facters, X/Q, of about 2.5x10*4 , o exist even very close to the release point, resulting in an average Ar.41 concentration of about 6.5% of the Part 20 limit for Unrestricted Areas j assuming a 10 m Ci/see release rate. This rate will be shown to be i y i higher by at least a factor of 20, than that which may be expected for this isotope, 1 For accidental relsases and other short term releases, Pasquill F l type conditions, a 1 meter /see wind speed, and 0 stack height will be 4 16 l l
.- . _ _ . _ - ~. _, . _ . . ~ . _ . _ _ _ _ _ . _ _ . . _ _ . _ _ - . . - - _ -.
r l l l conservatively assumed as described in TID 14844. A dilution factor at I (c.) l 70 meters of 0.039 for these conditions will be used to determine l l downwind isotopic concentrations, the consequence of noble gas release under stable meteorological (Type F) conditions will be evaluated by determining whole body dose from the cloud, as an undiluted line source, j 4.9.3 toecific cases In this section, several specific cases of gaseous release will be ] i considered and the consequences thereof evaluated. l i 4.9.3.1 Arnon 41 from Reactor Room. lenz Tera j The production and release to the reactor room of Ar 41 was j discussed in Section 4.9.1. The specific activity of this isotope in the [ reactor room has been monitored for several years. With the reactor room i t ventilated, the concentration normally does not exceed 1/30 of the MPC l for Ar 41 in Restricted Areas (2x10 6 p Ci/cm3. The normal exhaust rate for this room is 7,000 CIM (3.3x10 6 e,3/sec). If one assumes that an Ar. 41 concentration of 6.7x10 8 pCi/cm 3 exists continuously in the reactor room, the resulting effluent source is: l (3.3x10 6 e,3/sec)(6.7x10*8 pCi/cm3 ) - 2.2x10*1 pC1/sec [ i or about a factor of 45 lower than that assumed for the calculations of j Column 8, Table 4.5. , l Even considering the factor of 2 increase for 2 NW operation, the [ release associated with Ar 41 production in the reactor pool appears to ! i be of insir,nificant consequence. , t 4.9.3.2 Reactor Room Ar-41. Persistent. Stable Conditions The most extreme case for this type of release would be that of a long persistence of Type F dispersion conditions. As is standard ! 4 17
O O O i TABLE 4.5 ! PERCCIT OF 10 CFR PART 20 LIT 4T OF A %1 RESULTIEG FR0ff COIITINUOUS REEASE OF 10-5 C1/SEC i. t z i Bearing Site Site i Pasquill-D from f _X_ n s 5 of PTC fy A 41 Meters (Ref. 3) Reactor i q with q=10- C1/see No. Identification 1 North Fence" 50 2.6 N 0.10' 2.6 6.5
! 2 South Fence" 70 3.4 S 0.17* 2.5 6.3 i, '
Van de Graaf Bldg. 115 5.1 SE 0.12 0.72 1.8 i 3 f , b RLES Laboratory 160 6.5 NW 0.09 0.30 C.75 - 1 ! l 265 11.0 NE O.06 0.095 0.2% 5 Radioastronour Lab. 6 Observatory 300 13.0 SSw 0.1T 0.15 0.40 j 7 Building & Grounds 300 13.0 E O.10 0,009 0.23 8 Dormitory 3T5 15.0 ESE O.12 0.072 0.18 I l i i 9 Filter Plant b30 17.0 SE 0.12 0.058 0.15 ' I i i
- a) The North and South Fence sites represent average distances to the facility fence in these two directions, and the corresponding wind frequencies are averaged over all northerly and southerly directions.
l 1he University of Virginia also owns the land..out to at least 300 meters in all directions from the reactor, I and controls its use. Other than the sites listed above, there are no areas mornelly occupied within the j 300 meter distance. I ,
1 l i i E i ! I practice for noble gas exposure (Reference 3 at end of this Chapter),
]
this situation has been analyzed on the basis of the direct, whole body ; i ganna. ray dose to an individual standing at the center of the narrow f t plume that would result. For these calculations, the plume is considered to be an infinitely long, line source passing the receptor at a distance l of 6 inches. Since no dispersion in the atmosphere is accounted for, { this model is independent of source.to receptor distance. Assuming that ! the Ar Al release rate at 2 W is twice that calculated in 4.9.3.1 for 1 [ t W, the whole. body dose rate in the plume is about 0.006 mR/hr. For 24 hour persistence of this condition, the total calculated dose would be 0.15 mR. ! i 4.9.3.3 Exhaust from Reactor Room af ter Ar 41 Buildur f If the reactor room exhaust is stopped temporarily, the Ar 41 f i concentration within the reactor room could eventually build up to ! Restricted Area Limits (2x10 6 p C1/cm3 ). If the room is then exhausted, j i a short term source of 6.7 p C1/sec exists, but will quickly die away as the reacter room inventory is depleted (exhaust half period of about 10 ( l min.) The calculated dose rate in the stable plume under these ; conditions is only 0.1 mR/hr, and the total dose to someone in the plume , i of 0.02 mR. s : 4.9.3.4 bhause from Be==nort ; The standard UVAR beamports have a stction near the core which is ; filled with water when the port is not in use, and is drained prior to use. During port operations this section may be filled with at.r (although He or C02 is preferred), under neutron irradiation. If the port is refilled with water after operation, the Ar 41 produced when air 4 19
i , i L' f
)
is used is exhausted into the experimental area exhaust system, and hence to the stack and atmosphere. A pump unassisted beasport drain ] takes several hours, and the tefill (and thus the Ar 41 exhaust portion) 20 30 minutes. Flooding a beamport immediately after a long period of operation at full power with air in the beauport results in the most severe Ar 41 dose > i that can be reasonably postulated. This situation is evaluated using ; stable meterologicel conditions and the line source approximation. , i Constants and assumptions used in the calculation are listed below. l
- 1. Activation cross section of Ar 41 in normal air E,- 1.59x10*7 cm*1 l l
- 2. Volume of 8 inch beamport, for 4.7 foot length, -l V - 4.7x104 cm 3 ;
- 3. Thermal neutron flux, 4 - 2x10 12 n/cm 2 ...e f i
- 4. Flooding time 1000 see (16.7 min) ;
- 5. Saturation Ar 41 inventory, 0.4 Ci (product of I,4V/3.7x1010)
Using the line source approximation, the calculated dose rate in the l plume during the 1000 seconds of discharge is 5.8 mR/hr. The calculated ; total dose to an individual in the plume as a result of the Ar 41 l discharge is l.6 mR. During the spring of 1983 the North Neutron Seasport was modified into a closed loop system to eliminate the buildup of Ar 41 activity. The ' [ system incorporates domineralized water and a helium cover gas and a reversible peristaltic pump. When the front tube (pool side) is drained I it is replaced with Helium gas, thereby eliminating Ar 41 activity 4 20 (
.-,,-.---.--,---.n, - . .
.. . ,,-n ,.,_,.-,--..,.,,--.-.---n,, - - . - , . - . - . ,
l production in the tube. This beasport is being used extensively for i neutron radiography experiments. The drain fill operating mechanism is remotely located to reduce personnel exposure while the reactor is at power. 4.9.3.$ Evnerimental Failures ; In Section 5.4 the offsite consequences of gaseous release as a !
)
result of an experimental failure are evaluated for the case of a fission ; 1 gas release from a U 235 plate operated at I watt for a long time. Such , an incident represents an upper bound on the type of major experieental j failures which may be postulated, in view of the past and currently projected experimental programs. l 4.9.4 Effluent Monitorine j i I ; l Each of the ducts leading from the experimental facilities may be fitted with an internal C M tube radiation monitor. Such a monitor has ; been used on the duct leading from the beamports and access facilities for many years. (See Figure 4.2). , A thin-walled G M tube inside a duct has a count rate signal , proportional to the specific beta activity of radioactive materials in i the duct. In addition, the count rate depends on the beta energies, the diameter and length of the duct. The exact relationships defining these , dependencies are complicated but in general the count rate, for a given ; specific activity in the duct, will vary directly as the duct volume, and will be rather insensitive to beta energy unless the average beta particle range is less then the wall thickness of the G M tube. The best means for calibrating such a monitor for a particular isotope is by filling the duct with the beta emitting material in known O 4 21
I 1 l l concentration and measuring the count rate. This has been done l with Ar 41 in a section of duct similar to that in which the monitor is located in the exhaust line for the beanport and access facilities. The calibration of this monitor, thas obtained, was about 200 cps, for an Ar- ; 41 concentration of 10 6 p Ci/cm 3 in the duct. Once a monitor has been calibrated, it can be checked for proper operation by noting the G M : tube's response to a gamma ray source located outside of the duct, i 4.9.4.1 Monitorine Perfor=nnem for Ar-41 ; The function of the exhaust duct radiation monitor is to notify the j reactor operator of abnormal levels of airborne radioactive effluent in the duct. The monitor in the beanport and access facility exhaust duct l is normally set to alarm in the reactor room for a count rate in excess ; of 800 cpm. For a normal background of 100 cps, this count-rate would indicate an Ar-41 concentration in the duct of (700/200)x10 3.5 x 10 6 y C1/cm3 . The nominal flow in this duct is 150 cfs, or 7.1x10 4 e ,3/sec. Thus a monitor alarm corresponds to an Ar 41 release at the stack of about . 0.25 p Ci/sec. This amount, on the basis of the conservative analysis of ; i Section 4.9.3.2, would lead to a maximum offsite dose rate of 0.004 mR/hr. The expected Ar 41 production rate for any experimental facility I can be determined analytically, if the thermal neutron flux and volume of air irradiated are known. One cubic foot of air, exposed to a flux of . 109 n/cm2 -sec will yield about 0.013 p Ci/sec of Ar-41, 4.9.4.2 Monitor Performance for Other Isotones The G M tube duct monitors will respond to other radioactive gases or airborne particulates in a manner similar to its response for Ar 41, _O 4 22
, . , , - - , - - - . - , . , . . - , - .,....w,~ n, ,,-,----,,------a,,... --,--..,-,.,--,-,,-,,--.,,n., . - n,, ,, - - - - . ,- ,
I l l as long as their beta particle ranges are well above the wall thickness of the G M tube. The latter dimension for the current tube is 30 mg/cm 2 (0.0044' A1), equivalent to the range of an electron of
- 0.16 MeV.
Table 4.6 gives beta particle decay for various activation and fission product gases, and the rate at which the latter would be produced in a fueled experiment cperating at I watt. Inspection of the Table shows that the isotopes of Chlorine and 3romine, and the more intense fission ] product isotopes I 132, 1 134, and Kr-87 have beta energies siellar to l l < that of Ar 41, and would thus be expected to produce a similar monitor response. Thus, a fission product gas release of 0.25 y Ci/see would be i sufficient to cause a monitor alarm. This value amounts to 14 of the ; fission products being produced by a 1 wate, fueled experiment. 4.9.4.3 Consecuence Evaluation. Fission Gaa Release , O The consequences of a fission gas release that would produce a count - i rate just below the duct monitor alarm setting have been determined and , i are presented in Table 4.7 for the iodine isotopes. Two types of release are considered long term in which the dispersion factors at the site ; I fence (Table 4.5) are used, and a short term release under stable, , i Pasquill-F type conditions. l The results of Table 4.7 show that for long term operation at the l l alare point, the release of fission product gases produces site boundary concentrations of less than 24 of Part 20 limits for Unrestricted Areas. - The short term release at this level, using extremely conservative dispersion conditions yields concentrations which are only 2.5 times these limits. Thus, the duct monitor appears to provide adequate - protection in the case of accidental fission product gas releases. ; O 4 23 m l i l
- O O O
.l 4 TABLE 4.6 , DATA POR RADI0 ACTIVE OASES Major Beta Particle ' Production Rate j Maximum Fraction of for a 1 Watt W C, y C1/cm3 Part 20 Isotope Beta Energy Emission Fueled Experiment Appendix B, Table 2 e C1/see Ar k1 1.2 MeV 0 99 4x10
- l C1-38 >1.1 MeV 1.0 Tx10 '
Br-80 2.0 MeV 0.92 Br-82 0.hk MeV 1.0 6x10 ' i
- , Fi44 ion Producta
! A _to .
- 0.61 .0.027 1x10 l I-131 0 90 I-132 >1.0 0.79 3.2 3x10 '
i I-133 1.3 0 93 0 53 4x10 *' I-13h >1.2 1.00 14.90 6x10 ' , I-135 >1.0 0.65 1.53 1x10 '
-7 Ie-133 0.3% 0.99 0.086 3x10 j Ie-135 0 93 0 97 1.15 1x10-T l Kr-85m 0.82 0.TT 0.049 1x10 8
Kr-87 >1.3 1.00 3.30 2x10 - Kr-8B 0.85 0.23 2.10 2x10
- Total Fission Products 26.9 i l
1 i l 6 ' j s . i I I TABLE 4.7 , IODINE ISOTOPE CONCENTRATIONS FOR A 0.25 y Ci/SEC l FISSION PRODUCT RELEASE ; Long Term Short Term Release Fraction Fraction Rate Conc. at of Conc. at of Isotope y Ci/see Fence v Ci/cm3 10 CTR 20 Fence y Ci/em3 10 CTR.20 I-131 2 7x10' 6.8x10'u 0.00068 1.0hx10-II 0.01
-12 1.2x10~' O.k0 i I-132 3 2x10-2 3xto 0.0027 P
0.0033 2.ox10~30 0 50 Q I-133 5 3x10-3
-1 1 3x10-12
~II 0.006h 5 8x10 '
~
1.00 [ I-134 1 5x10 3.8x10 1 5x10'* 3.8x10'I* 0.0038 5 8x10-10 0.60 : I-135 ! i Totals 0.017 2.5 , F l l l O l 4-25 l
IT [q I L 4.10 Core scrav system i ) The WAR reactor is equipped with two independent core spray systems. I l' Their function is to provide protection against a postulated core damage in the I event of a very sudden and severe loss of coolant accident (14CA). Each system l consists of a pair of spray headers and an emergency water storage tank. The I i two water storage tanks are mounted on the inner side of two of the pool walls . and each holds approximately 200-foot 3 (1500 gallons). Either system is ; L designed to deliver an average spray flow of 10 gpm to the core for at least 1 1/2 hours, considered to be an adequate flowrate and time to prevent core : damage. The two core spray systems are illustrated (elevation and plan view) l in Figures 4 4 and 4-5. Recirculating water from the demineralizer continuously flows into each of , the emergency water storage tanks. The tank overflows are located about 2 p v inches above the highest operating level of the pool water. Accordingly, there is a slight head (2") of water on the tanks and a continuous flow of water i through the headers into the pool. This assures that the tanks are always full ; and that stagnant water and resultant corrosion does not occur in the spray headers. The material used in the construction, aluminum, and the flexible ' stainless steel couplers, serve to enhibit corrosion. The two sets of spray headers are mounted on either side of the core support structure, approximately 5-feet above the top of the core. A spray , I header consists of an approximately 1 inch diameter aluminum pipe, 2 feet long, { i with approximately 80 small holes drilled at a proper angle and spacing to provide a uniform spray over the top of the core. A mockup of one pair of core spray headers is shown in Fig. 4 3. (The flow rate was 10 gpm when this i picture was taken.) I f 4 26
\
l l
I I i j , Technical Specifications require a core spray flowrate of at least 10 gpm at 30 , t
~
/
[ minutes after LOCA onset. Actually, the calculated spray flowrate is 10.8 gpm l af ter 30 minutes, reaching 10 gpm only af ter 47 minutes. i k' hen the reactor bridge is returned to the full power position, the piping connecting the storage tank to its pair of spray headers is engaged with a remote coupler. After the coupling is made, a pressure test is conducted using an insert tube screwed into the pipe leading from the bottom of the storage tank, as shown on the right hand tank in Fig. 4 4. Air pressure is applied and a visual observation made to ensure that air bubbles are not lesking from the remote coupling. Air bubbles will emerge from the spray headers at approximately 7 1/2 psig air pressure. At 1/2 psig the coupler will be filled with air but the sprey headers will not. The absence of air bubbles from the coupler will be a positive verification that the coupler is securely engaged
.(^'y and is not leaking. %/
The pressure test insert tube is removed after it has been verified that there is no leak from the coupler. A coupler leakage test is made every time the reactor bridge is moved and returned to the high power position, after the coupler has been re engaged. A flowrate test, using a small water tank to provide a known head, is repeated at least once each year for each of the two spray systems, to demonstrate that at least 10 gpm are available for emergency core cooling for the first 30 minutes after a loss of coolant accident. In summary, the emergency core spray system is expected to provide an immediate supply of water to the core in case coolant is suddendly lost. The system has no moving parts that can fail and no automatic electronic or mechanical devices are required to function. SAR Section 9 discusses the loss of coolant accidents. 4 27 (~~' i
_.,_ _ _._._ _ _ _ _._._ _ __ ^' ' ' '
,- 8
,, : .s R
R - , 4,, \. , . ,& ,
5
,, j s
, - ,. 4 -
7
- s. , j
.r : ', .s, K.,
n- . . .a. a, . - -
...c., ,
ilg g s..1.1 g ti - ! a '- -
' ,e
, . - : .f-. 1
. M+ = . ,
, .g Ag ; .m l
4( t'. h ,.; y,t.; ,ish M ' l- * - t .. r aus , 4 ! i
, t i
9 ( ,
- e. i .$, .,, - .
. .,t .
~
.s ,
. ,. . . r - . ry y =; '
3 -. A . wy - 5 s : { ;.. . ,g +z. - . ,. - Qj, ' j .g . .
=
< . s
,v .
4
- _ ,x >
I y ) "' .'- i
..i ,L y)Q+ii ) -%, d %L. jj..' c t O
- y. - . .. y .. -
- r-- .
i ! 1 .. g 4,' .h . s .- -
< pr*. -
i- ,
'.. g,. ' .- .M.' _
% !ly}f} .'?'d L (- .
-- .' E, ,
jP y, i
, ., .' * ~,.
,,- 4 g s ,
- w. . . .
1 L , . il
- +
s
. : J. $. . f r.
TOPVIEW O FIGURE 4-3 IbCHUP OF @RE SPRAY HEADERS I 4-28 ) l
. . _ . . _ _ . _ . . . _ _ - . _ _ _ . _ _ . _ . _ _ _.___._________________________________3
CORE SUPPORT E STRUCTURE r.
( ,
l u t.1 M EACTOR BRIDGE r A = 1ae a mV i FROM
~
DEMINERALIZER
, 9N % g [" '
,A*
a 6 lll
- g ho ,
l Y. ', .' a. *a-l.* *,p l
. ,ee
...a*,l.
a4 :I a o + - 3' ll A i l ' s* -e - o ll
-4 Elf.RGENCY STORAGE TANK A
oA ll 4
'8 ll
- O j ll
! 4 8 E ll ' i
)$ $> PRESSURE TEST
# 0 y I INSERT REE
,6 Q 5 :n fREtCTE
- nx $ .'
4
\[
o r-COUP W i 4 In hPRAY ,s 4 4 HEADERS $s 4h A
,s A
y 4 ...... 4 4 e
*g a......
4 4 *a 44 4 ,, g *A 4 4 'A , J A A* j e 1 d 8 4 4 44 J 4 , s 4 . d < O FIGURE 4-4 , CDRE SPRAY SYSTEM ELEVATION VIEW 4-29
i 1 I - p* rr ;
..'*) l
- 64 - - o :
a' A- 4 s A ! { 4 * , , 8 ed 4 l
- s. * *
- i 4 ;
A b
- A . !
- a. .
' l
, , e A REmTE t [
, ;-' SPRAY COUPLER , 4 ;
' " " '*'"'a^' "'"
O *
- PN .,,
l $/_- CORE h #*I l j}i EFERGENCY l : ', STORAGE TANK l /' ' / a >p 1 l * ' I *~ * *#**, g i (
- I
= , , .
a . ,. ,e, I l
' '
- d l *0UTLINE OF REACTOR :
I _ _.._ _ -- - ._ .-.__..- BRIDGE t FIGURE 4-5 , CORE SPRAY SYSTEM PLAN Y!EW O 4-30 y - ., . - . - . , - , . - . ,,.,,.,-.-,,,.,---,,,,_,,.,,--,,,,...,-.,e.,,..,.. . - _ . - . _ . . . . , . . . , . . , , . , . . . . , . , - . ..
i l j I i i References for chanter 4.0 j ( l A) .
.1.
- Radiation Dose Evaluation Model for Maximum Credible Accidents," p.
)
23, Proposed DRL Safety guide .. Private Communication. l
- 2. Slade. David H., Edt.
- Meteorology and Atomic Energy, 1968," Section l 3 3.5.4, p.113. USAEC, July 1968. !
3', Rogers, Lester, and C.C. Camertsfelder, U.S. Regulations for the
~
Contro1 of Releases of Radioactivity to the Environment in Effluents L. from Nuclear Facilities,' I.A.E.A. SM.146/8, p. 8, paper presented at I.A.E.A. Symposium on Environmental Aspects of Nuclear Power Stations, U.N. Headquarters, New York City, N.Y., August, 1970. l. D ~ ( o
'O
).
- l. 1 i
I l- l
, x 5.0 Experimental Facilities :
) \
5.1 Beawnerts ) l Two 8. inch neutron beamports penetrate the concrete shield in i the southwest side of the pool. The ports extend through the wall and up to the side of the reactor. The ports on the pool side are normally ; filled with water when not in use. A blank flange aluminum plate : separates the aluminum port extension from the concrete shield j penetrations. Beam hole design is shown in Figure 5 1. ! The South Beam Port Extension (see Figuro 5 la) is vented to the duct exhaust system. The North Beam Fort is used for Neutron Radiography i experimenta and its extension is a closed system. The system uses deminera?ized water with a helium cover gas and a reversible peristaltic i pump as shown in Figure 5 lb. When the front tube is filled with water, the tank holds an amount of excess water and a large volume of helium (J)
% l t
gas. When the front tube is drained the tank is almost filled with water j
' r causing the front tube to be filled with helium gas. This eliminates the production and release of argon 41 to the atmosphere. The drain / fill pump circuitry requires switch actuation by both the experimenter and the i reactor operator.
There is a concrete shielding block wall around the beam ports on the ground floor (see Figure 5.la). The wall is roofed with steel and ; wood beams and plywood decking which holds graphite, paraffin and borax shielding. The back wall thickness is three blocks. The beam stop is a constructed of cadmium sheets, lead sheets, concrete and paraffin. Access to this structure is limited by a lockable door. Entrance to the , inside of the blockhouse is procedurally controlled at all times. When > 51 (OD
-e- .~ e- & - A-ma- *-a- kee+ *- .siA+A 3,-4a a"mw,, A - - -aL, AI- a ~.~" "
I
/ ,
t 1 J N ' * - I [.
<s .. m. ;
v v g ) {}n a '[<
]
i l i u{ g p_.. . . . _ q# , i l ,*, t , t..... 4
. l
.< i
' l r . . * . . v . , * . . . i '."'
. l
/ !:
- ! \\
i;
'* .?:
l4 e :
- i. .il l unu l. g *l e l? " . .l )
l< I i. im. ! [j l 'a . 'l - F. l ! u ;. . .. :: < n i
\
g J .* 5 qd * *C - i
, i .. , , . 7, . . !, l . . , . ,
d *) s 4; i ;
. 4l lixl4
>. :. : t
.. I l -a ;
I!!, M.... W1. [<r :.- B 0 - I !! kgrs. sh: I g i gw i g.l vu .
... ...b e . 2 um y 8
-a - k i
\ + . .. . .
z .. . . .. , : . . . e.
. m ,
m :
, k l
*y. e- A (
4 4, y A. .. . . A i .* .i. in .
\ ['" :.
i I. I h
' ' '. o. l.. . ,
, e
.a ug : e. t i. ': ip +- .
; ,ly d% . ".
,; s
- ' e ,
. . g}9 l l
e
*Ya ,
.'.4I e e
.:.I
,- n i
' r. * ~ .* ' * . ' ' *
. )
l 1 I l l 5-2
O O U
+ + + +
s
- POOL+ V/iLL +
+ NpRTF1 BEAA y,.
+ + + + /
PLATE CARRIER + + + /
, . . . p SYSTEA + f f
STORAGE e o SHUTTLE / ROOA TABLE y u PUMP NEUTRO O CAneRA l{TTT'Nd[N(xxx<J V W Q)Q BLOCK VALL 0
'///////> /
NlNNNN ==-= V////////////A il BEAM STOP h Figure 5-la Top View of North Beam Port Shielding and acesee re-tent u,ite
- - - . = - .
4 (m .(.s ,. m Q) ( v' 3,
- FRun DEMIN
- : SYSTEM i i R POOL i: VALL i -x g
- : df
- - He i
! $1d!O - :
o Ta Reect.or- 3
- i y Te w Nor-t.h Been Par-t. I rull 1: y.ng, . .M.
s gdf ele == i F1as a Drean i _g,q_ ) i i: ]( t
. e -
Soll w lw. A i i V
. I 4 : -
I Rev.- bI. Pump e 35 to 155
- we == 1
- : m cm b
N
! TD w
Figure 5-1b Morth Beam Port Drain / Fill Syste ,Sch N tic
. _ . _ _ _ .. . - , , _ . - ~ _ . . , . . . . - - . _ . . , . . - . . . - . - _ . . . - - . . - . _ . . _ . _ - . . . . - - _ _ _ _ . - - . . . . . . _ . . _ _ . _ . . . _ . _ _ _ , . _ - . _ . - . . .__ .._-._ __ _
+
b d g 4 6
- v.
4 .g Y Y s , 0-
? ' snsNETED e
- v > AluntNEM a e PLATE >
4 J V y , 8 e ll# , I # s'i 's'i 4 , D D
- MOVASLg. _I I i C ONCRE TE 3 I I I
- f. O SHIE LD PLtAG 3
', ! , _ ' , 3 g llll '
T u
- e , m pacxt9 Reacion CORG l A CoWcRETE
= 4 OR LEAD l "
- 4 Bak KS 1 4 .J.g l.
Gh0UdD LEVEL Floot l' l * ' 3 e a s A , e
* > d 3 5
- s i
& a
, y a A y * .
, s , -
, , roce. rtoog a .c s a a s s a - - , .
v b # d 4 & > w g'5 's b b g g y y 4 , fIsare 5-2 LARGE ACCESS FACILirY i _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ - _ _ - -- - - -r< , . - - - , - -. . . , ,- , - - . - - . - - , - - , - - - - _ . _ _ _ _ _ _ _ _ _ _ _ _ _ _ . - _ . .
E' , . {j i' < A. ' s , r j the front tube of the beamport is partially or fully drained, several I
- s. 1 alarm systems are activated. These include:
\
- 1. An audible = alarm on the opening of the blockhouse door based on q signals from either the LED water detector on the vent tube, or the LED water detector on the top sight glass tube, or a.BF3 .
neutron detector within the beam tube. This is a one of three t logic system.
- 2. An LED sensor on the top sightglass tubes of both neutron . ,
beamports which gives a drained / filled indication at the reactor console. The. north bermport LED de.tector enables an alarm and reactor scram upon breaking a light beam actuator placed inside the blockhouse corridor prior to entering the [ I L . north beamport beam path, p l p) Ds. 3. .A neutron detector located on the reactor side of any movable shielding or experimental apparatus enables a scram and a local alarm associated with the light beam actuator whenever neutrons are detected in the beam. The Radiation Area lights and the , door alarm and beam open light are also controlled by this , r detector. The setpoint of the detector is chosen based upon neutron and gamma ray sensitive portable survey instrument O readings in the beam. 5.2 Larne Access Facilities The south wall also accommodates two large access solid concrete , l
', facilities measuring 5 feet wide by 6 feet high, rolled in the wall recess on dollies. Each facility is closed off from the pool by a 56 a
f
-, .-- - _ - - , - - , - r.-.- . - e,- - - - . , , --v-- --c---,e-,,---,--. -, ..,r w. ,v-. + - - , - -,---,-.es-, -
4 Y V e , [, i b' .
' fN[ plate. A large-acesss facility is shown in Figure 5-2. A thermal column ,
may be incorporated.in a large access facility. Small diameter beamports are available for use in conjunction with the access facilities. 5 5.3 Rabbit Facilities e a)' Two pneumatic rabbit facilities are located on the reactor grid plate to accommodate thermal and epithermal irradiations, and they are used primarily for activation analysis experiments. The pneumatic
. transfer system may be connected to either the thermal or epithermal irradiation facility by manually changing the tubing at the top of the 5
reactor bridge. Samples may be loaded in a room next to the counting room on the lower level of the building and pneumatically sent to the top of the reactor pool. The reactor operator then takes control of the sample and can by computer control insert it in the rabbit facility for a
h preset amount of time. When the sample is removed it can be stopped (d -
approximately 5 feet under water and measured for radiation exposure. An i administrative limit of 240 mR/hr(Gamm'a) at one foot is placed.on samples transported through this' system. If the radiation is less than this the reactor operator can activate the system to transport the sample to the l counting room for analysis. This system is very useful for investigating 1' l short lived isotopes. 1 b) A hydraulic rabbit facility is also located in the reactor grid plate. Samples run in this facility must be loaded and unloaded from the top of the reactor pool. The overall configuration of possible experimental facilities is shown in Figure 5-3. All facilities shown are not necessarily in place at one time. k
> ) 5-7 l \)
1 v-- w - - - - - - , _ _ _ _ , w ---ae- e- -~ ,
n -.
. s& _ (% ) .
*; * '.e;...
. ,~.. > .:.
NUCLEAR RESONANCE .. F .:. FLUORESENCE , 2-
.. . e T-i EXPERIMENTAL AREA . ,1. 2 ,-
,=. .:- ',,.,-:.'...
I SHIELDING $2'.[j;.'j[ r*['IN,j'.k32*i!!$ '.*Y'.'.5h, EXPERIMENTAL 4.0.$:l.h5k/8:'U:[j
. 1 ...".iD....:.O..-.N.4,::.r,r.
. ", .,s.
' . .e 'f ,, >
$m.F....s.~~(?.If.e..
- e. . . . : . ~Fry..? 9. 7s .- -
AREA
. - , . .. ~ ,,
.."flRON'.
M : NEUTRON ' : . .j
- g yg, M.:;. ,':.7,gjjfd,'-)g[F>sgory$ 4 53'.}[$y5 S j,:cy
...s-
;6 7.'9 *ff
.-<.. .~,.-*..4..
., t T.s:;1'f., s
- s g:0l;h.7N *;:
5.p!f ... ... ,.:a
.~ ;,,
RADIOCRAPHY , *, . . -
- a. : .y . :, .,g.;* n..
... . - : *m
. . ..,.y**- .. s- .
2 c..s ,.ep:-1::.l:.:.: s 49.-:=.a
- e. y . c. , . . p :- , .- . , . t .1.,: . -r e. - fhys.,'.v;.;L*
;-O.f.:) . .s.r. . ' :- * ,.p. .*.. .. .::.L:
- y ..:,.:..
ea:?i Q: .m. s. r- . m>
- ~ : :s.. ~*
.*i.: .e:~.. '.'.;g ?
*1.
.:.o . . .: t.
.,.::e..e,.,....*:. :*.y: .- e. s s. .:.:.?- g.'.,,;' NW.. . ~.: .
I
. .,. :.s . . .. s- . .r..s.u.- 4 ; . . ...:..
.....:. a ,~.*~.. o. c. ... .J.g,. :.
o ?, ;;
..=. . .
. .. r.
..t'.y. . y; ?:.~. 2 ;
- I V. a.. ..J..s.$ -
(?6:.'d, ;:.
.:i.*.:;,.:., i:;ggy 9: :':s.~~:..t.
- n. - . :, < - a .-
$y:->: p:
. a.:
- . - . .z *W
'f'. .?. , ;l0.'.f, .M.. t... 9r.,3.p:s.
=. . .. . -::.: .. n '8.
%.,.*;i's
.6. .,i, l.Y, st.=9.en.
. -e..> :
=.:.=p%..
- y:.;p e.
- .r
- .:.M ny::.:-
r
.g.
U:'5:2'2::: .:. ..%'h:y::;.";.S: l : :.: -;'.li ::.o.;:h,' Y ,:.r.-s::#:'hll. 1 9.-kV:'
- .~,: s.... -: .., . . .. : .
-e.<. . -
.s . .3.,:
- - . y . .:.n.
.5s.;. -x..a. . . :..-s.: . :. ::t.4 l ;;"* ,~: (M... ' . ,c:^a. a. ..... h ... ,~.4.cm.q,::;:::p
.i'-97 f..
1 2n .t r
. : g
.::: y*. Q.
.? t -Q:i 2:- -. :s,-
s.r, n . t. r V:L. ^ , ,1.;::q:.. ~ .1. ; < a co .M.,:.;.;:..4.h
~.t
- . : c.9::::ej:.
- .-:.: : *:..'i. l
,, -;: .? Q -
- - ~-
&.;..,5..<:q,.u.:y
<..r..
=%/- -~ g- -
i, 3 - 4 .-'M d' , 7 h'* KEY *- y., 5-
~
"* d:
y-;>l I TANGENTIAL BEAM PORT (I) l ' j ' ; THERMAL $;j,' 2 HYDRAULIC RABBIT l ..
.'. COLUMN $,..j.i. 3 PNEUMATIC RABBIT ITHER) j* REACTOR "!.I.3 4 PNEUMATIC RABBIT IEPil l ^' v*~~
i :, .
^
POOL 2. *4 6 RADIATION BASKETS (21 g ff.' dil: 6 FISSION PLATE-U-235 6 s d "r *- gl 7 TANGENTIAL BEAM PORT e
" (
N
- o 1, 1 CD i
N FIGURE 5-3 UrlivERSITY OF VIRGINIA REACTOR OPTIMUM CONFIGURATION
! CORE IN FORCED CONVECTION COOLED POSIT!ON.
d
. .i.>-
-w w -
o a i .
/~i 5.4 Fueled Exoeriments j
) '
5.4.1 L Introduction The purpose of this analysis is to assess the hazard
- associated with the failure of a fueled experiment and subsequent
[ release of its fission product inventory. The Maxinium Permissible [ Concentrations of 10CFR Part 20, averaged over a 1-year period will be L used as a measure of the consequence of a fission product release. Exposure to an isotope, in the concentration shown in Table II, Appendix B of 10 CFR 20, for a year (3.15x107 sec) is considered to be the Part 20 limit. Accident exposures to concen*: rations of the same isotope, for i specific time periods, are evaluated according to the following equation. l (AccidentConcentration.hx(TimeofExposure)(BRR) Fraction Exposure (FE) - 7 MPC (Table II)(3.15x10 sec)
, n() (5.4.1) where BRR - breathing rate ratio, the ratio cf the breathing rate assumed during an accident to an average breathing rate.
Only the isotopes of Iodine and Strontium-90 are considered in this analysis. The reason for this is that these are isotopes having l (: relatively high fission yield and very low permissible concentrations.
- j. If the concentrations of these isotopes are within permissible levels, !
l' the concentration of others will also be tolerable. 5.4.2 Isotone Release j' The amount of radioisotope released from a fueled experiment and becoming airborne is assumed to be that specified by TID 14844: 100% Noble Gas; 50t Halogen; 1% Solid Fission Products. These values provide
,Q 59 V
l
[I i l j ~x, a very conservative upper limit for fission product release from a fueled ,
<} experiment over a broad spectrum of accidental conditions (e.g., fission 1
3 . plate clad' failure). Isotopes initially released to the reactor building in general (other than.to the reactor room), are assumed to be released
, subsequently to the atmosphere at a uniform rate, over a 2-hour period. [
This assumption maximizes the inhalation dose (or isotope concentration) for a 2-hour exposure time, by making the entire isotope inventory . available for inhalation. For material released to the reactor room, , which'is automatically closed upon receipt of a high radiation level signal, the isotopos are assumed to be released at a uniform rate over a
. period of 20-hours.* The retention properties of the reactor room, a vindowless structure. designed to provide confinement, have been measured, L t and the 50%-in 20 hours exfiltration value verified (see Section 6.1).
(O In case of fueled experiments which are operated in the reactor pool, a reduction of a factor of 10 will be assumed for iodine isotopes released to the atmosphere. This " partition factor" accounts for the , s 4
*A gamma ray sensitive chamber at the top of the reactor pool provides automatic closure of the reactor room (and a reactor trip) for radiation levels above the set point, normally 30 mR/hr. In addition there are 2 independent air monitoritig instruments in the reactor room which would sense a fission produce release and alarm. The operator ,
would have ample time to close the reactor room to prevent an excessive amount of exfiltration by the building exhaust system normal rate ~11% per minute. Note that forced exfiltration by the building exhaust would , give an elevated, puff-type release, resulting in lower site boundary doses.
.O 5-10
h. L u' L-f'N solubility of iodine gas in water. and the fact that any fueled L i) experiment operating in the pool, at a significant power level, must be located near the: reactor core- under approxinately 20 feet of water. In
. a fission gas release from a fission plate at the WAR on May 3,1968, isotopic concentrations of noble gases were measured in the reactor room, amounting to ~0.1% of the total fission plate inventory. Iodine isotope concentrations were too low to be detected. However. by inference from the instrument sensitivity, they were released form the pool in amounts smaller by at least a factor of 10 than the noble gases, i
5.4.3 Offsite Exposures I For isotope releases from the facility, the following
.information is used to determine fractional exposure to an isotope.
Accident concentrations are computed using the conservative [)
%J meteorological model of TID 14844, and a 70 meter exclusion radius. The time of exnosure is taken to be 2 hours (7200 sec). An accident breathing rate of 3.47x10 4 g3/see is assumed, and a yearly average breathing rate of 2.32x10-4 M3 /sec (i . e . , ER - 1. 56) . MPC values used are taken from 10 CFR Part 20, Appendix B, Table II.
5.4.4 In-Facility Exnos e ! i' Fractional exposures due to inhalation of isotopes by personnel I within the building are computed by assuming mixing of the isotope in .j question with the air in the space in which the release occurs. For calculational purposes, the exposure times are set at 5 minutes, a period considered long enough a detect a release and evacuate the affected ! areas. No correction is made for exfiltration from the building during exposure time.
' b) 5-11 l
i
e-f f rN "
\
)-
5.4.5 calculations ; l- 5.4.5.1 offsite Ernosures Fractional exposures (FEi ) for each isotope, and the doses due , to each may be calculated. The atmospheric diffusion expression of . TID 14844 (Section B) may be used to determine the downwind, site - l boundary, concentration of any isotope. Concentration: El (5.4.2) Xi (d) - xyCCd ygn where X i (d) - concentration of isotope "i" at a distance of d, Ci/M 3 Qi - isotope "i' source, Ci/sec - inventory /exfiltration time i Ei-windspeed,1 meter /see n Cy - horizontal diffusion coefficient, 0.40 meters /2 Cz - vertical diffusion coefficient, 0.07 meters"/2 d - distance downwind, 70 meters for exclusion radius l-n - 0.5 , Xf (70 M)~
~
1
= 0.039 Qi w l'(0.4)(0.07)(590)
Fraction exposures are determined using Eq. (5.4.1), in which the accident concentration, Xi (70) found by calculating a value of Qi for each isotope: 5-12
4 U , [- (Inventerv of Isotone "i") x (Release Fraction) [ '# Si ~ Exfiltration Time (2 or 20 hrs.) E . Table 5.4.1 presents fractional exposures calculated for the iodine isotopes and Strontium 90. The iodine inventories refer to L infinite time experiment operation at I watt, and the Strontium 90 concentration to 6 years of continuous operation at 100 watts. The period of isotope release (exfiltration time) and of inhalation exposure
- L-are both two hours.
For the case release from the pool to the reactor room, the iodine inventories specified are 100 times greater, corresponding to infinite operation at 100 watts. Here a water-to air iodine partition of 10 and a ' 20 hour exfiltration time are assumed. These factors combine to reduca () the iodine isotopic release to exactly that of a 1 watt building release i L U (2 hour), and the Fractional Exposures are the same as in Table 5.4.1 for the iodine isotopes. Sr-90 Fractional' Exposure for a reactor room l-
- i. release would be lower than for a building release by a factor of 10, due to the increased exfiltration time.
l l 5.4.5.2 Buildinc Exposures To calculate building exposures, the isotope concentrations are < l- found by' dividing the fraction of the isotope inventory released by the volume of the space into which the exposure takes place. Fractional i exposure is then calculated in the same manner as for offsite releases. l~ L The results of such computations are shown in Tables 5.4.2a and 5.4.2b below, for the reactor room (2.3x1033H ) and first floor experimental area (1.7x1033 M ). ) C 5-13
.( .
O
~
O .O-i : i l TABLE 5.k.1 i j EXCLUSION RADIUS FRACTIONAL EXPOSIRES 4 i 2-Hour 70 Meter lo CFR-20 ! Isotope Accident Release Concen- Appendix B. Rate Q . (7.2x10 3sec)(BBR=1.56) Inventery, Release, tration. TABLE II Isotope Curies 50% or 1% Ci/see 3 X C1/M , MPC, Values %()MPC, )(3.15x10"see ) _-i. a i
-s -1o I-131 0.0251 0.013 1.93x10-6 7 5x10 1x10 0.27
~7 2.6kx10-8 I-132 0.0381 0.019 1.03x10 3x10 ' O.012
- - -7 -l'
. u. I-133 0.0563 0.028 '3 9x10 ' 1.52x10 hx10 O.136 . b ~7 ~ I-13b 0.0658 0.033 b.6x10 1.8x10 6x10 ' , 0.010
-7 1x10-8 I I-135 0.0510 0.026 3.6x10 ' 1.bx10 0.051 Total rogo, 0.2363 0.49
{ i
-5 -11 sr-90 0 75 0.0075 1.ohx10 4.1x10-' 3x10 o.k9' l
j Sr-90 Fractional Exposure = 0.05 for a reactor room release. i ( I
. , ~. ._. -
~
t 4 n-l
, TABLE 5.h.2a REACTOR ROOM - 5 MIN. EXPOSUhE Accident y Fractional Isotope Release, i Exposures Inventory, (0.05 or 0.01) )(300 see)(BRR)
Isotope C1 Ci inRoga Ci/M , C(3.1Sx10/sec) I-131 2 51 .13 5.Tx10~5 85 I-132 3.81 .19 8.3x10-5 ,gy ; I-133 5.63 .28 1.2x10~" k.4 I-132 6.58 .33 1..hkx10~" .36 I-135 51 .26 1.1hx10-4 17 t Total Iodine 15.4" L Sr-90 75 .0075 3.3x10 1.63 l
. e The reactor room is a Restricted Area, thus the MPC values of Table II, Appendix B,10 CFR-20 do not apply strictly. Since yearly dose limits on such areas are N10 times greater than for unrestricted areas, the total iodine FE value of column 5 is the order of 15 times allovable.
l l r i 5-15
l l 1 1 l 1 (
~ j TABLE ' 5. k'. 2b - !
l EXPERIMENTAL AREA - 5 MIN. EXPOSURE l l 1 Accident y Practional I _ Isotope Release, i Exposures 1 Inventory, (0 5 or 0.01) in Area Isotope ci LX)(300see)(BRR) ci . Ci/M3 MPC(3.15x107see) l i I-131 0.0251 0.013 7.Tx10 1.15 I-132 0.0381 0.019 1.12x10-3 .05 i I-133 0.0563 0.028 1.62x10 5
.60 I-134 0.0658 0.033 1.95x10-5 ,o$
I-135 0.051 0.026 1 55x10-8 .23 I Total Iodine 2.0 St-90 0 75 0.0075 k.5x10-' 2.2 1 1 O 5-16
-a, , e. w y... ., , . , . * . . g.e.e-, e_ ,m-, g. +*.w-=g-
, , .9 *4g , .9+,. 9 yw,+w-y,,, ,-9.-,,-,,.wyw-y-pp,.. _ -m,,g ,,,.-yv.,w
,y.-w--wW e h em- p=w"wwwvvw-*ww'w*Pge-'twu,we* -w-w-'
I i l L i
! s 5.4.6 Conclusion a ).
The offsite Fractional Exposures for Iodines and Strontium 90 l
, (Table 5.4.1) are within the limits of 10 CFR 20, for very conservative j I
isotope' release assumptions (TID-14844). The Fractional Exposures for 1 spaces within the reactor building, considering the same conservative release fractions, and a 5 minute exposure, are only slightly greater i than the 10 CFR 20 limits for these isotopes.
, It is concluded that fueled experiments can be operated within power limits set by the Technical Specifications without undue hazard to the general public, the reactor staff or visitors. It is noted that the isotope inventories are related to fueled experiment thermal power or fission rates. These values can be ascertained by fission product radiation intensity measurements, or from fission neutron yield
(} measurements (G.I. Coulbourn and T.G. Williamson, NSE, 35, 367 (1969)). o l l
-(}
5-17 i l~
i
<~N 6.0. RADIATION HAZARDS V
6.1 Confinement The reactor room above tha pool level is of cylindrical construction 54 feet in diameter and 36 1/2 feet high. It is designed to withstand a differential pressure of 1/2 psi. The walls are of reinforced masonry,
. plastered on the inside for gas tightness while the roof is a concrete slab. The openings into the room are the truck door, the personnel door, ,
the escape manhole,.and the air intake and exhaust ducts. The truck door is of steel construction and is opened and closed by use of a manually operated chain fall and gear arrangement. It initiates a' scram condition to the reactor when moved from the fully closed
\.
position, thereby precluding reactor operations when confinement is broken at this point. I () x_- The personnel door swings closed against a rubber gasket by gravity. . l~ It is held open during operations by a magnet which releases automatically in the event of a high radiation level in the reactor room as detected by the monitor located on the reaccor bridge. Should an incident occur and pressure build up in the reactor room, the pressure will seal the door tightly. The personnel door is illustrated in Fig. 6-1. L In the event of an accident which seals the personnel door, the l operator would be trapped in the room. A simple underwater escape hatch is provided as an emergency exit. The emergency escape hatch is normally closed and secured with a slide bolt which will allow easy opening from 1 l-the inside in the event of an emergency. The escape system is shown in l Fig. 6-2. l
<-~. 1 t j 6-1 1
1 1 l l l L 1 l' . 1
.- _ _ . . ~ , . . . . _ _ . _ _ _ ..
..e 9
n
, P k
s N s P p l 2 6
-O 59 l g
'" (
GASKET 1 l- ' 1 L l FIG. 6-1 PERSON NE L DOOR l l 1 l.'. 6-2 ! l l
ia " '
~ e .
(r' _(s V
- A
- e. OUT5eDE. . WAt.t. -
1 - I 'l - -l l - ! Weed 4 t 4 REACTOR stocu r (FaMST) FLOOR ~ , f, -* *. ! WATER Lt.W L %
- 4. e , ....a..
6 m. d a n w a s ,* _ -. - s
}'
,. ~
i a .. ! 6 j i 4 4*-e' to w . s j . OF STANDRPE - = + J s .a P & q
..-a , .
e a . . - 4 i l
. t i
s s* .
. .s 7 A i r. A
{ e 4 - 3 ' , , s' A . 3.J.* , 4
, dy',
i l-
..r. ; .
. ..A . v. u. s. . . , ,
a b .) t i ! Fig. 6-2 E)(IT MANHOLE 4 4 4 1 1 (
I ]
. @' , i g<
R: ; (3 .. The. reactor room ventilation exhaust ducts are operated in a manner Ls t~) [ similar.to'the personne1' door. Each duct has a gasketed, solid, ' inside
!' door which is opened manually and held open by a magnet. As in the case '
l of the' personnel door, these magnets release automatically upon a high , I . t p radiation level at the reactor bridge. The pressure tight air duct is shown in Fig. 6 3. L- Past measurements of the leak rate from the reactor room verified that the exfiltration rate is less than 1006 in 20 hours. The leak rate of the reactor room was measured using 85 Kr ac a tracer. The measurement involves releasing a small amount of the tracer isotope into the f containment volume, and subsequently monitoring changes in the relative $ specific activity of the air in the room. Such changes can be related to the rate of air leakage, f) uj Krypton itself is one of the principal fission product gasses, so its l use allows the measurement to be made with an actual isotope of interest. ; 85 Kr has a half-life of 10.7 years, and decays with the emission of a 0.67 MeV beta particle (99+%), and a gamma ray of energy 0.52 MeV (0.7%). The two decay products allow two independent measurements to be made. To make the measurements, detectors are arranged so that they view
' P I
y the air space within the containment. The beta counting channel need only consist of relatively thin-walled G-M tubes. The range of 85Kr beta particles in air is about two meters, thus the count rate in this ! channel reflects the local specific activity and is indicative of average specific activity only if there is good air mixing. The gamma-ray counting channel must be able to detect selectively the 0.52 6-4 i
/\
v l' I l l-
~ J.
.O O O .
1 4
)
, ; i , ",* A b * ,
- P W D 4 *3
) T T
/ D &
/
i / s j # - , l ( j ' u o . 8 l M . I
. g LL-AAROUND SPONGE- *
, NEOPRENE 6AStGLT
~"
i
.r, i'
\
4 i ~
/ i
- i i i 1
I ELECTROMAGNET - e
't
! o l I i FIG. 6 -3 PRF_SSURE-TIGHT AIR DUCT ;l 1 i i 4 1 . __ . . _ _ . , , . . . - . . . _ ~. .. . . , - . . ,.- .. . .., ,
. .. ~ .,. . , . ~ . .-... . ..
E t: J~ MeV photon from 85Kr. 'The gamma ray count rate provides a direct
'~'
measurement of the average specific activity, because of the long range of these photons in air (relaxation length about 90 meters). The release of a concentration of 85Kr of 8x10 6 pC1/cm3 provided an ! initial count rate of 3000 cpm above background from a bank of 3 $ aluminum walled G M tubes, 3 inches long, and 3/4 inch in diameter. The L 0.5 MeV photopeak count was 50 cpm, for a 3 inch x 3 inch NaI(Tf) 1 crystal. . Figure 6 4 shows the count rate as a function of time for both beta ( and gamma rays. The leak rate of 26% per 20 hours was well within the ' Technical Specification 3.7 exfiltration rate of 50% in 20 hours. 6.2 Shieldine - t The pool is shielded by earth on three sides, but the fourth side consists of a massive concrete shield with thicknesses ranging from (')'N a maximum of 90 inches,at the bottom, near the core region, to a minimum - of 30 inches at the extreme top. This shield is penetrated by two 8 inch , team holes, a large access facility and a thermal column. These penetrations are filled with concrete plugs when not in service. Further descriptions of these facilities are provided in Section 5.0. A complete survey of the University of Virginia Reactor Shielding , was made by the Neutron Physics Division of the Oak Ridge National Laboratory. The results from that survey with the reactor et a power of one megawatt are described in the next paragraphs. The thermal-neutron dose rates observed over the surface of the shield are plotted in Fig. 6 5, in which the three accessible surfaces have been developed, or " unfolded" to lie flat on the page. f) N_/ 6-6 c
- , _ _ _ _ _ _ _ _ _ _ _ _ , .m
~
r , .. , ! ,- :e
?.
! i.
I L O - G#t% RAY CHAtNEL MTA - COWTS PER 200 MINUTES I O - BETA CHANNEL MTA - COWTS PER P4tNUTE 6 I , I I i 1 l ' l 5- - g .,,, f O > 4 - g - 3 ,,, pLEAK PATE 26.7% PER 20 HOUR _ O
.[ g 2 I '
f -
?
103 -- - - - 9 -
- g. - < LEAK RATE 25.9% PER 20 HOUR _ l 7 -
g - 6 - 5x102 l I l i I O 20 40 60 80 100 120 140 T!!E, HOURS FIGURE 6-4 WAR CONFINEMENT ROOM COUNT RATE MTA ba 6-7
"o- 9 as am . e
- 9-i 1
. .J. } q s **, ,-
g' ,-f.c," w g g j g l
"s
\, S, % 4 T I s '
\ '\ \, \ **, '\\ \ \
~s !V,, \a s 't ,
g
% \ \ '% g .
5, \k k \.. k*. H k k *
*S!!!
t a
%... '\ \ '\ *%. " SIS--h .l},---\- -
s o
% \ \\\ %
I % { +}Qg ( s-s g s,. Q
- a.
+
e 5 1.e g \; % %' \ \ 'h, 'Q f l
'- . J !i i ;
\ .\ g .*., .% .% \
y- % y g ; o - l 3 . 4 5
,j
%, , Sj
' * .+g Pm
*( g g '>,
j 1
" I !
lg: i ..
! 8 r -
2 glik " r)!ii - B o sya e
- l. - ,. .. .- -, ....-. .--- . ...- -. ....-.. . ..
- g i
f y _g _%.,_k. k. _ . . j i ly" ~N Y
\ '\\ g g )
Nl 1 3
\ \\ \ \ \\ \ \ W 3 :
W \\ h. \ \ 1 g U~s % \ \ 3 s y L. \ \ \\ \\ \ \ h ! e
, g-.)S,h,hd"5,-- '"g \-. k '
, g.
I 7'11 3 L O.
\\ \\ \\ A \
s L I g gg ; I i t 4 g% N g .
'd%" b'
%g L g g %g g' g' g' #
i L x \s w s s"~ \gj j ; u
- s. . .
s .s s ii j - s# : 4,4] ., g ! .
~
L
- 3
'.i l It i 2 i
li i -i
- f ra. E 80IE
_O - L 6-9
, . . _ . _ _ , _ . . . . . . . . . _ . _ . _ , _ , _ . . , . _ . . . _ , _ _ . . . . . . _ . - . _ . . _ _ _ . - . . . _ _ _ _ _ _ _ _ - _ . _ _ _ _ _ _ . . . . _ _ . . . _ _ _ . . _ ~
- p. .
q l
.li l
[l O s. t 41 si l lli - , i kilf\hY5'I s \
'N}! j x
j
%,. \ . \'s \ \ \'t,)g tie, <$1 ] n e l ;1 ,1,\ljj -
L
'Ofj\L_gg_IMl 'i ;
\lp; g \ l t
% \ \ \ ' 4 TA j , p >..[sql gs, g lI \
~
l
.Al__A_._%).!A \_._%\__\ _,}h_._\_.l5 3 .
O \\\ ~
\ % : s s '
s 1 g ' 4_! k k" tf il_h l ;
\r \ \ '( P!!! 5 r n 2
! % \ \ \ \ '$. \\\ ! s \ pi ! a 1 L I
'{ ~
\ \ *>J $
,! l l
- l ri -
3 1 l o
,4 - ,- 7 E Bj e
- Ij l -
B O l in - 6-10
1 ( ', i l l rS Fast neutron doss rates are plotted in similar fashion in Fig. 6-6, and gamma ray dose rates in Fig. 6 7. In connection with the fast neutron dose rates reported in the ; t[: present work, some comments are required. The modified 1cng counter with which the fast neutron data were obtained is strongly directional, responding in terms of correct dose only when the neutrons are incident upon its front face. This factor is inherent in its design. Its response to neutrons impinging upon its sides is considerably higher. , Thus, the dose ~ rates measured by the modified long counter must be i t considered to be an upper limit to the true fast neutron dose'at the point of the measurement. Against a shield face, or in measuring neutron streaming.through a crack or hole in a shield, the dose rate measured is probably very close to the true value. As the neutron field surrounding ()
%/
the counter approaches isotropicity, the dose measured is expected to be an overestimate, possibly a considerable overestimate. This latter i condition would apply to measurements made in the vicinity of the heat t exchanger. Fluxes and dose rates were-in general very low. Streaming around the thermal column and access facility closures was somewhat evident, amounting to a maximum fast neutron dose rate of 0.0652 mrad /hr, a thermal-neutron flux of 10 to 12 neutrons em-2 ,,c-1, and gamma ray dose rates of roughly 6 mr/hr. Data was also obtained at distances from 4 to 6-inches above the pool water surface. These values are plotted in Fig. 6-8. Gamma-ray dose rates as high as 7 mrad /hr were measured. I 6-11 I)N L e - , wen,-- -,. c g- , . , - , - , - - , , .
~1 s i', ;
-I 1
_.t: 1 D i e i asm
.am sse .eoe l
.ase .a.e .eem w.
g__ i te_ ._._. __ I . =__
%
- este l ,
.am .asm sue .some h--
I ,
.ase mas .ese .e i
t d e i GuuumB*GEp Gems .imus 0 6 mes. ums Bar.# of punL i J a
. ..>. p.
,>., > . , m T..
1. s>= * - - _-_e .e-o L _. _e . ._ _., . .> > ...L _ _
>. . ..L.
l I 7 I
- , Pup *etuusses tous Asums mea 6 sense tu.fase et penL r
I i i i. [' t.>e e>. .aw.i t.tl> .
. sere .ar>.s ..ep. .r.>.s
..>e ..'p t.>.. p.>m . ..
. . > . , ..>. .e. . ..>..
I. i %
- . .. w . L., -
L 1'
>.i .'
l i l i e i n n ri ..m Gamma-Ray and Fast-Neutron Dose Rates and Thermal-Neutron I O(_./ . Figure 6-8 Fluxes Near Surface of University of Virginia Reactor Pool. L l- 6-12 1 i
== *-t---W er--vyg enr ea,-w--e. e e- g -'r*--rw-e-lr-5ete--rp.--y-w,u-emwe g-,wge+-y.e w---g. g rw, y-.,g-,em- , . - -. - - - , - - - , -. ,, p- e,.r-+.-y- we
v .. 4 t ('~; 6.3 Hazards During Normal Ooeration vF Calculation of the concentration of Argon 41 in the beamport i extension after prc)onged operation at 2 MW predict a discharge to l, J the atmosphere that was approximately e factor of two below the maximum permissible concentration.- Comparison of experimental measurements by E the ORNL team and calculated values of Argon-41 concentrations demonstrated that the assumptions made in the calculation were conservative by an additional factor of three. These factors, combined with the . ease.with which the concentration of radioactive Argon-41 can be j reduced by delaying release after operation to permit decay, indicate 1 that this hazard presents no problem at a reactor power level of 2 MW. The measurements of radiation intensity by.the ORNL team at 2 MW
, indicate high gamma ray dose rates at four locations: (1) at the
/"'N i
V i surface of the pool immediately above the reactor core, (2) around the shield plug of the large access facilities, (3) in the heat exchanger room, (4) in the main demineralizer room and (5) in the source storage room. Current policies limit access to the heat exchanger room during 1 operation and the safeguards are adequate to control this hazard. When the limited amount of time spent at the other three locations is ! l considered, these do not appear to be a serious hazard. However, marking ' these locations will serve to remind personnel of the existing dose rate l l at 2 MW. l l l l l l fD g 6-13
( , M t
.(3 7.O HEALTH PHYSICS w Q,)
~
. 7 .1 ceneral Information J
The Reactor Facility is a research tool of the University and . C' as such,< subject to use by all of its schools. It is the responsibility . t of the operations staff and the Health Physicist to provide and maintain full use of this tool, yet prevent undue risks and hazards to the
+
t individual workers, the University and the Community at large. l The Health Physicist is responsible for assuring that those l measures and regulations pertaining to the Health Physics aspects of the reactor and its operatian are carried out and mantained. The Director of the Reactor Facility is advised by the Health. Physicist in all pertinent 1 matters. The close association but independence of the Health Physics and Reactor Facility operations has worked well at other reactor n4 4 installations, and the University has patterned its organization
.x_/
accordingly. 7.2 Education in Health Physics It shall be the duty of the Health Physicists, or his designee, to periodically instruct Reactor Facility personnel about the risks and l l; hazards of radiation and the means of lessening this danger to themselves and others. This shall be done as follows: L (A) Each individual with unescorted access to the Reactor Facility l i will be given an initial indoctrination lecture about Health Physics, followed by a question and answer period, so that the biologic aspects i and the genetic aspects of radiation change are understood, i (B) On the-spot lectures may be given by the Health Physicist, during a particular phase of operations, to emphasize the protection
,~ .
(' aspects of Health Physics. 71 _ _ . _ . _ _ - _ - . _. .- _. ~ ._- _ _ . .
p - P L , [ [ D'
'(C) Pre experiment evaluation of hazards associated with a j:, . L).
particular experiment will be performed first by the individual proposing k ' a new experiment and then by'the Health Physicist. (D) A radiation log will be prepared for each " permanent" worker at the facility. In this log the monthly dose data will be recorded. 'l 7.3 Personnel Monitorine and Protection The Health Physicist is charged with the procurement and maintenance of the detection equipment and the dose badges for personnel exposure monitoring. Badges: These will be used for monthly checking of personnel neutron'end/or gammma dose, with the dose evaluations performed by a commercial supplier. l Pocket chambers: Direct reading - will be worn by personnel working
,w
;] in suspected high radiation areas.
Finger badges - will be worn by personnel handling highly radioactive material. Protective clothing - street clothes are worn by the majority of workers at the facility; however if there is a possibility of personal or area contamination,' protective snd/or disposable clothing is provided and will be worn. l 7.4 Permanent Monitoring and Surveys li I Fixeo Radiation monitors are mounted in the following areas:
- 1) On the reactor bridge, !
1
- 2) Ground floor wall (reactor face area), f l
- 3) Outside the hot cell, j 1'
- 4) In the demineralizer room.
I The readings of these, monitors are displayed individually on the i 1 b
r l secondary console in the reactor room. The existence ci excessive t (w) _ radiation in any of these areas causes an audible alarm to sound at the
} console. The initiation of the reactor bridge monitor alarm and the ground floor monitor alarm isolates the containment, as described earlier, and scrams the reactor. In addition, numerous portable instruments are available for surveying all areas in the facility.
Calibration of these instruments on a regolar basis will be assured by the Health Physicist. Calibration records will be mantained. The initial run of a new experiment or the use of radioactive materials will be extensively monitored and a record maintained of the results, if the work is such that the Health Physicist determines a possible radiological hazard to personnel exists. Facility radiation surveys, air sampling, and contamination smears q
-( ) will be performed in work areas on a regular basis by the Health Physicist or his designee. Such official will be recorded. Informal surveys made by the experimenters of their work areas to confirm typical (safe) radiation levels may be made unofficially and need not be L recorded. However, if excessive levels not previously identified are found, the experimenter will notify the Health Physicist, who will then make an official survey.
7.5 Prohibitione and Sanctions The Reactor Safety Committee and the Reactor Facility Director are responsible for enforcing all applicable Federal, State and University regulations necessary to run the facility. The Health ; Physicist will report irregularities and recommend necessary steps for ,
?
73 , (~ l-k, - l t ll;
l l r ! [-) their correction to the Director and the Safety Committee. The Director v . i. will determine corrective actions to be taken in such cases. However, ! the Health Physicist may issue emergency orders if necessary, on his own responsibility. Safe areas for eating and drinking will be designated by the Health .i i Physicist. Smoking is permitted in private offices and specially ; designated common areas where unsealed radioactive materials are not g used. ! The University of Virginia has established a whole body personnel dose limit of 0.5 ren/ year which is 106 of the limits in 10CTR, Part 20. If an individual receives a radiation dose in excess of t.hese lirits as j determined by ionization chambers, dosimetry badges or other methods, the i Health Physicist will notify the Facility Director. The llealth Physicist - will provide information concerning the e. mount and type of exposure and i recommend actions to be taken by the individual to avoid further, similar exposures, j 7.6 Waste Discosal The Health Physicist will check radioactive waste and refuse i from the Reacator Facility, monitoring it prior to legal disposal. The t public water system will be separate from water drainage systems used to l L collect and discard radioactive material. The disposal of liquid radioactive wastes is discussed in Section 4.8 of this document. Dry ! litter and waste will be stored until it han decayed to safe levels or I can be shipped to a licensed burial site. Radioactive materials may be j released only with the approval of the Health Physicist. L : l \ 74 s
- - - - ,v- - en--, -
1 i 7,7 Shiening and Trananort L .(x} .. s ' I Radioactive material produced or purchased by the Reactor l l Tacility will generally be used on site. }!owever, in instances where l material must be shipped from the reactor to one of the schools on the University grounds, or to national or international recipients, f applicable DOT regulations governing shipments of radioactive material, i k. t t as outlined in 10 CFR Part 71, will be followed. Radioactive material i t i will not be allowed to leave the reactor site unless the recipient is ; i qualified to receive the materials under NRC or international { 1 regulations, as applicable.
- i f
i [
> I f
I f i l l t i { 7-5 l
e- ] o , l l l
/ 'y
~ '
8.0 ADMINISTRATION I N l
-- )
8.1 General organization i The reactor is operated under NRC License R 66 granted in 1960. The Reactor Facility organization responsible for assuring safe reactor operations and use of radioactive materials at the University of Virginia is shown in Fig. 8 1. This organization includes two major committees: { the University Radiation Safety Committee and the Reactor Safety t i Committee. t 8.2 Radiation Saferv Committeo The Radiation Safety Committee is appointed by the President of , the University and must approve the possession and use of radioactive i materials at the University of Virginia with the exception of those ! associated with the Reactor Facility. Production, possession and usage f t
/ ) of radioactive materials at the reactor comes under the reactor license K/ [
and is reviewed by the Reactor Safety Committee. 8.3 Reactor Safety Committee i As required by tl.e license, a Reactor Safety Committee is j active. As a minimum, the Reactor Safety Committee is composed of at least five members and includes the University's Radiation Safety Officer, the Reactor Director and a professor from a department other than Nuclear Engineering. Senior members of the Reactor staff attend ; i committee meetings in an advisory capacity, but there is only one vote by ; i the Reactor Staff which is cast by the Reactor Director. This is to ; i prevent domination of the Committee by members of the operating ! t organization of the reactor. The Reactor Health Physicist is also welcome to attend Reactor Safety Committee meetings. O 81 V 1 [
p ;r~s C8 (a) i
% ./
PRESIDENT OF T T U. OF VIRGINIA I DEANf SCHOOL of ENG. 4 APPLIED SCID4CE REACTOR SAFETY RADIATION SAFETY C0amlTTEE l .ConWITTEE OtAlmeANf DEPT. of IEADf 0FFICE of NUC. M . 4 N. . EWIR0* ENTAL EAL N , , , , , , PHYSICS tleveIII oed SAFETY j 1 1 :
~
I I I DIRECTORf U.VA. I U.vA.' RADIATime REACTOR FACluTY . . . . . . . . . . .... HEN.TH PHYSICIST . . . . . . ........... a tlevel 28 8 M I I i REACTOR stPERvisoR sADeflNISTRATIONI .. ..... .. .. .. tSERVICESI (l*8 38 tsevel 33 , 1 > I I FACILITY OPERATIONf REACTOR SBtVICES a ImINTENANCE a RESEARCH PROGRARIS ENO. SLPPORT t ievel4 8 tievel 43 OWfELS of RESPONSIBILITY
...... .. . CHMfELS of coeWUNICATION Figure 8.1 Organizational Structure of
~
U.Vo. Reactor FacIIity
/
1 m h The Committee reviews and passes on new experiments that could affect the safety of the reactor. These include critical erperiments, as well as experiments in which the reactor is used as a radiation source. Written standard operating procedures (SOP's) and emergency plan and j k implementing procedures (EPIP's) approved by the Committee are in effect ) t for reactor oporations. Experiments involving the reactor are run under l, written and Reactor Safety Committee approved precedures. While the Reaccor Safety Committee approves the procedures to be followed for safe operations and experiments, the detailed routine enforcement of reactor i safety is the responsibility of the operating staff. Also, while the Reactor Safety Committee has the authority to require and approve i l specific procedures to prevent unacceptable exposure of personnel to - radiation, the immediate responsibility for compliance with Title 10,- I O Code of Federal Rerulations, Part 20, rests with the Reactor Staff and, f in particular, with the Health Physicist. l ( : i 8.4 Procedurqa ! The reactor is operated in accordance with written standard operating procedures (SOP's) approved by the Reactor Safety Committee. These procedures include normal startup, operation and shutdown of the reactor. Emergency plan implementing procedures (EPIP's) exist to 1 implement a Reactor Safety Committee and NRC approved Emergency Plan ; (EP). General procedures for the handling of radiation experiments are j promulgated by the University but these are supplemented by special L procedures which apply only to the experiment under consideration. ; Procedures dealing with the operation of the reactor and associated ; experiments must have the approval of the Reactor Safety Committee. ( O 83 ! 1 i
i Changes to these procedures require the approval of this committee, [J however, ainor deviations not changing their intent may be made by the Reactor Director, and is permitted by the SOP's. When deviations to SOP's occur, the RSC is informed by memo from the Reactor Director. It is to be recognized that procedures notwithstanding, the safe operation of the reactor is dependent .ipon the Reactor Staff and their exercise of good judgement. All personnel, including students who work routinely at the Reactor Facility, wear dosimetry badges. Occasional visitors are issued self. reading pocket dosimeters. For large groups of visitors, two pocket l dosimeters are worn by the staff cember acting as a guide. This permits tours to be conducted without issuing large number of individual pocket I dosimeters. !. n l C 8.4 EtsrIda The reactor log book and all other records are open for inspection by the Reactor Safety Committee. These records, plus all the records of the activities of the Safety Committee, are also available for review by NRC compliance inspectors. An annual report is submitted to the NRC listing changes made to the facility under 10CFR50.59 and describing minor accidents pertinent to safety. Major incidents are reported to the NRC as required by the Technical Specifications. 84
---s ,
j l i l
! t i i
() 9.0, Safety Analysis l The UVAR's MTR type LEU fuel elements contain plates with low [ enriched uranium fuel clad with aluminum. Standard elements contain 22 ! flat fuel plates, control rod elements contain 11 fuel plates, and f t partial elements contain 11 fueled and 11 non fueled plates in an { alternating array. The overall external dimensions of these three element types are essentially the same, so that they may fit anywhere on f the 8 by 8 reactor core grid plate. Details of the physical arrangement , are contained in Section 3.0. The limiting conditions of reactor power, coolant water flow rate and inlet pool water temperature are given in the UVAR Technical , Specifications. The bases for the Safety Limit and Limiting Safety , i System settings and the thermal hydraulic analysis are presented in this j
',_)
k' chapter. References used in the analysis are listed at the end of this chapter, i 1 i 9.1. Thermal Hydraulic Analysis of the UVAR , To establish the safety limit, limiting safety system settings, limiting condition of operation, and normal operating conditions for pool l reactors it is necessary to examine factors such as burnout, or an , l ; excursive hydraulic instability resulting in subsequent burnout, which ; 1 may lead to the disruption of fuel plate integrity. In addition, it is j necessary to determine by calculation the operating conditions at which f the onset of nucleate (incipient) boiling will occur in a coolarit , t [ channel, so that reactor operating conditions may be set at levels which l l 7 prevent there phenomena. ; i
\
91 I
r l ( An analytical procedure was developed to predict: 1) burnout, 2) ! , hydraulic instability, and 3) incident boiling, in the hot channel of a WAR LEU 22 plate fuel element. The effects of the following parameters ! were studied: (1) channel inlet (pool) temperature, (2) heat flux distributions, i.e. uniform, chopped cosine and exit peak, and (3) r coolant flow rate. l L The Gambilll correlation was used for estimating burnup heat fluxes for subcooled boiling. Only downflow was considered for forced cooling in the analysis. An equation proposed by Rohsenow and Bergles2 was used for predicting the unset of nucleate boiling and a comparison was made with experimental data obtained from the Oak Ridge Research Reactor.3 Three vertical power profiles were considered in the analysis: 1) uniform power profile, 2) chopped cosine power profile, and 3) exit peak (% Q power profile. The exit peak power profile illustrated in Figure 9 1 was found to be the most limiting and was used throughout this analysis. The exit peak power profile is based upon neutron flux neasurements in the hot channel of the WAR HEU core which is located in a control rod element and is next to the water filled gap left by the withdrawn control rod. Numerous empirical correlations were used in the thermal. hydraulic analysis of the LEU WAR. Because of the importance of justifying the use of these correlations, the following sections provide a brief discussion of the correlations and the ways in which they were employed in the calculations. Where possible, comparisons of analytical calculations with experimental data are made. l 9-2 \ i J
~
O O O i I l 2.6 _ e Experimental I i l ! 2.2 - l , x < a ; i s
- m. I.8 -
! p ! ks I.4 CHOPPED COSINE ,7 y, EXIT PEAK i us l 7 u N
- g l : i 4 - 'N
- 21.0 - UNIFORM ------ ------ - - - - - - - - - - - - ------ -
- E -
t o i l z ' 1 l i .6 - - I ! i i l i .2 - i . j a a e a a a a a a a O .2 .4 .6 .8 1.0 1.2 1.4 1.6 1.8
- CHANNEL POSITION, FT BON FIG. 9-1 HEAT FLUX DISTRIBUTIONS I* i
, . . - - - . . . - . - . . ~ . - . - . . . . . . . _ . . . . . . - _ . - - . . . . - - - - . . - . .. - _ . _ . . - - - - _ _ - - - _
l l i j g Forced Convection Heat Transfer
) 9.2.
In the thermal hydraulic analysis of the LEU UVAR, the flow l velocities are small and the channel length to equivalent diameter ratio ' l is 130. As a consequene.e. part of the analysis involves flows in the l, transition Reynolds number range and entrance effects exist over a substantial portion of the channel length. The Hausen 4equation accounts for these factors and was therefore used. The expression for the local 4 heat transfer coefficient is given by i r D h(z) - 0.116 ( ) N - 125 N pI 1 + f (j) / (#b/#w) * ' (9 1) f where: h(z) - local heat transfer coefficient, BTU /hr ft OF 2 k - fluid thermal conductivity, Btu /hr ft OF l, D, - channel equivalent diameter, ft NRe - Reynolds number I NPr - Prandt1 number . z - axial location in channel, ft. l
#b - fluid viscosity at local bulk te.aperature, Ib./sec ft py - fluid viscosity at local wall temperature, Ib./see.ft I
Except for py, the fluid physical properties in Eq. (9.1) are ! evaluated at the local bulk temperature. Since the wall temperature is ,, i needed to use Eq. (9.1), an iterative technique is required in its application, j l i l. 94 ( l
- -- - - , - , , , , , , e . - ..,---,v -
, --m,,, -----mr- ,, , , , - , - - - - , ,
i I / 'l 9.3. Prediction of Incinient Boiline L-l ; Based upon a semi empirical approach, Bergies and Rohsenow obtained ! a correlation for predicting incipient boiling that is dependent only on ' pressure and wall temperature. Although some of their experimental data
- indicated that bulk fluid subcooling had an effect upon incipient f S
boiling, this factor was not incorporated in the correlation. For water, over a pressure range from 15 to 2000 psia, the heat flux at incipient boiling is given by 2 qID l
- 15.6 p .1 6 (Tw -
TSat)-
. O jp 0.0234 (9.2) where q{g - incipient boiling heat flux, Btu /hr f t 2 p - pressure, psia ;
<~ Ty - local wall temperature, OF x >
/
T sat - fluid saturation temperature, OF ' For pool reactors with low velocity flows, Eq. (9.2) generally predicts incipient boiling at lower wall superheat temperatures than is actually the case. This is attributed to an absence of very large cavities which would support nucleation at the lower heat fluxes in such a system 2. . The wall temperature must be determined in order to use Eq. (9.2). This quantity is calculated using the heat transfer coefficient given by Eq. (9.1). The solution q1g requires an iterative procedure since h is dependent upon wall temperature, which is related to the local heat flux. v 9-5
.[ i <
') In order to test the validity of the use of Eqs. (9.1) and (9.2) for x._/
pool reactor thermal analysit, a comparison was made between calculated
- f. incipient boiling and experimental results obtained from the Oak Ridge Research Reactor (ORR). F15 ure 9-2 shows the results of the comparison, and as seen, a very good agreement exists. For the experimental curve, the anomalous behavior shown by the dashed line has not been explained.
However, the Oak Ridge investigators thought that gas bubbles other than steam may have been produced in the core during this part of the experiments. 9.4. Burnout Heat Flux Numerous correlations have been proposed for the prediction of subcooled and quality burnout. Unfortunately, most of these correlations
-(.v n) . are not applicable to the WAR parameter range. Selection of burnout correlations should be made on the basis of burnout data that fall within the pertinent parameter range. Data collected by Gambill were at conditions representative of the WAR parameter range.
Cambill's correlation for subcooled burnout was used for the WAR analysis because it'is quoted as being applicable for the WAR parameter range and its predictions are reasonably accurate when applied to the available data that is closest to the WAR parameter range. Cambill proposes a superposition approach for treating burnout which is given by l qj' - q y(pool boiling) + q y (forced convection) (9,3) b 96 i
~
O O O i-1 1 ! I4 - i 12 - P= 0.0244 Fas i F > 1500 3 2
- 10 P~0.0225 Fc.s , F> 1700 -
- a
! o e
! w 8 -
EXPERIMENTAL / CALCULATED 1 E / i o / i o / \ to / ! O d ' i F 6 -
- o F /
/
/
4- / l 2 - l 3 a I a a a f I A B i 800 1200 1600 2000 2400
- TOTAL CORE FLOW, GPM 4
l FIG. 9-2 -CALCULATED vs. EXPERIMENTAL INCIPIENT j B6lLING IN TE ORR l
- - _ _ . . --- ._. .. .- . . - . . . . -. .. . . . .- ..- .- -. . .---......--_...-_.-._-.=,._.-....._...._.-1
I I I X The pool boiling term for conditions representative of pool reactors is 5 q y - 0.15g h [ [ ogg, (pgp I/' x 1 1+( *3/'
/ *P( Sat b (9,4) g 9.8 h g _
l i l i where hgg - latent heat of vaporization, Stu/lb , a - liquid vapor interface surface tension, Ibr/ft 2 g - acceleration of gravity, ft/sec lb a A ge - 32. 2- Ib g ,,,2 i t f pg - saturated liquid density, Ib /ft 3 i t 3 p g - saturated vapor density, Ib./ft 0F - cp - liquid specific heat, btu /lb , Tsag - saturation temperature, OF l Tb - local bulk fluid temperature OF The forced convection term is . qy - h(Ty3 -T) (9.5)
- b l'
where h is the Hausen local heat-transfer coefficient given by Eq. (9.1) and Tyg is the Bernath critical wall temperature given by 6 Tg3 - 102.5 in p - 07.2 (p 415 ) - 0.45u + 32, O F (9.6) , O - 98 , t l 1.. - , . _ . _ - .. . . . .-
i , ! t i F [ r) LJ where p is the fluid pressure in psia and u is the mean fluid velocity in { ft/sec. j Because radiation exposure to personnel from N 16 activity in the pool water is a concern, the UVAR uses coolant downflow which has an l adverse effect upon burnout. Buoyancy effects reduce bubble velocities and as a result burnout heat fluxes are reduced. Cambill has studied the r effects of downflow at low velocities and pressures and recor. mends the , i following corrections be applied to his burnout correlation.5 l 1 i q(downflow) , 1 (9,7) q c 6.58 (NGr/Nga)0*39
}
The ratio of Grashof to Karman numbers if given by ) 2 I Pr g $(Tb T1 )D, 49,g) NGr/NKa
.(~} \s fyG 2
- where B - liquid volumetric coefficient of expansion, OF*1 Th - local channel bulk temperature, OF Ti - channel inlee temperature OF fy - Darcy Vaisbach friction factor G - mass velocity, Ib m/ft The evaluation of fg is addressed later in the discussion of flow stability. All fluid properties in Eq. (9.8) are evaluated at the average channel temperature up to the point of consideration. The range of applicability of Eq. (9.8) is for 0.008 $ (NGr/NKa) $ 1.0, for NGr/NKa
> 1.0, and it is recommended that the value 1.0 be used for the ratio in Eq. (9.8).5
() c
' 99 l
l
)
7, i i c 4 Some of the downflow burnout data collected by Co.mbill were at :
; conditions representative of the UVAR parameter range. These data were t taken in a 0.9 cm diameter tube of 25 and 30 cm lengths. The pressure !
was 16.2 psia, the fluid inlet velocities ranged from 0.29 to 7.7 ft/sec ; and the fluid bulk inlet temperatures ranged from S10F to 710F. Figure . i 9 3 shows Gambill's ratios of calculated burnout heat fluxes (from Eqs. ' (9.3) ti. rough (9.8)) to experimental burnout heat fluxes. As shown, the agreement is quite good with 90 percent of the predictions falling within ! 1 30 percent of the experimental values. This agreement causes a ! i considerable amount of confidence to be attached to the analytical procedures. Two further uncertainties exist in using Gambill's burnout ; correlation for the UVAR burnout analysis. These uncertainties are: (1) t
) the effects of non. uniform heat generation in the UVAR and (2) the '
(~'/ N_ validity of applying a burnout correlation based upon circular tube 4 burnout data to rectangular channels. As regards the first uncertainty, l 7 experiments by Todreas indicate that for subceo?.ed burnout the effects , of non. uniform heat generation are not substantial. A review of the Non-Uniform Heat Generation Experimental program of the Babcock and Wilcox ! Company indicates that the total channel power at burnout is essentially ( i independent of heat flux distribution. As concerns the second ! T l uncertainty, it appears the effects of geometry are small or advantageous ! for rectangular channels 8s9, when burnout data with tubes and rectargular channels having the same equivalent diameter are compared. Some L researchers have observed up to 40 percent 8,10 increases in burnout heat ' fluxes for rectangular channels. For the UVAR thermal analysis, (d 9 10 l l l l
O O O N ! x L6 - l - 3 en. i, F l.4 - W i Z .
- - - - ~ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
l o i. -
- z g
i . e 1 E - J * * { 1.0 - g g . . . .
= . -
? W 3 9 E ".
4
- ; 2
- l E OA -
w i 8L x - _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ __.___ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ J ans ! o 0.6 - t w b o e-0.9cm a 25 cm TOSE w [ 0.4
*- 0.9 cui a 30 cui TUBE l
a i I t g g g g g
- O I 2 3 4 5- 6 7 8 9 m ,
COOLANT VELOCITY, FT/SEC FIG 9-3 -COMPARISON --OF CALCULATED WITH EXPERIMENTAL BURNOUT HEAT FLUXES ; !, i i 4
. , . _ + ..% --w w.. .e*r ~+,+++a -e=*-o=-n vy>.-*---=1 e - ew- . - - - , - + - -+wei-w*- e , 4ww--r= s.-,-em - --,,,,=+w+----e--r=*e-.. _ .-*---------.--a-+ms- _ _ . _ - - - +
4 1 i 7 ) burnout or a flow instability usually occurs before the water reaches t V' L its saturation temperature and burnout is assumed to occur when it i
. reaches its saturation temperature at the channel exit.
i 9.5. riow Instability ( Due to the effects of increased pressure loss with subcooled i nucleate boiling and adverse buoyancy from coolant downflow, an excursive hydraulic instability can occur in pool reactor hot channels. Such an instability occurs with flow in parallel channels when the pressure loss in one or more channels increases with decreasing flow rates. The end result of this effect is a drastic flow reduction or a flow reversal in the unstable channels. A flow instability is determined by calculating chaur.el press.ute ;
<~x
( )
% ,/
lossos l'or successively decreasing flow rates, A ainimum on the channsi .
\
pressure loss flow rate characteristic curve lecates thu minimum stable , flow rete. Pressure loss across pool reactor element coolant. channels is due to i the following: (1) inlet and exit losses, (2) channel friction loss, (3) L loss due to changes in fluid somentum, and (4) loss due to change in i elevation. Since burnout was assumed to occur when the fluid reaches its I saturation temperature at the channel exit, only single phase pressure f losses vera considered. : Pressure losses were calculated incrementally down the channel in . order to account for fluid property changes and transition from regimes l of forced convection heat transfer to subcooled nucleate boiling. l t f) U/ 9 12 l l 1 l
r-()
~
The total pressure loss of the channel inlet was calculated usingll 2 l Kpug t-op(inlet) - (9*9) 2g c where op (inlet) - total pressure loss at channel inlet, Ibf /ft2 K - loss coefficient, dimensionless
#1 - fluid dentity at channel inlet, Ib /ft3 ul - fluid velocity inside channel inlet, ft/aec The loss coefficient K depends upon the shape of the channel inlet and the ratio of the downstreas to upstream channel cross section areas, A 2/A1, For shape edged entrances K is given by ll (3
K - 1.5 0.4 (A2 /^1) * (^2/^1)2 (9,go) for A2 /^1 < 0.715, and K - 1.75 - 0.75 (A2 /A 1 ) - (A;/A1)2 (9.11) for A2 /Al 2 0.715. 1 For a sudden expansion in flow area, the pressure rise due to decreased fluid momentum is larger than the pressure loss due to l friction. Consequently, there is a pressure rise outside the channel exit. The total pressure change at the channel exit was calculated f- using ll-
\1 9 13 1
1
l ti l l , , i i [ i
- ~ ,
g Ap (exit) - . Ic
*( )2 (9.12)
,2 2j i t
2 where Ap(exit) - total pressure loss at channel exit, Ibr/ft l
- p. - fluid density at channel exit, Ib./ft 3 u, - fluid velocity at channel exit, ft/sec j A1 /^2 - ratio of upstream to downstream channel cross section areas at exit.
6 Friction pressure losses across c.hannel increments were calculated with the fellowieg equation ll l l u Ap (frictien) .4 fg # - 0 (9.13) f 2 ge D. { r where fp - Darcy Weisbv.h frJction factor , p= fluid denrity, Ib /ft 3 , u - menn fluid velocity, ft/sec [ AL - axial length of channel increment, ft , i D - channel equivalent diameter, it i i Experiments have shown that, for isothermal flow, rectangular l channel friction factors are close to the Moody curve for smooth tubes.12 , The isothermal friction factors on the Moody curve are well approximated by the von Karman equationI3 2 1 f-g (9.14) Iso f 1/2) 0,8 2 log (N"Is8 Q 9 14
I l 7 Friction factors are reduced with heating because of the decreased fluid viscosity near the wall. Therefore, to account for the decreased friction factors with heatingl4 the isotherisal friction factors are multiplied by (pg/Fb)0.14, or fy-fy (pg/pb )0'14 (II) ISO Equation (9.14) is only applicable for fully developed turbulent flow, which implies its range of applicability is for entrance L/D, ratios greater than 50. Since the L/D. ratio for the WAR is 130, the flow is developing over a large portion of the chonnel 1snph. The l' friction factora in devaloping flows are letSer thta those predicted by p Eg, (9.14). Sir.cc friction pressure losses add to the stability of V tiaraflol channel systems, the use of Eg, (V.14) over the non belling i
'- chimnel length is a conservative approach f.n a stabil1R, analysis, j
.With the eccurrence of subcooled nucleate boiling in channe1% l friction pressure losses increase. A correlation proposed by Reynolds 15
! i l was used to calculate subcooled nucleate boiling pressure loss in the WAR hydraulic analysis. The correlation is given by [ (dp/dL)Ma b* b
- cosh (4.6 x 10 6 q" + 1. 2) ('T y3 ) (9.16)
( (dp/dL) ISO T Sat b y g_. ( i where (dp/dL)NB - local friction pressure gradient with subcooled nucleate boiling, Ibg/ft 3 l (dp/dl.)tso - isothermal pressure gradient, Ibr/ft 3 I 1' l
, ' - wall heat flux, Btu /hr ft2 ,
?
9 15 , 1 i L
s [ , ) T b - fluid bulk temperature at incipient boiling, OF
'J Ib Tb - local bulk temperatute. OF Ts g - fluid saturation temperature. OF The isothermal pressure gradient is calculated using Eqs. (9.13) and (9.14) with all fluid properties evaluated at the average channel temperature.
The approach used in the analysis was to let q" be the average heat flux in the channel from the position of incipient boiling to the position under consideration. Hrvever, because of the low heat flutes involved with pool reactors, the contribution of the term 4.6 7 10 6 q. in Eq. (9.16) ta saml1 compar$d to 1.2. Conteruently, the cpproach ueed g f q in evaluating q" is nor too irtport ar.t. It shr>uld be noted that at l conditious in which flow inntrbiitties c'/c:3r in pool reactor 2, the heat ' fluxes are smallar than the heat fluxes used in developing Eq (9.16). I In heated channels, the fluid undergoes a slight density decrease ! i which in turn causes an acceleratior. The increased fluid momentum produces a pressure loss which is quite small compared to the other losses in the channel. However, this loss was incorporated in the ; i hydraulic analysis and the pressure loss across a channel increment is , given by f op (momentum) - (9'17) 2 e i where a - correlation factor to account for non uniform velocity in the f I channel 1 x 9 16 1 1
L -
, t i
1 j $ - fluid volumetric coefficient of expansion,
, N_J ;
AT - temperature rise across the channel increment OF i Since the flow is turbulent 'a" is close to 1.0, which is the value i i: used in the analysis. , i l With downflow the elevation pressure loss is given by j t Exit Ap (e evation) ~ . E- pdz l (9.18)
! Ec . Inlet where g is the acceleration of gravity. For single. phase flow in a ;
channel increment, Eq. ( ? .16') is us\1. approximated by o Ap (elevatiori) , . A* pgAh(1-$4Tg ) (9.19) ;
. , ~) fe o (/-
where p1 = liquid density at inlet to channel increment, lb /fg t AL - length of channel increment., ft B - liquid volumetric coeffic.ient of oxpansion, OF*1 AT - temperature rise of fluid in channel increment. OF The adverse buoyancy effect due to a decreasing fluid density ; downstream in a heated channel is the major contribution to flow instabilities in single. phase downflow. I i 9 17
=
W ,
}
i t s. 9.6. Burnout Ratio ; P ( .) i'
~
Because of non uniform heat generation in the channel, it is ; possible to calculate two types of burnout ratios defined by the , following equations: f e i l !
~
i q" (predicted) Local Burnout Ratio - q"(local) (9.20) , f q" (predicted)
}
[' Average Burnout Ratio - q" (average up to position conridered) (9'21) [
- s In the WAR safety analysis both burnanc ratios sere esiculeted A oug the ch.snnel lendth.
4' The smaller of these retics was used in -[ l Jeter:nining the burnout ileit.
)(N '
Equation (9.10) 1anplies that local conditions detorr.ine burreat. l 1 tiowe,ver, expuf mental d6ta cor paring uniform and non uniform heat ! generation burnout indicates that' burnout depends on upstream heat . flux conditions 16,17,18 The results of experiments performed by Babcock f and Wilcox for uniform, chopped cosine, inlet peak, and exit peak heat
- flux distributions, indicate that the total channel power at burnout or t
the average channel burnout heat flux is essentially independent of axial f heat flux distribution 17,18 Since burnout correlations are developed ! from uniform heating burnout data, it appears that the non uniform heat flux burnout data indicates that the use of Eq. (9.21) for calculating l burnout ratios is probably a better criterion for determining the burnout ' i E safety margin. Equation (9.21) usually predicts smaller burnout ratios
- than Eq. (9.20) for positions far devnstream from the peak heat flux in a '
t 9-18 i
, --n, -- -- . , , , - . , - - - - , - - . - , - . ~ . - - -
f .. a [
.ce
'B
( ) channel, The results of calculations on the WAR showed that Eq. (9.21) LJ was the limiting criteria for determining the WAR safety limit. ; For relatively high power levels and flow ratea, the burnout ratio is the limiting criteria for establishing the safety limit. At lonr power levels and flow rates, an excursive flow instability is the limiting critoria. The flow instability in the WAR hot channel is , predominately caused by the adverse buoyancy effect due to downflow.
- 9. 7 . Nome nc l a ru rg,,.p rJui.1;dhp rnl .lf;digulie Analve,i s i
A2 /Al p.atio of downstrecar to+vpstresur channel creas seccif>n are#5 e'; channal inlet [ ep Lpteifte heat, btu /)b, - %' ; D, Equivalent diameter, it fu Darcy Veisbach friction factor i FyISO Isothermal Darcy-Weisbach friction factor g Local acceleration, ft/sec2 lb ge Conversion factor, 32.18 gf see a G Mass velocity, Ib m/ft sec h Heat transfer coefficient, Btu /hr ft2 .op , hg Saturated liquid enthalpy, Btu /lb , , h Saturated V8Por enthalpy, Btu /lb , ! 6 h fg Heat of vaporization, Btu /lb , k Thermal conductivity, Btu /hr ft OF K Inlet pressure loss coefficient L Channel length, ft , p 9 19 i l l
. _ _~ . _ . _ - - - . - .
I q t
. (] AL Incremental channel length, ft ; i._/
NGr Grashof number, dimensionless NKa Von Karman number, dimensionless NPr Prandit number, dimensionless s NRe Reynolds number, dimensionless p Pressure, Psia q" Local wall heat flux, btu /hr ft2 q" Burnout heat flux, Btu /hr.ft2 l qFC Fo ced constion heat flux, Btu,Ard ; qgg Incipiont boiling heat flux, Bru/hr ft2 q)3 Pool boilin5 borrout heat flux, 'Bta/hr ft2 : (dp/dL)ygo 1sc themal pre.nure gradient, 1.b f /ft 3 (dp/dL)NS Subcooled nucleate boiling pressure Bradient, Iq/ft 3 Tb Local bulk temperature, OF i T1 Channel inlet temperatire, OF f Ty Local wall temperature, OF i AT Fluid temperature rise in a channel increment, OF l Tg,e Fluid saturation temperature, OF .; 1 TVB' Bernath's wall temperature at burnout, OF ! l T bg Fluid bulk temperature at incipient boiling, OF u Mean coolant velocity, ft/sec ' j ut Coolant velocity outside channel exit, ft/sec i z Axial position in channel, ft l o Correction factor for non uniform flow in a channel B Fluid volumetric coefficient of expansion, OF'1 9 29 \ i l l l
w - - i ! i h ! M l {~') p Local fluid density, Ib /ft3
, v i I
pg Saturated liquid density, Ib /ft3 l i t l' pg Saturated vapor density, 1b ,/ft 3 p1 Fluid density at inlet to channel increment, Ib /ftm j o Liquid vapor interface surface tension, Ibf/ft ;
#b Local fluid viscosity at bulk temperature, Ib /ft see .!
pg Local fluid viscosity at wall temperature, Ib /ft sec !
, . P.8. []ct Channel, and MinigvL Core leadine yn -
The hot channel is p lateo to the ucminal channel by the radial , penking fretor. Thr hot channel considered in thin analysis war ; i calculatt>d for a 4 x 4 core which is the minimum core loading. Larger , t cores tiere analyzed *, however, their lower power densities and lower nou. -l
- l
.( e'.etaental flows more than compensste for their higher radial peaking {
factors, resulting in less limiting conditions, j The radial peaking factor for a 4x4 core was determined to be 1.66 f using a two dimensional diffusion theory computer model of the LEU 22 plate per-element WAR core 27 This radial peak was found to be in the l channel adjacent to a control rod water hole as shown in Figure 9 4. ' Sternberg I9 performed a complete map of a WAR 4x4 HEU 12 plate per. -l t element core. The location of the measured radial peak flux in ( Sternberg's core corresponds to the location of the computer calculated l peak flux in the 4x4 LEU 22 plate per element core. , Sternberg's work provided an exit peak axial distribution which was l fitted to a polynomial for the analysis of the 22 plate / element core. Figure 9-5 shows Sternberg's axial flux map in the hot channel. For O 9-21 i t i
4...... . _ . .
,r
, e myvvvvvvvvv0yvvvvvvvvv l , , , ,J vvvvvvvvig_wwvvvww wavv
' N -
l l ! l ;
'l 1 ,
l 4 {l .
'i +l l l . ,
<il l ii ! l ll -
'I: ;jl il ll R .
i t' ll l l= l!jjl!!' E G l l l l- ; l! ' L
- l '
iill ' i
, Ili l lI1 l ll t ,
i <+ . i I, , , ,
= ,..l .
l l r
\l
\
)l \ ' !l\\l l \
C : ,jj i i . }l R ' l l i j , 1 Ii ,i l l l l l
! ' ' I I I r ll I
i l l, ! [ i I l I l 5
, i 1 t )
....,i I r i !i i i! i : i I i i i I. . , , , , , .,If., , . .I
\
C R 2
- 4. , ,ill.,i... .
.i
, r ..<.. , ,, ..,..,n.. , . . .i l
l C l . R
' 3 L ,
( d i , il '
'l 1'
l l' Peak Flux (RPF=1.66) 1 1 1 FIGURE 9-4 Core Loading Soowing Peak Flux Location f I l I w 1' l l 9-22
\ l
' y._,.,_,, _ _ , . . . . . . , _ _ _ _ _ _ . , _ , _ , _ , , ,_
..m. _ - . -__..._....___m_ _____._,_._,_.........,_._.,,_r..., _ , _ ., _ , , , , . _ , _ , , _ . . _ _ . _
L, , I ANS AN18 AC4.. AN2 i AN17 AC6 . AN10 -AN14
,, 1 l ze r & AN9 AN6 AC5 AN13 ;
FOIL LOCATION IN FUEL , ELEMENT ACS AN1 AC1 AN20 AN 8 t l ELEMENT ICCATION IN THE CORE -
- 1. 3 - -
6.0 ( 1. 2 -- 1.1 - t
.i 0 ' - 50
- 1. 0 . - - ,
09- -
} ,
l' d 7 m"g0
; 0. 8 . R e 5 6 i 07- -
5
. 8. 30 0.6 -. g. t u w 05 .u S g .
O.h ..E l- 2.0
, 0. 3 .. g L
W - ( y- . 0.2 - - 1.0 0.1 .
! +30 +25 +20 +15 +10 +5 0 -5 15 25 -30 DISTANCE FROM MID-PLANE (CD'TIMETERS)
FIGURE 9-5 VERTICAL FLUX TRAVERSE IN POSITION OF AC-5
, 9.23- . . . . . . . -
7 .. ..- _ L c
, ' ~f , these measurements, the rods in channel rod elements AC-1 and AC-4 were
/
p completely withdrawn and the tips of the rods in control rod elements AC-i 5 and AC 6 were at about six centimeters above the midplane. This i results in th'e most severe axial flux peaking possible in the reactor. The curve'in Figure 9-5 is the basis for the Exit Peak Heat Flux distribution used in this analysis. l L. t
;. 9.9. Allowance for Error in the Burnout Determination In Figure 9-3, a comparison is given of the ratio of calculated to experimental heat fluxes. With the assumption that the points in this figure form a normal distribution (Gaussian error curve), an analysis was performed to find the standard deviation, o, from the expression a 2= (X1 E)2 (9.27)
J where n is the number of points, and i is the mean value of all points. j A value of a - 0.21 was obtained. For 2.32 standard deviations above and below the mean, the probability of occurrence of ratios outside this range is 2%. Half of these occurrences will be above and half will be below this range. Accordingly, for 2.32 a or a burnout ratio (BOR) of l l 1.49, there is a 99% confidence factor that burnout will not occur based on Cambill's experimental data, i 9.10. Safety Limit i l \ A steady state heat transfer code for downflow in rectangular l. channels, THERHYD, that incorporates the correlations discussed in the g preceding sections, was developed by Dahlheimer25, U> I 9-24 i o . W
c; I 6 [ The limiting core power to system flow derived from THERHYD channel . power and channel flow define the burnout limit curve for the WAR with forced convection cooling. The WAR Safety Limit is set so that the fuel and cladding do not molt. If power and flow conditions that can be reached during various transients are always below the burnout limit l curve (lower power for a given flow) the safety limit will not be ' exceeded. Thus the safety limit is actually based upon the burnout ; criteria since if burnout (either departure from nucleate boiling or dryout) does not occur, the fuel and cladding will not melt. The Limiting Safety System Settings (LSSS) for power and flow are chosen (see section 9.11) so that the reactor will always operate below the burnout , limit curve during any analyzed transient. The LSSS's also take into
- account uncertainties in measuring the core power, system flow, and pool !
'O V temperature.
THERWD was run using the channel and plate dimensions for the 22 plate / element LEU fuel to find the limiting channel flow for a range of channel powers. The exit peak axial flux distribution measured by t Sternberg3 and fitted to a fifth order polynomial by Dahlheimer25 was used for all analysis. Additional THERHYD runs were performed to obtain differential pressure information for a wide range of powers and flows. The primary output from THERHYD is the channel flow for various channel powers at a specified limiting condition. The limiting condition used throughout the calculations was a Burnout Ratio (BOR) of 1,49 which section 9.9 demonstrates allows for errors in the correlations used in THERHYD. Radial peaking factors from 2DB-UM26 computer models of various
'I ' a I fT proposed LEU cores were used to convert individua1' channel power Q ,1 calculated by THERHYD to full core power. Reference fuel element flow I parameters experimentally determined by Brunot 4 along with calculated differential pressures from THERHYD were used to convert the individual channel flow calculated by THERHYD to system flow. The.effect of tolerances in fuel loading, fuel width, flow distribution within a fuel element and channel size (Summarized in Table 9-1) were all individually converted to a corresponding increase in channel flow needed to account for each effect. The total increase in flow for these four effects was determined by taking the root of the sum of the squares of the individual effects. The resulting core power vs. system flow is the burnout limit curve for the particular core. In the calculations, the bulk pool water temperature limit of 1110 F is used as
<s '
the inlet temperature and the depth of water to the center of the core is lv! I 20.36 feet. ! Brunot measured the single element flow in a 4x4 core using 12 plate HEU elements to be 48 CPM when the system flow was 940 CPM. The dynamic differential pressure across the core for this flow condition was i calculated using THERHYD to be 0.0776 psi. These conditions are used as a reference point to calculate non-element flow in other core configurations. Non-element flow is any flow that does not go through i the fuel portion of the elements. Examples of non-elemental flow are the flow through the 49, 3/4 inch holes in the grid plate used to cool the ! reflector and channels between elements, and the flow through the rod l channels. ' i
<~) 9.gs i
m.t e t j: , [ \ 3<A~), ^ L TABLE 9-1. ' p Leu 22 Data and Parameters C
' Core size 4x4 '4x5 Number of Plates 308 396 ,
i Number of Channels 294 378 Radial' Peaking Factor 1.66 1.71 ;
& l Axial Power Distribution. Exit Peak as defined by 5th order , polynomial in Dahlheimer.
t- [' Inlet Temperature: 1110 F
) Pool Depth to Core Center: 20.36 Ft. ,
Flow Distribution Factor: 16.5 % rg . Reference Non-Element Flow Parameters:
~ N "# Core:
4x4 12. Plate HEU System Flow: 940 CPM Element Flow: 48 CPM Dynamic D.P.: 0.0776 psi Element Dimensions: i Tolerances Channel Cap 0.0927" (+/- 0.007") Channel Width 2.621" (+/- 0.013") l Fuel Width 2.395" (+/- 0.075") Plate Thickness 0.05" l
*1 ate Loading U-235 12.5 g (+/- 0.35 g) l
.b) 9 27 l ..
l, 1 . t
e. l' 'O THERHYD was run for the nominal channel dimensions as well as for L ,' the minimum channel dimensions (both gap and width) to calculate the minimum channel flow with hign resolution (0.01 GPM) for a BOR of 1.49. Additional runs were performed with the BOR limit set at 1.00 to obtain differential pressure data for a wide range of flow rates at both nominal l . and minimum dimensions. 1 In the computer runs, the resulting channel power for the nominal channel dimensions, with BOR 1.49, were converted to core power using the appropriate radial peaking factor (RPF) and number of plates in the core, according to the relationship: Core Power - (Channel Power * # of Plates)/ RPF (9.28)
' /3 l j For each limiting channel flow rate from the above runs, the corresponding dynamic differential pressure (DP) across the coro was calculated and the channel flow was converted to system flow, using:
1 Sys Flow - Chan Flow * # of Channels + Non-Element Flow (9.29) where Non-Element Flow is given by 6 (940 GPM/sya - 48 GPM/elem
- 14 elem/sys)
- SQRT(DP/DPref) (9.30) where DPref is the dynamic differential pressure calculated for the conditions at which 48 GPM per element at 940 GPM system flow was measured and DP is the dynamic differential pressure for the core O
D 9 28
- . .,.,_..m
i I c , l
i. conditions under study. Dynamic differential pressure is the total '
)
differential pressure across the core (or channel) minus the difference i in hydrostatic' head between the inlet and outlet of the core. Use of equation (9.30) assumes that the non elemental flow area does not vary from core to core. The 49 small holes are never plugged and the number of rod channels remains fixed at four. The resulting function of Core Power versus System Flow is the nominal burnout limit without considering any tolerances. The slope of the nominal burnout limit curve was calculated to convert the tolerances ' i on fuel width, fuel loading and power to tolerances on flow. Separate THERHYD runs were performed to determine the limiting flow with a BOR of 1.49 in a channel with the minimum gap.and width. The total diffsrential ! pressure (dynamic and static) required to maintain safe flow in the l -[~') smallest channel must be applied across the entire core. Therefore, the v (. flow rate in a nominally sized channel at each power level was calculated using the differential pressure required for safe flow in the smallest [ channel. This results in a significant increase in nominal channel flow. Each increase in flow to account for the tolerances was normalized to the ! nominal channel flow and added to the nominal channel flow in quadrature, l .' The dynamic differential pressure for the adjusted channel flow was calculated and again using equations (9.29) and (9.30), the system flow for each core power was calculated. The resulting power to flow curve is l the burnout limit for a particular core configuration and set of l dimensional tolerances. The burnout curve in Figure 9-6 is based upon the minimum core l Q' 9 29 l
. , - - , ,- ,..e w .w - - - , , --n, ,- - , - -
. = ~ . - . -
* ?l
-8 Nominal Specifications; BOR = 1.49; 111 F
-7 Worst Tolerances; BOR = 1.49; 111 F 6
-5 g Q Burnout Limit 5
g -4 3 Sec. Transient
'/////////// 3S e
N 3.45MW Limiting 8 \ #" **
~
+ LSSS, 3MW, 900 GPM Q
o P
-2 4 Normai Operation 2MW, 1030 GPM
-__837 GPM Limiting True Value
-1 xxxxxxxxxxxxxxxxn Loss-of-Flow 200 400 600 800 1000 1200
' 8 I i ,
Total System Flow (GPM) FIGURE 9-C- LEU Core Power vs System Flow, 4x4, RPF=1.66 _= _ - . .. - _ , . . ~. -
.. .. . , - , .. . . _ . .-~ . . - . _ - - _ =_______ __ ____.-_-__-_-_..-.-_.:
( -. F i I l
~
) loading of 14 normal elements and 4 control rod elements in a 4x4 array, j V u/
~ Larger cores were analyzed; however, their lower power densities and j i
lower non elemental flows more than compensate for their higher radial I peaking factors, resulting in less limiting conditions. i i; 9.11. Limitine Safety System Settings and Measurement Errors
'I .
L a) Cnolant, Inlet Temnerature The Limiting Safety System Setting (LSSS) on coolant inlet f
-temperature (pool temperature) is 1080F, Normal operation is at 1050F or less. The manufacturer's specified probable error on the measuring instrument is i 0.750F, The standard deviation would then be i 1.1250F and using 2.32 standard deviations for a 99% confidence factor, the I temperature can be determined within 2.60F. For an LSSS of 1080F, the (f
limiting true value on the coolant inlet temperature is therefore less than'1110F. All calculations for the curves in Figure 9 6 are based on an inlet temperature of 1110F, ' b) Flow Rate The primary coolant flow is detected by measuring the - ! differential pressure across an orifice in the primary piping. The ' manufacturer's specified probable error is i 2% or a standard deviation of i3%. For a 99% confidence factor that the flow rate will be no less
~
than a specified amount we take 2.32 standard deviations, or 7%. The l limiting safety system setting on flow rate is set at 900 gpm. Hence, ; l l the true value should then be no less than 0.93 x 900 - 837 gpm. This is the limiting true value on flow rate shown in Figure 9-6. A-V 9 31 4
, - , ,- ,- n -
e ,
W
.. - -l ' .
J t l' l I- -
. ('")s ,
c) Reactor Power l The reactor power is determined from the product of flow rate x AT, where AT is the differential temperature across the core. The probable error with the AT measurement could be as much as 10.40F. At a power of two' megawatts the AT is about 130F, for a probable error of ; 13.2%. The safety level scram point is set by adjusting the neutron chambers to read the same power level as the thermal power. Here, the , limiting accuracy is the readability of the indicator which can be read with a probable error of 12%. There are then three independent probable errors in determining the power level setting: a) Flow rate i2% b) AT 13.2% [~) N. s c) Chamber Setting 12% The combined probable error is obtained from the square root of the sum of the squares and corresponds to i 4.27%, with a standard deviation , of 16.4%. For a 99% confidence factor, 2.32 a is 14.8%. Therefore, with a measured value of 3 MW for the LSSS on reactor power, the true value should be no Sreater than 3 x 1.148 or 3.45 MW. This is the limiting true value on reactor power shown in Figure 9 6, 9.12. Short Period Transient A limiting condition for operation is that the magnet release time be less than 50 milliseconds. The minimum setting on the period scram is 3 seconds. From the rod calibration curves, the three rods must drop about 31/2-inches from the fully 9-32 1 1
. q;
(j withdrawn position of 26. inches above the bottom of the core to overcome this period. This is the least reactive position of-the rods, and normally the rods are cperr.ted at a level well below this. However, to allow for variations in core loading, it is assumed that the rods must drop 5. inches from the fully withdrawn position of 26-inches, to overcome a'3.second period. This is a very conservative assumption, i A rod drop time cannot be measured unless the drop terminates on the l seating switch at the fully inserted position. Accordingly, drop times 1 were measured from 3, 4, 5, and 6. inches above the fully inserted position as well as from the fully withdrawn position (26. inches). The l results obtained are presented in Table 9 2. TABLE 9 2 cm (-) TIME TO DROP FROM A PREDETERMINED POSITION (All times in milliseconds) i Initial Position j E E E E 2E j i L Rod 1 160.0 185.0 205.0 225.0 481.2 i Rod 2 170.0 195.0 210.0 227.5- 506.0 j Rod 3 162.5 112.1 203.0 212 Q 482.5 i Average 164.2 187.5 206.2 227.5 489.9 i True Gravity 124.6 143.9 160.9 176.2 366.8 (
.............................................................................. l The times for True Gravity (last line) are calculated from: ;
S- fgt 2 (9,31) l
/~'s - !
wJ 9 33 4 J l k l~ 1 i i l 1 1
m
' l i
1
}
The actual time for the rod to drop is greater than the True Gravity time 1
. \s ;
because of the friction of the water as the rod drops through it.. Of more significance is the slowing up of the rod as it enters the dash pot during the last few inches before it is seated at the bottom of its travel. The action of the dash pot in slowirg the drop time is illustrated L by the times to drop from' 26 inches (last column). The measured values T l of the drop times from 3 inches, 4 inches, 5 inches, and 6 inches include , some delay from the dash pot action. The drop time for any one of these distances from the fully withdrawn position would be less. For example, a 5 inch drop from 26 inches to 21 inches should be less than 206 milliseconds. On the other hand it would be greater than the true gravity time of 161 milliseconds as there in some frictional resistance (n) from the water as soon as the rod starts to move. For conservatism, it is assumed that the time for the rods to drop 5 inches from the fully withdrawn position is 3/7 of the time to drop the full travel of 26 inches. Accordingly, for a measured free drop time.of 490 milliseconds for full travel, the time to drop 5 inches will be no i greater than 210 milliseconds. A maximum free drop time for the full 1 travel is established at 700 milliseconds in the Technical Specifications. Therefore, the time to drop 5 inches will never be
- greater than 300 milliseconds. Adding the 50 millisecond release time to the 5-inch drop time, the time from the initiation of a scram until the rods are inserted 5 inches will never be greater than 350 milliseconds.
1' l- It is not easy for the reactor to go on a short period at high power l due to the negative temperature coefficient, which has been determined to l
*' # 9 34 u
- 1. .
~
77%. us , ,
'f(
H f ; ("'y be about -1x100 Ak/k OF27 Neglecting the temperature coefficient, and I %J
;- assuming the reactor is on a period just' greater than 3 seconds, and goes through the maximum true value of the LSSS at 3.45 megawatts, the power will rise to a value no greater than i
3 L ..
'P - 3.45 exp [0.350/3) - 3.88 MW. (9.32)
The range between 3.45 MW and 3.88 MW is the allowance for a 3 second period transient in Figure 9 6. Thus, 3.88 MW is the maximum power for any transient because transients shorter than 3 seconds will ' scram the reactor at a lower power on period while longer transients will scram the reactor on high power with less power overshoot.
. s-[ ') 9.13. Loss-of-Flow Transient and Natural Convection I
A loss-of-flow transient is illustrated by the shaded area extending to the lef t of the LSSS in Figure 9 6. It is assumed for this transient that the reactor is operating at the LSSS of 3.0 MW and 900 gpm, with the flow header jammed in the UP position, and there is a power failure to the pump which then results in a reactor scram signal. The following sections 9.14 and 9.15 are the original analysis 24 i for loss of flow and natural convection. NATCON28 runs for 22 plate, 18 !~ plata, and 12 plate fuel elements show that the margin of safety l increases with the increase in heat transfer area and lower power densities associated with the increase in the number of fuel plates in the 22 plate LEU core over the 12 plate HEU core used in sections 9.14 1 and 9.15.
\>
9-35 i
, _ . _ , , - . . . _ . , , , ,, _ ,m.,.- . _ . _ _ , . . , , . _ . .. _ . , _ , ,
1, - . [l i
]
~
[) ( /- To determine the coolant flow coastdown curve, measurements were ! i made for the HEU SAR of the decrease in flow rate versus time, starting l l with a flow rate of 800 gpm and followed by pump cut off. For these l measurements the output signal from the differential pressure cell across the orifice plate was used after this signal had been calibrated against various flow rates through the primary system with the flow header in the UP position. For the LEU SAR, the results were linearly shifted to a starting point of 900 gpm and are shown in Figure 9-7. The flow . coastdown curve is expressed by the following equation (shifted from equation (9.34)). , . Flow (gpm) 900 - 636 x t(sec) (9.33) 1 9.14. HEU Analysis for Loss-of-Flow Transient
<~s (s.-) A loss of flow transient is illustrated by the shaded area extending to the left of the LSSS in Fi 5ure 9 6A. It is assumed that the
[HEU) reactor is operating at the (previous HF.U) LSSS of 3.0 MW and 800 gpm with the flow header jammed in the UP position. There is a power failure to the pump and the loss of power to the pump then initiates a scram signal to the reactor. Measurements were made of the decrease in flow rate versus time i- starting with a flow rate of 800 gpm, and then cutting off the pump. The l measurements were made using the output signal from the differential l pressure cell across the orifice plate, after calibrating this catput signal versus various flow rates through the primary system with the flow header in the UP position. The results are shown in Figure 9 7 and the flow coast down curve is expressed in the following equation: f~)
\,,)
L 9 36 1 i
.-.,n _ . . , _ . . .
(
~
[
- CURVE 1; NOMINAL SPECIFICATIONS; BOR = 1.00; 111 F x i
'0 CURVE 2; NOMINAL SPECIFICATIONS; BOR = 1.%9; 111 F CURVE 3; WORST TOLERANCES: BOR = 1.%9; 111 F T
6 i ! 5
-5 SAFETY LIMITS t
ie - 0 3.88 W
'a
.~ s 3 SECOND TRANSIDT - 1 . m ' ', 3.%5 W LIMITING TRUE VALUE
- 3 p
! 188S, 3 m , 800 spa, 110*F l Flow / / W l
! INSTABILITY /
I 2 / /
, /
! / / f e NOIWEAL OPERATI0tt 2 W ; 900 gym, < 105,F ;; f / , g
- 1 // / 7 EPsLN NYM l M7< . . . . . , i i o roSS or Frow TRANsIn T i 200 %00 600 800 1000 1200 r I i a l _
9 1 l FIGlRE 9-6a HEU Core Power vs System Flow, 4x4, RPF=1.37 1 y .$ 7~ ~ ---- e __ -_____________________-_.__.____z = - 4-_____c__[
g,, f '
.g .
v^ '
\ / (.1,N3D83d) H2 Mod g
L 8 E S S 2 g. L 3-
/
l~ m
/
l, / L
/- .
/ -
/
/
- l. G. / . $
' U /
\
/ 1 l '
l
/
L */ $ l-
/ -
b
/ g 3 L /
1, :3 s' E 7
/
u- ; a i
/ C 3 a*l !
/ ,
l l / , "g 1 K / , I /
- 1. .
2 = j g g l
/ .
/ " - -
g i
/
l 5 _5 ;
., y v
, i i i l (ads) glyg M(TIJ
- 1. >
l
%5#
(\
)
x_/ ; r Flow (gpm) - 800 636 x t(sec) (9.34) The power coast down curve was calculated by solving the kinetic equations with six groups of delayed neutrons on a digital computer using 0.01 second time increments for the first two seconds and 0.1 , second time increments thereafter to 100 seconds. It was assumed that the rods did not move for 50 milliseconds until after initiation of the scram and that they moved with constant acceleration until they had traveled the full length of 26 inches in 700 milliseconds. These assumptions are most conservative but are consistent with the limitations on rod drop times given in the Technical Specifications.
- O xe The complete reactivity insertion was assumed to be only 3%Ak/k and the rods were assumed to drop from the fully withdrawn position (also very conservative assumptions). By comparing with rod calibration curves determined experimentally in the reactor, a curve of negative reactivity insertion versus time was plotted, and this data was input to the computer.
It was assumed that 6% of the power at the instant of scram was from the decay.of fission products and this was held constant in time l thereafter. The power coast down curve is plotted along with the flow I coast down curve in Figure 9 7. Because of the same amount of negative reactivity inserted, the power does not approach the 6% in fission products until about 100 seconds. At the instant the flow rate reaches l zero (1.258 seconds), the power is still at 21.7% full power. By picking l (~'\ i (~ / 9-39 l 0 1
t- 1 I
.Y l off powersJ and flows at equivalent times the curve of power versus flow '
r [^J x- { extending to the left from the LSSS in Figure 9-6A was determined. This is the expected result of a loss-of flow transient. However, since the L true values of power and flow at the LSSS could be 744 gpm and 3.45 t megawatts, the procedure was repeated starting from this point in Figure , 9-6A. The shaded area between the two curves represents the range of uncertainty for a transient resulting from a loss of flow after steady L . state operation at the LSSS. Not included in the figure are the results i 1 as the flow rate decreases to zero and then reverses due to buoyancy ' effects. At a downflow rate of 225 gpm and a power of 790 kW, the loss of flow transient curve intersects Curve 3 in Figure 9-6A. The i si6 nificance of the point of intersection is that if the reactor were l operating in the steady state at this power, and this flow rate, a flow t rg ( ,/ . instability would occur in the hot channel. (See Section 9.5). , However, the reactor is not in the steady state and the flow rate is changing rapidly. Also, it is impossible for the flow header to jam in the fully UP position. The header is reised by flotation with air as described in Section 4.3. After the header is raised, the pump is started and the header is held up by the pressure drop of the flow through the core. The air pressure which raised the header is vented and the reactor cannot be started until this pressure has dropped below 2 psi. If there is a loss of coolant flow, gravity will cause the header i to drop. However, it is possible for the header to jam in a cocked posi. tion. The worst possible position in w'nich the header could jam is y shown in Figure 9-8. In this case, the header would be wedged with a minimum Sap of 0.956-inches between the header and grid plate on one (s
\w/
9-40
I i
<~N side. 'd )
v A transient analysis was made for a loss of flow incident from 3.45 megawatts reactor power with power and flow coast down as shown in Figure 9-7 [HEU). The header is assumed jammed in the position shown in Figure 9-8 and the transient response of the average flow, average fuel i temperature, and average coolant temperature are shown in Figure 9 9. As f will be shown, a safety limit can be established, using the maximum fuel plate temperature as a criterion. The response was calculated with a computer code which treated transient heat transfer at the average power location. Heat transfer coefficients were calculated as a function of flow rate, starting in the l turbulent flow. regime at rated flow and switching to a laminar flow correlation for Re < 2300. Upflow resulted from buoyancy, which was set [) u equal to the combined effect of (1) the inertia of the water in the core flow channel and the header below the core, and (2) the friction loss through the gap between the wedged header and the grid plate and friction loss in the core-coolant channel. The pressure drop through the 0.956-inch gap relative to the It pressure drop through the core after flow reversal was calculated in order to solve the transient flow equations, as follows: 2 9 y AP - f h. E-D 2ge + 1.5 #2geEEE (9.35) where: first term - friction pressure drop through core 9 - average velocity in core second term - pressure drop through gap V gap - velocity through gap
,,s
's l 9 41 1 1
i f,
. l I
'; m .l
\
MmR i
^
MINIMUM GAP GRID
, 0 956 IN. PMTE i
f N /-
\ / FLOW HEADER ;
N- IN COCKED g / POSITION g / . s / , N y _ _ _ _ _ _ ,/
'l e
O t t 4
~ '
I g , ll ll IlI ll l ll ll I I l gh lI I I w" ll l- 2 lI 18 ' Il U l l l l O i i i I l \ /l l FIG. 9-8 FLOW HEADER JNTED IN COCKED POSITION 9-42
s l l 350 j 7 (2)
'u J 1
~
+---- PEAK FUEL TDfERATURE 1 300 - ;
L
- 150 250 ,,, 4 A N GE N TMPERATURE ;
J ' r O _ s x l 100 200 - AVERAGE FIcW RATE ; l UFFLOW
~
l 50 150 - 4--- AVERAGE COOLANT TEMPERATURE
. 120 ..
l TDE (SECONDS) 0 ' ' ' ' I f I o 2 h 6 8 10 12 1h 16 FIGURE 9-9 FLOW REVERSAL AFTER LOSS OF Fl.fH TRANSIENT FROM 3.h5 th 9-43
7
^'
V 1 I L V From' continuity, it was found that, for the gap width of 0.956 inches, Vg ,p.- 1.86 V. The transient heat transfer code has the standard v point reactor kinetics and heat transfer equations. The only unusual feature is the momentum equation which controls the coolant flow rate. The time dependent momentum equation was:
- 2 p V pV 2"*" -
dV (p.po) h E - f + 1.5 + ph d2+pL - AAE (9.36) Sc De 2g c 2g c 8c de g, de where: p - average density in heated channel in core p,- density of water at pool temperature h - core height L - equivalent length of flow path between the gap above the funnel and the core. r~' ! (,,)s The flow velocity in the peak channel can be related to the flow velocity in the average channel by setting the pressure drop across both channels equal. For laminar flow and steady state, this results in a value of velocity in the peak channel 1.17 times the velocity in the average channel together with a temperature rise across the core, AT, of j 1.17 times the AT for the average channel. These values assume a radial 1 peaking factor of 1.37. The basis for the factor of 1.17 is as follows: The pressure drop, AP, across the hot channel and the average channel are equal, or Buoyancy AP - Friction AP in both hot channel and average channel. l l Define x such that: L Buoyancy AP in hot channel - x times greater than Buoyancy AP in average channel (9.37) G 9 44 l l l t
y;
\
l. P l' , Since buoyancy AP - Friction AP at steady state, then Friction AP 1 L q: l l? in hot channel - x times friction AP in average
; channel (9.38) 2
' Friction AP -ef f6 cf (9.39) f' In Laminar flow, f- -
(9.40) Dp 4
. Friction'AP -
- cV (9.41)
VD Ec e l where* c - constant, independent of channel. Since Friction AP in hot channel - x times friction AP in average channel and since Friction AP - cV
./*
- u. Then V(hot channel) - x . V(average channel) (9.42)
Return to buoyancy AP to obtain a relation between AT (hot channel) and AT (average cht.nnel) . Buoyancy AP'-(p, - p) L (9.43) where: p, - density at inlet temperature (1110F) p - average density in channel l From Eq. (9.37)
~
(#o' ) hot channel e
~
*( o #} avg. channel ( )
s f
'I 9-45 ,
c:
. - . . . - .~
i [~'y Since the change with density is about linearly prcportional to a j
, change in temperature over a limited temperature range, p,-p can be l written.as follows for the hot channel: I
~
-(t o p) hot- a(Th d hot" *( exid) oho ~t *O hot I )
p
;. Similarly, for the average channel, using the same a, (p,p) ,y - a(T-T y ,y - 2a AT ,y (9.46)
Substituting Eqs. (9.45) and (9.46) into (9.44) gives: AThot " **O avg ( } We now have two relations, relating Vhot t0 V avg Eq. (9.42) and AThot to AT avg Eq. (9.47). The energy equation allows us to evaluate x as follows:
,-s, t
) The Enercy Eaustion is:
([ncpAT) avg. channel - (Btu /h, heat production rate in average s channel) (9.48) (m cp AT) hot channel - 1.37 (where 1.37 - Radial peaking factor) (9.49) But m - pVA, where p and A are essentially the same for both the hot and average channels. l. L- Hence, Eq. (9.48) can be rewritten as pac V AT -Q (9.50) p avg avg (3 I' \w/ 9 46 1'
i F' - Pi I i
~
Substituting Eqs. (9.37) and (9.47) and V hot and AThot int .Eq. (9.49) 4
.gives: .
Acp (xV,yp (xAT,yj - 1.37 Q (9.51) -' L Dividing Eq. (9.51) by Eq. (9.50) gives b x 2 - 1.37 or x - 1.17 (9.52) Hence, from EQ. (9.42) and Eq. (9.47), , i i V(hot channel) - 1.17 V (average channel) and AT(hot channel) - 1.17 AT(average channel) (9.53)
.The safety limit [HEU) during a loss of-flow transient is established such that the maximum fuel temperature shall not exceed p) ,
; 3500F. The average fuel temperature does not exceed 2470F in Figure 9-9 as calculated by the transient heat transfer computer code. This occurred at 5.2 seconds where the average coolant temperature was calculated to be 140 0F. I L
To compare with the steady state heat transfer code it was assumed l l that the reactor was in steady state at the conditions existing at 5.2 seconds. The steady state code gavs an average fuel temperature of 2140F and an average coolant temperature of 1430F. The coolant temperatures are in good agreement, but as to be expected, the fuel temperature is considerably higher in the transient than in the steady state. ! The transient code will not give the peak fuel temperature in the L $ hot channel, but a conservative approach is to calculate this temperature for the steady state and ratio up accordingly. From the steady state
.A
> 1
\_ /~
9 47
mz ', t [: p , l l ( l , '- , N [G code, the peak fuel temperature in the hot channel was 2520F, This l occurred near the mid point, axially along the channel where the coolant-temperature was 1470F, The' heat t. msfer coefficients do not vary greatly, so for a given rate of coolant flow in a channel, the channel power,.P, is closely proportional to the temperature difference, T between the fuel plate and i the bulk coolant. Accordingly- l
'(9.54)
P(hot channel in transient) ~ P(hot channel in steadv state) P(average channel in transient) P(average channel in steady state) and to a very good approximation (9.55) AT(hot channel in transient) AT(hot channel in steady state) ET(averagechannelintransient) ~ AT(average channel in steady state) , 1 1 l r3 or abbreviating l, AT(he.tri AT(he.ss) AT(ac,tr) , AT(ac,ss) * (9*56) Using the data from the steady state code at 5.2 seconds given above AT (he.ss) (252-147) AT (ac,ss) , (214-143) - 1.48 (9*57) We assume this ratio is constant at time other than 5.2 seconds. At times far from 5.2 seconds the assumption is poor, but at times near 5.2 ! 1 seconds where the peak temperatures are encountered, the assumption is l quite valid. Accordingly, AT(he,tr) - 1.484T(ac,tr). (9.58) j AT(ac,tr) can be obtained at any time by subtracting the average f I coolant temperature from the average fuel temperature shown in l l l n.
- Figure 9-9. At 5.2 seconds l' t 9-48 i
L l L
- t: ,
1
> 1 L r-1 r~~p AT(ac,tr) - 247.- 140 - 1070F (9.59)
QJ '
;, and therefore, (still at 5.2 seconds) 1 AT(he,tr) 1.48 x 1070F - 1580F (9.60) [
As mentioned before in laminar flow, to obtain equal' pressure drops across both channels, the hot channel must have a flow rate 1.17 times ' M the flow rate in the average channel. The temperature rise of the coolant along the hot channel must also be 1.17 times the temperature rise of the coolant along the average channel. (These two factors of
, 1.17 combine to give the radial peaking power factor of 1.37 [HEU)).
-From Figure.9-9, at 5.2' seconds the average coolant temperature is ;
1400F, and subtractin6 the 1110F entrance temperature, gives a temperature rise of 290F to the mid point of the average channel. For the hot channel the temperature rise is 1.17x290F - 340F giving a (} temperature at the mid point along the hot channel of 1450F. Therefore, the peak flux fuel temperature in the hot channel in the L transient-is T(pf,he,tr)-1450F + 1580F -3030F This is at 5.2 seconds after the start of the loss-of-flow transient. In general, it is assumed that at any time during the transient, the peak fuel temperature in the transient is T(pf,he,tr) - 1110F + 1.17 T(aw,ac,tr) - 1110F) + 1.48 AT(ac,tr) (9.61) l l where T(aw ac,tr) is the average water temperature in the average channel l . ( V) l 9 49 l l 1 1
~ - . . . . . .
j i l 1 l
/ in the transient and 1110F is the coolant entrance temperature (pool temperature). T(pf,he,tr) is the peak fuel temperature shown in Figure 9 9. The maximum is 3030F and is well below the [HEU) safety limit of I 3500F, This method of analysis predicts a peak fuel temperature of 2790F i I
at the start of the transient (time zero). The more accurate steady { i state code gave a peak fuel temperature of 2$20F in the hot channel prior ! to the transient, indicating the conservatism of the method. 9.15. HEU Analeair for Natural Convection [This section, as is the previous section, is the analysis performed for the stEU SAR. Refer to section 9.13 for justification for the validity of including this analysis for LEU) According to Figures 9 6A and 9 7 and also Figuro 2.1 in Technical ( Specification 2.1 [HEU Technical Specification Ref 29) a loss of flow a transient from 3.45 megawatts will result in a flow coast.down followed by natural convection cooling at a reactor power of 750 KW. This power is due to fission product decay and cfse off gradually with time. To be consistent with the Loss of Flow Transient Analysis (Section 9.14), 750 KW was chosen as the maximum power for the Safety Limit 29 in the natural ., convection mode of operation. Q The transient heat transfer code (see Section 9.14) was used to L determine the equilibrium flow rate established at 750 KW of power with natural convectfon flow. A flow rate of 129 CPM through 168 channels was established in about 25 seconds and remained steady in time, thereaf ter. With 168 channels (Sec 9.4 of Ref. 24) the average power per channel is 9 50 t
l i ( ') 750/168 - 4.46 KW. Multiplying bv the radial peaking facter of 1.37 l v gives 6.12 KW as the power in the hot channel. The flow rate in the average channel is 129/lv8 - 0.768 CPM. The
, velocity in the peak channel is 1.17 times the velocity of the average channel (see Section 9.14). Therefore, the flow rati in the hot channel i f5 0.768 x 1.17 - 0.90 CPM. Using the power of 6.12 KW and a flow rate i
of 0.90 CPM in the hot channel as input to the steady state thermal hydraulics code with natural convection upflow, the maximum fuel plate temperature was found to be 2590F with an inlet coolant temperature of i i 1110F, 7$0 KW and 1110F were conservatively chosen as the [HEU) Safety himits in the Natural Convection Mode of Operation, Specification 2.1.2 + [HEU Technical Specifications Ref. 29). A maximum fuel plate temperature of 2590F is well below the temperature at which fuel clad damage could F\ r I ( occur. 9.16, Maximum trU 22 Plate Puel Terrocratures ro113 wing thCA 9.16.A. Introduction e l The maximum fuel temperature reached after loss of cociant ; 6 ( from the limiting WAR LEU 22 plate core was calculated using semi. l empirical relationships similar to those developed for the safety analysis of the Omega West Reactor (OWR) at Los Alamos 22, which is an l MTR. plate type pool reactor. The fuel clad melting temperature for the , WAR is the melting point of aluminum alloy 6061 at 10800F, but l l structural integrity may be lost if the fuel is held above the softening l temperature of for a sustained period of time. For the purposes of these l 1 ; analysis, a softening temperature of 8400F33 was chosen as the limiting ' criteria for the analysis of a lhCA. 9 51 i
l l I l 7i For several potential coolant loss mechanisms, the time required to l V i uncover the core was calculated because it is an irportant parameter in ! calculating the maximum fuel temperature following a LOCA. However, the WAR relies upon the Emergency Core Spray System (ECSS) (analyzed in Section 9.17) to cool the reactor following any IhCA that takes less than 90 minutes to uncover the core after the reactor scrans due to low pool level. Therefore, the minimum time after a shutdown to shift from water i cooling to air cooling is taken to be 90 minutes. I i i The fission fragment heat source used in the analysis was the l
" Simplified Method for Determining Decay Heat Power and Uncertainty" given in Section 3.6 of ANSI /ANS Standard 5.131 This decay heat power l.
can be 1.5 times as high as the power predicted by the Way Wigner l relation during the limiting IhCA transient. A modification to the heat transfer correlation developed by OWR was made such that, when used with i the ANSI /ANS 5.1 heat source, it safely enveloped the temperatures ; 1 cciculated with the original OWR heat transfer correlation and the Way-Wigner heat source. For the benchmark transient based upon uncovering the OWR core 30 minutes following rhutdown, the modified heat transfer I > l correlation with the ANSI /ANS 5.1 heat source gave a peak temperature 1400F higher than the original OWR calculations and was higher throughout I the transient. Figure 9 10 compares the benchmark OVR transient when calculated using the WAR correlation and the ANSI /ANS 5.1 heat source I l l with the same transient calculated with the OWR correlation using the ! t i Way Wigner heat source. This comparison, along with similar comparisons
)
i in reference 22 between the OVR correlation and experimental data, 1 ( l O 9 52
e'"'~% O p
-i/ .o 7
) i- U)
- n 6
- k Z 1 5 :*
O , l m L . Q) U) .I L L v E L C / OL O 4 L O - uou 7
=
0 - i U # ! 'Y Tu O T$
>o s 8
~
9 m l D pu ~- O c o ! mi e m U '
- e EF;.
a s x . _x w a w 1 - e > : m
- ' CD D ,
E = r. c a F i ._ H e e
.o. i 0 I U) a t ,
O \ B E 2
=
,8 O g u
f y _ g U O
'I e e i e i e a e i i 7 hk_b
_ _ __hhkb$hkkhb E e cd (d Sea) eJnioJedu;e.L leng >joed 9-53
. . .~ - -
( i j f ( ) indicates that the new correlation together with the ANSI /ANS 5.1 heat i
\)
source will overestimate the maximum temperature reached following a f. LOCA, 9.16.B Calculation of Peak Fuel Temnerature rollowine a IECA After loss of water cooling, the reactor is cooled predominantly by ambient air flowing through the core by natural convection. Since heat ; conduction to the grid plate and to the upper part of the fuel element occurs together with natural convection and some radiation, the
- calculation of maximum fuel temperature from first principles is too '
complex to be reliable. Hence, experimental data from LITR was used by ; OVR22 in their safety analysis to develop a heat transfer correlation i that, together with the Way Wigner decay heat relation, safely enveloped [ the maximum axial temperatures measured in the LITR experiments. In the LITR experiments, an element was actually removed from an l operating core and the maximum fuel temperature along the axial profilo , was measured as a function of time while the element was cooled by air convection. As described in Reference 22 the LITR experimental data l was safely enveloped by the following transient equation which is solved for the difference, f between the peak fuel plate temperature Ty, and the ambient air temperature T . ac p h=Q(t)-hAf (9.61) where
# = Ty Ta (OF) ,
Q(t) = time dependent heat source (MW) f I l 9 54 l l l . i
~
('T
\s/
hA = product of the natural convection heat e transfer correlation and the effective heat transfer I L area of one fuel element (NW/0F) mc p - heat capacity of a fuel element, with a being the mass and pc being the specific heat of the associated fuel element material. r t Mcp for the UVAR 22 plate fuel element was calculated by determining ; the mass of aluminum and 3U Si2 in the standard element from the fuel ! plate specification and the preliminary construction drawings for the ; elements and applying formulas determined by Argonne National i 1.aboratories30 for the pc of these materials as a function of temperature. c p A1 .892 + .00046T (J g*l K*l) (9.62) Mass A1 - 5195 g per element ; cp U3 Si2 .199 + .00010T (J g*I K*1) (9.63) i Mass of U3 Sig - 1505 g per element [ i Therefore, the total acp for a standard element is ' acpElement - 4933 + 2.54T (J K*1) (9.64) i For the heat source, Q(t), the method given in Section 3.6 of
]
ANS1/ANS Standard 5.1 for heat produccion from fission products after shutdown was used. The ene sigma uncertainty associated with the decay l heat power was included as a positive bias. 1 9 55 P l b
, . + - - - - , -, ------n,
\
)
I p i [) s _- P(t.T)-Pj(t.T)*G(t)+APd d (9.65) j l l and i Pj(t.T)-1.02 EEA3[F(t,=) F(t + T,=)) (9.66) where: ., P d
- decay power !
O G(t) - a correction factor for the activation of fission ' fragments given in Table 10 of ANSI /ANS 5.1 U ! APd " one sigma error associated with Pd (see ANSI 5.1) l 1: r Q - 200 Mev/ fission (conversion factor) [ Pmax - the operating power of the hottest fuel element i F(t,=) - the fission fragment heat source for thermal fission of U 235 given in Table 4 of ANSI /ANS 5.1 i time since shutdown from operating at Pmax t - T - the time operated at Paax i Non linear interpolation between time steps given in the ANSI /ANS , 5.1 tables for Eq. (9s65) was done using the time dependency in the Way-Vignor expression for decay heat generation. 4 The value used for T was 120 hours, or 5 days. The value used for Pmax was 0.209 MW, which is the power of the highest power element in a 4x4 LEU core, at 2 MW total core power, as determined from 2DB UM26 flux maps. Larger cores were evaluated and found to be less limiting than the 4x4 element core. I The heat transfer parameter hA was found at OWR by fitting l coefficier.ts in the form : l' hA-C(af" + b) (9.67) to get an expression for hA which was then used in Eq. (9.61) to fit the l i data from the LITR experiments using the Way Wigner decay heat source and the appropriate acp . The coefficient C includes the heat transfer area (}}) 9-56 j l-l 1 l ,
c a., f- : l t m ( of the fuel plates. The expression for hA used for the present analysis ! was found by searching for a new constant C in Eq. (9.67) (retaining the I t other constants found by Ok'R) that when used in Eq. (9.61) with the ANS1/ANS 5.1 heat source and the case specific input data from Ok'R, j safely enveloped the transients calculated with the original Ok'R correlations in their safety report. The new expression for hA which incorporates the higher h per plate and includes the number of plates per i element as a variable is: 0 hA - 9x10 8(PN)(6.4x10' # .72 + 0.5) E (9.68) ! F where PN is the number of fuel plates in the standard element. Substituting Eqs. (9.64), (9.66), and (9.68) into Eq. (9.61) gives a transient expression which was solved for #, with the initial conditions ' ( of e - 1120F at the time at which air cooling starts. The peak fuel I temperature. Tg, was obtained from # - Tg Ta assuming that the ambient air temperature, T a, was 1000F, or Tr(peak) - # + 100. The peak fuel-plate temperatures versus time after shutdown, with time to shift to air cooling in the core as a parameter, are plotted in Figure 9 11. Transients assuming no ECSS and a shift to air cooling 0.3 hours and 1 hour following shutdown are plotted as dotted lines whereas the transient for the minimum specified time to shift to air cooling with ECSS (1.5 hours for flow dropping below 7.5 gpa, see Table 9.4) and a later transient (2 hours) are plotted as solid lines. The maximum fuel temperatures corresponding to these transients are listed in Table 9-3. For the limiting case where the ECSS is used to provide cooling for 90 minutes follouing a reactor scram due to low pool level, the maximum 9 57
, . ':1 .
' t
,.,ep: , .
z< ,
; .. -;c' .
, temperature reached during the transient is 775'F. This t's well below the aluminum cladding softening temperature of 840'y assumed for this analysis'.
4 4 r,(
' s.
k t, , .<lO s 7 4 (- I,'. + . h-
.j f. - ~
r .. h r s i
' l jY
. /
9 58
/ '. Ae i' ,
g{ .'
.. .. ei ,
l
,,I
~\
=
.o l
i l i I
...~ ' l TABLE 9 3 !
i P MAXJNUM FUEL TEMPERATUltES a , j (Operation Time = 120 h at 2 MW) j
! i k
[ (- ' 4 F Time to Shift to Maximum Tannerature (07 1 I
- h. Air coolinn (min.) w/o ECCS w/ ECCS '
- r. - ,
r c 20 975 775 60 835 775 , p
- v 90 775 i t
i 120 730 ! t t o j
........................................................................ l
[t k b a f. i
.r
.p 1
1 4
?
k 6 6 I r 1 6 i
, 6 l ./
I s e f i 9 59 ! E i f I t. , i 1- t l~ , ,, --,...,_..:_.,-... _,....~...._-._,_.m.--. _ - . , , , . - _ , . , - . . - . . _ - . . _ _ ....-.._,......--~~-...I
\ . .. -
s Peak Fuei Temperatures for LOCR n u_ 22 Plate 4x4 LEU 2 MW for 120 hr
,1200 E1100 _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _10_8 0_ _F_ M_e_i_t j n_g_ _q f_ _OJ _ _6 06] _ _ _ _ _
o
" 1000 -
g g 900 ~... f ... 840 F Sof tening of Al
! ~~%
o 800 - L m 700 l l X e -
- e ~
G-6 : l
$ 600 -
!0.3 hr ; Time to
" ! ! Uncover Core 500 -
! il hr O :
2 400 l
- l. 1.5 r MINIMUM TIE SPECIFI8 TO SHIFT TO RIR C0Q_ING USING ECSS
.x 300 -
! i o i ~l '
f 200 - 2 hr 100 - 0 0 2000 4000 6000 8000 10000 12000 14000 16000 Time Rf ter Shutdown (Seconds) Fig'are 9-11 Peak Fuel Temperatures for Various LOCA Transients I
~
1 ( i ! 9.17 Emergency Core Sorav Svaten Analvais j s (/ The emergency cora spray system is described in Section 4.10. Using the decay heat source in ANSI /ANS 5.1 as described in section 9.17 the decay heat power in the hottest element 0.3 hours after ! shutdown (minimum credible time to uncover the core) is 3.5 kw. Assuming i the cooling water is initially at 1000F, a water flow of 0.184 pounds per f t minute or 0.0221 gal / min is required to remove the decay heat from the hottest element (assuming complete evaporatin of the water). A coolant flow of 10 gallons per minute over the entire grid plate will provide an I average flow of 0.156 gallons per minute to each of the 64 fuel element positions. Using a flow maldistribution factor of 1/232, the minimum l flow to any fuel element position is 0.078 GPM. This flow is capable of i removing 12.3 kw per fuel element. Thus, the rated flow of 10 gallons .{
.- s per minute to the core area is greater than the flow rate required to i remove the decay heat of the hottest element after 0.3 hours by a factor of about 3.5. From 30 minutes to 90 minutes, the rated flow of the ECSS ic 7.5 CPM which can remove 9.2 kw per fuel element. During this time !
the decay heat in the hottest element falls from 3.4 kw to 1.9 kw ; providing a minimum safety factor of 2.8 for the time the ECSS is relied upon to cool the reactor core. . At five second after shutdown, the decay heat power is 11 kw. This value is well below the 12.3 heat removal capability of the spray system. ! Even in the incredible situation where all water is lost from the pool in ! a matter of seconds, the spray system should be able to cool the reactor l t and prevent core damage. No operator action, automatic electronic, or automatic mechanical mechanism is required for the ECSS to function. As the water level drops below the level of the spray headers, the b 9-61
E . _
; i spray system begins to spray water on the core. Using the mockup of the
[V] core spray system shown in Fig. 4 3, tests were run of the flow rate versus head of water above the spray headers. Theoretically, if W is the flow rate (gpm) and Z is the head of water above the spray headers, then W - KZO .5 (9.69) l Using the spray headers in the mockup it was found experimentally that i W(gal / min) - 0.48 70 45 (9,70) vhere Z is in feet. From this relationship the characteristics of the ; I spray system can be predicted as shown in Table 9 4. , LEU Technical Specifications for the spray system are based on this , table, and require that each of the two systems shall be capable of { delivering at least 10 gal / min for the first 3v minutes and at least 7.5 f (")
\ r gal / min for the next 60 minutes after a lhCA onset. Ideally, this flow r would be distributed equally to each of the 64 fuel element positions, but in actuallity this is not practical. It is practical to have (1/2 x '
F 1/64) or 1/128 of the total flow delivered to each element. At the initial installation, measurements 32 were made to verify that each of the two spray systems is capable of delivering 10/128 or 0.078 gpm to each of the 64 fuel element positions at the end of 30 minutes of flow. Figure 9 12 gives the decay power of the hottest element as a function of time for the first 2 hours after shutdown, the specified heat ' removal capacity of one ECSS sub system, and the actual heat removal capacity of each ECSS sub system based upon actual flow testing of the L ECSS. l l 9 62 l
w c . s~ '
- I
- ~ -) . r TABLE 9 4 .
CHARACTERISTICS OF SPRAY SYSTEM (FACH TANK) ! Time Head of Water Flow Rate Volume of Water Used i (mini (feet) (cal / mini
- ( m_ al ) ;
O 14.92 12.1 0 (Tank full) t 7.8 14 11.8 103 16.6 13 11.4 215 P 25.7 12 11.0 328 35.2 11 10.6 440 45.1 10 10.1 552
$5.6 9 9.7 664 66.7 8 9.2 776 !
o f 78.4 7 8.6 889 i 91.0* 6 8.1 1000 I 105 5 7.4 1113 i 120 4 6.7 1225 i~
. 137 3 5.9 1337 158 2 4.9 1450 l 168 1.58 4.4 1497(Tank empty) .
i
- Flow beyond 90 minutes may be less than the specified 7.5 gal / min and therefore credit is not taken for this flow in the calculations.
k 6 l f G. 9-63
~q i
d V ' .- ( Power of Hottest Element and ECSS Cooling Capacities 20
" West ActuoI Performance
----..%t----.......,----........__
e a 3 x ECSS Specificotions
- 10 F v
L e 3 O Q_ j Heat Production 0 O 1000 2000 3000 4000 5000 6000 7000 Time After Shutdown (Seconds) Figure 9-12 Power of the Hottest Element and ECSS Cooling Capacities f
L f i i r I
) 9.18. Time to Uncover the Core Followina A LOCA {
(x_/ l
!. Table 9 5 shows the time it takes to uncover the core for the i f 1 i various mechanisms of pool leakage calculated in Section 9.18. '
L i, Table 9 5 l l Time to Uncover Core j For Various Leakage Mechanisms j Leakage Path Time to Ur. cover Core
- 1) Double ended break at lowest point in 6. inch primary piping....................................... 19 minutes '
- 2) Rupture of 8. inch beam port and flange . . . . . . . . . . . . . . . 22 minutes [
r^x t' 'j ' 3) Double ended break in the 6. inch primary pipe at the level of the floor in the heat exchanger roem.... 23 minutes i
- 4) Quarter. inch, vertical crack extending the entire !
depth of pool, assuming no frictional losses......... 36 minutes , i It is postulated that a guillotine pipe break in the six. inch t diameter pipe between the reactor and the heat exchanger is the most ; rapid way for the UVAR core to uncover. Without subsequent closure of the isolation valves, such a double ended pipe *areak would allow flow I from both ends if the pipe. The geometry of the system is shown in Fig. 9 13. Calculations were made assuming that a pipe break would occur at - heat exchanger room floor level and at the lowest pipe level. It is noted that this pipe is surrounded by either concrete or earth wherever n. l 9 65 i t
-a, -
-,, - . . . - - - - , , . . - - - - - . . a ,, ,. , , - , - , . . . , - - . - .
~
)
L i v [ it is below the floor level. Since a guillotine pipe break is not likely r to occur at this location and water could not freely flow away if it did, i ! i l the calculations are considered to be conservative. Because the break is ' assumed to occur close to the pool, there would be little frictional ! resistance to flow through one end of the break. However, flow through the other end of the break must first travel through the pipe and the ! heat exchanger, and consequently the resistance would significantly decrease this flow rate. This was taken into account in the calculation of the time to uncover the core. 9.18.A. Flow Rate without Frierional Loss : The flow velocity V2 through the pipe end for which no frictional loss was assumed was calculated using Bernoulli's equation, h ( With positions 1 and 2 defined on Fig. 9 13, Bernoulli's equation for flow between 1 and 2 is. 2 2
+ z1 b- + r 2 gcL (9.71) 2 g, + p g, 2g,+ p ,
From Fig. 9 14 V) - 0 (i.e., Vi is velocity before the water enters the pipe) El " 22 , P1
- Po + pz (where z is the water level above the pipe break)
EC P2 ~ Po - 1 atm i Hence, for the present case, Bernoulli's equation reduces to - V Po+ FE ' P2
'S c p O
9 66
- -- - . ... - . . . - . - - ~ - . . . .. . ..~-- - _ .
- . .... .I i
t l
.i i O !
v I I ' ','
- f.*e ,. !
'j* g _ __
h . e- :
') . )
e ', i
> 4*.)
' i l e' A !
q.* ' i s'. *6 , . l
. ,. /
t)i b. . '. 1
}
i l'g.' 22'3 " / .'* . i 1 I .6' . %. IA l 1') ! .; \ ;. l
> e*. l !
l', I e ',p'. A .' j'*. .
>' e l
[$ ...
}
( ! *b
, ,. =
. am c, ., "
i
/
GRID * ,5,, i g7.'3 f PLATE .b _ _ e r excg493tn ; p. aJ.4 l j'
, .. t
- q. . d i
(,3 6I (
, . . _ _ . _._-)
i p* *
.> a 6 .
t .. .. p
- h. .
,,;. 3 .
, ,3, , .. .c .. j t
Y'y. , 's . ' , 5 . ',p ,' , le
. ; a '. ' '
t EXTERNAL PIPING Length =70 feet i
's f
I l i h FIGURE 9-13 PRIMARY PIPING
- l.
- I f
6 O 9-67 7 F 9-
~...,..----,.,.,,,.-,,-,,-e,,-n-,-. ,,,,..,...,.,,-...,,,,----,,_.--,,-._-,.--,-+-,,.,--,,-,n.---,_ene.,- -
..-,---,-mw, , , -.
4 h O Yo h l Z t i O PIPE BREAK U P1 y, 4 I h /- 2P
- . -N2 l'
I i 1 2 l flare 9-14 GEN POR CALCt1ATING Ya 1 h o O 9-68 !
, . , ,,-m-, -. .- .,% -
I ( ) or N ../
- V2 - /2gz (9.72)
! 9.18 B Flow Rate with Frictional Loss The effect of the frictional loss in the pipe and the heat exchanger is to add an energy loss term to the Bernoulli equation, as follows: 2 2 2g g
+ + zy A-gc
+ + z2 ES + H I (9.73)
P 2E e e l e vhere Hj - (and where Apr is the frictionai pressure drop) The subscript f is added to V2 to denote the velocity at position 2 with friction loss taken into account. Equation (9.73) reduces to: 2
+ - z I-2 g, p g, (9.74)
In order to solve for V 2
.f. it is necessary. to express opf in terms of V 2 ,f. This can be done by recognizing that the form of the frictional pressure drop is 2
L t2 APf~I D 2g c By neglecting the dependonce of the friction factor "f" on velocity, this expression for Apr has the form, 2 opf - aV (9.75) where "a" is a constant. Neglecting the dependence of 'f' on the o velocity is sufficiently accurate for the present calculation, as can be 9 69 4
i j ') shown as follows. The velocity through the 6 inch pipe during normal l n-operation is 11.4 ft/sec, and the associated friction factor is 0.0144. The maximum velocity in the pipe at the beginning of a pipe break, with f no resistance to flow, is 41 ft/sec, and the friction factor for this ! velocity in a 6. inch pipe is 0.0136. The velocity remains between 11.4
- and 41 ft/sec until the water level in the pool is 2.0 ft above the pipe l break. j The constant "a" in Eq. (9.75) can be evaluated from a knowledge of the pressure drop and velocities for the piping and heat exchanger during ;
rated flow conditions, as follows: Mass flow rate during normal operation - 500,000 lbs/h (900 gal / min) ' V (through 6" dia, pipe) - 4.1 x 104 ft/h
~
opf (Heat Exchanger) - 9.5 psi - 1370 lbf/f t2 (during normal operation, as experimentally measured) Length of pipe in external line, exclusive of heat exchanger ; 70 ft. Re (in 6" pipe) - 7.7 x 105 f - 0.014 apf (pipe) - 1.7 psi - 250 lbf/ft2 f Apr (pipe) + apf(heat exchanger) - 11.2 psi - 1620 lbf/ft2 ge - 4.17kl0 8 2 l i Using the form for op from Eq. (9.75), together with continuity, one can . express the sum of the pipe and the heat exchanger pressure loss as a function of the velocity in the pipe as: j opf - Ap pipe + apheat exchanger " "Y I o 9 70
B~ l
, / 's i" ) or >
l
' 2 i
, ,1620 lbf/ft 9,7g j (4.1 x 100hrL) I i
i Substituting Eqs. (9.75) and (9.76) into (9.74) gives the following !
, result for V2 .f: '
I V 2 ,f - - 0.378 5 (9.77) i i {+6.5 2 i
, This value compares to V2-/2ge-1.41M [Eq. (9.72)) :
r where V2 is the velocity without friction loss, or V, = 0.27 V2 (9.78) 2 .l f 9.18.C. Time to Uncover core with Double Ended Pine Break ' To calculate the time to uncover the core with a guillotine ; pipe break, we equate the following two expressions for the rate at which
- water is flowing out of the pool:
(1) Flow rate out of the broken pipe $
- PSV2 + pSV2 ,t
- 1.27 pSV2 [from Eq. (9.78))
(9.79) I where S - pipe cross sectional area I (2) Since h - the rate of change of water level in the pool, the flow out of the pool is also i
- pA (9.80) t where A is the surface area of the pool, i Equating Eq. (9.79) and (9.80) for flow rate gives sig , 1.27S y l dt A 2 i i G l
I V 9 71 1 f
.. _ 4 . . _ . ~
( ) Substituting Eq. (9.72) for V2 gives L J
, .1.z?S g g, , Finally, the time, to for the water level to change from (zo + zi)
(i.e. the time to uncover the core), is n , t
-fz o4tg .
1.27S (9.81) jp , MS (/zo 4 zi . k) Values used in the calculations reported in Section 9.18 are: g - 32.17 ft/sec2 A - 32 ft x 12 ft S A (f ft) ex zo - 22.25.ft
]
for pipe break at floor level of heat exchanger room, sg - 1.92 ft, for pipe break at lowest point in pipa, z1 - 4.83 ft. (See Fig. 9 14). 9.18.D Time to tineover core with crack in Pool Van Tor a crack in the pool wall of width 6 running the full height of the wall, the time necessary to uncover the core, assuming Bernoulli's j equation with no frictional resistance, is I 1 65 (k I t .
) (9.81% ^
/zo + z1 where si - distance between bottom of the pool and the bottom of the grid .
i' plate and zo - same as in Fig. 9 13. p i
't' ) 9 72 i
j- ._. P l-t p) Q
,< REFERENCES FOR CHAPTER 9 l
l 1. Cambill, W.R. , " Generalized Prediction of Burnout Heat Flux for , Flowing, Subcooled, Wetting Liquids " Chemical Engineering Progress Symposium Series, Vol. 59. Houston, 1962. !' 2. Bergies, A. and Rohsenow, W., "The Determination of Forced. Convection Surface Boiling Heat Transfer," ASME Paper 63.HT 22 (1963).
- 3. Cox, J. , "0RR Operations for Period April 1958 to April 1959," CR.59 39, Oak Ridge' National Laboratory (April 1959).
- 4. Eckert, E.R.C. and Drake, R.M., " Heat and Mass Transfer." McGraw.
Hill, New York, 1959.
- 5. Private Communication from Wallace R. Cambill, ORNL.
- 6. Bernath, L., " Theory of Local Boiling Burnout and its Application to Existing Data," Ch.E. Progress Symposium Series, Storrs,1960.
- 7. Todeas, N., "Effect of Non Uniform Axial Heat Flux Distribution on s the Critical Heat Flux," Ph.D.. Dissertation, M.I.T., Sept. 1965.
'~ 8. DeBortoli, R.A., " Forced. Convection Heat Transfer Burnout Studies for Water in Rectangular Channels and Round Tubes at Pressures Above 500 psia," WAPD 188, Oct. 1958.
- 9. Cambrill, W.R., " Forced Convection Burnout," Nuclear Safety, Vol. 5, No. 2. Winter 1963 64,
- 10. Tong, L.S., "DNB Studies in an Open Lattice Core," WCAP 3736, Aug.
1964.
- 11. Perry, J.H. (ed), " Chemical Engineer's Handbook," McGraw Hill, New York, 1950.
- 12. Esselman, W.H. , 31_al, " Thermal and Hydraulic Experiments for Pressurized Water Reactors," Proceedings of the 2nd United Natiens International Conference on the Peaceful Uses of Atomic Energy, Vol.
7 pg. 758, United Nations, Geneva, 1958.
- 13. Von Karman, T., NACA Technical Memo No. 611, 1931.
14 Cambill, W.R. and Bundy, R., "HFIR Studies of Tutbulent Water Flow in Thin Rectangular Channels," ORNL 3079, June 1961.
- 15. Reynolds, J. , " Local Boiling Pressure Drops," ANL.5178, March,1954.
O 9 73 9
i l ') 16. Barnett, P.C. , "The Prediction of Burnout in Non Uniformly Heated Rod
'/ ~
Clusters from Burnout Data for Uniformly Heated Round Tubes,' AEEW R- i 362 November 1964. !
- 17. 'Non Uniform Heat Generation Experimental Program " Quarterly .
Progress Report No. 5, July September 1964, BAW 3238 5, Sept. 1964. !
- 18. 'Non Uniform Heat Generation Experimental Program," Quarterly l Progress Report No. 6, October December 1964, BAW 3238 6, December !
1964. . i
- 19. Sternberg, H.I. , ' Thermal Power Celibration and Correlation of UVAR :
by Toil Irradiation and Heat Balance," Master's Thesis, University of l Virginia, June 1964 l
- 20. Brunot, W.K. , 'An Analysis of Fuel Element Surface Temperature and '
Coolant Flow Rate in UVAR," Master's Thesis, University of Virginia, i November, 1963, i
- 21. Custer, G.A., "The Experimental Determination of the Thermal Neutron l Flux in the University of Virginia Reactor Using an Aluminum !
Hydraulic Rabbit, Master's Thesis, University of Virginia, August i 1960,
- 22. "1969 Status Report on the Omega West Reactor, with Revised Safety 7-~ Analysis," Los Alamos Scientific Laboratory, LA-4192, July 1969.
' 23. DiNunno, J.J., et al. ' Calculation of the Distance Factors for Power and Test Reactor Sites," TID 14844, 1962. !
24
- University of Virginia Reactor Safety Analysis Report," '
University of Virginia, UVAR 18, Revised January 1979. !
- 25. Dahlheimer, J.A., ' Thermal Hydraulic Safety Analysis of Pool Reactors", Master's Thesis University of Virginia, 1967.
- 26. University of Michigan, "2DB UM Appollo Version". Version #10, September 1986.
- 27. Freeman, D.L. , *Neutronic Analysis for the UVAR HEU to LEU Conversion Project *, Master's Thesis, University of Virginia, January 1990.
- 28. Smith R.S. and Woodruff U.L., 'A Computer Code, NATCON, for the Analyses of Steady State Thermal Hydraulics and Safety Margins in ;
Plate Type R$ search Reactors Cooled by Natural Convection", ; ALN/RERTR/TM 12, December 1988. - L 29.. " Technical Specifications for the University of Virginia", September 1982 Amended April 1988, and December 1938,
- 30. Telecopy from Argonne National Laboratory, " Properties of Fuel Heat g-~) Materials", August 19, 1988.
9 74
. - - - - - +-- r - - - - --,-
w
f < i (~- / 31. "Docay Heat Power in Light Water Reactors" American National Standard ANSI /ANS.5.1 1979, Reaffirmed July 1985 American Nuclear Society, La Grange 1979. L
- 32. Meem, J. L. , " Emergency Core Spray System Installation And Testing",
Report to University of Virginia Reactor Safety Committee, September, 1971.
- 33. Nagler, A. , Gilat, J. , Hirshfeld, H. , " Evaluation of LOCA in a l
Swimming. Pool Type Reactor Using the 3D AIR 1DCA Code," paper presented at the XII International Meeting on Reduced Enrichment for Research and Test Reactors, Berlin, Federal Republic of Germany, r *. 0 13 September 1989. I U t J l 9-75 1 e I
(: e S a fe ty Analysis for the tiniversity of L Vi rci ni n Itenc tor 1.rtl Conversion . [ R.A' Rydin, D.W. Freerran, B, llosticka and R.U. Mulder Dept. of Nuclear Engineering and Engineering Physics t
. University'of Virginia j Charlottesville, Virginia, USA . ; ABSTRACT 6
[ The University of Virginia is preparing to convert the 2MWT UVAR reactor from 18 curved plates /elernent high enriched fuel to comparable flat plate low enriched fuel. Depletion studies of low enriched cores suggest s~ that the~use of 22 plates / element in a 4 by.$ array will give a
' considerably longer core life and a generally higher thermal flux than direct replacement with 18 plates /olement. The excess reactivity of this larger and more highly loaded core can be regulated by selectively changing
- i. portions of the reflector from graphite to water, Calculations of control
- rod worths, moderator and Doppler temperature coefficients, void F coefficients, and kinetics parameters indicate that the LEU cores will have performande. characteristics that are not very different from the 110U cores, Thermal hydraulic calculations indicate that these cores are acceptable with only small changes in safety systern settings.
L; b DESCRIPTION OF THE UVAR FACILITY f l' ~ The UVAR is.a 2 MWT swimming pool type research reactor. It is made up of plate type MTR fuel elements mounted on an 8 by.8 grid plate that is {h p suspended frorn a movable bridge above a large open pool of water. reactor can be moved to either end of the pool while the other pool half is The ! drained for maintenance purposes, However, the core can only be operated E
- at full power when it is mounted on the South end of the pool, directly above a coolant funnel that provides forced down flow circulation. This
! -position is shown in Figure 1, which also shows the location- of the experimental bearn ports. L The original UVAR design was done by J.L. Meem (1) et al. , circa 1960, [ using analytical two group theory. The Technical Specifications (TS) require maintenance of a minimum shut down margin of 0.4% Ak/k with the largest worth shim' rod fully withdrawn, and a maximum excess reactivity of , 5% Ak/k. Any core arrangement that will fit on the grid plate, and that r meets.this TS, can be used, providing that the control rods are experimentally recalibrated each t.ime a new core arrangement, which was not previously tried, is assembled. The UVAR has been operated for more than twenty years using experimental techniques, essentially without benefit of computational modeling. During this time, both 12 flat plate fuel elements (- and 18 curved plate fuel elements have been used in separate cores, and arrangements having anywhere from 16 to 27 fuel elements have been operated. Some cores have been entirely water reficcted, others graphite reflected, while most cores have had water on some sides and graphite on the others. FL_ _ _
~
It is not. really to our advantage to allow the WAR core size to become as large as 27 elements, because the thermal flux availabic for experiments is correspondingly reduced. We believe that a more practical future strategy would be to try to operate with a core arrangement that is essentially fixed in a 4.by 5 array, with the shim rods placed close to the core center, as illustrated in Figure 2, to provide a high shutdown margin, Ve also believe that a planned fuel shuffling pattern should be used in place of our current ad hoc procedure of trying to obtain equal burnup for each element in inver. tory. L LW c3 cJ
* /
nevaim W. -e .
..._ sfl meessue , a
- I a o
- -l ' mesmane/
a" r I-x
~~
w vu 1AN0tNTIAL
; I BE AMDOX 1ARGET POR T t .-+]
} '
"DE l
t -FULL ELLMENT
.T ANCE Nil AL BEAMTUDE
<a :IIxr.ta- A B AM l
= ,-
sOUT HE Ast F Li y South ACCESS i ACilli Y Figure 1. Sketch of WAR Pool Showing 8-by.8 Crid Plate l J
+- ,
V - C N*i5-.-. r v
] an: _1 -
.i?*tN,**U!!NTI!*k*I=I"i ;; N .vz '-~~"Li" PI
- K m -s h -.
pv -Qm'1 l
- s .. 4~ +~m d r-- - N
- k. I _ 'b!"E h- [i;;* it=++h ~ **'
t T
- =;w : -m p =st -- , &
gmg. - [3 ,..,.,.*p
?' ;
F- - - - - 2lv y f' l
<~
J-n:. 3.M,- 5 Jute ,
===n z yh. lh t--
~J.: - 1 y
n=iE3
- =t '
fgs. .' . 4 -. =: j h="q f "ja!!E $ fiEEEEP is:::n m n - IEEssssss::xrn.:===i=:v;12i l me===m " ~ Lgunsp=EEEEmirn;iinuz=u=tyg mns:E nit ;=.n~ p: : -] :======= Esse =mE=EsssE ..i
~ u m
- .- ---:e.--.a=..=.-
-e
_g [ ;pnEEEEiii~5E,!.E_EEE.!EEEEEEE.E.!, I g g . -4 -~- - q FIEE
- nt GRAPHITE :fi WATER h:t GRAPHITE : WATER i
- Figure 2. Ideal LEU Figure 3.1975 Texas A91 Core F. UVAR Core Configuration i
BENCitMARK CALCULATIONS l l' The most recently measured UVAR. core configuration that met the conditions of being clean, unburned, and fully documented, was an' initial 1975 loading of 18. plate Texas A91 fuel elements. This arrangement, shown p g' in Figure 3, was a 4 by-4 element array, asymmetrically surrounded by i graphite. Almos*: all other recent UVAR core arrangements have contained j fuel elements having only an approximately known individual burnup history. L making them unsult.able for benchmark purposes. Availabic data for the Texas A&M core included measured individual i control rod worth curves, an implied measured excess reactivity worth with l 3 all control rods withdrawn from the critical position, an approximate i temperature defect worth obtained from singic warm up and cool-down swings, ; and an implied measured equilibrium xenon worth. Since the control rod f s worth curve- are measured by repositioning any three control rods at equal j depth to ic. remontally measure the fourth, some control rod interaction effects are built into the measurements. ; i i a 1
v - Effective Control Rod Cross Sectingg Control'~ rods are strongly absorbing bo.fies having relatively small planar dimensions. They must be treated using transport theory which fp *' ' h'
- must.also.take into account therma 11 spectrum hardening. We have developed
't . effective diffusion theory cross sections in two groups, for' both the boron no steel, shim rods and the stainless steel regu1~ating rod, by applying the p following procedure. .
A ' transport theory model of e control rod surrounded by fuel material'
'was'made in slab'goometry using the THERMOS thermalization code. . Thermal i ' group absorption fractions'in both the fuel and absorber regions, vere obts.ined:from this model. ,Past group absorption fractions'for the same
, regions.wcre;then obtained using the CAMTEC slowing down and u thermalis:ation code in cylindrical geometry; the CAMTEC thermal group result'was cross checked against THERMOS. A cell model was then made'
^
tusing the 2D diffusion theory code EXTERMINATOR, and the fast group and h', ' thermal group absorption cross sections were iteratively varied until both
.the' thermal and fast group absorption fractions matched those from THERMOS and CAMTEC. . Finally, these effective cross sections were used with similar 7
mesh spacings in the'2DBUH and 3DBUM diffusion' theory codes.for our reactor b
^: design studies.
p Control Rod Worths [ [ j,1 l The procedure for obtaining effective control rod cross sections was applied separately to both HEU and LEU models. Within'the uncertainties ? of the iterative process. : the results were essentially the same for both, i' When these values were usedLin 2DBUM models of the'4 by 4 Texas A&M core (called IIEU-18) the. results shown in- Table 1 were obtained. F Table'1. Control' Rod Worths for the 4 by 4 Texas A&M Core and - Replacements b Case Rod 1* Rod 2* Rod 3* Rod 4* 4 6 $ $ ,
.S $ r c ,
y, HEU-18 Expt. 4.75 5.00 3,06 0.57 HEU-18 4.71 4.96 2.86 0.73 i' LEU-18 4.61 4.86 2.79 0.83 . LEU-22 4.69 4.91 2.90 0.84 ,
/ , b
- Experimental and Computational Uncertainties are 15% -
y -e' ' 7 The. integral rod worths for all three shim rods were predicted within i the experimental accuracy of the measurements, while the regulating rod [ was' predicted slightly high, probably because the boundary conditions on ; the cell model were noc quite correct. Experimental uncertainties of 15% are' attributed to inaccuracies in period measurements and to uncertainties : in the precise value of beta effective, which was taken to be perg - 0.C38. ! 1: S > _ , . , , _ .,
m g .;. ( - - . c ; L y D ; 4 N -Calculations were also'made for LEU replacement cores having 18 y L plates / element' (LEUg18) and 22 plateshlement (LEU 22). The predicted * [ : control rod worths'are also.given in Table 1, where it is-seen that they p i do not ' differ markedly from the HEU 18 results. , k ' When the critical control' rod positions for a core are entered into l [ the : experimentally. measured integral control rod worth curves, the excess j
' reactivity available for removing the rods entirely can be determined.
- h( ^ ,
- The' sum of these values for the Texas A&M core experiment is reported in.
Table 2. The value of k org that is'obtained from the corresponding 2DBUM , L model of the same core, using axial B2 values obtained from an ANL 3D. . model,;is~approximately 0,026 higher. {
'n . . .
Three-dimensional calculations for this core were recently completed -
! - using 3DBUM. The rodded and unroddedok rt values (Table 2) agree well' l, with 2DBUM. The value of ek rg with the control rods placed at their.
F <
. critical positions is k gg e - 1.026, thus confirming that the difference
{o between the measured and calculated ek gg values is real. Table 2. Beginning of-Life Unrodded ko rg for Texas A&M UVAR Core - [ !
'ONDITION C kegg 3
EXPERIMENTAL,* HEU 18 1.036" 0.004 ] 2D'B CALC, 6 - 7.8 CM, B2 .0017 . 1.062 60-by-60 mesh n 3DB' CALC 1.062 ! h :60 by-60-by 21 mesh
- Implied From Control Rod Worth Curves _
a i [ A series'of 3DBUM calculations was performed where the control rods were moved together in a bank in order to simulate a composite integral ~' i control rod worth curve. The results are given in Figure 4, along with a corresponding experimental integral control rod worth curve constructed " from the individually measured control rod worth curves. The data agree quite.well, with the computed curve shifted only slightly towards the > > bottom of the core. The maximum uncertainty in the measured curves is therefore of the order of Ap - .01. f We conclude that the initial burnup status of the Texas A&M fuel
'> elements was not known as accurately as we thought. The elements probably E had somewhat more than the reported 0.84% average burnup during their many
- years of periodic operation at 100 KW. Indeed, the fuel elements used in L
.the 1975 UVAR core may have had greater than average burnup. Furthermore, '
residual camarium could account for as much as Ap - -0.005 in this otherwise cold clean configuration. l 1 i
'i
- r .O
- .y ;
p.7 r1 . t POSITIVE REACT VITY INSERTED IN UVAR I 4-ST-4 88U CORE TIA ROD Wit!!DRAWAL 0.18 U 0.10 - /
, [j } -
0.00 - ,./ 0.04 - / ! o +
/
0.07 - 0.00 -
/ ' N 0
0.03 - /
+
0.04 - '
/
I 0.02 - / ' i- 0.04 - 4
/
o / I 0.01 - 0.00 -- i , , , , , , , , , , , , , ! O 4 8 ' 12 18 20 24 20 CONTROL R0D WIT!!DRAWAt (tNCRES)
+ CALC COMPOSITE g . O Ray composite Figure 4. Composite Control Rod Worth
, Curves for the 1975 Texas A&M Core Feedhnck Ef fects f. Experimental values are also available for the Texas.A&M core for the
; worth of equilibrium xenon and for the moderator temperature coefficient.
The xenon worth was obtained from the differences in the critical control. i rod positions for the no xenon case as implied from the measured intogral i control rod worth curves. The temperature coefficient was implied from a . single heat up experiment and a corresponding cool down experiment whose l worths differed by about 50%. The AT used for the experiment was one half j the average core temperature rise; the effective AT is undoubtedly higher. 1
. Xenon buildup was also ignored. lle nc e , this measurement cannot be i' considered to provide anything more thin an' order of magnitude estimate.
The experimental Texas A&M core feedback results are given in Table 3, where they are compared with calculated results for both IIEU and LEU , models. The calculations of temperature coefficients were made from ! LEOPARD LINX 2DBUM models of the full cores. ! The experimentally derived xenon worth is about 0.004 lower than the calculated worth, which implies that samarium was already included in the ! f initial criticality. The eclculated temperature coefficients are lower j than.the measured value by more than a factor of two, but this is not ; considered bad agreement due to the inaccuracy of the measurements. The 1 LEU cores have very slightly lower moderator coefficients than the llEU j core, but they pick up a Doppler coefficient due to the increased 238U loading. 1 i
- n. 'l
k E ' ~ [ f h .s+ [ s. lt' 'h f..' , h > R Table 3. Feedback Ef fects for the 4-by 4 Texas A&M UVAR Cores and (g : ..
. Replacementsc ;~
n, , N Case Xenon Samarium Temperature
- Moderator
- Doppler Worth -Defect Coefficient Coefficient Ap' Ap *C 'C (x 104 ) op/AT(4).)
(x 10 .Ap/AT(5))
.(x 10 V n.;g u.
?E HEU 18 Expt ' 1.9%:(Xe) 19. -5.2 HEU 18 2.3% -7.6 1.9' -0.09 R LEU 18 _ 2,3%
-7.3 1.8 1.0 1.EU 22 . - 2'. 4 % '
6.8 -1.7 ' 1.2
- Experimental Measurements are i50%
- g. ..
;, : 1 LEU DESIGN STUDIES
.In order to make the design problem tractable, we picked three fixed garrays (4 by 4, 4-by 5 and 5 by-5) as the bases of comparison, and did' core lifet calculations with IIEU-18 plate fuel and LEU 18 and LEU-22 plate replacement' fuel. All LEU cores start out with a somewhat lower'k ogg-than the corresponding HEU cores because they have a. harder neutron spectrum and' -
a' consequently greater leakage. The burnup curves for LEU are Jesu steep
. than. for. HEU, and therefore the excess reactivity curves eventually cross ; as depletion' increases. However, for the 4-by-4 core, the lower initial L
M kegg of LEU-18 fuel cannot be made up by the decreased burnup slope before the excess reactivity crosses zero, and therefore LEU 18 fuel will not. last
,o as-lo'ng as HEU 18 fuel. On the other' hand, LEU-22 fuel will have comparable performance to HEU 18 fuel.
6c .]
]
,s M For a 4 by-5 array, as seen in Figure 5, one finds that both the LEU- (
,, 18 and HEU-18 fuel reach an asymptotic behavior, and these cores attain I essentially equal burnup at the same point in life where the excess reactivity crosses zero. On the same basis, LEU-22 fuel lasts about 50%
longer than LEU 18 fuel, even though the uranium loading is only 20% 'j
. greater.
{ l.. - e A similar behavior is seen for the 5-by 5 core models. Again, the ] HEU-18 and LEU 18 cores have essentially the same endpoint, while the LEU-a ' 22 core lasts-about 50% longer. But the most interesting result is'that y h *
. the. LEU-22 core in a 4-by 5 array lasts almost as long as an LEU-18 core in a 5 by 5 array. This means that an LEU 22 core can be kept in a 4-by 5 e configuration, with attendant higher average thermal flux for experimental !
purposes, and still operate almost as long as our previous larger cores. ( [ l 4 .> y . l
W> -- k l ' l. . k [ ANL[2] has independently calculated all of t.he 4 by-S cases using a 3-
- e. dimensional model, and has verified that a 2 dimensional model gives
[ similar results when the correct spatially dependent axial B2 values are used. Their results are shown in Figure 6. We find that our HEU results F are in almost exact agreement with theirs, while our LEU results are offset low by about 1% in kegg. Only minor differences exist between the UVA and ANL 4 by S LEU models. UVA included a small amount of 2340 and 236U in thn b LEU fuel specification, whereas ANL used only 238U, This change
- essentially accounts for the difference in results. The qualitative t.
conclusions are still the same: LEU 22 fuel is a superior replacement, relative to LEU-18 fuel, for use in the UVAR. t I
'.' a v./r vs. DEDLELON b- io , ._-_.'+*5 ._ . . _ _ _ _ _ ,
l 1 e I
*1 8,
I k i .y . 4-g f ,.~ +,,,- s i-
,,+s ,
e -s
-i , . 5, ,
0' 200 400 600
;;c'/k
< /s. DEPLELON Figure 5. Unshuffled 4 by-5 ,,
""5"'"*
Depletion of HEU and LEU Cores Using 2DBUM ' " t-L 7 7-k I D 3 3-x N,\ s- ,. s n N i-g .-
-i , --r . . , - -
0 300 .00 600 0 ntu e u8 6 LtU 32 Figure 6. Unshuffled 4 by 5 Depletion of HEU and LEU Cores by ANL in 3D
m ... 4 3u . c:7 Wa 1
' .r j f, ' Re flec tor ' Ef fect s .
- i. -!
e -y, The UVAR. currently has a Technical' Specification' limit of 5% excess !
, . re ac tivi t'y . Examination of Figure 5 reveals that the new 4 by 5 LEU cores-
{ q , will=have,between 7 and-10% excess reactivity. Therefore, some external p, action is needed to meet this specification, 1 We have done a series of calculations whereby various' rows of - hy reflector graphite have been replaced by pool water. This is: accomplished f ii
- in practico by inserting aluminum plugs in the grid plate. As shown in i Table 4,~ selective une of water on one or more faces of the core provides p?
an adequate means to. cope with excess reactivity above TS limits. Fine
$ tuning can he accomplished by replacing individual rows-of graphite by' 1 wate'r. rather= than. replacing all graphite on a core face.
~
p g< < ,
- Table 4, UVAR Reflector' Effects for 4-by-5 LEU Cores
[ < WATER ON FACE; top V L
- Bottom (B)- 1.2 i
Top-(T)' -2.0' i t L~ .Left (L) or Right (R) 2.3 _ TD /3.4 TLB -5.4 TLRB 7.4
-i l
Temnerature and Vold Feedhnck I Temperature coefficients.were calculated for 4-by 5 element cores (
'using.both HEU and LEU fuel. In all cases, the normal moderator t 0 temperature was taken to be 21'C, whereas the, nominal fuel temperature was taken to be 33*C for HEU 18 and 30*C for LEU 22 fuel. Doppler coefficients were evaluated at 75, 100 and 200*C. Moderator coefficients were evaluated Eat 50 and 75'C.
( , The results are shown in Table 5. It should be noted that small , variations of the temperature coefficients occur with temperature. The , numbers quoted are estimated to have an uncertainty of about i 10%. It is i seen that the Doppler coefficients are comparable to those given in Table 3 L , for'4-by 4 cores, whereas the moderator coefficients are slightly lower. 1 w E /m : - r
y -. -.
.J - h g.
~
~.,.
p.. .. y , .k Table 5. Temperature Coefficients for 4 by 5 Cores x ; bU , Situation , Moderator Doppler ap/AT(*C) p Ap/AT(*C) t [ LEU-22 1.4x10'4 -1.1x10-5 4< Graphite HEU-18 -1.5x10-4 -1,3x10 6 Graphite
. Void feedback can be simulated by recalculating cross sections in' / 1.EOPARD.using reduced water densities, and using these cross sections:in
[ 2DBUM. ,There'is a limit to which the procedure'can'be applied due to
, limited.-data libraries, the range of validity of the LEOPARD treatment, and p .the. applicability of diffusion theory. On the other hand, the order of-magnitude'of the effect is important to estimate.
A series of 2DBUM calculations were performed assuming both uniform voiding and relatively complete but' local voiding near the hot spot in the-core. Not surprisingly, the void coefficient changes-with the percentage
,of core voiding due to neutron spectrum changes. Voiding will also take
<. place locally near the hot spot, which tends *.o be in a position of high
- . Imporcance'.
The results of the calculation are shown in Figure 7. For uniform voiding, HEU.and LEU are comparable. On the other hand, for local voiding, llEU provides more feedback than LEU due to neutron spectrum effects. The void coefficients are also several times larger than for uniform voiding, showing the importance of the position where voiding is expected to take place. UNIFORM AND LOCAL VolD COEFF FOR 4-pf-S CORES r-I -u -
.I m
" ' " ~ g ," Nu
. :'i;1 1 -u - N <'
j,
-o.4 -
- " 'N %
! N
> ju -u - #
, -u -
( . rwmMI m n == na ! l 8 -u -
!- . E%%l .... m. no g .o - ;
,I
-u - , , , ,
{ PE AC8NT WODERATOR Y010 Figure 7. Uniform and Local Void Coefficients { y. I l 9 y l,; l v, l
f
< ', ,1 ~
; ~
?
e , l
' Kinetics parameters, i I:I *'
V , l Although'a.value of beta' effective of pegg .008 was in use for UVAR g , We , in 1975, soon afterwards it.was changed to egg - .0075. .'A series-of new I; calculations have now been completed to determine Berg and the neutron lifetime, 1, for both LEU and HEU cores under various. conditions. These !.(, ' eniculationsmwere;made using the EXTERMINATOR code, with 4 group cross 6 sections taken from both LEOPARD and EPRICELL. The first step in the process was. co compare the 4 group EXTERMINATOR
~
models-to the' corresponding 2 group 2DBUM models. Calculated values of
~ke rg egree quite well~when identical 60 by 60 mesh models are used.
p.
~
Values for neutron'lifetimo are given in Table'.6. In general.'t.in. E, about 15C lower for LEU corestthan for IIEU cores, due to greater neutron p V: absorption'in.the core. :In addition,1: is about 20% smaller for water p"i -- 2 ' reflectors! than' for graphite reficctors,. due to increased neutron leakage.. 6 Changes 'of this order of magnitude will not Icad to significantly dif ferent dynamic behavior for the new LEU cores as compared to the existliin. IIEU UVAR
- - cores.
a u-I
< Table 6'., Neutron Lifetime for 4 by 5 UVAR Cores
. Situation f(psec)
R ' V. i/ ' LEU-22 67.0 g CRAPilITE ' LEU 22 66.5
'CRAPHITE RODDED t
c HEU 18 78.8 CRAPHITE
- LEU-22 53.0 WATER llEU 18 64.4 WATER i r 4 CROUP, EXTERMINATOR
- -l L :
i h 1 l rj s i o: . i
4 - r.-
. 'b
~
< ... .. Beta effective calculations were made using recent delayed neutron
/' data published by Brady et. al. [3]. For the present.4 group model, the delayed neutrons.are ensentially all born into the second group. iloweve r ,
.in order to test the sensitivity of these results, some calculations were made with a fraction of the neutrons introduced into the fast group. The results are shown in Table 7, Table 7. Beta Effective for 4-by 5 UVAR Cores Situation x Berg LEU 22 .2,.8,0,0 .00713 GRAPilITE
" .1, . 9,0 ,0 .00724
" 0,1,0,0 .00736
.00741 RODDED LEU 22 "
.00752 WATER .
IIEU 18 "
.00738 GRAPilITE ;
I 4 GROUP, EXTERMINATOR, .0065 l Segg was found to be relatively insensitive to reasonable variations !
.in spectral assumptions, or to the fuel or reflector compositions. A value j of egg .0074, not much different from the value in current use, is suitable for all cases. Unfortunately, this value slightly worsens the
' agreement between measured and calculated control rod worths, mainly by over-estimating the worth of the regulating rod.
6
f ~q ,
- f D, g j c; _' .;'. , .
s, ; .
, 1;
~
% Beam port ' Effects n An effort has also been made to simulate the reactivity effects of.
n'. typical beamport next to the core,- The actuni beam port is a
~
i ' inserting
, cylindrical tube 8 inches .in diameter, ' entering at an angle, whose nosepiece cont. acts one. core face roughly positioned in tho middic~of the face, LThree cases were simulatAl: .'l) a black region, . representing'an open beam' tube; 2)'n'6 inch thick tank of D2 0 followed by water, representing a promoderated but' closed beam port; and 3).6 inches of D2 0 followed by a blackeregion,; simulating a premoderated and open beamport. The. beam. tube 7 , =was modeled by a rectanniev 6.78' inches by.7 23 inches, having.an area essentially the same as the 8 inch pipe, but extending outward
. perpendicular.to'the core,- The reactivity effects of.these combinations are shown in Table'8, s
Table 8. 3D Calculations of the Effect of Using an 8 inch Beamport s l j. . 1 Situation Ap
, No beam port 0,0
, 6" D 2 0, closed +0.0005 6" D 2 0, open -0,0027 Open up to -0.044 c.
The simulation of an open beamport by a black region gives a considerable overestimate of the reactivity effect of an actual beamport. '
- Nevertheless, the size of the reactivity loss could be significant, so that this; method of operation would not be recommended. The effect of .using 6 g inches of D2 0 in'a tank to replace graphite is not large at all, and <
presents no problems, And the effect of opening the beamport behind this ! D2 0, while not insignificant, is also not a major problem, although some limits should be placed on the rate at which the tube is filled and emptied, < THERMAL HYDRAULIC ANALYSIS
-l C
't The UVA thermal hydraulic analysis makes use of three basic computer code packages, PARET, TilERilYD and NATCON. The PARET code from ANL was used to calculato an envelope of maximum achievable power transients, based upon . Jpump coastdown and period trips, all accompanied by Scram. The net resuit i of all of the PARET analyses is the fact that the control rod release ant , insertion times are the limiting factors for the UVAR, and that temperature feedback plays only a minor role. The responses for both ilEU and LEU corns are quite similar. Transients not acco-pnnled bv v ram nre nre.ned to fall y within the'SPKRT Experimont envelope. c q-
, i
) ~ x
F e The NATCON code from ANL was used to calculate the limits of core performance under natural convection conditions. The primary results are that adequate natural convection cooling exists for LEU 22 cores that are operated beneath our allowable TS limit of 200 KW. The use of additional fuel plates in our fuel elements increases the amount of natural convection cooling available. The main tool for our thermal hydraulic analysis is the THERHYD code- [4), developed at UVA in 1967. This code-is used to calculate limiting-power'versus system flow envelopes for the UVAR, below which all PARET transients must lie. The code handles forced convection down flow, using-an axial power distribution fit and planar peaking factors from 2DBUM. The limiting condition'is given using a burnout.rntio of 1.49 (99% confidence-that burnout will'not occur)'and taking into account channel and loading tolerances and bypass flow. The peaking factors obtained from 2DBUM are shown in Table 9 for,all of the cores that have been analyzed. Also shown is an older experimental measur,ement [5,6), scaled up to an 18 plate HEU fuel element. In general, the calculated peaking factors are a bit larger than the measured value, but lie within the experimental uncertainty. Table 9. Planar Power Peaking Factors CORE CONFICURATION CALCULATED EXPERIMENTAL * [ t HEU-18 4 BY 4 1.59 1.45 1 0.15 l LEU 18 4 BY 4 1.64 LEU 22 4 BY-4 1.69 l l LEU-18 4 BY-5 1.73 l l LEU-2 4 BY-5 1.78
- Scaled From 12-Plate Measurement
'{
L When these data are employed 'In THERHYD, we obtain limiting-power versus system flow curves, such as shown in Figure 8. In general, the inclusion of reasonable tolerances in fuel element manufacture causes the j limiting envelope to approach the actual transient results from PARET. We ; conclude that reasonable control must be exercised on the process of making q LEU fuel, especially for the LEU-22 assemblies. ]
)
8
gm . , , - .,
;f 8 I
Q - jf h + psG , PMR M' l 7 Y p 4 , 11 22 PLATE 4x5 LEU 10 " d NO TOLERAtlCES -
+
9 -- GAP t .002" - - - - - . - ./ ,- b' GAP'i .007" / # 8 -- a s' ,# i
/ s s
. 7 - f,' '
- 6 - / '
i / $"~- / ENTS
- 5. - A %,y(m 3,83 gy o'
3-- 9 .......
., 2~ ! Gil0RMAL ' 0PS. -t . .- IDOWNS g l (2 MW 9 1020 GPM) t .
- . ; ;. l l l '; .;;i: i . .i k 100 300 500 700 900 1100 1300 1500 L t SYSTEM FLOW (GPM) ;
figure 8', Limiting Power vs. Flow, 4 hv 5 LEU 22 Plates / Element I The overall conclusion is that the LEU fuel is only slightly worse than llEU fuel from a thermal hydraulic standpoint. This result will require a small revision in the minimum safety system settings for the UVAP when it is converted. Otherwise, the small 4 by-4 core is more limiting than the larger'4-by 5 core (due to a higher average heat' flux), .and the 22-plate fuel element is more limiting than the 18 plate fuel element (due i F to manufacturing tolerances). - ANL have also suggested that we consider using an equilibrium cycle !
, shuffling pattern, based upon the adoption of a fixed 4-by S core array. '
This cycle appears to be very attractive for the UVAR, if LEU 22 fuel is indeed used to replace our present ilEU 18 fuel'. This option is under serious consideration, although we would Itko to retain the right to use ;
~
' bigger cores, if needed.
CONCLUSIONS t We conclude that the LEOPARD LINX 2DBUM package of computer codes is in good working order at UVA and provides a reasonable basis of predicting the future performance of LEU replacement cores in the UVAR. A 3DBUM , modeling capability is also now operational. It has been used to predict the effect of partially inserting control rods into the UVAR, thus confirming the previous 2D models and verifying the measured integral t control rod worth curves. The 3D model is also useful for estimating the reactivity effects of proposed experimental facilities mounted near the core, t The best replacement option for the UVAR appears to be the use of 22- , plate LEU fuel assemblies in a fixed 4 by-5 core array, We will seriously consider the adoption of the shuffle pattern recommended by ANL. On the other hand, it is to our advantage to retain the flexibility of loading , UVAR cores te meet experimental needs.
,77 1 -
, qEs l-w' t
,~
l; ; m' . ? .
- f. >
f'<, t None' of .the ;important feedback or kinetics parameters change greatly, L in going from HEU to LEU fuel. Furthermore, these parameters do not, vary
- greatly with c'epletion, core size or, reflector changes.- We' conclude.that the dynamic performance of'our new LEU core will be quite similar to that.
. of our : existing HEU cores.
[ -Finally, all of the postulated LEU UVAR replacement cores meet'the L required thermal hydraulic conditions'for safe operation. However, we will' L haveLeo make slight changes in current UVAR. limiting safety system: settings
, and pay close attention to the manufacturing tolerances placed'on the new p ' fuel. ;; A D' REFERENCES t- ,
- 1. J.L. Meem, Two Groun Reactor Theory, Appendix'A, Gordon and; Breach Science Publishers, New York, N.Y., 1964.
- 2. J.E. Matos, Private Communication from Argonne National Laboratory, i' August 1988.
,3. M.C. . Brady, R. Talmadge and W. Wilson, "Few Group Analysis of Current .
Delayed Neutron. Data", Transactions of the American. Nuclear Society, 11, 1988.
- 4. 'J.A. Dahlheimer, " Thermal-Hydraulic Safety Analysis of Pool Reactors",
M.S. Thesis, University of Virginia, 1967.
- 5. W.K. .Brunot, "Ar Analysis of. Fuel Element Surface Temperature and Coolant Flow Rate in the University of. Virginia Reactor,".M.S. Thesis',
University of Virginia,.1961.
- 6. R.L.J . McGuinness , " Thermal-Hydraulic Safety Analysis of 18-Plate Reactor Fuel' Elements", B.S. Thesis, University of Virginia, 1973, i
-i i
t$. i
- )
k 1- ) i l i
]
- i u -i ,,,
l}}