ML19309C551

Chapter 14 to TMI-1 PSAR, Safety Analysis. Includes Revisions 1-11

ML19309C551
Person / Time
Site:	Three Mile Island
Issue date:	05/01/1967
From:	JERSEY CENTRAL POWER & LIGHT CO., METROPOLITAN EDISON CO.
To:
References
NUDOCS 8004080753
Download: ML19309C551 (131)

Text

__ _ - _ _ _ _ _ _ _ _ _ _ _ _ - __ _ _ _ _ _ _ _ _ _

l l

l l O s

a O

80040so 753 O

'

0001 283

.

i

.

4 TABLE OF CONTENTS O

Section M 14 SAFETY ANALYSIS 14-1 14.1 CORE AND CCOLANT ECUNDARY PROTECTION ANALYSIS 14-1 14.1.1 ABNORMALITIES 14-1 14.1.2 ANALYSIS OF EFFFCTS AND CONSEQUENCES 14-3 14.1.2.1 Unccmpensated Operating Reactivity.

Chacges 14-3 14.1.2.2 S,tartup Accident 14 4 14.1.2 3 Rod Withdrawal Accident Prem Rated Power Operation 14-6 14.1.2.4 Moderator Dilution Accident 14-7 14.1.2 5 Cold Water Accident 14-9 O 1'126 t - <-c 1 == r1== 1'-1o 14.1.2 7 Stuck-Out, Stuck-In, or Dropped-In Control Rod 14-12 14.1.2.8 Loss of Electric Power 14-13 14.1.2 9 Steam Line Failure 14-15 14.1.2.10 Steam Generator Tube failures 14-18 14.2 STANDBY SAFEGUARDS ANALYSIS 14-19 14.2.1 SITUATIONS ANALYZED AND CAUSES 14-19 14.2.2 ACCIDENT ANALYSES 14-20 14.2.2.1 Fuel FaM11cg Accidents 14-20 14.2.2.? Red Ejection Accident 14-21 14.2.2 3 Loss-of-Ccolant Accident 14-27 14.2.2.4 Maximum Hypothetical Accident 14-51 14 3 REFERENCES 1k-56 ik-1 0001 284 l

l

.

w LIST CF 'J" GLES O

Table Iro. Title Page 14-1 Abnoz nalities Affecting Core and Coolant Bourda / 14-1 14-2 Unccmpansated Reactivity Disturbances 14-3 14-3 Situaticna Analyzed and Causes 14-20 14-4 Reactor Building Structural Heat Capacitance Se6ments 14-33 14-5 Core Floodic6 Tank Perfor=ance Data 14-37 14-6 Reactor Operating Conditiens for Evaluatien 14-41 14-7 Reactor suilding Structure Data for Analysis of Time-Dependent Reactor Building Pressure 14 42 14-8 Summary of Reactor Building Pressure Analysis for Four Reactor Buildi:6 E=ergency Cooling Units 14-45 14-9 Sensitivity Analysis Shcwing the Effect of Param-eters on the Two-Hour Iodine Dose at the Exclusion Area South / Foll:nric6 an MHA 14-53 O i a-u 0001 285

,

! I

LIST OF FIGURES, O (At rear of Section)

Figure No. Title 14-1 Startup Accident from 10-9 Rated Power Using a 1.2% a k/k Rod Group; High Pressure Reactor Trip Is Actuated.

14-2 Startup Accident from 10-9 Rated Power Using All Rods with a Worth of 9 5% a k/k; High Flux Reactor Trip Is Actuated.

14-3 Peak Themal Power versus Rod Withdrawal Rate for a Startup Accident frem 10-9 Rated Pover.

14-4 Peak Neutron Power versus Rod Withdrawal Rate for a Startup Accident from 10-9 Rated Power.

14-5 Peak Themal Power versus Trip Delay Ti=e for a Startup Acci-dent Using a 1.2% ak/k Rod Group at 5 8 x 10-5 (ak/k)/see from 10-9 Bated Power.

14-6 Peak Themal Power versus Doppler Coefficient for a Startup Accident Uting a 1.2% ak/k Rod Group at 5 8 x 10-5 (a k/k)/

see from 10-9 Rated Power.

'

14-7 Peak Thermal Power versus Trip Dela

'

'

dent Using All Rods at 5 8 x 10-4 (y Ti=e for a Startup Acci-a k/k)/see from 1 Fever.

14-8 Peak The=al Power versus Doppler Coefficient for a Startuu )

Accident Using All Rods at 5 8 x 10-4 ( a k/k)/see from 10-9 l Rated Power.

14-9 Peak Pressure versus Trip Dela Using All Rods at 5 8 x 10-4 (yak/k)/see Time forfrom a Startup Accident 10-9 Rated Power.

l 14-10 Peak Pressure versus Tripped Rod Worth for a Startup Accident 1 Using All Rods at 5.8 x 10-4 ( ak/k)/see frem 10-9 Rated Power.

I 14-11 Peak Pressure versus Doppler Coefficient for a Startup Acci-dent Using All Rods at 5 8 x 10-4 (ak/k)/see from 10-9 Rated Power.

14-12 Peak Pressure versus Moderator Caef dent Using All Rods at 5 8 x 10-N a(ficient forfrom k/k)/sec a Startup Acci-10-9 Rated Power.

l 14-13 Rod Withdrawal Accident from Rated Pcver Using a 1.2% ak/k Rod Group at 5 8 x 10-5 (a k/k)/sec; Eigh Pressure Reactor Trip Is Actuated.

0001 286 3 111

i FIGURES (Cont'd)

Figure No. Title 14-14 Peak Pressure versus Rod Withdrawal Rate for a Rod Withdrawal Accident fro: Rated Power ik-14-a Max 1=u= Neutron and The=al Power for an All-Rod Withdrawal Ac-cident fro = Various Initial Power Levels  !

l lk-lk-b Peak Fuel Temperature in Average Pin and Hot Spot for an All-Rod .l Withdrawal Accident from Various Initial Power Levels l 14-15 Peak Pressure versus Trip Delay Ti=e for a Rod Withdrawal Acci-dent fro = Rated Power Using a 1.2% ak/k Rod Group; High Pressure

- Reactor Trip is Actuated 14-16 Peak Pressure versus Doppler Coefficient for a Rod Withdrawal 1 Accident fro Rated Power Using a 1.2% ak/k Rod Group l l

14-17 Per Cent Reactor Coolant Flow as a Function of T1=e after Loss of Pu=p Power ik-13 Mini =u= DNER Which Occurs during the Coastdown for Various Initic Power Levels 14-19 Reactor syste= Cooling Rate for 4 in.2 stea= Line areak 14-20 Per Cent Core Experiencing DNB as a Function of Ejected Control Rod Worth at Ulti= ate Power 14-21 7r-H2O Reaction as a Function of Ejected Control Rod Worth at Ultimate Power 14-22 Reactor Neutron Power Variation with Ejet.ced Control Rod Worth 14-23 Reactor Tier =al Power as a Function of Ejected Control Rod Worth 14-24 Enthalpy Increase to Hottest Fuel Rod versus Ejected Control Rod Worth 14-25 The Effect on Reactor Neutron Power of Varying the Doppler Coefficient - Rod Ejection at 10-9 Ultimate Power ik-26 The Effect on Reactor Neutron Power of Varying the Moderator Coefficient - Rod Ejection at 10-9 Ulti= ate Power 14-27 The Effect on Reactor The=al Power of Varying the Doppler i Coefficient - Rod Ejection at 10-9 Ultimate Power l

'

1k-28 The Effect on Reactor The=al Pever of Varying the Moderator l

Coefficient - Rod Ejection at 10-9 Ultimate Power ik-iv (Revised 7-21-67) l

.

FIGURES (Cont'd)

Figure No. Title 1L-29 Reactor Ther=al Power versus Trip Delay Time - Rod Ejection at Ulti= ate Power 1k-30 Enthalpy Increase to the Hottest Fuel Red versus Trip Delay Ti=e

- Rod Ejection 1h-31 LOFT Semiscale Slowdown Test No. 546 - Vessel Pressure versus Time 14-32 Predicted Per Cent Mass Remaining versus Ti=e - QFT Test No. Sk6 14-33 Core Flow versus Time for a 36 in. ID Double-aded Pipe Rupture ik-3L Hot Channel Clad Surface Heat Transfer Coefficient after DN3 versus Time for a 36 in. ID Double-hded Pipe Rupture IL-35 Reactor Vessel Water Volu=e versus Time for 36 in. ID, Double-2ded Pipe Rupture for 600 psig Core Floeding Tank Operating Pressure 14-36 Reactor Vessel Water Volu=e versus Time for 36 in. ID, Double-Ended Pipe Rupture for 400 psig and 1,000 psig Core Flooding Tank Operating Pressures

'

'

O 1L-37 Core Flooding Tank Analysis; Maximu= Clad Temperature versus Ti=e to Quench for a 36 in. ID, Double-Ended Pipe Rupture 14-38 Max 1=u= Hot Spot Clad Te=perature versus Maxi =u= Heat Transfer Coefficient after DNB for a 36 in. ID, Double- hded Pipe Rupture 14-39 Hot Spot Clad Temperature versus T1=e for 36 in. ID, Double-Ended Pipe Rupture and Variable quench Coefficient lL-LO Hot Spot Clad Te=perature versus Ti=e for 36 in. ID, Double-2ded Pipe Rupture and Variable Sink Te=perature 14-L1 Mass Released to Reactor Building for the Spectru= of Hot Leg Ruptures 14-k2 Reactor Coolant Average Pressure for the Spectru= of Hot Leg l Ruptures '

14-k3 Reactor suilding Pressure versus Ti=e - 36 in. ID Double-Eded Rupture ik-43-a Reactor Building Pressure versus Time for a 36 in. ID Double- I Eded Rupture with and without Cooling of the Recirculated Spray Water

• 1 L

14-v (Revised 7-21-67)

J 0

l

.- _. _ _

FIGURES (Cont'd)

,

Figure No. Title ik-k3-b Reactor Building At=osphere and Su=p Coolant Temperatures Follow-ing a 36 in. ID Double-Ended Rupture ik-kk Reactor Building Pressure Versus T1=e after Rupture - 8.5 ft2 Rupture ik-k5 Reactor Building Pressure versus Ti=e after Rupture - 3 0 fta Rupture -

ik L(2 Reactor Building Pressure versus T1=e after Rupture - 2.0 ft2 Rupture ik-k? Reactor Building Pressure versus Ti=e after Rupture - 1.0 ft2 Rupture ik-k8 Reactor Building Pressure versus Ti=e after Rupture - 0.k ft2 Rupture ih-k9 Reactor Building Energy Inventory for 36 in. ID Double-Ended Ruptt re 14-50 Reactor Building Energy Inventory for 3 0 ft2 Rupture 1k-51 Reactor Building Vapor and Su=p Te=peratures for 36 in. ID Double-Ended Rupture as a Function of Ti=e af ter the Rupture 14-52 Reactor Building Vapor and Su=p Temperatures as a Function of T1=e after Rupture - 3 0 fta Rupture ik-53 Criterion 17 Case for 36 in. ID, Double-Ended Rupture lk-5h Reactor Building Zr Reaction Capability for 55 psig Design Pressure ik-55 Thyroid Dose frem loss-of-Coolant Accident Hour, 2k-Ecur, and 30-Day Doses ik-56 Maximum Hypothetical Accident Thyroid Dose Assu=ing 100 Per Cent '

Core Meltdown ik-57 Integrated Direct Dose Follewing MEA vith 3-1/2 Foot Reactor Suilding '4all Thickness O

14-vi (Revised 7-21-67)

p 14 SAFETY ANALYSIS v

14.1 CCRE AND CCOLANT BOUNDARY FROTECTION ANALYSIS l

14.1.1 ABNCRMALITIES In previous sections of this report both no=al and abnc=al operations l of the various systems and components have been discussed. This section summarizes and further explores abnc=EJ.ities that are either inherently )

teminated or require the normal protective systems to operate to =ain- 't tain integrity of the fuel and/or the reactor coolant system. These ab- l normali:1es have been evaluated for a rated power of 2,452 MWt. 'wtenever  !

a fission product release to the environment occurs, the release is based '

upon the fission product inventory associated with the ulti= ate reactor I core power level of 2,535 MWt. Fission product dispersion in the atmo-

!

sphere is assumed to occur as predicted by the dispersion =odels devel- I oped in 2 3 Table 14-1 su=marizes the potential abnor=alities studied. .

Table 14-1 Abnomalities Affecting Core and Coolant Boundary Event Cause Effect Unccmpensated Oper- Fuel depletion Reduction in reactor system O atine aeactiv1er rhanges or x =c= eu11d-up.

averaee te=re=ature. Ausc=atic reactor trip if unccmpensated.

No equipment damage or radio-logical hazard.

Startup Accident Uncontrolled Power rise teminated by nega-rod (*) with- tive Doppler effect, reactor drawal. trip frc= short period, high reactor coolant system pres-sure, or overpover. No equip-ment damage or radiological hazard.

Red Withdrawal Acci- Uncontrolled Pcver rise terminated by over-dent at Rated Power red withdrawal. pcVer trip or hi6h pressure trip. No equipment damage or radiolc61 cal hazard.

O

(*)Centrol red, red, and control assembly (CRA) are used interchangeably in this section and elsewhere in the report.

' " '

-

1,_1 0001 290

.

Table 14-1 (Cont'd) i

,

Event Cause Effect l

l Moderator Dilution Equipment mal- Slow change of power terminated i

Accident function or by reactor trip on hi 6h te=per-operator error. ature or pressure. During shut-l down a decrease in shutdown =ar-gin occurs, but criticality do'es not occur. No radiological ha:-

l as.

(

'

Ioss-of-Coolant Flow Mechanical or None. Core protected by reac-

'

electrical tor low-flow trip. No radio-j failure of re- logical hazard, actor coolant pump ( c.) .

Stuck-Out or Stuck- Mechanical or None. Suberiticality can be In or Dropped-In electrical achieved if one rod is stuck-Control Rod failure. out. If stuck-in or dropped-in, continued operation is permitted if effect on power peaking not severe. No rsdio-logical hazard.

Loss of Electric Miscellaneous Possible power reduction or re-Power faults. actor trip dependin6 on condi-tion. Redundancy provided for safe shutdown. Radiological hazard within limits of 10 CFR 20, even for multiple occurrances.

Steam Line Failure Pipe failure. Reactor automatically trips if rupture is large. No damage to reactor system. Integrated doses at exclusion distance are 0.00k rem whole body and 0.88 rem thyroid.

t Steam Generator Tube Tube failure. Reactor automatically trips if l

Failures leakage exceeds no mal =akeup capacity to reactor coolant I system. Integrated doses at exclusion distance are 0.h8 rem whole body and 1.6 x 10-b rem thyroid.

O l

w

-

-

~,

14-2 0001 291

14.1.2 ANALYSIS OF IFFECTS AND CCNSEQUENCES 14.1.2.1 Unecmpensated Operating Reactivity Changes 1k.l.2.1.1 Identification of Cause During normal operation of the reactor, the overall reactivity of the core changes because of fuel depletion and changes in fission product poisen concer tration. These reactivity changes, if left either uncompensated or overcem-pensated, can cause operating limits to be exceeded. In all cases, however, the reactor protective system prevents safety limits frcm being exceeded. No damage cccurs from these conditions. -

14.1.2.1.2 Analysis and Results During normal operation, the automatic reactor control syste.s senses any reac-tivity change in the reactor. Depending on the direction of the reactivity change, the reactor power increases or decreases. Correspondingly, the reac-tor coolant system average temperature increases er decreases, and the auto-

=atic reactor centrol system acts to restore reactor power to the power demanc level and to reestablisa this temperature at its set point. If manual correc-tive action is not taken or if the automatic control system malfunctions, the reactor coolant system average temperature changes te ecmpensate for the reac-tivity disturbance. Table 1h-2 su=marizes these disturbances.

Table lk-2 Uncompensated Reactivity Disturbances Maximum Rate of Average Reactivity Bate, Temperature Change Cause (ak/k)/sec (Uncorrected), F/see Fuel Depletion -6 x 10-9 -0.0006 Xenon Buildup -3 x 10-8 -0.003 These results are based on +6 x 10-3 (ak/k)/F moderator coefficient and -1.lh x 10-3 (ak/k)/F Doppler coefficient. The ncminal value of +6 x 10-5 (Ak/k)/F is representative of the moderator coefficient at the beginning of core life for an equilibrium cycle. This value is also valid at 30L for the first cycle after 15 days. A higher value (+10 x 10-5 (ak/k)/F] exists at the start of the first core cycle. However, the effect of this slightly higher value has been shown to be of minor importance by the evaluation of the sensitivity of the reactor to moderator coefficient variations. These reactivity changes are extremely slow and allow the operator to detect and ecmpensate for the change.

'

O

,[t

'

'

,

14-3 (Revised l'.-6-67) 0001 292 .

.

.h.l.2.2 Startuo Accident

.h.l.2.2.1 Identification of Cause 7ue objective of a normal startup is to bring a suberitical reactor to the crit-

. cal or slightly supercritical condition, and then to increase pcwer in a con-

. rolled manner until the desired pcVer level and system operating temperatures tre obtained. During a startup, an uncontrolled reactivity addition could

ause a nuclear excursion. This excursion is. terminated by the strong negative soppler effect if no other protective action operates.

te following design provisions minimize the possibility of inadvertent contin-tous rod withdrawal and limit the potential power excursion:

a. The control system is designed so that cnly one control rod group can be withdrawn at a time, except that there is a 25 per cent over-lap in travel between two successive red grcups. This overlap oc-curs at the minimum worth for each group since one group is at the end of travel and the other is at the beginning of travel. The =ax-imum vorth of any single control red group is 1.2% ak/k when the reactor is critical as specified in 7.2.2.1 3

b. Control red withdrawal rate is limited to 25 in./=in.

c. A short-period withdraval stop and alarm are provided in the source range.

d. A short-period withdrawal stop, alarm, and trip are provided in the inter =ediate range, h

e. A high flux level and a high pressure trip are provided in the pcVer range.

he reactor protective system is designed to limit (a) the reactor thermal over to 11h per cent of rated power to prevent fuel da= age, and (b ) the re ac-or coolant system pressure to 2,515 psia.

i i

1 k.l.2.2.2 Methods of Analysis 1

n analog medel of the reactor core and coolant system was used to determine he characteristics of this accident. This analog model used full reactor

'

colant flow, but no heat transfer out of the syste= and no sprays in the pres-urizer. The rated-power Doppler coefficient [-1.lh x 10-5 (ak/k)/F } vas used 1though the Doppler is much larger than this for the principal part of the l

'

ransient. The rods were assumed to be moving along the steepest part of the od-worth vs red-travel curve. A reactor trip on short period was not incer-orated in the analysis. The nominal values of the principal para =eters used ere: 0 3 see trip delay, +6 x 10-5 (ak/k)/F moderator coefficient, and -1.lk 10-5 (ak/k)/F Doppler ccefficient. The total vorth of all the centrol reds nserted into the reactor core following any trip is 8.h% ak/k without a stuck entrol rod, or 5.L% ak/k (the nominal case in this study) with a stuck rod.

'

-

... II>

> b

'

ik-k (Revised 11-6-67)

1h.1.2.2.3 Results of Analysis

'

Figure ih-1 shows the results of withdrawing the maximu= vorth control rod group at a red speed of 25 in./=in frc= 1 per cent suberitical. This group is worth a =axi=um value of 1.2% ak/k. *his rod velocity and verth result in a

=axi=u= reactivity addition rate of 5.8 x 10-5 (ak/k)/sec. The Ocppler effect begins to sicv the neutron pcVer(*) rise, but the heat to the coolant increase:

the pressure past the trip point, and the transient is ter=inated by the high pressure trip.

Figure lk-2 shows the results of withdrawing all 69 centrol red assemblies (with a total vorth of 10.0% ak/k) at the maximum speed frc= 1 per cent sub-critical. This results in a maximum reactivity addition rate of 5.8 x 10 h (ak/k)/sec. About 15.3 sec after passing through criticality, the neutron power peaks at 1kT per cent, where the power rise is stopped by the negative Doppler efrat. The high neutron flux trip takes effect 0.25 see after the peak power is reached and ter=inates the transient. The peak ther=al heat flu:

is only 16 per cent of the rated pcuer heat flux.

A sensitivity analysis was perfor=ed en both of these startup accidents to de-termine the effect of varying several key para =eters. Figures ik-3 through lk-6 show typical results for the single group, 1.2% ak/k startup accident.

Figures ih-3 and ih-k show the effect of varying the reactivity addition cte en the peak ther=al pcuer and peak neutron power. This reactivity rate was varied frc= one order of magnitude belev the nc=inal single rod group case (1.2% Ak/k) to = ore than an order of =agnitude above the rate that represents all rods (10.0% ak/k) being withdrawn at once. The sicver rates - up to about

%) 0 5 x 10-3 (ak/k)/sec - vill result in the pressure trip being actuated, whereas only the very fast rates actuate the nigh neutron flux level trip.

Figures lk-5 and lb-6 show the peak ther=al power variation as a function of a vide range of trip delay times and Doppler coefficients for the 1.2% ak/k rod group. Only a small change in power is noted. Figures 1h-7 and lh-8 are the corresponding results from the withdrawal of all rods (10.0% ak/k). Since this transient inserts reactivity an order of =agnitude faster than does the single control rod group case, there is considerably = ore variation in the peal thermal power over these vide ranges. At high values of the Doppler coeffi-cient, the neutron power rise is virtually stopped before reaching the high flux trip level. Reactor pcVer generation continues until sufficient energy 1:

transferred to the reactor cool' ant to initiate a high pressure trip. This re-sults in a higher peak ther=al power.

Figures lh-9 through 1h-12 show the peak pressure response to variations in several key parameters for the case where all reds are withdrawn. It is seen that the safety valve is opened when these parameters are changed considerably frc= the nc=inal values, except in the case of the =oderator coefficient which has little effect because of the short duration of the transient. Again for a high Doppler coefficient, the high pressure trip is relied upon.

~he neutren power is defined as the total sensi'cle energy release frc= fissi p ,

th-5 mevisee u-e-sn 0001 294.

.

None of shese postulated startup accidents, except for reactivity addition rates greater than 2 x 10-3 (ak/k)/sec, which is three ti=es greater than for withdrawal of all rods at once, causes a thermal power peak in excess of LO per cent rated power or a nominal fuel red average te=perature greater t..an 1,715 F. The nominal 1.2% ak/k rod group withdrawal causes a peak pressure of 2,515 psia, the safety valve set point. The capacity of the safety valves is adequate to handle the maximum rate of coolant expansion resulting from this startup accident. The 10.0% ak/k withdrawal - using all 69 rods - causes a l3 peak pressure of only 2,h65 psia because the flux trip is actuated prior to the pressure trip.

It is concluded that the reactor is completely protected against any startup accident involving the withdrawal of any or all controi rods, since in no case does the thermal power approach 11b per cent, and the peak pressure never ex-ceeds 2,515 psia.

1h.1.2.3 Rod Withdrawal Accident From Rated Power Oeeration 1h.l.2.3.1 Identification of Cause A rod withdrawal accident presupposes an operator error or equipment failure vnich results in accidental withdrawal of a control rod group while the reac-tor is at rated power. As a result of this assumed accident, the power level increases; the reactor coolant and fuel rod temperatures increase; and if the withdrawal is not terminated by the operator or protection system, core dam-age would eventually occur.

The following provisions are made in the design to indicate and terminate this accident:

a. High reactor outlet coolant temperature alarms.

b. High reactor coolant system pressure alarms.

c. High pressurizer level alarms.

l3

d. High reactor outlet coolant temperature trip.

e. High reactor coolant system pressure trip.

f. High power level trip.

Ik.l.2.3.2 Methods of Analysis An analog computer model was used to determine the characteristics of this ac-cident. A complete kinetics =odel, pressure =odel, average fuel rod model, steam demand =odel with turbine coastdown to 15 per cent of rated load, cool-ant transport model, and a simulation of the instru=entation for pressure and flux trip were included. The initial conditions were nor=al full p'ower oper-ation without aute:datic control. Only the moderator and Doppler coefficient

of reactivity were used as feedback. The nominal values used for the main pa-I rameters were 0.3 see trip delay time, -1.lh x 10-5 (ak/k)/F Doppler coeffi-cient, +6 x 10-5 (ak/k)/F mederator coefficient, 25 in./ min control rod speed, h snd 1.2", ak/k control rod group vor.h. The total vorth in all the control rods l

'

w4 '. .

'

.

0001 295

'

.

ih-6 (Revised 11-6-67)

_ __ __ __ . - --. __

inserted into the reactor core following any trip is 8.k% ak/k without a stuck control rod, or 5.h% ak/k (the nominal value used) with a stuck red.

The foregoing red speed and group rod vorth give a maximum reactivity addition rate of 5.8 x 10-5 (ak/k)/sec, which is the nominal case. The reactor protec-tion system is designed to limit (a) the reactor power to lik per cent of rate power to prevent fuel damage, and (b) the coolant system pressure to 2,515 psi to prevent reactor coolant system damage.

14.1.2.3.3 Results of Analysis Figure 14-13 shows the results of the nominal red withdrawal from rated power using the 1.2% k/k rod group at 5.8 x 10-5 (ak/k)/sec. The transient is ter-minated by a high pressure trip, and reactor power is limited to 108 per cent, much less than the design overpower of 11h per cent of rated power. The changes in the parameters are all quite small, e.g., 5 F average reactor cool-ant temperature rise and 200 psi system pressure change.

A sensitivity analysis of important parameters was performed around this nomi- 1 nal case, and the resultant reactor coolant system pressure responses are shov  :

in Figures 1k-14 through lk-16. l i

l Figure ik-lk shows the pressure variation for a very vide range of rod with-  !

drawal rates - more than an order of magnitude smaller and greater than the t nominal case. For the very rapid rates, the neutron flux level trip is actu- l ated. This is the primary protective device for the reactor core; it also pro-tects the syste's against high pressure during fast red withdrawal accidents.

The high pressure trip is relied upon for the slower transients. In no case does the thermal power exceed 108 per cent rated power.

An analysis has been perfor=ed extending the evaluation of the rod withdrawal accident for various fractional initial power levels up to rated power. This evaluation has been performed assuming simulated withdrawal of all 69 control rods giving a maximum reactivity addition of rate of 5.8 x 10-h (ak/k)/sec.

This rate is a factor of ten higher than used in the cases evaluated at rated power.

ik-1k-b. The results of this analysis are shown in Figures ik-lk-a and Figure As seen in Figure ik-lk-a the peak ther=al power occurs for the rated power cas and is well below the maximum design power of 11L per cent. The peak neutron power for all cases is approximately 117 per cent of rated power and repre-sents a slight overshoot above the trip level of lik per cent. Figure ik-lk-b shows that the maximum fuel temperature reached in the average rod and the hot spot are well below melting. Even in the most severe case at rated power, the average fuel temperature only increases by 26 F. It is therefore readily con-cluded that no fuel damage would result from simultaneous all-rod withdrawal from any initial power level.

1 1

O l hf

• r.,,. , ., , c ,

\\s ..o*;. lk-7 (Revised 11-6-67) 2 I

.

gures 14-15 and 14-16 show the pressure response to variations La the trip lay ti=e and Doppler coefficient. For the higher values of the Doppler coef-cient, the pressure trip is always actuated, and, therefore, the pressure vels off.

is analysis shows that the high pressure trip and the high flux level trip equately protect the reactor against any rod withdraval accident from rated ver.

.l.2.k Moderator Dilution Accident

.l.2.4.1 Identification of Cause e reactor utilizes boric acid in the reactor coolant to control excess reac-vity. The boron content of the reactor coolant is periodically reduced to 7pensate for fuel burnup. The dilution water is supplied to the reactor cool-system by the =akeup and purification syste=. This (l>

i l

l l

l l

O

.*

h k! ' -

• lk-Ta (Revised 7-21-67) 0001 i!97

,

!

-

syste= is designed with several' interlocks and ala=s to prevent i= proper cperation. These are as follows:

O .a. Flev of dilution water to the =akeup tank =ust be initiated by The dilution water addition valve can be opened

,

the operatcr.

I only when the centrol rods have been withdrawn to the preset position (95 per cent) and the timing device to 11=1: the inte- l grated flev has been set. Dilution water is added at flow rates up to 70 gym.

b. Flev of dilution water is autc=atically stepped when either -'

the flow has integrated to a preset value or when che rods have been inserted to a preset position (at about 75 per cent full stroke).

l

c. A varnir4 light is on whenever dilution is in progress.

The =akeup and purification syste= ner= ally has one pump in operation which supplies up to 70 gp= to the reactor ecolant syste= and the re-quired flew to the reacter coolant pu=p seals. Thus, the total =akeup flow available is li=ited to 70 gpm unless the operator takes action to increase the a= cunt of =akeup flow to the reactor coolant syste=. When the =akeup rate is greater than the maximu= letdown rate of 70 gym, the net water =akeup will cause the pressurizer level control to close the

=akeup valves.

' The ac=inal =cderator dilution event considered is the pu= ping of water with :ero boren concentration frc= the =akeup tank to the reactor coolant syste= by the =akeup pu=p.

It is also possible, however, to have a slightly higher flev rate during transients when the syste= pressure is lover than the nc=inal value and the pressurizer level is belev no mal. This flov =ight be as high as 100 gpm.

In addition, with a ec=bination cf =ultiple valve failures or =alopera-tiens, plus = ore than one =akeup pu=p cperating and reduced reactor cool-ant system pressure, the resulting inflow rate can be as high as 500 spm.

This constitutes the =axi=u= dilution accident. A reacter trip would teminate unborated water addition to the =akeup tank, and total flew into the ecolant syste= vould be teminated by a h16h pressuriter level.

The criteria of reactor protection for this accident are

a. De reactor pcVer will be limited to less than the design over-pcVer of 114 per cent rated power to prevent fuel damage,

b. De reactor protection syste= will 11=1 the reactor ecolant system pressure to less than the syste= design pressure of 2,500 psig.

g c. S e reacter =ini=u= suberiticality =ar61n of 1% Ak/k vill be Q =aintained.

l

• 0001 298 lLS (Revised 1-8-c8)

__

l l

!

Ik.l.2.k.2 Analysis and Results  !

i The reactor is assumed to be operating at rated power with the maximum initial Os boron concentration (1,650 ppm) in the reactor coolant system. The dilution I

water is uniformly distributed throughout the reactor coolant volume. Uniform )

distribution results from a discharge rate of 70 - 500 gym into a reactor cool-ant flow of 88,000 gpm. A change in concentration of 110 ppm produces a 1%

ak/k reactivity change. The effects of these three dilution rates on the rese-tor are as follows:

l Average Reactor Dilutibn Water Reactivity Rate, Coolant System Flev. gom (ak/k)/see Temo. Change , F/see 70 +3.0 x 10-6 0.3

-

100 +h.h x 10 0.3 500 +2.2 x 10 -5 0.5 The fastest rate of dilution can be handled by the automatic control system, which would insert rods to maintain the power level and reactor coolant system temperature. If an interlock failure occurred while the reactor was under manual control, these reactivity additions would cause a high reactor coolant temperature trip or a high pressure trip. In any event the thermal power vill

(,) not exceed 114 per cent rated power, and the system pressure vill not exceed the design pressure of 2,500 psig. Therefore aoderator dilution accidents will not cause any damage to the reactor system.

During refueling or maintenance operations when the reactor closure head has been removed, the sources of dilution water makeup to the makeup tank--and therefore to the reactor coolant system--are locked closed, and the makeup pumps are not operating. At the beginning of core life when the boron concen-tration is highest, the reactor is about 9.5% ak/k suberitical with the maxi-mum worth rod stuck out. To demonstrate the ability of the reactor to accept

=oderator dilution during shutdown, the consequences of accidentally filling the makeup tank with dilution Vater and starting the makeup pumps have been evaluated. The entire water volume from the makeup tank could be pumped into the reactor coolant system (assuming only the coolant in the reactor vessel is diluted), and the reactor vould still be 6.5% ak/k suberitical.

14.1.2.5 Cold Water Accident The absence of individual loop isolation valves eliminates the potential source of cold water in the reactor coolant system. Therefore, this accident is not credible in this reactor.

'

i 14-9 (Revised 11-6-67) 0001 299

!

.

14.1.2.6 Loss-of-Coolant Flov i 14.1.2.6.1 Identification of Cause A reduction in the reactor coolant flow rate occurs if one or = ore of the reactor coolant pu=ps should fail. A pu= ping failure can occur from me-chanical failures or from a loss of electrical power. With four indepen-dent pu=ps available, a =echanical failure in one pu=p vill not affect op-erstion of the others.

Each reactor coolant pu=p receives electrical power from one of the elec-trically separate busses of the 6,900 volt system discussed in 8.2.2 3 A singular loss of either outside power or the Station generator vill not cause a loss of electrical power to the pu=ps. Faults in an individual pu=p motor or its power supply could cause a reduction in flov, but a complete loss of flow is extremely unlikely.

In spite of the icv probability of a complete loss of power to all reac-tor coolsnt pumps, the nuclear unit has been designed so that such a fail-ure would not lead to' core da= age.

The reactor protection criterion for loss-of-coolant flow conditions start-in6 at rated power is that the reactor core vill not reach a Departure from Nucleate Boiling Ratio (DNER) smaller than the DNER in the hot channel at the stesdy state design overpower. This corresponds to a DNER of 1 38 at lik per cent rated power (Table 3-1) .

14.1.2.6.2 Methods of Analysis h

The loss-of-coolant-flov accident is analyced by a combination of analog and digital computer progrsms. Analog s1=ulation is used to determine the reactor flow rate following loss of pu= ping power. Reactor power, coolant flow, and inlet te=perature are input data to the digital program which determines the core ther=al characteristics during the flow coastdown.

The analc6 model used to determine the neutron power following reactor trip includes six delayed neutron groups, control rod vorth and rod insertion characteristics, and trip delay time. The analog model used to determine i flow coastdown characteristics includes description of flow-pressure drop relations in the reactor coolant loop. Pump flow characteris-ics are de-termined from manufacturers' cone =aps. Flow-speed, flow-torque, and flow-head relationships are solved by affinity lavs.

A transient, thermal-hydraulic, B&W digital computer program is used to compute channel DNER continually during the coastdown transient. System flow, neutron power, fission product decay heat, and core entering enthal-

, py are varied as a function of time. The program =aintains a transient l inventory of stored heat which is detemined from fuel and clad te=pera-l tures beginning with the initial steady state conditions. The transient core pressure drop is determined for avera6e channel conditions. .The representative hot channel flows and correspondin6 DNER are obtained by usin6 the average core pressure drop. The hot channel DNER as a function of ti=e is cc= pared with the design DNER at maximum overpower to deter-mine the degree of heat transfer =argin.

op. '

.-

• 0001 300 14-10

. - . - - - _ _ _ _ _ _ _ _

The loss-of-coolant-flow analysis has been carried out in the power range 3 between 102 and 114 per cent rated power. Conditions utiliced in the anal-ysis are as follows:

a. Initial core inlet te=perature for given power level is assumed to be plus 2 F in error.

b. Initial system pressure is assumed to be minus 65 psi in error.

c. Trip delay time, i.e., ti=e for sensor detection for low flow condition until initial downward =ove=ent of control rod, is 300 milliseconds.

d. The per cent of initial reactor power as a function of time after loss of pumps is as shown in Figure 3-6.

2

e. The pu=p inertia is 70,000 lb-ft ,

14.1.2.6 3 Results of Analysis The results of this analysis show that the reactor can sustain a loss-of-coolant-flow accident without damage to the fuel. The results of the evaluation are presented in Figures 14-17 and 14-18. Figure 14-17 shows the per cent reactor flow as a function of ti=e after loss of all pump power. Figure 14-18 shows the =ini=um DIGR's which occur during the coastdown for various initial power levels. The degree of core protection during coastdown is indicated by comparing the DfGR for the coastdown with h" the design value of 1 38 at 114 per cent rated power. This DIER (138) in the representative hot channel corresponds to a 99 per cent confidence that 99 5 per cent of the core will not experience a departure from nu-cleate boiling under steady state conditions at the design overpower

('3 2 3 1).

Under normal conditions, the maximum indicated power level from which a loss-of-coolant-flow accident could occur is 102 per cent ratea power (as indicated by reactor instru=entation). This power level represents an allowance of plus 2 per cent rated power for transient overshoot. This power level also represents the maximum power demand that will be permitted to the reactor control system. The 102 per cent rated power is an instru-ment-indicated value and is subject to the following maximum errors:

(a) 2 per cent heat balance and (b) =4 per cent nuclear instrumentation.

The true power level could be as hi6h as 108 per cent at 102 per cent indi-cated power. As shown in F16 ure 14-18, however, the D GR at 108 per cent is 1.k2, which is significantly larger than the design DIGR.

The reactor coolant system is capable of providing natural circulation flow after the pumps have stopped. The natural circulation characteristics of the reactor coolant system have been calculated using conservative values for all resistance and form loss factors. No voids are assured to exist in the core or reactur outlet piping. The following tabulation and Figure 9-7 show the hatural circulation flow capability as a function of the decay heat 6eneratien. This material is presented in greater detail in 14.1.2.8.3

-

,.

'

ik-11 (Revised 7-21-67) 0001 301

,

-.

Time After Decay Heat Natural Circulation Flow Required for Loss of Core Power, Core Flow Available, Heat Re= oval, Power, sec  % 4 Full Flow  % Full Flov 0 36 x 102 5 k.1 23 2.2 x 102 3 33 1.2 1.2 x 104 1 1.8 0 36 1 3 x 105 1/2 1.2 0.20 The flows above provide adequate heat transfer for core cooling and de-cay heat removal by the reactor coolant system.

The reactor is protected against reactor coolant pu=p failure (s) by the protective system and the integrated control syste=. The integrated cen-trol system initiates a power reduction on pu=p failure to prevent reac-tor power from exceeding that pe. issible for the available flow. The reactor is tripped if insufficient reactor coolant flow exists for the pcVer level. The operating limits for less than four pu=ps in operation have been presented in 4.3 7 14.1.2 7 Stuck-Out, Stuck-In, or Dropped-In Control Red 14.1.271 Identification of Cause The control rod drives have been described in 3 2.k.3 The results of continuous control rod withdrawal have been analyzed in 14.1.2.2 and 14.1.2 3 In the event that a control red cannot be =oved because of electrical faults or =echanical seizure, localized power peaking and sub-critical margin =ust be considered.

14.1.2 7 2 Analysis and Results Adequate hot suberitical =argin is provided by requiring a suberiticality of 1% A k/k suberitical with the control rod of greatest worth fully with-drawn from the core. The nuclear analysis reported in 3.2.2 de=onstrates that this criterion can be satisfied.

In the event that an un=ovable control red is partially or fully inserted in the core or a single rod is dropped during operation, its location and effect on local power distribution dete. ine whether continued power op-eration is pe:21ssible. The location of a stuck rod in the core vill be studied further to define permissible conditions cf operation. The cri-teria for these studies are (a) operation with a stuck rod will not in-l

' crease the DNB probability above the probability specified for design conditions, and (b) a hot suberitical margin of 1% a k/k vill be =ain-l tained with the stuck rod in its inoperative position and the operating

!

rod of greatest reactivity worth in the full? vithdrawn position.

Af a control rod is dropped into the core during power operation, the l

l same consideration of localized power peaking as for a stuck rod vill apply.

,

l

U Mh 14-12 0001 102

....

,

s

..

, 14.1.2.8 Loss of Electric Power O 14.1.2.8.1 Identification of Cause The Three Mile Island Nuclear Station is designed to withstand the effects of loss of electric load or electric power. Two types of power losses are considered:

a. A " blackout" condition, caused by severe interconnected grid upset.

a

b. A hypothetical condition resulting in a cocplete loss of all station power.

The reactor protection criteria for these conditions are that fuel da= age vill not occur from an excessive power-to-flow ratio and that the reactor coolant system pressure vill not exceed design pressure.

14.1.2.8.2 Results of " Blackout" Conditions Analysis The net effect of a " blackout" condition on the Station would be openin6 of all 230 kv transmission line breakers, thus disconnecting the Station from the entire transmission system. When this occurs, a runback signal on the integrated master controller causes an auto =atic power reduction to 15 per cent power. Other actions that occur are as follows:

/ a. All vital electrical loads, including reactor coolant pu=pe,

\]_/ condenser circulating vater pumps, condensate and condensate booster pumps, and other auxiliary equip =ent, vill continue to obtain power from the Station generator. Feedvater is supplied to the steam generators by steam-driven feed pu=ps,

b. As the electrical load is dropped, the turbine Generator accel-erstes and closes the governor valves and combined inter =ediate valves. The Station frequency will peak at less than the over-speed trip point and decay back to set frequency in LO-50 sec.

c. Following closure of the turbine governor valves and combined intermediate valves, steam pressure increases to the turbine bypass valve set point and may increase to the steam system safety valve set point. Steam is relieved to the condenser and to the at=osphere. Stes= ventin6 to the atmosphere occurs for about 2 min. following blackout from 100 per cent rated power until the turbine bypass can handle all excess steam gen-erated. The capacity of the =odulating turbine bypass valve is 15 per cent of the valves vide open (VWO) steam flow, and that of the safety valves is 100 per cent of VWO steam flow. The first safety valve banks are set at 1,050 psig with additional banks set at pressures up to 1,104 psig (5 per cent above de-sign pressure as allowed by code). Stea= venting per=1ts en-ergy removal fro = the reactor coolant system to prevent a high I

() pressure reactor trip. The initial power runback is to 15 per cent power which is greater than the Station auxiliary load.

This allows sufficient steam flow for regulatins turbine speed

s

. %'

-

n

,

,u-13 0001 303

!

s.

control. Dcess power above Station auxiliarj load is rejected by the turbine bypass valve to the concenser.

g

d. During the chort interial while the turbine speed is high, the vital electrical loads connected to the Station generator vill undergo speed increase in proportion to the generator frequency increase. All motors and electrical 6 ear so connected are de-signed for the increased frequency.

e. After the turbine generator has been stabili:cd at auxiliarf load and set frequency, the Station operator =ay reduce reactor power to the auxiliarf load as desired.

The blackout accident does not produce any fuel da= age or exces-sive pressures on the reactor coolant system. There is no re-sultant radiological hacard to Station operating personnel or to

• he public frc= this accident, since only secondary cyste= stea=

.

is discharged to the at=osphere.

Nuclear unit operation with failed fuel and steam generator tube leakage is shown to be safe by the analysis presented in 11.1.2 5 2 and 14.1.2.10.

For the same conditions, the steam relief acco=panying a blackout accident would not change the whole body dose. The whole body doce is pri=arily due to the release of Xe and Kr. Release of these gases is not increased by the stesm relief because even without relief, all of these gases are re-leased to the at=osphere through the condenser vacuu= pu=p exhaust. The rate of release of iodine during tne approx 1=ately 2 =in. of relief would be increased by al=ost a factor of 10k, because the iodine is released di-rectly to the at=osphere ratner than througn the condenser and Station vent. However, tne quantity released during this snort time is s=all, and it would be less than 0.05 MFC at the 2,000 ft exclusion distance.

1k.1.2.3 3 Analysis Results of Cc=plete Loss of All Station Power The second power loss considered is the hypothetical case where all Sta-tion power except the Station batterf is lost. The sequence of events and the evaluation of consequences relative to this accident are given belev:

a. A loss of power results in gravity insertion of the centrol rods.

l

b. The stea= generator safety valves actuate after the turbine trips and prevent excessive te=peratures and pressures in the reactor coolant system.

c. '"he reacter coolant syste= ficv decays without fuel da= age cc-curring. Oecay heat re=cval after coastdevn of the reacter ecolant pumps is provided by the natural circulation character-istics of the system. This capability is discussed in the loss-cf-coclant-flev evaluation (lk.l.2.c).

O 0001 304

b. b;i ( ik-lk (Revised T-21-c7)

. _. .

.

d. Two turbine-driven e=ergency feedvater pu=ps are provided to sup-ply feedvater any ti=e the main feed pu=ps cannot operate. The pJ emergency feed pump takes suction from the condenser hotwell and the condensate stors6e. The emergency pu=p supplies feedvater to the steam generators. The emergency feed pu=p is driven by steam from either or both steam generatore.

The controls and aux 111 arf systema for the emergency feed pump operate on d-c power from the' Station batter /.

A recirculation line from the emergency pu=p discharge back to -

the condenser is provided to permit periodic testing.

e. The condenser hotvell and the condensate storage tank provide coolin6 vater in the unlikely event that all power is lost.

The nomal condenser hotwell inventor / is 165,000 gal, and the minimum condensate stora6e tank inventory is 200,000 gal. This total inventor / of 365,000 gal provides sufrietent vater for decay heat cooling (assu= ins infinite irradiation at 2.535 MWt) for a period in excess of two days.

The features described above permit decay heat cooling of the nuclear unit for an extended period of time followin8 a complete loss of electric power.

The foregoing evaluation de=enstrates the design features incorporated in the design to sustain loss of power conditions with just the Station bat-ter/ to operate system controls. I:=ediate operation of the emergency feedvater pu=p is not of critical nature. The reactor can sustain a com-plete electric power loss without emergency cooling for about 25 min. be-fore the steam volume in the pressurizer is filled with reactor coolant.

These 25 min. are derived as follows:

a. Steam generators evaporate to dr/ ness 10 cir.

b. Pressuriser safety valves open 5

c. Pressuriser fills with water 10 25 min.

Beyond this time reactor coolant vill boil off, and an additional 90 =in.

vill have elapsed before the bo11off vill start to uncover the core. The emergency feedvater pu=ps can be actuated within this period of time. Ac-cordingly, core protection is insured for the unlikely ecndition of total loss of Station electric power.

14.1.2 9 Steam Line Failure 14.1.2 9 1 Identificatica of cause Analyses have been performed to determine the effects and consequences of loss of secondarf coolant due to failures in the steam lines between the stear 6enerators and the turbine.

The criteria for Station protection and the release of fission products to the environment are as follows:

ye +

. *> -

.

"il.

a-15 0001 305

!

l l

a. The reactor shall trip and remain suberitical *1thout boron l addition until a controlled rate of system cooldown can be effected . .

b. The potential environ = ental consequences from radioactivity in the secondary coolant system shall not exceed those specified by 10 CFF 20.

14.1.2 9 2 Analysis and Results The rate of Icactor system cooling following a steam line break accident is a function of the area of the failure and the steam Generator water inventory available for cooling. Tne steam generator inventory increases with pcVer level. The inventory at rated power is L6,000 lb and decreases linearly to 20,000 lb at 15 per cent power. The steam line break acci-dent analysis is performed at rated power in order to determine max 1=um cooling and inventory release effects.

The inmediate effect of any steam line creak accident is a reduction in

'

steam pressure and a reduction in steam flow to the turbine. These effects initially cause the reactor control system to act to restore steam pressure and load generation.

A steam line rupture of a small area causes a relatively slow decrease in steam pressure. This places a de=and on the control system for increased feedvater flow. In addition, the turbine control valves vill open to main-tain power generation. Iccreased feedvater flev causes the avera6e reactor coolant temperature to decrease, and the resulting temperature error calls lll e for control rod withdrawal. The limiting action in this condition is the 102 per cent limit on power demand to the rod drive control system. If the l1 moderator temperature coefficient of reactivity is small or slightly posi-tive, the reactor power vill decrease when the centrol system reaches the power demand limit because of continuing temperature decrease. The reac-tor vill then trip on icv reactor coolant Syste= perssure. A reactor trip will initiate a reduction in tne feedvater flow to the stes= generators.

When the =oderator temperature coefficient is negative,, the reactor power vill tend to increase with detressing average coolant te.~pe rature . This vill cause control rod insertion to limit reactor power to 102 per cent. 1 With power limited at 102 per cent, additional cooling causes a reduction in reactor coolant pressure, s=d the reacter trips on low reactor coolant pressure. Turbine trip occurs when the reactor trips. Upon turbine trip the unaffected steam line is isolated by the stes= line isolation valves l as shown in F16 ure 10-1. The unit with the ru'tured; stea= line continues l to blev down to the atmosphert.

1 The maxi =um ecoldown of the reacter coolant system vould be that resultin6

,

from the blevdown from ene steam generater. A typical cooling rate fol-loving reactor trip for a a in.2 steam line rupture is shown in Figure

) 14-19 The, tabulatien below lists the approximate time required to blev down the h cenients of the steam generator with a ruptured steam =ain.

!

0001 306 i

i

[h , 5 5'!' li-16 (Revised T-21-67)

. k Leak Area, in.2 31ovdown Ti=e, see o u 32 ee0 110*

u8 27 A steam line failure of large area results in high stess flow with result-ing rapid pressure decrease in the reactor coolant system and steam sys-tem. The reactor trips on low reactor coolant system pressure or high flux. Reactor trip causes turbine trip and reduction in feedvater flow to decay heat level. The turbine trip and the two steam line isolation valves isolate the stess lines and prevent blevdown of the other steam generator whose secondary side does not have a rupture. The steam generators are de-signed to maintain reactor system integrity upon loss-of-secondary-side pressure. Therefore, this accident will not lead to a reactor coolant sys-tem failure.

Assuming the blevdown from one steam generator results from a secondary steam system rupture, -the maximum cooling rate during this accident occurs during the first 10 see after the break. The msximum cooling rate is ap-proximately 3 F/see and a low pressure or high flux trip occurs. The net cooldown of the reactor coolant system, assuming total blevdown of one steam generator and accounting for transfer of core stored heat and decay heat, is less than 50 F. This results in an average coolant temperature of 530 F which is about 10 F lover than the normal ::ero power average coolant te=perature.

O The minimum shutdown margin at 540 F vith the most reactive rod stuck out is 2.9% Ak/k. The reduction in reactivity shutdown margin associated with coolin6 the moderator temperature 10 F below its normal shutdown temperature of 5ho F vould be 0.30% Ak/k. Using the =aximum value for the moderator temperature coefficient (-3 0 x 10-4 ak/k/F), *,he shutdown margin at 530 F would be 2.6% Ak/k, which is adequate to prevent return to criticality.

In addition, high pressure injection can be actuated during the cooldown period following a large area steam line failure. This system supplies borated water to the reactor coolant system to increase the shutdown mar-gin further. Boron addition to the reactor coolant during the controlled cooling of the system to atmospheric pressure vill prevent criticality at lover temperatures.

The effect of a steam line nipture inside the reactor building has been evaluated by conservatively assuming an instantaneous release to the re-actor buildin6 of the energy associated with this accident. The mass and energy releases per steam generator is this analysis are i

O J

hI f ,

14-17 (Revised 7-21-67) 0001 307

) .

Mass, ib. Energy, Stu x 10-6 Steam Generator k6,000 26.0 1 O

Feedvater Flov 6 see full flow plus coastdown to 7.5",

flow s 16 see 12,800 5.6 Reactor Coolant System Inergy Transferred 17.6 ,

Total 58,800 51.2 Based upon the above, a single steam generator release vould result in approx 1=ately 10 psis pressure rise in the reactor building. This is well below the reactor building des 16 n pressure of 55 psis.

The environ = ental consequences from this accident are calculated by assum-ing that the nuclear unit has been operating with steam 6enerator tube leakage. The reactor coolant activity assu=es prior operation with 1 per cent failed fuel rods. With these assu=ptions, the steam generators con-tain a total of 0.09 equivalent curies of iodine-131. It is further as-su=ed that steam Generator leakage continues for three hours before the nuclear unit can be cocied down and the leakage ter:inated. This addi-tional leaka6e corresponds to 3 4 equivalent curias of iodine-131. The iodine is assu=ed to be released directl-/ to the at=osphere where it =ixes in the wake of the reactor building. With these assu=ptions an integrated dose to the thyroid at the site boundary of 0.88 rem is obtained. The corresponding dose to the whole body during this same ti=e period is 0.00k h rem. The total release of all activity when averaged over a year is 55 per cent of the allovable limits of 10 CFR 20.

14.1.2.10 Steam Generator Tube Failures 14.1.2.10.1 Identification of Accident In the event of a reactor coolant leak to the secondary system, such as a complete severance of a steam generator tube, the activity contained in the coolsnt vould be released to the secondary syste=. Radioactive Bases and so=e of the radioactive iodine would be released to the at=osphere throu6h the condenser air re= oval system.

14.1.2.10.2 Analysis and Results In analyzing the consequences of this failure, the following sequence of events is assu=ed to occur:

a. A double-ended rupture of one stesa generator tube occurs with unrestricted discharge from each end.

b. The initial leak rate, approxi=ately L35 gpm, exceeds the nor-

=al =akeup of 70 3pm to tne reactor coolant syste=, and system pressure decreases. No operator action is assu=ed, and a lov

-

i,g reactor coolant system pressure trip vill occur in about 8 =in.

'

=; m' . 0001 108

\

ik-18 (Revised 7-21-67)

.

c. Following reactor trip, the reactor coolant system pressure continues it decrease until high pressure injection is actuated at a pressure of 1,800 psig. The capacity of the high pressure injection is sufficient to compensate for the le

• ge and main-tains both pressure and volu=e control of the reactor coolant system. Thereafter, the reactor is consertatively assu=ed to be cooled down and depressurized at the normal rate of 100 F per hour.

d. Following reactor trip, the turbine stop valves vill close. ,

Since a reactor coolant to secondary system leak has occurred, steam line pressure vill increase, opening the steam bypass valves to the condenser. Each bypass valve actuates at a lower pressure than do the safety valves. The reactor cool-ant that leaks as a result of the tube failure is condensed in the condenser. Only the fission products that escape from the condensate are released to the atmosphere.

e. The affected steam 6enerator can be isolated by the steam line isolation valve when the reactor coolant system pressure falls below the setpoint of the secondarf system safety valves, i.e.,

1,050 psig. Cooldown continues with the unaffected steam gen-erator until the temperature is reduced to 250 F. Thereafter, cooldown to ambient conditions is continued using the decay heat removal system.

(

\

f. At the design cooling rate for the proc--tzer of 100 F/hr, depressurization to 1,050 psig rgqu.'.res appre..; ately 1 7 hr.

During this time period 1.6 x 10o ec (5,650 ft>) of reactor coolant leaks to the secondary systum. This leakage corre-spends to approxt::ately 45,800 curies of xenon-133 if the re-actor has been 0 prating with 1 per cent failed fuel.

The radioactivity released during this accidant is discharged through the turbine bypass to the condenser and then out the Station vent.

tion factor of 104 is assumed for iodine in the co idenser.(1)(2)ANobleparti-gases are assumed to be released directly to the Station vent. The total dose to the whole body from all the xenon ari krypton released is only 0.48 rem at the 2,000 ft exclusion distance. The corresponding dose to the thyroid at the same distance is only 1.6 x 10-4 rem. Thit, calcula-tion conservatively assumes that the Station vent discharge mixes in the wake of the building structures rather than remaining at its elevated release height.

14.2 STANDBY SAFEGUARDS ANALYSIS 14.2.1 SIrUATIONS ANALYZED AND CAUSES In this section accidents are analyzed in which one or = ore of the pro-tective barriers are not effec *,1ve and standby safeguards are required.

All accidents evaluated are based on the ultimate power level of 2,535 Wt rather than the rated power level of 2,k52 Wt. Table ik-3 summa-rizes the potential accidents studied.

-

-

. - . 0001 309

-

,, ik-lo

.

.

Table lh-3 Situatiens Analyced and Causes Event Cause Effect Fuel Handling Mechcnical da= age Integrated dose at exclu-Accidents durin6 transfer, sien distance is 0.73 re:

thyrcid and 0.7h re: whole body.

Red Ejection Failure of control Ec=e clad failure. Thirty-Accident red drive pressure day dose at exclusien dis-housing. tance is 2.65 re: thyroid.

Loss-of-Coolant Rupture,cf reacter :Tc clad melting. Thirty-Accident coolant system. day dose at exclusion dis-tance is 10.7 re: thyroid.

Maxi =u= Relea'e s of 100% rare Two-hour dose at exclusion Hypothetical gases, 50% iodine, distance is 36 re: thyroid.

Accident and 1% solid fission Thirty-day dose at low pop-prod'tets . ulation distance is 72 re=

thyrcid.

1L.2.2 ACCIDE:IT ANALYSES 1h.2.2.1 Fuel Handling Accidents 1h.2.2.1.1 Identification of Accident Spent fuel asse=blies are handled entirely under water. 2efore refueling, the reactor coolant and the ruel transfer canal vater above the reactor are increased in boren concentratica so that, with all centrol rods re-

=oved, the ke rf of a core is no greater than 0.98. In the spent fuel storage pool, the fuel asse=blies are stored under water in storage racks havin6 an eversafe gec=etric array. Under these conditions, a criticality accident during refuelin6 is not considered credible. Mechani:al da= age to the fuel asse=blies during transfer operations is possible but i= prob-able. This type of accident is censidered the =axi=u potential source of activity release during refueling operaticns.

l 1h.2.2.1.2 Analysis and Results The fuel asse=bly is conservatively assu=ed te have operated at 29 MWt, twice the pcver level of an average fuel asse=bly. The reacter is assu=ed to have been shut down for 2h hr, which is the =inimu= time for reactor l cooldevn, reactor closure head removal, and re=cval of the first fuel as-I sembly. It is further assu=ed that the entire outer rev of fuel reds , 58 of 208, suffers da= age to the cladding. Since the fuel pellets are cold, only the gap activity is released. The fuel red gap activity is calculat-ed using the escape rate coefficients and calculational =ethods discussed i= 11.1.1.3 llh

'

90 t .."?

0001 310

• ,\,

lh-2C (3evised 7-21-57)

,-

,

,o

The gases released frem thei fuel assembly pass through the spent fuel e stora6e pool vater prior to reaching the auxiliary building at=osphere.

k As a mini =um, the gases pass through 10 ft of water. Althou6h there is experi= ental evidence that a portion of the noble gases vill re=ain in the water, no retention of noble gases is assu=ed. Pased on the data in References 3 and 4, 99 per cent of the iodine released frem the fuel assembly is assu=ed to re=ain in the water. The total activity released to the building at=espht:re is therefore Iod.ine 28.4 curips No'sle gases 2 79 x 10 curies The auxilia:7 building is ventilated and discharges through 90 per cent efficient charcoal filters to the Station vent. The discharge from the Station vent is assu:ed to =1x in the wake of the building structures rather than re=ain t.t its elevated release point. This assu.ption pro-duces less favorable dilution and, therefore, higher ground concentra-tions at the exclusion distance.

The activity is ansumed to be released as a puff frem the Station vent.

Atmospheric dilution is calculated using the 2-hour dispersion factor of 5 x 10-4 developed in 2 3 The total integrated dose to the whole body at the 2,0C0 ft exclusion distance is 0 74 rem, and the thyroid dose at the same distance is 0 73 rem. In evaluating the sensitivity of this analysis, the thyroid dose at the site boundary is directly propor-tional to the quantity of iodine released. For example, if only 90 per cent retentica of iodine is assu=ed by the spent fuel storage pool water, the dose at the exclusion distance is increased by a factor of 10. The dose from this increased iodine release is still a factor of 40 below the 10 CFR 100 guidelines.

14.2.2.2 Rod Ejection Accident 14.2.2.0.1 Identification of Accident Reactivity excursions initiated by uncontrolled rod withdrawal (14.1) were shown to be safely terminated without damage to the reactor core or reactor coolant system integrity. In order for reactivity to be added to the core at a more rapid rate, physical failure of the control rod drive housing or control rod drive noccle . ust occur. Failure in the drive upper pressure housing can cause a pressure differential to act on a con-trol rod assembly and rapidly eject the assembly from the core region.

The power excursion due to the rapid increase in reactivity is limited by the Doppler effect and terminated by reactor protection system trips.

The criterion for reactor protection, should this condition occur, is that the reactor vill be operated in such a =anner that a control rod ejection accident will not further damage the reactor coolant system.

a. Accident Bases

a The bases for the rod ejection accident are as follows:

+s .

• y ', ,.

.

"n. .

'

la_21 0001 3ll

Worth of ejected rod 0.3% ak/k Rod ejection time 0.150 see Ultimate power level 2,535 MWt Reactor trip delay 0.3 see The severity of the red ejection accident is dependent upon the worth of the ejected rod and the reactor power level. *he control red group of greatest worth is the first of the entire rod pattern,to be with-drawn from the core. The vorth of this rod can be as high as 30 per 3 cent of the total pattern verth of 10.0% ak/k, i.e., 3% ak/k. Hov-ever, the 3% ak/k value exists only when the reactor is suberitical.

The details of control rod vorth calculations and the methods of se-lecting the number of control rods in each group are presented in 3.2.2 and 7.2.2.1.2.

When the reactor is suberitical, the boron concentration is saintained at a level whereby the reactor is at least 1 per cent suberitical with the centrol rod of gr atest worth fully withdrawn from the core.

Therefore, rod ejection, when the reactor is suberitical and all other rods are in the core, does not cause a nuclear excursion. As criti-cality is approached, the worth of the remaining control rods de-creases. At criticality, rod ejection vould result in a saxim m re-activity addition of 0.6% ak/k.

l3 At rated power, but before equilibrium xenon is established, the total rod pattern verth remaining in the core is 2.6% ak/k. At equilibrium xenon the pattern vorth is 1.6% ak/k. Before establishing equilibrium xenon, the greatest single centrol red worth is 0.46% ak/k. A single l3 rod worth of 0.3% ak/k is assumed for analysis of this accident at this time.

In order for any one rod to have this much vorth, it would necessarily be fully inserted in the core. Assuming that a pressure housing fail-ure occurs in such a manner that it no longer offers any restriction for rod ejection, the time and therefore the rate of reactivity addi-tion can be calculated. Further assuming that there is no viscous drag force limiting the rate of ejection, control rod travel ti=e to the top of the active region of the core is calculated to be 0.176 sec. To account for the S-shaped reactivity worth versus position

• of the rod,'an ejection time of 0.150 see (75 per cent of active core height) is used in the analysis.

b. Fuel Red Damage Criteria Power excursions caused by reactivity disturbances of the order of magnitude occurring in rod ejection accidents could lead to three potential modes of fuel rod failure. First, for very rapid and large transients in which there is insufficient time for heat transfer from fuel to cladding, fuel melting folleved by vapori:,ation can generate destructive internal pressures without increasing cladding tempera-tures significantly. The second mode occurs when the internal vapor pressure is not sufficient to cause cladding rupture, but subsequen*

heat transfer 0001 H2

,, ;jj  ! [/i lh-22 (Revised 11-6-67)

.

'

. - -

,

- _ _ _ . _ _ _ _ _ _ _

raises the temperature of the cladding and weakens it until O failure occurs. The third mode occurr. When the nuclear excur-sion has insufficient energy to cause significant =elting of l

'

the fuel, but subsequent heat transfer to clad frem fuel may cause excessive cladding te=peratures. In all three cases there is a possible occurrence of =ctal-vater reacticas. Ecv-ever, only very rapid and large transients vill generate a rapid pressure buildup in the reictor coolant system.

The energy required to initiate U02 fu81Sitin6

• 220 to 225 cal /gm, based on an initial te=perature of 68 F.( The heat of fusion requires an additional 60 cal /g=. Any further energy addition vaport:es the fuel and produces a buildup of vapor pressure within the fuel rod. The effect of the vapor pressure is dependent upon the temperature and ulti= ate strength of the cladding. Ecergy additions of up to h20 cal /g= have been cal-culated to be necessary before the bursting pressure of cladding is exceeded. The lower limit for produciq significant fuel vaporpressure(lk.7 psi)is325 cal /gs.l01 The potential clad-ding failure is a function not only of the fuel vapor pressure, but also of fission product gas pressure, cladding and fuel ir-radiation exposure, and zirconium hydriding. As a lower limit, l1 the potential for bursting of claddin6 and release of molten fuel to the reactor coolant is conservatively set at a fuel en-thalpyof280 cal /gminthisevaluation. 1 For power excursions with energy bursts belov 280 cal /gm, =ir-coniu=-vater reactions are possible. A correlation of the TREAT experiments presents a method of correlating the poten-tial =1rconium-vater reaction as a function of fission energy input.(7) These data are based on initially cold (rocm tempera-ture) fuel rods, but are also correlated as a function of peak adiabatic core temperature. This correlation can be used either by computing the core temperature or by adding the initial steady state fuel enthalpy to the nuclear energy burst and obtaining an equivalent final fuel enthalpy. Accordingly, a tirconium-vater reaction requires a minimum fuel enthalpy of 125 cal / p . In-creasing fuel enthalpies cause a linear increase in the percent-a6e of the reaction, which may be approximated by

%Zr-E2 0 Reaction - 0.125 (Final Fuel Enthalpy - 125).

It is assumed that DNB vill take place when the clad reaches a heat flux of 6 36 x 105 Stu/hr-ftd. At this heat flux the hot fuel rod enthalpy would be approx 1=ately lk0 cal /gm at EOL and 130 cal /gm at 30L. Applying the peaking factors described in 3 2 3 to the results of these analyses, the per een of the core havin6 an enthalpy greater than the values above can be calculated. Any fuel rod exceeding the enthalpy values above is assumed to fail from overheating and releases the gap activ-ity of that fuel red.

O

's , s

...! r,!

'

-

Ik-23 0001 3;3

,

(Revised 7-21-67)

14.2.2.2.1 Method of Analysis g

The hypothetical control rod ejection accident was Mvestigated usi 6 the exact 1-d1=ensionalVIGL2digitalcomputerprogrs=.h) It was found that the point kinetics analog model results agreed with the WIGL2 results +4 within 10 per cent for rod worths up to 0 75% ak/k. The point kinetics

=odel assu=es an initial flux distribution which is undisturbed by local control red asse=blies. The space-dependent model, however, has signifi-cant flux depressions in the vicinity of control rods. Althou6h the flux throughout the core begins to increase shortly after the start of the rod ejection, the flux increase in this depressed re61on rises = ore quickly so that by the ti=e the average power has reached a level just a few per cent above the initial power level, the flux shape has al=ost no pertur-bation in the region previously occupied by the ejected rod. The entire reactor flux then rises unifor=ly until the Doppler effect ter=inates the excursion. Thus by applying the peak-to-average flux factors of 2 92 for EOL and 3 24 for BOL to the point kinetics results, the peak and inte-grated flux at any point in the reactor can be accurately assessed.

D.2.2.2 3 Analysis and Results

a. Source power

! This analysis was perfor=ed with the core 0 5% Ak/k suberitical so that a total rod vorth of 1% A k/k was withdrawn in 0.150 sec.

'

The reactor power was initially at 10-9 of the ultimate power level. The low pressure trip occurs at 1 7 see after the ejec-tion starts, and the reactor power is teminated at a peak value of 39 per cent ultimate power. This peak neutron power value is not reached until about 15 see after the rod is ejected because Doppler feedback controls the rate of rise and =agnitude of the neutron power. Therefcre, a lov pressure trip vill ter=inate the accident before significant power is generated owing to the loss of coolant throu6h the rupture.

An analysis vas perfor=ed for the accident above without a lov l

pressure trip to de=enstrate the capability of the reaccor to

accept the accident.

In this case the neut:on pover reaches 1,,000 Mwt (39 per cent I ulti= ate pover), and the peak fuel te=perature is 990 F. This is far belov the =elting te=perature of U0 2. and the resultant j ther=a*. power is only 16 per cent of ultimate pover. Hence, no fuel dama6e vould result fro = the rod ejection accident at source power level.

b. Ulti= ate Power For the ulti= ate power case at beginnin6-of-life (30L), the ejec-tion of a single control red worth 0 3% Ak/k would result in virtually no Zr-E20 reaction and approx 1=ately 1% of the core g experiencing DNB (see F16ures ik-20 and ik-21). The hot fuel W rodwouldreachapeakenthalpyofabout166=al/g=.

W ' "~ 3., 0001 314

For the end-of-life case (ECL), the reactor neutron power peaks at 6,190 Wt, 200 =1111 seconds after the start of ejection of a 0 3% A k/k contm l rod. The pro. pt negative Doppler effect teminates the power rise, and control red insertion from high flux signal ter::inates the excursion. The total neutron energy burst during the transient is approximately 3,200 MW-sec. The final fuel enthalpy of the nominal red is 113 cal /sm, i.e., the enthalpy of the hot rod is 163 cal /g=. This enthalpy is con-siderably below the mini =um range (220 to 225 cal /sm) for een-tral fuel =elting. As a result of the excursion, approximately <

13 5 per cent of the core vould have DIG (see Figure 14-20).

The power distribution at the beginning of core life, with the higher power peaking factors shown in 3 2 3, was used to de-termine the distribution of the energy of the excursion. With this distribution of fuel enthalpies, and using the TREAT cor-relation, 0 53 per cent of the circonium elsading =ay react (see Figure 14-21) to contribute an additional 677 W-see of energy. The resultant te=perature increase is spread over a relatively long period of ti=e. Consequently, the metal-water reaction energy is liberated over a long period of ti=e, and no damaging pressure pulses are produced in the system.

As a result of the postulated pressure housing failure, which produces a rupture size of 0.04 sq. ft., reactor coolant is lost frem the system. The rate of mass and energy input to the reactor building is considerably lower than that for the 3 sq. ft. rupture discussed in 14.2.2 3 This lower rate of energy input results in a lower reactor building pressure than that obtained for the 3 sq. ft. rupture.

The environmental consequences frem this accident are calculated by conservatively assuming that all fuel rods that undergo a DIG result in clad failure and subsequent release of the gap activity. Actually, =ost of the fuel reds will recover from the DIG, and no fission product release vill occur. For the case of a 0 3% Ak/k red ejection frem ulti= ate power at the end of life,13 5 per cent of the fuel rods are assu=ed to fail, releasing 177,000 equivalent curies of I-131 to the re-actor building. Fission product activities for this accident are calculated using the =ethods discussed in 11.1.1 3 Using the environmental =odels and dose rate calculations discussed under the loss-of-coolant accident, the total integrated dose to the thyroid at the site boundary from this accident is only 2.65 rem in 30 days, which is r re than a factor of 100 belev the guideline values of 10 CFR 104.

c. Sensitivity Analysis The results of a sensitivity acalysis perfo =ed on the control O .

'

red ejection accident are shewn in Figures 14-22 through 14-30.

,, ,,. Figure 14-22 shows the variation in the peak neutren power as  ;

a function of the verth of the ejected centrol red. For the i nominal 0 3% a k/k case frca ultimate power, the peak neutron

'"'

o 0001  !)15

,

_ - - _ _ _ .

power is less than 300 per cent, agai assu=ing thac a low pres-sure trip does not occur. The rod etretion from source level g

resulth in a Doppler turn-around befere %e flux trip is reached.

Figure 14-23 shows the variation in e corresponding the=al power with control red vorth.

Figure 14-24 shows the corresponditr athalpy increase of the het fuel rod versus control rod'vorth. ! Tote the very s=all spread in values for the 30L and EOL ulti-ate power conditions.

As expected, the enthalpy increases with rod vorth.

Figures 14-25 through 14-28 show the peak reactor neutron and ther=al powers as a function of changes in the positive =odera-tor te=perature coefficient and negative Doppler coefficient for the nominal 0 5% ak/k control rod ejection from source level. There was insignificant variatio:. of the peak neutron and ther=al power vith changes in the two reactivity feedback coefficients..

Figure 14-29 shows the change in nominal the=al power with variations in the trip delay ti=e for the nominal 0 3% ak/k rod ejection from ulti= ate power (the variation from zero power is negligible). The trip delay ti=e does not affect the peak neutron power because the Doppler effect controls the power transient. Figure 14-30 shows the corresponding change in the total enthalpy increase of the hot fuel rod versus the trip de-lay.

The thermal power never exceeds 114 per cent ult 1= ate power for any of the variations studied using the nominal rods (0 3% a k/k for ultimate power and 0 5% 4 k/k for source level). The hot fuel rod avera6e te=perature never increases by = ore than 310 F above the ulti= ate power peak value (4,090 F). It is therefore con-cluded that each of these para =eter variations has relatively little effect on the nominal results.

O

'

-

. .

e' ,' :'

1

,. o . n . 0001 316 14-26

_ _ _ _ _ - _ _ _ - _ _ _ - - _ - - -

14.2.2 3 Loss-of-Coolant Accidsnt 14.2.2 3 1 Identification of Accident Failure of the reactor coolant system would allow partial or ec=plete release of reactor coolant into the reactor building, thereby interrupting the normal

=echanism for removing heat from the reactor core. If all the coolant were not released i==ediately, the remaining amount vould be boiled off eving to residual heat, fission product decay heat, and possible heat from che=ical re-actions unless an alternate means of cooling vere available. In order to pre-vent significant chemical reactions and destructive core heatup, emergency Core cooling equipment rapidly recovers the core and provides makeup for decay heat removal.

14.2.2 3 2 Accident Bases All ecmponents of the reactor coolant system have been designed and fabricated to insure high integrity and thereby minimize the possibility of their rupture.

The reactor coolant system, the safety factors used in its design, and the special provisions taken in its fabrication to insure quality are described in Section L.

In addition to the high-integrity system to minimize the possibility of a loss of coolant, e=ergency core cooling is provided to insure that the core does not melt even if the reactor coolant system should fail and release the coolant.

This emergency core cooling is provided by the core flooding system, the makeup and purification system (high pressure injection), and the decay heat removal syste= (low pressure injection). These systems are described in detail in Sec-tion 6, and their characteristics are summavi:ed in the following paragraphs.

}

The perfor=ance criterion for the e=ergency core cooling equipment is to limit the te=perature transient below the clad melting point so that fuel gec=etry is =aintained to provide core cooling capability. This equipment has been censervatively sized to li=it the clad temperature transient to 2.300 F or less as te=peratures in excess of this value prc=ote a faster circoniu=-water reaction rate, and the ter=ination of the transient near the melting point would be difficult to de=cnstrate.

The fuel rods may experience cladding failure during the heatup in the less-of-coolant accident. This could be due to fission gas internal pressura and weak-ening of the clad due to the increase in clad temperature. The =echanical strength of the Zircaloy cladding is reduced as the temperature exceeds 1,000 F such that the highly irradiated fuel rods, with high fission gas internal pres-sure, may fail locally and relieve the gas pressure unen the temperature ex-ceeds 1,200 F. Some local ballooning of rods is likely to occur. However, cooling would still be effective siace the fuel reds are submerged, and cross-channel flow around the balleoned area vill cool the rod. At vorst a local hot spc t =ay occur.

It is calculated that a s=all number of fuel rods operating at peak power vill experience a cladding te=perature transient to 1,950 F in about 18 sec. The injection of emergency coolant, at a time when vne claddin6 is at a te=perature of about 1,950 F, may also cause distortion or bowing between supports. As a result so=e of the fuel rods =ay crack and allow relief of internal pressure.

Ik-27 (Revised 7-21-c7) i C.'

i i .'i.>.-

.

000$ $

s

However, the claddi 6 is expected to re=ain sufficiently intact to retain the solid fuel =aterial and to prevent gross fuel shifting. The transient vould be li=ited to regions of the core which operate at peak power. The =ajor portion lll of the core vill not experience as severe a transient.

Heatin6 of the fuel can and the fuel red spacer grid requires heat flow frc=

the clad to the structure by conduction and radiation; therefore, the st:veture temperatures vill lag the cladding te=peraturn transient. As the fuel red te=-

perature rises, the fuel rods are expected to experience sc=e boving between supports due to the te=perature differential existing between the fuel red and the can. The cans and spacer grids are =ade fron ::ainless steel and have sub-stantial =echanical stren6th, even at the maximuct expected te=peratures. The supportin6 stainless steel structure vill therefore retain sufficient strength to assure spacing between fuel rods to allev e=ergency ecciant to reach the=,

and vill keep the fuel reds in the sa=e location in the core to prevent 6:088 fuel shifting.

The core ficeding syste= has tvc independent core flooding tanks, each of which is connected to a different reactor vessel injection neccie by a line containing two check valves and a nor= ally open, re=otely operated isolation valve. Since these tanks and associated piping are =1ssile-protected and are connected di-rectly to the reactor vessel, a rupture of reactor coolant syste= piping vill not affect their perfor=ancs. These tanks provide for autc=atic ficoding when the reactor coolant syste= pressure decreases below 6CO psi. The ficoding water is injected into the reactor vessel and directed to the bot:0= of the reactor vessel by the ther=al shield. The core is flooded frc= the botto: upward. The co=bined contents of the two tanks (1,880 ft3) rapidly reflood the core i==edi- g ately after the bicwdown to provide ecoling until coolant flev can be estab- W Iished by low pressure injection.

High pressure injection, actuated by low reactor coolant syste= pressure, sup-plies coolant at pressures up to the design pressure of the reactor coolant sys-te= and at a rate up to 1,000 gp=. Lev pressure injection actuated by low reac-tor coolant syste= pressure supplies coolant at p sssures below 1CO psig and at a rate up to 6,000 gp=. Both of these syste=s can operate at full capacity fror the on-site e=ergency electrical power supply and can be in operation within 25 see after the accident. In the reactor vessel, decay heat is transferred to the injection water.

l

,

1 1

O

$, !

%} '}h

'

U!:

l IL-2Ta (?.evised 1-c-c5)

.

Injection water is supplied frc= the borated vater storage tank. When this tank e=pties, vater is circulated frc= the reactor building surp through heat exchangers and returned to the reactor vessel. g Engineer-d safeguarus are also provided to cool the reactor building environ =ent following a loss-of-coolant accident and thereby li=it and reduce pressure in the building. Reactor building sprays, actuated on a high building pressure signal of 10 psig, deliver 3,000 g;= to the re-actor building at=osphere. This spray water reaches ther=al equilibriu=

vithin the building at=osphere during its passage fro = the no::les to the st p. Spray vater is supplied fro = the borated vater storage tank until it is e=ptied. Thereafter, water collected in the su=p is recirculated to the sprays. Cooling is also provided by the reactor building e=er-gency cooling syste= in which recirculating fans direct the stea=-and-air =1xture throuGh e=ergency coolers, where stea= is condensed. Heat absorbed in the e=ergency coolers is rejected to the nuclear services cooling water syste=. The heat re= oval capacity of either of these two reactor building cooling syste=s is adequate to prevent overpressuriza-tion of the buildin6 during a loss-of-coolant accident.

This analysis de=onstratec that in the unlikely event of a failure of the reactor coolant syste=, both of the other two boundaries that prevent fis-sion product release to the atmosphere, i.e., the reactor core and the reactor building, are protected frc= failure. Accordingly, the public vould be protected against potential radiation hazards.

In order to evaluate this accident, a range of rupture sizes frc= s=all leaks up to the co=plete severance of a 36 in. ID reactor coolant syste=

line has been evaluated. A core cooling analysis is presented for the co=plete severance of the 36 in. ID reactor coolant piping.

Since the large rupture re= oves the least a=ount of stored energy fro =

the core, this represents the =1ni=u= te=perature =argin to core darage and, therefore, places the = cst stringent requirerents on the core flood-ing syste=.

The reactor building pressures have been evaluated for a ec=plete spectru=

of rupture sizes vithout the action of core flooding tanks and, therefore, are conservative. The peak pressure occurs for a 3 ft2 rupture rather than for a 36 in. ID (14.1 ft2 ) rupture. Rupture sizes s= aller than the 36 in. ID leak result in longer blevdown t1=es, and the a=ount of energy trannferred to the reactor building atmosphere is increased. As a result the inter =ediate leak size results in a reactor building pressure greater than that produced by the 36 in. ID rupture.

1k.2.2 3 3 Accident Si=ulction

a. Hydraulic Model 31ovdown of the reactor coolant syste= following an assu=ed rupture has beaa e'-"'ated by using a =odified version of the FLASH.9) f code. This code calculates transient flows, coolant g

= ass and energy inventories, pressures, and te=peratures during W lu-as 0001 3

.

l

'

a loss-of-coolant accident. The code calculates inflov fro = the e=ergency cooling and calulates heat transferred frc= the core to the coolant.

Modifications were =ade to FLASH to =ake the :odel more appli-cable to this syste=. The chan6es are as follows:

(1) The calculation of reactor coolant pu=p cavitation was based on the vapor pressure of the cold leg instead of the hot leg water. .

(2) Core flooding tanks have been added. Water flow from the core flooding tank.s is calculated on the basis of the pres-sure difference between the core flooding tanks and the point of discharge into the reactor coolant syste=. The line resistance and the inertial effects of the water in the pipe are included. The pressures in the tanks are cal-culated by asst =:ing an adiabatic expansion of the gas above the water level in the tank. Pressure, flow rate, and = ass inventories are calculated and printed out in the co=puter output.

(3) Additions to the water physical property tables (=ainly in the subcooled region) have also been =ade to i= prove the accuracy of the calculations.

(4) A change in the stec= bubble rise velocity has been =ade from the constant value in FLASH to a variable velocity as a function of pressure. The bubble velocity ter= deter-

=ines the amount of water re=aining in the system after depressurication is co=plete. For large ruptures, this change in velocity shows no appreciable change in water re=aining fro = that predicted by the constant value in the FIASH code. For s= aller ruptures, an appreciable differ-ence exists. The variable bubble velocity is based on data in Reference 10 and adjusted to correspond to data fro = the LOFT se=iscale blevdown tests.

Test No. 5k6 frc= the I M r se=iscale blowdown tests is a typical case for the blevdown through a s=all rupture area.

A co=parison of the predicted and experi=entally observed pressures is shown in Figure 14-31. Figure 14-32 shows the per cent = ass re=aining in the tank versus ti=e as predicted by the code. At the end of blevdown, the pre-dicted = ass re=aining is 13 per cent. The =essured = ass re=aining is approxi=ately 22 per cent.

The FIASH code describes the reactor coolant syste= by the use of two volumes plus the pressuri:er. The syste= vas grouped into two volu=es on the basis of the te=perature distribution p)

%.

in the syste= as follows:

Volu=e 1 includes half of the core water volu=e, the re-actor outlet plenu=, the reactor outlet piping, and ap-prox 1=ately 55 per cent of the stea= generators.

B- W 1,_ ,

0001 320

.

Volu=e 2 includes half of the core water volu=e, the re-actor inlet plenu= and. devnec=er section, the reactor in-g let piping, pu=ps, and 45 of the steam generators.

Volu=e 3 represents the pressuricer.

The resistances to flov vere calculated by breaking the reactor coolant system into 24 regions and calculating the volume-Weighted resistance tc flew for a given rupt"re location based on nomal flow resistances. For the double-ended ::ptures, all of the leak was assu=ed to occur in the volu=e in which that pipe appeared.

The reactor core power was input as a function of ti=e as deter-

=ined by the CHIC-KIT code in conjunction with the FLASH output.

Steam generator heat re= oval was assu=ed to cease when the rup-ture occurred.

While the ::xidified FLASH code nov has the capability of si=ulating injection flov from the core flooding tanks, the calculations shown in this report were made prior to the ti=e that the core ficoding simulation was added to FLASH. Core ficoding vr.s cal-cu.'.ated by taking the reactor Vessel pressure as predicted by FLASH vithout core ficcding and using this pressure as input to a separate progrs= to get the flow frem the core ficoding tanks.

Reactor vessel filling was calculated by adding the = ass re-

=aining in the vessel as predicted by FLASH to the = ass injected from the core flooding tanks. This =ethod of calculation is con-servative in that condensation of steam by the cold injection water is not taken into account. A more recent analysis using the FLASH code confirms that conservatis= used in this analysis.

Pressure, te=perature, mass and energy inventories, and hydraulic characteristics as determined by FIASH are input into the core ther=al code (SLUMP) and the reactor building pressure buildup code (CONTEMPT).

b. Core Ther=al Medel The core heat generation and heat transfer to the fluid are dependent upon the blevdown process. The FLASH program in-cludes a core ther=al =odel and the re3dbacks of heat trans-fer and flow on each other. While the FLASH therral =cdel is acceptable for detemining the effect of core heat transfer on the blevdown process, a =cre extensive s1=ulation is necessary for evaluation of the core te=perature transient.

Additional analytical =cdels and a digital ec=puter progr m were developed to si=ulate the core the=al transient for the period beginning with the initiation of the leak and ending after the core te=perature excursien had teminated.

The model includes the effects of heat generation frem neutrons

.

before reactor trip, neutron decay heat, and fission and

'

mx

'

a. 0001 321 ik-30

.. -_

.

activation product decay neat; the exother=1c zieroniu=-vater O reaction based on the parabolic rate law; heat t'tnsfer

.

the fuel rods, li=1ted heat convection frc= the fuel clad sur-within face to any fluid within tne cors region, heat transfer frc=

reactor vessel valls and inte= als to the coolant, and heat transfer fro = fuel reds to the stea= necessary to sustain a

=etal-water reaction; and e=ergency injection flow and boiloff.

The basic =odel structure provides 50 equal-volu=e core regions with input provisions to allow any choice of power distribution.

The model =ay be used to s1=ulate the entire core or any sub-division of the core. Therefore, tus core geo=etry =ay be de-tailed to the degree consistent with the results desired.

'

/ The following parabolic law for the zirconiu=-water reaction j equation (ll) with the following constants is simulated for each p of the regions:

l .

I _dr = K A?y

-

exp -

/ dt (rg - r) ,

RT 1

/

1. vhere:

r = radius of unreacted metal in fuel red rg = original radius of fuel red t: ti=e K = rate law constant (0 3937 c=2/3 e)

AE : activationenergy(45,500 cal / mole)

R = gas constant (1 987 cal /= ole K)

T = te=perature, K The zirconium-vater reaction heat is assu=ed to be generated cc=pletely within the clad code. The heat necessary to increase the stea= temperature frc= the bulk te=perature to the reaction te=perature is transferred frc= the clad at *2e point of re-action. The above equation i= plies no steam li=iting. Howaver, the program does have provision for stes= rate-li=1 ting to any degree desired, but no stes=-11=1 ting of the reactions has been assu=ed. All heat fro = neutron, beta, and ga==a sources is as-su=ed to be generated within the fuel according to the pre-accident power distribution and infinite irradiation.

Within each of the regions there is a single fuel node and a single clad node with si=ulation of ther=al resistance accord-ing to the nomal fuel red geomet:7 Provision is =ade to simu-O late four different modes of heat transfer frc= the clad node to O the fluid sink node by specifying the ti=e-dependent surface coefficient.

.

.

.,

....

0001 322

, ,

.

i!!'

t. .

ik-31

l l

!

i l

The surface heat transfer coefficient input data are deter =ined from calculations which are based on flow and water inventory gl as furnished from the blevdown and the core ficoding tank per-for=ance analysis.

In the event that insufficient cooling is provided, the program vill allow clad heating to progress to the =elting point. At this point the latent heat of zirconium =ust be added before the clad =elts. Provisions are also incorporated to allow the clad to be heated to te=peratures above the =elting point be-fore slump occurs.

As each region slu=ps it may be assu=ed to surrender heat to a water pool or to some available =etal heat sink. If water is available an additione.1 10 per cent reacticn is assu=ed to occur. .

The program output includes the following (as a function of ti=e unless otherwise specified):

Average fuel te=perature of each region.

Average clad temperature of each region.

Per cent =etal-vater reaction in each region.

T1=e for the clad of each region to reach the metal-vater threshold, the beginning and end of melting, and the slu=p temperature.

Heat tri nsferred to the reactor building from the core.

Heat generation by hydrogen and oxygen recombination.

Total irconium-vater reaction.

Total heat stored in =etal sinks.

c. Reactor Building Pressure Model The reactor building pressure-temperature analysis is perfor=ed using the digital co=puter code "CONTEMP!" developed by Phillips Petroleum Co=pany in conjunction with the ILFT project. This program and its capabilities are described in Reference 12.

With minor =odifications this program was adapted for use on the E&W Philco-2000 co=puter.

In this model, the reactor building is divided into tvo regions: <

the atmosphere (vater vapor and air mixture) and the st. p region (liquid water). Each region is considered to be well mixed and in ther=al equilibrium, but the te=perature of each region =ay be different. The reactor building and its internal structures g

are Jubditided into five seg=ents, as shown in Taile 14-h, and w.,x,

,

0001 323

treated as slabs with 1-di=ensional heat transfer. Each seg=ent, divided into several heat conductin6 subregions, =ay act as a heat source or sink. The program includes the capability of cooling the reactor building at=osphere by air coolers (reactor building emergency cooling units) and spray coolers (reactor building spray syste=), and cooling the liquid region by su=p coolers (decay heat re= oval coolers).

During blevdown, mass and energy are added directly to the at=o-sphere where the liquid vater present is assu=ed to fall to the

liquid region. After blevdown is over and e=ergency injection has been initiated, = ass and energy are also added directly to the vapor region as stea=. When the water level in the reactor vessel reaches the noccle height, all = ass and energy are added directly to the liquid region since no boiling of the injection water occurs after the core has been covered. When the supply of injection water is depleted, recirculation and cooling of sump water is simulated.

The reactor building calculations are begun by computing steady-state results using initial atmospheric conditions as the input.

Following the rupture, the mass and energy addition is deter-mined from the energy input rates for each ti=e step. Heat losses or gains due to the heat-conducting slabs are calculated.

Then the pressure and te=perature of the liquid and vapor re-gions are cale "sted from the cass, voluce, and energy balance equations.

Table 14-4 Reactor Building Structural Heat Capacitance Segents Segent Description 1 Reactor Building Walls and Doce 2 Refueling Cavity (Type 304 SS Liner - One Side) 3 Reactor Building Floor 4 Inte:=al Concrete 5 Internal Steel The model has been developed so that the effectiveness of the natural heat sinks and the engineered safeguads can be clearly demonstrated. The model can readily produce the reactor build-in6 pressure history for any assu=ed combination of operable safeguards. Therefore, the effectiveness of any given arran68-

=ent can be analyced.

O 6 w. 3_g 0001 324

...

e

.

14.2.2 3 4 Accident M alysis a.

h Core Flooding Tank Design Base Accident 7he 36 in. ID, double-ended pipe rupture produces the fastest blevdown ar4 lowest heat re= oval from the fuel. This case therefore represents the =ost stringent e=ergency core cooling requirements. Results frem the = edified versien of FLASH in-dicate that the core flooding tank simulation provides for the retention of all injection plus a portion of the original re-t.ctor coolant that would otherwise have been released. Thus, the cool injection water provides a cooling ard condensing ef-feet which reduces overall leakage. For the present analysis, no credit has been taken for the extra accu =ulation of water due to the condensing effect.

The blevdown was analyced using the version of FLASH vithout core flooding tank si=ulation. This resulted in higher tran-sient reactor vessel pressure than vould have occurred if core flooding tank flow feedback effects were included. The core flooding tank transient analysis was then perfomed using re-I actor vessel back pressure which was provided by the FLASH a nal ysis . To detemine the fluid re=aining in the reactor ves-sel at any point in ti=e during the blevdown transient, the integrated flow from the core flooding tank is added to the fluid re=aining which is predicted by FLASH. The inventory obtained by this =ethod is conservative because it neglects the condensing effect which leads to an additional accu =ulation of water.

The SLUMP digital ecmputer program, as described in 14.2.2 3 3.b above,is used to evaluate core flooding tank perfomance in tems of core cooling capability. In the analysis, the hottest 5 per cent of the core was s1=ulated in seg=ents of 1/4 of one per cent each. The hottest seg=ent was assigned a peaking fac-ter of 3 1 ti=es the average of the total core power density.

l l The reactor is assu=ed to be initially at the ulti= ate power level (2,535 MWt). The core analysis s1=ulates reactivity ef-fects which correlate with the results which vere obtained from a detailed analysis of void shutdown without control rod inser-tien. The detailed analysis was =ade with the digital co=puter program CHIC-ICT vhich included positive =oderator and initially positive void coefficients. The results frc= the void shutdown calculation yield a total integrated neutron energy generation of approxi=ately 2.1 full-power seconds.

The transient core flow frc= the FLASH analysis of the 36 in.

ID, double-ended rupture was used to dete=ine the core cooling

=echanism used in SLLHP. The very high flow rates during the initial blevdown period provide nucleate boiling conditions.

HovcVer, the time for Departure from Nucleate Boiling (DNB), h especially for the hot regions, is extre=ely difficult to de-temine. Therefore, a conservative approach was adopted by w n,o.. .

14-34 0001 325

_ _ _ _ _ _ _ _ _ - _ _ _ - _ _ _ _ _ - _ _ _ _ _ _ _ _

w assuming DNB at 0.25 sec. Nucleate boiling surface coefficients at high flev rates =ay exceed 50,000 Btu /hr-ft 2 -F. A nucleate bctling surface coefficient of 25,000 Btu /hr-f:2-F was used in the analysis. However, the series heat transfer frem the clad node to the fluid sink is limited to 6,500 Btu /hr-ft -F 2 by the relatively low corductance of the clad.

After DUB the surface heat transfer was calculated using the flow provided by FLASH results and Quinn's modified version of the Sieder-Tate (13) correlation: a

-

-0.8 0.14 b

k 1t 1-X IP B *B g = 0.023 g (NRe)O.8(Ng)1/3 X P Fs "W where hp* = two-phase fib heat transfer coefficient, Btu /hr-ft2-F k = fluid cenductivity, Btu /hr-ft 2_p Dh = hydraulic diameter, ft ITRe = Reynolds neber Np , = Prardt1 number x = qtality p = density

= viscosity subscript B = " Bulk" subscript F = " Film" subscript W = " Wall" With this correlation, bulk steam properties are used in the basic form, and the last two bracketed terms are modifiers which correct for quality ard different corditions at the vall.

Figure 14-33 shows the core flow vs time for the 14.1 ft2 1 ,g as calculated by FLASH.

Figure 14-34 shows the clad surface heat transfer coefficient versus time based on the flow of Figure 1h-33 and the modified Sieder-Tate equation. The straight line in Fitre 14-34 indi-cates the surface heat transfer values which vere used in SLLMP, and which are conservative as ccmpared to the results O obtained frem the Sieder-Tate equation.

,

,. e

v. ; '

e- ,-

m . *,;

,

0001 326 1k-35

!

.

In applying the Sieder-Tate equatica constant values of bulk stes= quality and te=perature corresponding to the =est conser- h vative asst =2ptions were used.

A sensitivity analysis was =ade for =axi=u= coefficients in SLUMP ranging fro = 400 to 2,000 Btu /hr-ft2 -F initially and de-creasing to zero at the end of blevdown. Results are discussed below.

After blevdown no core cooling is assu=ed until after core re- .,

covering starts. When the water level reaches the core botto=

and starts to rise up on the core, the sub=crged portion v11]

be cooled by pool boiling, and any stes= thus produced vill provide so=e cooling for that portion of the core above the water line. However, in deter =ining peak clad ts=peratures no cooling is assu=ed for that portion of the core which is above the water line.

At the point of initial contact of cool water against hot clad-ding the heat flux and te=perature differences vill be such that fil= boiling is the probable =ede of heat transfer. This mode provides the lowest surface coefficients which would be in the range of 100 to 300 stu/hr-ft2 -F. However, in evaluat-ing the core flooding tank design a conservative approach was used by assuming a value of 20 Btu /hr-ft 2 -F. This value is adequate for ter=inating the te=perature excursion in the clad.

The core flooding tank analysis incorporated the study of per-for=ance sensitivity to three significant core flooding ' M parameters: (a) gas pressure (k00 to 1,000 psig), (b) ratio of nitrogen gas volu=e to total volume (1/3 and 1/2), (c) and size of piping between the core flooding tanks and the reactor vessel (12 in and 14 in. ID). Figure 14-35 shows the reactor vessel water level versus ti=e for core flooding tanks operat-ing at 600 psig with different ec=binations of volu=e ratio and line si:e. This figure includes an allowance for boiloff and also shows the effect of the flow provided by high pressure and low pressure injection beginning at 25 see when e=ergency power is available. Si=ilar curves for 400 psig and 1,000 psis i core flooding tanks are shown in Figure 14-36. Figure 14-37 shovs the max 1=u= clad te=perature reached by the hot spot and by the 1, 2, 3, 4 and 5 percentiles of the core as a function of quench ti=e.

Se quench time for a given percentile is taken as that ti=e when the water level reaches the highest point in the core at which the peaking factor corresponding to that percentile

-

exists. The fact that the same peaking factor =ay exist at l some lever point in the core provides an inherent conservatis=

in the data as plotted. De axial peaking factor profile for the beginning of core life was used as it i= poses the =ost stringent require =ents on the ccre flooding tank design.

'

h m ,

y ,w.n a 0001 327 lh-36

.

Peak clad te=peratures for the core ficcding tanks perfor=ance

] described abcre are also shcun en Figure lk-37. *hese curves de=enstrate that all of the syste=s presented are capable of keeping the peak te=perature at the het spot =cre than 1,000 F belev the =elting te=perature of the clad. The a= cunt of cir-coniu=-water reaction which cecurs for each of these core ficod-ing syste=s is shewn in Table ik-5 While this prel1=inary anal-ysis indicates scme difference in the performance of the sys-tems, it is not considered to be a significant difference since the analysis was perfor=ed without censidering the effects of condensation by the core ficoding coolant or of pessible bypass ,

to the leak of part of the ecolant.

The prel1=inary core f1 ceding tank design selected is for a 600 psi charge pressure, 9hc ft3 vater, 470 ft3 cf nitregen, and a 14 in. supply line. The perfomance of this syste= in li=1 ting core temperatures is approxi=ately in the center of the range for the syste=s described. The para =eters selected for the final syste= design vill be based on the results of core =elting analyses to be conducted as part of the final de-sign of the reactor. For this 600 psi charge pressure, Figure 14-37 indicates that the hot spot clad te=perature vould reach 1,950 F at 17 5 see and that less than 5 per cent of the core would exceed 1,690 F. For this same case calculations indicate less than 0.005 per cent total zirconiu=-vater reaction for the whole core.

O Table 14-3 Core F1 coding Tank Perfor=ance Data Line Nitregen Total Metal Sice, Volu=e, Water Reaction, Pressure in. T,of Total  %

400 14 33 .C22 400 1k 50 .009 600 14 33 .005 6CO lb 3C .002 600 12 33 .022 ,

600 12 50 .010 l 1,C00 12 33 .003

'

1,0C0 12 50 =0 Additienal analysis was peric=ed te evaluate de sensitivity cf de =axt::= clad te=perature tc tree i=pcrtant de=al paraneters. All cases discussed belev have in ec==ca the fel-icving ;s. s=eters:

l

.

'

.

' "' .

0001 328

'

l g,3 , 1--3~ 3.*"is e i 1-i-f i . '

.

<

,

i;l'il

I l

Leak size: 14.1 ft2 g l

Time of ' DIG: 0.25 see i

I Ti::.e at ulti= ate pover; 2 see T1=e that blevdown cooling ends: 9 5 sec l

Core region: Hot spot Time to initiate quenching: 18 see Dependent variable examined: Clad temperature for hottest 5 per t cent of core.

l l Figure 14-38.shows the clad maximum temperature sensitivity to I

the initial surface heat transfer coefficient after the 0.25

sec nucleate boilin6 period. The coefficient is linearly de-l creased to zero at 9 5 sec. Zero cooling is =aintained until quenching is initiated with a clad surface coefficient of 20 2

Btu /hr-ft-F. Previous discussion indicated justification for assuming 1,000 Btu /hr-ft 2 -F for the clad surface at 0.25 sec.

Figure 14-38 shows that a value of 1,000 is not on the most sensitive part of the curve and a 20 per cent decrease in h g would only res i a increasing the peak clad temperature 120 F. W F16ure 14-39 shc,s het spot elad temperature transients for a ran8e of injection cooling coefficients. All cases have a clad surface coefficient of 1,000 Btu /hr-ft 2-F at 0.25 sec, decreas-ing to zero at 9 5 sec. Heat removal is then zero until the effect of injection cooling is simulated. Figure 14-39 shows that without any cooling the temperature reaches the =elting point in approxi=ately 50 sec.

An h value of 15 stops the fast temperature excursion and al-icvs only a low rate of increase thereafter. Since the contin-uously increasing depth of covera 6e provided by the floodin6 tanks and the pumped flow injection systems provide additional coolin6 capability with ti=e, an initial cooling value as lov as 15 is probably adequate.

i An h value of 20 provides immediate quenching action and a slov l coolins rate thereafter.

t An h value of 100 provides very fast cooling. Even thou6h the

100 is a realistic value for film boiling in a pool - the prob-able mode for the sumerged portion of the core - a more con-servative value of 20 has been used as the reference for eval-uating core floodin6 tank performance.

g I . Figure ik ko shows hot spot clad temperature transients for a p s ?, {yi1!rangeofpoolfluidsinkte=peraturec. Para =eters for heat l

1k-3s 0001 329

.- . - _

,

transfGr prior to 18 sse are tha sam 2 as discussed in the prgceding paragraph. At18seeasurfacecoefficientof20Stu/hr-ft-Fwas 2 applied with sink te=peratures as indicated. All results reported O herein previously have hcd a sink te=perature of 280 F during the quenching period. Prior to quenchin6 the sink te=perature in all cases is based on the transient fluid pressure which results from the FIASH analysis. Figure 14-42 show that any sink te=perature below approximately 500 F is adequate for holding or reducing the clad tem-perature which existed at 18 sec. The core riceding tanks vill pro-vide a high flow of cool vater. Although some heating vill occur from centset with hot metal before the injection water reaches the core, the temperature rise could not be over 50 F rasuming that the water came in contact with all reactor coolant system metal below the no::le level before it contacted the core. Using a reference value of 281 F provides an added conservatism to the analysis.

In conclusion, the analysis has shown tnat tne pre 11=inary design of the core flooding system v111 provide for covering approximately 80 per cent of tne core at 25 see after the double-ended rupture of the 36 in. ID pipe .first occurs. Beyond this ti=e high pressure and low pressure injection vill provide a continuous increase in the water level.

The clad hot spot te=perature excursion is terminated at 1,950 F and less than 5 per cent of the total cladding exceeds 1,690 F. Only a minute amount (0.005 per cent) of circonium-water reaction occurs, and the maximum te=perature is at least 1,400 F below the clad :relt-ing point.

O The temperature transient in the core can produce significantly highe.

than normal temperatures in components other than fuel rods. There-fore a possibility of eutectic formation between dissimilar core ma-terials exists. Considering the general area of eutectic formation in the entire core and reactor ve;sel internals, the following dis-similar metals are present, with major ele =ents being in the approxi-

= ate proportions sho'.m.

Tyne 30L Stainless Steel 19 per cent Chromium 10 per cent Nickel Balance Iron Centrol Rod 80 per cent Silver 15 per cent Indium 5 per cent Cad =1um Zircaloy-k 98 per cent zirconium 1-3/4 per cent Tin 002

/}

.

jl;; , ik-39 (Revised 7-21-67) 000I$30

-

, ....+

1 these ele =ents have relatively high =elting points, i.e., greater than 1 700 F, except those fer silver, cadmiu=, and indiu which, in the case of g dium, is as lov as approximately 300 F. W e binary phase diagrs= indicates that circenium in the preportion of 75 to 80 r cent has a eutectic point with either iren, nickel, er chrc=iu: at the te:-

ratures of approximately 1,710, 1,760, and 2,370 F, respectively. If these ssimilar =etals are in centact and if these eutectic points are reached, the

.terials could theoretically =elt even though the te=perature is belev the iting point of either material taken singularly.

"

.e point of such dissimilar =etal contact is between Zirealcy-clad fuel rods

.d stainless steel spacers. The analysis of the performance of the core flood-

.g tanks during a less-of-coola=t accident indicated that only h per cent of

.e cladding would ever exceed the circonium-iron eutectic point. Since the acers are located at 21 in. intervals along the asse=bly and each grid has a ry small contact area, only a fraction of the h per cent vculd be in contact th stainless steel. The approxi= ate time period that the h per cent of the

. adding is above the eutectic point is 30 sec. Becacre of the relatively small

'

ea of contact , the condition could not progress very f ar and fuel gec=etry aild be =aintained. Unless the proper ratio of metals is available, the =elting

> int is higher than the eutectic point.

.cther area of dissimilar =etal contact is that of a circoniu: guide tube with te stainless steel cladding of the control red. Felleving blevdovn, heat can generated in the centrol rods by absorption of ga-~n rays. Beta ray decay eat will be deposited in the fuel rods where generated. Since ga=na decay heat i only about ene-half the total decay heat, and the control rod is shielded c= the fuel by a guide tube, heat generation rates in control rods vill be oss than ene-half the rates in the fuel. As a result, the peak heat generation tte in centrol rods adjacent to het spot fuel vould not exceed an esti=ated te-half times the rate in these fuel rods which have a 3.1 power ratio. The

ntribution frc= radiant heat transfer frc= higher powered fuel rods vould be elatively small. The analysis of core =elting shows that, with core ficoding ads, fuel rods with a 1.5 power ratio will not exceed 1,500 F. This is well d mr the eucectic =elting point.

e cesctor core vill re=ain suberitical after ficoding withous control rods in ae core because the injection water centains sufficient boren (2,270 pp=) to

>1d the reactor suberitical at reduced temperatures. The =osw stringent boren squire =ent for shutdevn without any control rods is at the beginning of core

'fe when the reactor is in a cold, clean ccndition and 1,620 pp= boren are re-

.

l:

Aired to maintain k,ff of 0.99. (See Table 3-6, Soluble Boron Levels and 'Jorth. )

le concentratien existing in the reactor building su=p after a less-of-ccclant

cident frc= cperating power at the beginning of core life is 2,1~h pp: bcron.

<

11s concentration represents a boren =argin cf 3Sh pp= above the suberiticality

(

I rsign value =argin of 1 per cent.

.

O p ', - io t

ik-39 a (Revised 11-6-(T) 0001 331

4

b. Reactor Building Design Base Accidert A range of leak sizes between 0.h ft2 and ik.1 ft2 has been evaluated The 2,L.1 ft2 is equivalent to a double-ended rupture of the 36 in.

• ID reactor outlet piping. The reactor operating conditions used in this analysis are listed in Table ih-6.

The basis for this analysis is that only the =akeup and purification system and the decay heat removal system are working. It was as-sumed that the makeup and purification system (high pressure injec-tion) had one of the pumps available for operation and that the de- 4 cay heat removal system (low pressure injection) had both of the two pumps available for operation. These syste=s are assumed to operate on e=ergency power and can be in operation to deliver a total injec-tion flow of 6,500 gym within 25 see after the accident occurs.

This approach is conservative as any combination of core flooding tank operation and minimum flev fro,the high and low pressure pu=ps vill provide a. lover energy release rate and peak reactor building

.

pressures than those resulting from the above flow of 6500 gym.

During blevdown mass and energy releases to the reactor building are calculated by FLASH. Figure ik kl is a plot of mass released to re-actor building and Figure ik h2 is a plot of reactor coolant average pressure, each calculated by FLASH for the spectrum of hot leg rup-tures. Following blevdown a 20-O

!

i O' .

1., i

'#' 4 1h-3cb (Revised 1-3-63) l

!

• s.

~

region SLCMP :odel was used to st::ulate the core ther=al tran-sient. Bis si=ulation includes fuel heat 6eneration, =etal-water reaction, and quenching when the injection vater pro-h vided cooling by contact with the core.

As any given se5=ent reached k,800 F it was assu=ed to drop into vater below the core and release all heat down to a datum of 281 F. Also, it was assumed that 10 per cent additional zirconium-water reaction occurred. '4 hen the water covered ap-prcximately 25 per cent of the core, the surface heat transfer coefficient from all the core clad to the water was assu=ed to be 100 Stu/hr-ft 2 -F. The determination of water level vas based on injection flov ard included the effects of boiloff.

Assuming a pool boiling ccefficient of 100 for the whole core when only 1/4 was covered ves conservative for reactor build-ing pressure analysis because it ec= pressed overall energy transport into the shortest credible pericd.

Heat was also released from the hot =etal of reactor coolant system and the reactor vessel internals. During the blevdown period a surface heat transfer coefficient of 1,000 Btu /hr-ft2 .

F was used. After blevdown this coefficient was changed to 100 Stu/hr-ft2_y for the metal telow the leak and 3 Stu/hr-ft2 .

F above the leak. D e coolant sink te=perature was provided by FLASH for the blevdown period and assu=ed to be 281 F there-after. The internal heat transfer of the =etal was based on a g multilayer finite difference model. The whcle process of reac- w ter coolant system =etal heat transfer was simulated with a digital ce=puter program.

All heat transferred from the core and the reactor coolant sys-tem =etal was assu=ed to generate steam without takira credit for the subcooled condition of the injection water (except for that portion which was boiled off) until the reactor vessel vas filled to the leak nei6ht. Thereafter all energy was removed by lov pressure injection flow of subecoled water, and the l energy release to the reactor building at=csphere terminated.

No delay was assu=ed in transportin6 steam to the reactor build-ing. The heat l' rom hydro 6en burnin6 was added directly to the reactor building as hydrogen was evolved frem the =etal-water reaction.

Both reactor inlet (cold) and reactor outlet (het) line breaks were analyzed with FLASH. However, a ec=plete analysis was

=ade only for the hot line breaks since they provided for the

=ost rapid heat transport from the core. This was true because the hot line breaks had longer blevdown and better heat trans-fer during blevdown than did the cold line breaks.

The results of calculations of fluid and heat transport to the

'

.

reactor building as dete: ::ined by FLASH, SLUMP, and other ana-lytical =edels were used as input to the reactor building pres-g sure analysis pregram, CCIC"iMPr.

r.r ' " '

• 333 1k 40 0001

_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

Table 14-6 Reacter Operating Conditions for Evaluation O Parameter Value Reactor Ccolant System Pressure, psig 2,185 Reactor Ccclant Average Temperature, F 584 Reacter P:ver Level (ultimate), mit 2,535 1 Reactor Ccolant System Mass, lb 519,173 Initial Reactor Building Temperature, F 110 Initial Reactor Building Relative Humidity, ) 0 .

Initial Reacter Building Pressure, psig , O In calculating the reactor building pressure. it was assumed that the average temperature of t!".c building atmosphere and 1 structural =aterials was 110 F. Upon release of hot reactor coolant, the steel and concrete act as heat sinks which re-duce the reactor building pressure. The heat sinks considered in this analysis are specified in Table lk-7

,

.

-

l 1

!

i O

0001 334 l

')

-

{ ;,s i( 14 kl (Revised 7-21-67)

....-

r

l i

l

!

l O

l

,

Table 14-7 Reactor Building Structure Data for Analysis of Time-Dependent Reactor Buildir4 Pressure l

Parameter Value Reactor Building Free Volume, ft3 2,000,000 Exposed Liner Plate Surface, ft 2 87,220 Mass, lb 1,238,000 Dcme and Wall Liner Sickness, in. 0 375 Refueling Cavity Liner Thickness, in. 0.250 1

Reactor Building Ccncrete Enclosure Consisting of a 3-ft-Thick Dcme and 3-ft.,6-in.-nick Valls and a 2-ft-Thick Floor &

Wall and Dcme Surface, ft2 81,700 W Wall and Dome Mass, lb 41,100,000 Exposed Floor Surface, ft 2 11, coo Exposed Floor Mass, lb 3,190,000 Structural and Miscellaneous Steel Exposed to Reactor Building At=osphere Surface, ft2 go, coo Mass, lb 500,000 Internal Concrete Surface, ft 2 102,280 Mass, lb 23,398,5co

Refueling Cavity Concrete Surface, ft 2 5,520 Mass, lb 3,001,500

\'

l 1

.

..,}

l . O

t. ; '. : on',-

14 h2 0001 '335 no

Heat transfer frem the reactor buildin6 atmosphere to the steel liner vas calculated using a condensin6 coefficient of 620 Etu/hr-ft2 .7 (V3 until a totcl heat input of 110 Btu /ft2 had been achieved. There-after, a condensing coefficient of 40 Stu/hr-ft -F2vas used.

For heat transfer from the reactor building at=osphere to the con-crete, a condensin6 coefficient of k0 Stu/hr-ft -F 2 vas used. For heat transfer f em the sump water to the concrete floor a coefficient S

of 20 Stu/hr-ft -F was used. No credit was taken for heat transfer to reinforcin6 steel in the internal concrete structures.

For structural and miscellaneous steel, one heat transfet section with an equivalent thickness of 0.15.1 in. was uced. Condensin6 coef-ficients of 620 and LO Stu/hr-ft 2-F vere used.

Followin6 a loss-of-coolant accident, the reacter building is cooled 1 by three reactor buildin6 emer8ency cooling units and a spray system.

Each cooling arran6e:ent has a heat removal capability of 240 x 106 Btu /hr at a vapor temperature of 281 F. Two ecoling units plus 1,500 Cpm sprays, or 3,000 gpm sprays, provide ceoling that is at least equivalent to the three reactor bu11 din 6 emergency cooling units.

Each system is designed so that it alone can protect the reactor buildin6 against overpressure. Each syste= van assumed to operate on e=ergency power and was delayed until 35 see after the rupture occurred.

(DELETED)

Figure 14-h3 shows the reactor buildins pressure for com s ) anceofa36in.IDreactorcoolantsystempipe(14.1ftgletesever- ruptu e area) with 6,500 gym of borated water injection into the reactor cool-ant system beginning 25 see after the rupture. Reactor building cool-

.in6 is provided by three emergency cooling units. The peak pressure !I resultin8 frcm this accident occurs 181 see after the rupture at a value of 52.1 psig.

An analysis of the reactor building pressure for the 36 in. ID pipe 1 rupture and spray ecolin6 of the building has also been perfor ::ed to de=enstrate the effectiveness of this system. Initially coolant for the building sprays and for injection to the cere is pu= ped from the borated water storage tank. 'a* hen water from the borated water stor-cge tank is depleted, the water collected in the reactor buildin6 su=p is recirculated throu6h the reactor building sprays and threugh the decay heat removal coolers to supply the low pressure injection water ne result is an increased injection and spray water tempera-ture. No boilin6 of the injection vater results frem this decrease in subecolin6 The reactor buildin6 spray effectiveness vill decrease.

The net result is a decrease in the energy removal rate from the reac-tor building at=osphere.

Se requirements for coelin6 the water recirculated from the reactor 1 l

building sump to the reactor building spray system are set by the de-l sign basis of this system. The design basis is to maintain the post-t

!

(~ accident reactor bu11 din 6 pressure below the design value. 31s criterion can be =et by spraying the sump water directly into the re-d (

• actor building atmosp2.ere without additional cooling, other than that provided by the decay heat removal system.

1h.h3 (Revisei 7-21.C) bl)

!

1 1

The water te=perature in the reacter building su=p during the recirculation phase of a loss-of-ccolant accident is =aintained belov saturatien te=perature by the decay heat re=cval ecclers.

lh I These coolers reduce the te=perature of water recirculated to  !

the reactor vessel and returned to the reactor building su=p.

The heat transfer surface of these ecolers is set by the nor-

=al operating conditions under the decay heat re= oval opera-tion = ode. The ecoling capability of this = ode of operation vill =aintain the reactor coolant at ik0 F cr less at 20 hours2.314815e-4 days <br />0.00556 hours <br />3.306878e-5 weeks <br />7.61e-6 months <br /> after extended rated power operation and is in excess of that required under accident conditices. The perfor=ance of these coolers at varicus inlet te=peratures is shcvn in Figure 6 h.

Figure ih-h3-a shows that the reactor building pressure decays 6 to less tnan 5 psig in 2h hours. For ec=parisen purposes and to show that the effect of spraying cooler vater into the re-actor building is sma'1, a second curve is presented on Figure IL-k3-a which is based upon a spray recirculatien cooling rate of 100 x 100 Stu/hr (approxi=ately equivalent to tvc decay heat re= oval ecolers ) at a su=p te=perature of 195 F. (This is the te=perature of the su=p when recirculation to the sprays begins.) Figure lk k3-b shows the te=perature of the reactor building and su=p coolant for the two conditions.

These curves de=cnstrate that cooling of the recirculated spray water has no effect on peak building pressure and only a =iner effect on the rate of pressure decay during the first 2k hours.

Accordingly, it is concluded that no cooling of the recirculat-g T

ed sprs/ vater is required for this accident.

Figures ih-kh through lk-h8 show the reacter building pressure for the other rupture sized analyzed with the sa=e cooling

,

l l

337 000l %g

,-

<

t

['r .

• • I *. ;t ' ;, '

lk L3a (Revised 1-2-63)

capability as the 14.1 ft2 rupture above. A su==ary of the g input para =eters and results for :he spectrum analysis are V tabulated in Table IL-8.

A 3.0 ft2 rupture area results in the highest postaccident re-actor building pressure (see Figure ik L5).

Figures ik k9 and ik-50 shav the reactor building energy in-vertory g as a function of time after rupture for ik.1 and 3.0 ft rupture areas with three e=ergency coolers opernting. 1 These curves show the effectiveness of the reactor building structures e.nd emergency cooling units.

Figures 14-51 and ik-52 shev the reactor building vapor te=-

peratures and su=p te=peratures folleving ik.1 and 3.0 ft2 ruptures..

The peak reactor building pressure shown in this evaluation for the spectru= of leak sizes results is 52.1 psig and is the result of a 3.0 f 2 rupture in 1:he reactor outlet piping. The reactor building design pressure is 55 psig and a design =ar-gin of = 3 psi exists. '41th core flooding tank cperation this

=argin would be increased further.

The above analyses conservatively assume that the hydrogen 1 liberated vill burn at the rate for=ed, and that no core flood-1 ing tank operation occurs. The core flooding tanks limit the O a=ount of :irconiu=-vater reaction to 0.005 per cent for a 36 in. ID pipe rupture, and the potential hydrogen energy release is approximately k,000 Btu. This a= cunt of energy vill not significantly affect reactor building pressure if ignition is delayed or if the hydrogen burns as formed.

For the case of no core flooding tanks, as used in the above reactor building design pressure evaluation, the amount of

=etal-vater reaction is sc=ewhat greater. The zirconiu=-vater reaction begins at 40 see and stops at 130 sec, by which ti=e

'

the 6,500 gp= of injection flev provides sufficient coolant inventory to the reactor vessel to recover the hot spot and quench the reaction. The stea= flow during this period is as-su=ed to provide the transport =echanis: for the hydrogen gen-ersted. The' resultant concentration of hydrogen (at time of

=aximu= =etal-vater reaction rate) in the stea= leaving the reactor vessel is 7.2 volume per cent. This concentration is belev the fla== ability limit. Further dilution vill occur as the stea= enters the reactor building, and cc=bustion vill not occur, even as the reacter building is depressurized.

Criterion 17 of the AEC General Design Criteris states that the contain=ent (reactor building) be designed to accc==odate the largest, credible energy release including the effects of cred-ible metal-vater reactions uninhibited by active quenening sys-

[]

'u te=s. Although the evaluation of the emergency injection sys-tems de=custrates that only a s=e.11 a= cunt of =etal-vater 9.'

.

" '

0001 338-IL kk (Revised 7-21-67)

reactica can occur, the case of no injection flow has been evaluated in response to the above criterion. This case as-su=ed that, after blevdevn, the reactor vessel vould have g water up to the bottc= of the core. The core was allowed to heat up by decay heat and =etal-vater reactica heat.

Stea= flow rate-li=iting of the reaction was not considered so long as any water was assu=ed to be in the vessel. If and when the clad reached the =elting te=perature, it was assu=ed that the whole region slu= ped into the bottc= of the vessel with the attendant reaction of 10 per cent = ore of the re=ain- ,

ing :1rconiu= and with the release to the reactor building of all sensible and latent heat above 281 F.

Upon ec=pletion of boiloff, beat input to the reactor building was assu=ed to cease, Figure ik-53 shows a reactor building pressure of 53.2 psig at 220 seconds, the ti=e at which the reactor vessel boils dry. This peak pressure is belev the 55 psig design pressure of the reactor building.

O

.

B

.. .

O p 3 'i

, ,

, . ~,as.

. 0001 339 ik kha (P.evised 7-21-o7)

O Table lh-8 Su:: nary of Reactor Building Pressure Analysda for Reactor Building E=ergency Cooling (2ho x 108 Btu /hr)

Rupture Size, ft 2 14.1 85 30 2.0 1.0 0.4 Reference Figure No. IL h3 lk kh ih h5 lh ,h6 lh h7 lh h8 Time 310vdevn Ends, see 15 20 h6 68 141 351 Time Icv Pressure In-jection Begins, see 25 25 39 59 121 321 Fraction of Core Zr-reacted 0.08 0.05 <0.01 zero zero zero Time Zr-reaction Begins, see ho 50 130 -- -- --

Time Zr-reaction Ends, see 130 130 131 -- -- --

Time to Reach Peak '

Pressure, see 181 181 kl 67 181 261 Peak Building Pressure, psig 52.0 51 3 52.1 50 7 h8 9 43 2 Vapor Temperature at Peak Pressure, F 278 277 278 276 274 266 Sump Temperature at Peak Pressure, F 232 230 221 215 210 196 Conditions for All Cases

a. 500 gpm high pressure injection

b. 6,000 gpm low pressure injection

c. Reactor hot leg rupture

d. No core floodin6

e. No reactor building sprays

f. Three emergency cooling units start 35 sec after the rupture.

n-

?j,). ,4,i.

00hl 340

, ,

lk h5 (Revised 7-21-67)

_ , - .

_ _ _ _ _ _ _ _ _ _ _

c. Reacter Building Zireenium Reaction Capabilitz _

In order to dete mine the theoretical ulti=cte :ir:eniu= reaction capability of the reactor building a series of hypothetical accidents h

was investigated.

31cvdown was based on the 14.1 ft2 leak case. Heat transfer frc= the core and all reactor ecolant system =etal belev the leak height was assu=ed to transfer to a 261 F sink based en a surface coefficient of 50,000 Stu/hr-f t 2 -F. For reactor ecolant system =etal above the leak height 5 Stu/hr-ft 2-F vas used.

Available core heat censisted of the initi:1 stored heat, the equiv-alent of two rated pcVer seconds, decay heat, and =etal-vater reaction heat, which was added at arbitrary linear rates. The total heat transferred frc= the core and reactor ecclant system =etal was as-su=ed to produce steam frc= vater initially at the saturated condi-tion. Rydrogen recc=bination energy was added to the reactor build-ing as superheat at the rate of hydrogen production from the 21r00-nium-unter reaction.

A series of calculations for each of the various ecolin6 capacities was =ade varyira the energy input rate, i.e., Zr-H 2 O reaction rate.

A ; per cent per second zirconium-water reaction produces 1.173 x lob Btu /see of =etal-vater energy and 0 902 x 106 Stu/see hydrogen recc=bination energy. In all cases the energy was input at a linear rate beginning 10 see after the rupture. The e=ergency ecoling units and spray ecclers were started 35 see after the rupture. The "ti=e to ec=plete reaction" is the ti=e it takes to reach reactor building g

design pressure (55 psig).

S e results of this study are presented in Figure IL-5k. This a= cunt of allevable :irceniu= reaction at any time after bicudown depends upon the a=ount cf reacter building ecoling in operation. The 02-pability curves show that at approximately 10 see, when the blevdevn pressure peak cecurs, the reactor building Oculd accept an instan-taneous circonium-vater reaction of k per cent. This capability in-creases greatly af ter the blevdown pressure peak vich reactor build-ing cooling eqtipment in operation.

'41th three emergency ecoling units in operation a 100 per cent reac- 1 tien in 4,200 sec will not exceed the design pressure of 55 psig.

~41th three e=ergency cccling units and two sprays operating, a 100 per cent reactica in 1,420 seconds will not exceed the design pres-sure.

-

mp. . ; . m 0001 3 o ik k6 (Revised T-21-67)

.

14.2.2 3 5 Environ = ental Analysis of Loss-of-coolant Accidents t Safety injection is designed to prevent si6 nificant clad =eiting in the event of a loss-of-coolant accident. The analyses in the preceding sections have de=onstrated that safety injection vill prevent clad =eltin6 for less-of-coolant accidents resultin6 from reactor coolant system ruptures ranging in sice from s=all leaks to the eccplete severance of a 36 in. ID =ain coolant pipe. Without clad melting, only the radioactive material in the coolant at the time of the accident plus some gap activity is released to the reactor buildin6

<

The environ = ental consequences from a loss-of-reactor-coolant accident are ana lyced by assu=ing that 1 per cent of the fuel rods are defective before the re lease of reactor coolant to the reactor buildin6 Table 11-3 lists the total activity in the coolant. In addition to the coolant activity, the activity associated with the Gap of all fuel rods is also assumed to be released. Cal-culations indicate that 77 per cent of the fuel rods vill have some point alon their lengths with te=peratures in excess of 1,200 F at the ti=e of core flood-ing tank injection. While perforation of fuel cladding vill require some ti=e

~

$ t is conservatively assu=ed that all of the fuel rods release their gap activ ity during the accident.

Half of the iodine released is assumed to plate out on exposed surfaces in the reactor buildin6 The other half is assumed to re=ain in the reactor buildin6 at=csphere where it is available for leakage. The sodium thiosulfate in the reactor building spray reduces the airborne iodine as described below. Of the iodine available for leaka6e, 5 per cent has been conservatively assumed to be unavailable for removal by the spray.

The rate at which the ele = ental iodine can be removed from the reactor buildin, at=osphere by the reactive spray is calculated using Griffith's methods.(14)

This method is based on the work of Taylor,(15) who showed that the rate at which elemental iodine can be transferred into reactive solutions is controlle:

by the gas film resistance, and on the work of Ranc and Marshall,(16) who showed that the equation below can be used to calculate the = ass transfer co-efficient when the rate of transfer is controlled by the gas film resistance:

DpM 7

-

dvp 1/2 1/3 '

k., =

"

2 +0.6 -

MedP ,

F Dp ,

where k*, = gas fils = ass transfer coefficient , gs/c: 2 -see-atmos 2

D = diffusivity of iodine in air, em /sec c = density of air, g=/cm M7 = =clecular weight of iodine, gn/g=-=cle Ms= =ean =olecular weight of the air-iodine mixture in the boundary layer g P = partial pressure of air in the gas fi1=, at=cs b d = drop diameter, em

.

'

l. la + 0001 14 kT (Revised 7-21-67)

• 3 4:2 1

l

.

v = relative velocity between the drop and the gas phase, or approximately the terminal velocity of the drop, em/sec g

= viscosity of the air Since the mass transfer of iodine is gas-film-centrolled, kg is approxi-mately equal to Kg (below), and the foregoing equation can be rewritten in terms of the velocity of deposition, V g:

'

DM dv 1/2 '/3' V g= KG= 2 t 0.6 p where V verall vel ity fdeposition,em/see 6=

R = universal gas constant = 82.057, at=os-em3/x-gm-mole .

T = absolute temperature, X Kg = overall = ass transfer coefficient, g=/cm2-sec-Stmos Since the maximum possible iodine concentration in the large volu=e in .

the reactor buildin6 is less than 10-7 gm/ce, the partial pressure of air in the gas film, P, can be taken as the total pressure, and the mean molecular weight, 4, can be taken as the molecular weight of air, M

• A h

If the gas equation is used, the equation may be simplified somewhat by substitutin6 MA /RT for p/P, as follows:

D dyp / /3' V 2 + 0.6 6=d,

-

4 Dp ,

The surface area of drops available for iodine abscrption can be calcu-lated frem the next equation, which is based on the assumption that all the drops are spherical and have the same dia=eter.

S= = =

d dv

,

'

Ed3 o

,

2 s suspended in the 6as phase, em S = spray F surfaceflowarea of em3 rate, drop /see l e = drop fall time, see I d = drop dia=eter, em

'

H = drop fall height, em

! v = drop fall velocity or ter=inal velocity, e:/sec

!

i s p. .y % 0001 343 l ik 48 -

l l

.--

_

If there is a large excess of che=1 cal reagent to react with the iodine O and convert it to a nonvolatile form with little or no tendency to return to the gas phase, then the iodine removal rate can be expressed by

/V S jd , ,; 1 7,,s7 dt (V e where Ve = free volume of reactor building, em3 or ft3

.,

As = iodine removal time constant, hr-1 The fraction remainin6 in the reactor building atmosphere is expressed as a function of time by the solu

• ion of the equation above as follows:

I "' As t

-=e lo

.

vhere 1I = fraction of initial inventory remaining o

t = spray time, hr When the specific parameters for the Three Mile Island Nuclear station are used:

F = 3,000 sPm v = 397 cm/see H = 90 ft V 6 = 5 06 cm/sec V = 2 x 10 6 ft3 6V FH As = VedV = 25 3 hr-1 d = 1,000 microns These iodine removal calculations have conservatively corrected the iodine deposition velocity (V ) to the peak temperature and pressure in the reactor buildin6 Asens!tivityanalysiswasperformedontheiodine removal calcula1, ions, and the results are shown in 14.2.2.4 3 in terms of the 2-hour iodine dose at the boundary of the exclusion area follov-ing an MEA.

Although the reactor building leaka6e rate vill decrease as the pressure decays, the leakage is assumed to remain constant at the rate of 0.2 per cent per day for the first 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />. Thereafter, since the reactor build-ing vill have returned to nearly atmosphere pressure, the rate is assu=ed to be reduced to 0.1 per cent per day and remain at this value for the

.

duration of the accident.

The atmospheric dispersion characteristics of the Station site are de-tj

' scribed in 2 3 The site dispersion factors for the duration of the ac-cident are listed in Table 2 -L. A breathing rata cf 3 47 x 10-4 m3/sec s

.

, s.a 0001 344 c . tu u9

is assumed for the 2-hour exposure. For the 2h-hour exposure, a breath-ing rate of 3.kT x 10 h 23/sec is assu=ed for the first 8 hours9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br />, and a g rate of 1.Th x 10-k/see is assu=ed for the remaining 16 hours1.851852e-4 days <br />0.00444 hours <br />2.645503e-5 weeks <br />6.088e-6 months <br />. For the w 30-day exposure, a breathing rate of 2.32 x 10 4 m3/ see is assu=ed. l1 The iodine doses to the thyroid per curie inhaled are obtained from the values given in TID-lh8hh:

I-131 1.h8 x 10' rem per curie I-132 5.35 x 10 re= per curie ,

I-133 h.0 x 105 rem per curie I-13h 2.5 x 10 rem per curie I-135 '1.2h x 10 5 y,,p,y cuyg, Figure ik-55 shows the total integrated dose to the thyroid as a fanction of diatance from the reacf.or building for 2-hour, 2h-hour, and 30-day ex-posures. 'The total thyroid dose at the 2,000 ft exclusion distance is 2.0 rem for a 2-hour expost re, 6.6 rem for a 2h-hour exposure, and 10.7 1 rem for a 30-day exposure. These doses are vell below the guideline val-ues of 10 CFR 100. The direct dose from this accident is insignificant since it is several orders of =agnitude below 10 CyR 100.

The environ = ental consequences of the loss-of-coolant accident were eval-usted without taking credit for ter=ination of the reactor building leak-age by the penetration pressurization system and the fluid seal system.

g Operation of these systems vould reduce the total integrated dose at the 1 exclusion distance for the entire duration of the accident to only 0.1 rem.

Ik.2.2.3.6 Effects of Reactor Building Purging At times during the nor=al operation of the reactor, it may be desirable to purge the reactor }uilding while the reactor is operating. In the event a less-of-coolant accident vere to occur during purging operations, activity would be released to the environ =ent. The purge valves vill be ec=pletely closed in 5 sec. During this time, assu=ing a 36 in. ID, double-ended rupture, essentially all of the reactor coolant vill have l1 been blevn down. The activity in the reactor building is due to the re-actor coolant activity aft er operation with 1 per cent failed fuel. For i this case, 0 53 per cent o f the reactor building atmosphere vill escape l1

!

through the purge valves bitfore they close, corresponding to a release of 3.0 etuivalent curies os' iodine-131. This analysis assumes unrestrict- 1 ed flov through the purge line for the full 5-second closing time. No reduction in flow is assumed as the valve closes, and therefore the re-suits are conservative. The release of this iodine results in a tot:1 1 integrated thyroid dose of 0.8 rem at the exclusion distance. This dese, when added to the thyroid dose for a loss-of-coolant accident without O

.

,

. .

M.. ".*

lh-50 (Revised 7-21-67) Ol'11

' 345

.-. _ ._ --

purgi 6, is well below the 10 CFR 100 guidelines. Therefore, purging operations can be perfor=ed during reactor operati;n.

14.2.2.4 Maximum Hypothetical Accident 14.2.2.4.1 Identification of Accident The analyses in the preceding sections have demonstrated that even in the event of a loss-of-coolant accident, no signi;'icant core melting will occur. However, to demonstrate that the opers. tion of a nuclear power station at the proposed site does not present any undue hazard to the general public, a hypothetical accident involting a complete core melt-down is evaluated. No mechaMsm whereby such a core =eltdown occurs is postulated since it would require a multitMe of failures in the eD61-neered safeguards provided to prevent its cccurrence. As a result of the meltdown, fission products are assumed to be released frem the core as stated in TID-14844, namely, 100 per cent of the noble gases, 50 per cent of the halogens, and 1 per cent of the solids.

Further, 50 per cent of the iodines released to the reactor building are assumed to plate out. Other parameters, such as meteorologien1 condition::,

I

'

iodine inventory of the fuel, reactor building leak rate, reactor building iodine removal rate, etc., are the same as those assumed for the loss-of-coolant accident in 14.2.2 3.6. The average iodine inventory, in terms of equivalent curies of iodine-131 available fer leakage at different time periods after the accident, is as follows:

0 to 2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br /> 28.7 x 106 curies O to 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> 22.8 x 106 curies 1 to 30 days 5 1 x 106 curie, 14.2.2.4.2 Analysis and Results of Environmental Analysis Within one minute following the accident, the isolation fluid block sys-tem and the penetration pressurization system have acted and tecnicated leaka6e from the reactor building. Durin6 this one minute time period approximately 2-dose-equivalent curies of I-131 have leaked frem the re-actor building. This value is determined in the follevin6 manner. A SLLMP computer case was run in which no injection water was added to the reactor vessel. During the first minute followin6 initiation of the acci-dent, 5 per cent cf the fuel has reached 3,000 F. At this temperature, it is conservatively assumed that all of the iodine that vill ultimately be released is released to the reactor building. It is further conserva-tively assumed that this quantity of iodine is available for leaka6e over the entire one minute period. No iodine reduction by sprays is assumed l during this initial one minute period. The release of 2-dose-equivalent curies of I-131 produces an integrated dose to the thyroid of 0 51 rem at the 2,000 ft exclusion distance.

As an upper 11=1t on the consequences of this accident, it can be postu-lated.that the leakage prevention systems are not completely effective in

! terminating all leakage. For this condition, leakage is assumed to l

i 6: W 0001 346

.

continue at the design leak rate. F16ure 14-36 presents the total inte-grated dose to the thyroid as a function of distance frem the reactor building for 2-hour, 24-hour, and 30-day exposures. It can be seen that h the 2-hour thyroid dose of 86 rem at the exclusion distance of 2,000 ft and 72 rem at the 2-mile low population :ene distance are less than the guideline values of 10 CFR 100.

The direct dose to the whole body follov.in6 the accident is shown in Figure 14-57 No significant dose exists frem this source at the exclu-sion distance.

The dose to the whole body from the passing cloud has been calculated using the same meteorological conditions used for detemining the thy-raid dose. The 2-hour whole body dose at the exclusion distance is only 3 1 rem, and the 30-day dose at the 2-mile icv population zone distance is 2.15 rem.

14.2.2.4 3 F.frects of a Sensitivity Analysis of the Reactor Building Sprays for Iodine Removal A sensitivity anstlysis on the calculation of iodine removal was performed using the reactive chemical sprays in the reactor building. The results are shown in Table 14-9 in tems of the 2-hour iodine dose at the exclu-sien area boundary following an MHA.

O

,

.

O

g.  ; m 0001 347

.... lu-52

._ . _ _ _ _ __ _ .

O O C

.Q s;e

.-

we

.

_

• '

htle 14-9 Sensitivity Ar.alysis Showing the Effect of Parameters on the

'IVo-Itotr Iodine Dose at the Exclusion Area Boundary Following an DetA latine Iodine -

Removal Removal Drop Drop Fs11 Velocity of t' e rodine III . Time los tne(2)

Case Size, Velocity, Ieposition, Temp, Press., "onstant, lue, ' Constant, Dose, No. microns en/sec en/sec F palg Is*I rem hr*I rem Hemarks 1 1,000 39rf 5 06 281 55 12.65 110 25 3 M operetton of the reactor building spray system at anstuum building tempera-ture ami pressure.

2 1,000 397 6.44 212 25 16.05 100 32.1 & operettoo of the reactor building spray-system after 8"

partial cooling, about 1 v.

7 hour.

W 1,000 100 0 28.8 3 397 11 55 83 57 6 72 operation of the reactor building spray system after coo.ing to ambient cond1-tions.

4 1,000 3,'J10 14.8) 281 55 37 228 7 39 145 Effect of drop falling at 10 times its te minal velocity.

5 2,000 649 4.26 281 55 3.25 256 6.50 156 Effect of large drop stae.

6 200 76 7.24 281 55 471 61.5 942 61.5 Effect of small drop stae.

F3r all cases, reactor buildtog free volum 2 x 100 ft3 , and drop fall height

• 90 ft, 50Lest (1) Flow rate of sprays .1,$00 gym.

C (2) Flow rate of sprays = 3,000 gpe.

C C

-

V:

A CD

'

>

7 -.

.

14.2.2.4.4 Effect of Rainout During the Max 1=um Hypothetical Accident g

To further evaluate the suitability of the site, the effects of rainout on surrounding drinking-water reservoirs following the marimum hypotheti-cal accident were analyzed. Calculations were =ade for the case of a continuous 24-hour rainfall covering the general area of the reservoir and the site. The max 1=um rainout rate as a function of distance is cal-culated frem(171 2 2 Qne

-(yja,Y) xery V2 x where um = -av bum rainout rate, curies per see per m, x = downwind distance, meters ey = horizontal dispersion, meters y = crossvind distance frcm plume axis, =eters Q, = release rate, curies per sec e = 2 718 h m e equation above is conservative, since the results do not consider the vind speed or vertical distribution in the cloud. The vind direction is assumed to remain toward the respective reservoirs listed in F1 ure 6 2-12 for the 24-hour period with the plume centerlines unifomly distributed over this section. Rainout is assumed to occur under Pasquill "D" condi-tions, which is typical for a rainy day.(17)

The average release rate frcm the reactor building durir4 the 24-hcur period following the accident is 0.027 equivalent curies of iodine-131 per sec. Using the foregoin6 equation, the maximum iocine rainout is calculated by assuming that all of the iodine that has rained out re=ains in the reservoir and is not affected by rum'ff. De iodine rainout was calculated for the reservoirs listed in Fl ure o 2-12. In all cases the iodine concentration in the reservoir averaged over one year is less than 1/9themaxi=umpermissibleconcentrationsof10CFR20.

The effect of rainout on the Susquehanna River has been calculated by as-su= ,ng that all the iodine released frem the reactor building 12 deposited unifomly in the river. Assuming that this activity =1xes in the river and that a child with a 2-gram thyroid continually drinks 300 ml per day of the contaminated water, the total dose to the thyr-id has been calcu-lated using the methods of TID-14844. For an adult with a 20-gram thy-roid, a drinkin6 rate of 1200 ml per day is used. If the contaminated river water is consumed for the life of the accident durin6 a period of g minimum flov (2,000 cfs), the total integrated dose to a child's thyroid B i" i "

• 14- 5a 0001 349

___________________-.______-----_a

.

O, i

is 2 7 rem and to an adult's thyroid is 1.1 rem. These doses are well below the li=its of 10 CFR 100.

14.2.2.4 5 Effects of Engineered Safeguards Leakage During the Maxi =um Hypothetical Accident An additional source of fission product leakage duri 6 the maximum hypo-thetical accident can occur from leakage of the engineered safeguards external to the reactor building during the recirculation phase for lon6-teng core cooling. A detailed analysis of the potential leaka6e from these syste=s is presented in 6 3 nat analysis demonstrated that the max 1=u=leaka6eisabout5,000cc/hr.

It is assumed that the water being recirculated from the reactor bu13dir4 sump throu6h the external system pipin6 contains 50 per cent of the core saturation iodine inventory. This is the entire amount of iodine release frem the reactor cooling system. The 50 per cent escaping frem the reac-tor coolant system is consistent with TID-lh844 The assu=ption that all  !

the iodine escaping frcm the reactor ecolant system is absorbed by the  ;

vater in the reactor building is conservative since =uch of the iodine re-leased from the fuel vill be plated out on the bu11 din 6 valls. The activ- l ity in the recirculation water is equal to 0.037 equivalent curies of I-131 per cc of water. The iodine is chemically bound to the sodium thio-sulfate, c'.. v111 not be released to the atmosphere. However, it is con-

.

servatively assu=ed that icdine release does occur. Since the tempera-ture of water in the reactor buildin6 sump is less than 200 F vhen recir-q culation occurs, the iodine release is calculated using a gas / liquid par-V tition coefficient of 9 x 10-3.

Leakage frem the auxiliary building is caused by exfiltration. The most restrictive case for a ground release occurs during inversion conditions.

It is assumed that the building leaks at the rate of 100 per cent per day with atmospheric dilution occurrire in the vake of the building. For this building leak rate and the inversion condition, the iodine vill pro-duce an integrated dose to the thyroid of 0.009 res in 2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br /> at the 2,000 ft exclusion distance.

O

, n.

.

i c4 3_;3 0001 350

~

.

14 3 REFERE*TCES (1) Watson, L. C., Bancroft, A. R., and Hoelke, C. W., Iodine Contain=ent

,

by Dousing in NPD-11, AECL-ll30.

i (2) Styrikovich, M. A., et,al_., " Transfer of Iodine from Aqueous Solutions l to Saturated Vapor", Soviet Journal of Atomic Enermy 17, July 1964.

I (3) Dispersion of Soluble Radioactive Material in Water, CF-58-3-109

~

l (4) International Sy=posium on Fission Product Release and Transport

! Under Accident Conditions, Oak Ridge, Tennessee, April 1965 l

!

(5) Lii=atainen, R. C., et al., Studies of Metal-Water Reactions at High Temperature, ANL 76250.

,

'

(6) Acker=an, R., et al., "High Temperature Vapor Pressure of UO "i 2 I Jou m al of CheH eH Physics, December 1956.

l (7) Reactor Development Program Progress Report, ANL-6912, June 1964.

l (8) AEC Research and Develop =ent Reports, WIGL2 - A Program for the l Solution of the One-Dimensional Two-Group, Space-Time Diffusion Equations Accounting for Te=perature, Xenon and Control Feedback, WAPD-TM-532, October 1965 (9) Margolis, S. G. and Redfield, J. A., FLASH: A Program for Digital Simulation of the Loss-of-Coolant Accident, WAPD-TM-53k, May 1966. h (10) Grenda, R. J. and Patterson, J. F., "The Velocity of Rising Steam in a Bubbling Two-Phase Mixture", Transactions of The ANS 5, No. 1, p 151, June 1962.

(11) Possible Zirconium Water Reactions in Water Reactors, AEC Regulat e/

Staff Sy=posium, April 27, 1965 (12) Wagner, R. J. and Finnegan, L. J., "An Analytical Model for Predict-ing the Pressure-Te=perature History Within a Contain=ent Vessel in Response to a Ioss-of-Coolant Accident", Phillips Petroleum Cc=pany, l Atomic Energy Division, Idaho Falls, Idaho, Presented at ANS Meeting,

, Wa:hin6 ton, D. C., November 1965 l

l (13) Quinn, E. P., Forced-Flow Heat Transfer to High-Pressure Water Beyond the Critical Heat Flux, ASME 6 CWA /HT-36, Nove=ber 27, 1966.

(14) Griffiths, V., The Removal of Iodine from the At=osphere by Sprays, AFSB (S) R 45, 1963 (15) Taylor, R. F., " Absorption of Iodine Vapor by Aqueous Solutions",

Chem.Eng.Sa.,X,,No.1/2,pp68-80, April 1959 O

T s"

• 0001 351

,,,

. _ __. . . . . - . _ _ _ _ ._ __ ._ _- _ _ _ . _ _ _ _ . _ __ . _ . . ..

.

]

,

.

j (16) Ranz, W. E. and Marshall, W. R., Chem. Eng. Progress, M , 141, 173,

!

1952. -

!

1 i

'

(17) Culkowski, W. M., Deposition and Washout Calculations Based on the Generalized Gaussian Plume Model, ORO-599

a 1,

.,

3 b

I

.

i

?

t i

O I I 1

4 h

!

i i

.

i i

• 5 E .

!O

4 ..

l 0001 352

,

_ 1k-57 i

_ . . - . _ . _ _ . . . . . . _ . . . _ , . _ - - , _ _ . _ - . . . . . . _ _ ~ _ _ - . . . . . _ _ , . . . . . . -

, , , , , , , , , ,

I i i r , i i , e i ,

O NEUTRON 30 i

i

i i i

, ,

. , i i

i i

,

,

i Al vi i

, -

1 ! i

,

, ,

i i i i

, , ,

i I

, , , , , f i , ,

i i i POWER,% 20 , , , , , , , ,

._7... ,g.  ; , j g . 3, _. . . ._7_ 7 7 ,g . .

,

, ,

10 i , , , , y i i i t i i i , , i i

' ' ' '# ' ' ' ' ' ' ' ' ' '

o

>

i , i i i i i i i , , i i i i i 20 , , , , , , , , , , i , , . . ,

i i 4 , i i . , , ,fNi , i , i i i

'U ' ' ' '

' ' ' ' ' 'N ' ' ' '

THERMAL i i i . .' i / . . i i i i , i-POWER,% o , , . , , , ,  ; , ,  ; i , i i , i i i i i i i i i i i e i ! i i ' i i;i i i i i ! ' I i i i i i ie i <

i ii  !

, , , , i , . , , , . , , i , i i i i i a e i t i i e i t i 4 i i I ' i i

' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' '

200 i t i i ! i i i  !'iiiiii!

i ! !i e i i i i i i i i I,M i i i i i i l FUEL loo i i , , i,,i,,f , ,%, i , , ,

~

' ' ' '

TEMPERATURE o , , ', ' 'i' ' ! '

i i i ! i i i i i i i i i i CHANGE , F i i ! ' '

i i

• i i i i i ' i

, , i i iiii i , !

i iii l i i t i i l i e i I ' i i ! I i i i i i t i i lo e  ! i .

t i e i i i i ! ! i iiii

,  ! . i , i i i i . . .

AVERAGE i i , , i i i if i i i

'

CORE MODERATOR i l, l! C K illlil ,

l TEMPERATURE O - - ' ie ' ' ' i 'i'ii i i ii i,ii i , ii. i ,i CHANGE , F ii i i i , i ! i ,,iiiii ii i i i i e i i 1 i i i ii

! i4 .  ! ,,,, , , , , i i, i ii,i r*- 5 SEC*i i , i ii iiii , , i , , , , .

2500 , iii,,ii .f , , i i i e iiii i i ri i i i i 2350  ; , ,,,, ,j , , , ,

REACTOR i i i i im ,, i SYSTEM 2200 , , i i i i e i i i i i e i i i PRESSURE  ! i i i ! , i i i i

'

PSIA , , l ! b, ! H  ! -

r ! i Fh!

START UP ACCIDENT FROM 10*' RATED USING A 1.27.OK/K RCD GROUP; HIGH PRES $URE REACTOR TRIP 15 ACTU

.

NM FIGURE THREE MILE 00115LADD 353 NUCLE

. _ _ _ . ._.

O V

120 l

N I Maximum Neutron Power ( l 17#.)

100 g

/

a -

-

Maximum Thermal s y Power /-

E E

/

/

/

80

/<

70

/

/

/

60 0 20 40 60 80 100 Initial Power. T. of Rated i

.

I 0001 354 MAXIMUM NEUTRON AND THERMAL I

,

FOR AN ALL ROD WITHDRAWAL ACC FROM VARIOUS INITIAL POWER LE NM FIGURE 14-THREE MILE ISLAND NUCLEAR STA l AMENC.1 (7 2167)

i 6 i i

-

4 r

)

_ >

-

-

%

-

-

l

-

Maximum Temp. in Hottest Fuel -

-

[ Rod (Hot Spot)

!

l i

a )

'o w

• 30

-

/

[/ -

-

g - -

? -

-

2 -

-

1 -

_

l 5 -

-

'

.- 1

.- -

- \

o

-

-

-

O

-

.i 20

-

-

i _

-

-

-

-

- Max. Temp. in Average Fuel Rod -

- _

-

-

-

I _

1

' '# -

-

-

, _

10

' ' ' ' '

O 20  % 60 80 100 Init i Power, 8 l

.

0001 355 PEAK FUEL TEMPERATURE IN AVERAGE I AND HOT SPOT FOR AN ALL ROD WITHDRA ACCIDENT FROM VARIOUS INITIAL POWER LI

,

l N[W FIGURE 14 14.h l THREE MILE ISLAND NUCLEAR STATION AMEND.1 (7 2167)

_ _ _

_ _ _ - - ___ _ _ _ _ _ _ _ _ _

_

O

'

'

t. '

' '

-

! l NEUTRON 15 0 .

r'  ! -

l! l POWER , % ggn _[ l ,ll

' '

l l 50 ,  ; ll,'!,l l _

0 I * ' ' ' '

THERMAL 20 l; [l,ll,,l ll POWER %3 lo iI'N ' ' ' ' ' '

i I i/l i %d i i i l 0

' ' t --

1 i  ! 4 I i 6 i 8 i i it l t i i i e i I i i i iie i l t i I I l l I 1 I l l 1 i i i e I ii i i i i i l I

' ' ' '

200 ' '

!'

'

'

FUEL 'A ' '

100 TEMPERATURE /i N ii 0

CHANGE , F iil ii llU i,,i i i i i i i l

ii iii 4 l t O l'i l

'

.

'

.

' ' ' ' '

10 AVERAGE ' ' ' ' '

5 l

CORE MODERATOR '

/ i i i i TEMPERATURE O i i e i i CHANGE ,F l i l ll l i i i ,,,4 i i i ,

f I I i i l i l 'i l 1 i s e s .

'

~5 SEC*

2500 I ' ' '

REACTOR l^' ' J '

2350

,

' ' '

SYSTEM ' iI ' ' ' ' ' I  !

PRESSURE 2200  ; ; fl l l

'

lll l PSIA ' '

i i

,' ' i

, ,

'

'

i,'.',,',

i .. .i

, , , ,., 1 t i . i i ie i i e i .

0001 356 START UP ACCIDENT FROM 10 4 RATED F USING ALL RODS WITH A WORTH OF 9.5P.t HIGH FLUX REACTOR TRIP 15 ACTUAT

!.Aicatsy"F FIGURE 14 THREE MILE 15 LAND NUCLEAR STATI

-_. _

1 ' J J 1 l 1

. _

_

2 l -

0 I

i I

/ - I e

i 1

l 1 6 O i I

i i

4

.

i i

i 2

l x

up a n

li i Fr h

gl T

[ N

= 3

-

0 ie

/

l i 1

i Hv i n

l i 1 I

-[

g$

.

1 l

6 .

_

AlI l

I

_

i = i 4 c e

-_

s i e

/

)

_

m up k

/ _

-

si k i sr 2 A

(

eT r

I

,

Pl e e t -

h v a _

gA R -

i I H g 4 l

-

a I

0 w i

a

/

I i 1 r I

i d

h O I 6 i t

I 1

h i i 4 W

d R

o 1

i l

o rp

i. t u l n no or l i

2 cG ued no oR 5

-

i o I

l l

l I

i I

i 6

I I

_ i 4 o

s 0 0 0 0 6 4 2 w 1 gg

1 ooc 6(JN O 5 >X

!go

!

3 r>{Qm" jm cm m

~. m> m mom p mlm1-

>Qg t*.

n =@

- i =$m* u9

'

n5c=

-

Y" m m-lC go2Cn{>"et

. _ _ _ _ _ _ _ _ _ _ _ _ - _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

O O O h

• 10 -

l l l l l111 1  !  ! l llll l l 1 l 1111 l l l l llIt 6 2 -^

_

-

_

h -

-

.-

_

g -

1h -

_

e  :  :

g -

_.

m 6 -

-

-

y -

S 4 -

4 -

All M 8 _-

o

, 2 -

-

C }

k 10 Single , /

6 -

Control Rod , -

W 2 Group \ Z y 6 -

h _

1, -

1 _

3('A*'

$ 6 yN  %

A 4 -

-

/

-

-

r- 5 mI g m

m;g 5 2 _

_

E '>' q G

E z ESZ

_Hm 8

O O

CD 1O 1 l l

I ltill i I I Ilill l I l t till I i l i 1 l ll

~

2 4 2 4 6 2 6 2 3 10 6 10-5 yp-4 4 10-3 4 6 yg -2 n yN tA m j>m (J1 Rod Withdrawal Rate, (Ak/k)/sec

$ hc OYN o>C CD M 9aU

_ _ _ _ _ _ _ _ _ _ _ _ - _ - _ _ _ _ _ - _ - _ _ _ - _ _ _ _ _ _ _ _ _ _ _ -. ._ _

-

_ ___ _ _ _ _ . _ _ - _ _ _ _ _ _

l

, O 1

l l

U

/

/ i

/

• 19 e 18 1 17 > /

a 1 16 /

/

A

/A w4,..i is /  ! I i 0 0.4 0.8 1.2 1.6 Trip Delay Ti:ne, sec 0001 359, PEAK THERMAL POWER VERSUS TRIP DELAY O FOR A START UP ACCIDENT USING A 1.2T.4 ROD GROUP AT 5.8 x 10 5 (AK/K)/SEC.

FROM 10-' RATED POWER NM ~ FIGURE 14-THREE MILE ISLAND NUCLEAR STATIO

..

_ __

. -

I

• l l

l

~l 30 25 l

7 \ - 1 l20 15 '

-

!N '

'

l hg Nominal ] 'M l

'

it: 10 l

• U 2 5 0

0 -0.4 -0.8 -1.2 -1.6 -2.0 Doppler Coefficient, (ak/k)F x 10 5 0001 360 s PEAK THERMAL POWER VER$US DOP

' s COEFFICIENT FOR A START UP ACCI U51HG A 1.2". OK/K ROD GROUP AT 5.8

~

(OK/K)/SEC. FROM 10 ' RATED PO NM FIGURE THREE MILE ISLAND NUCLEAR STA

- - _ _ _

• l l

l l

I O

i l

l

<

l

,

l 40  ;

s ,/

k 30 >

I 1 O l / l i /

5 20 /

e /

Nominal ,

/[/

10 l

0 0.4 0.8 1.2 1.6 Trip Delay Time, sec . 0001 361 PEAK THERMAL POWER YERSUS TRIP DELA)

'

FOR A START UP ACCIDENT USlHG ALL ROI 5.8 x 10 4 (aK!K)/SEC. FROM 10 ' R ATED PC l NM FIGURE 14-THREE MILE ISLAND HUCLEAR STAT 101

.

l O l

.

l l

1 i

I l l l High Flux High Pressure

\ Imvel Trip ' Invel ' hip

\

S 30 \ '

-

3

'

O l i

>

.' \

20 -

'

a N No i 5 ,_4 . . ,

E 10 0 -0.4 -0.8 -1.2 -1.6 -2.0 Doppler Coefficient, (ak/k)/F x 10 0001 362 O PEAK THERMAL POWER VERSUS DOI COEFFICIENT FOR A START.UP ACC U$lHG ALL R005 AT 5.8 x 10 4 (OK/K)/SEC. FROM 10 ' RATED POV ME FIGUR E THREE MILE ISLAND NUCLEAR STA

_ ._ .

O

.

2600 2550 2

A '

2500 <

,

- - -

I

= 8

l s Safety Valve g 2 +50

/ Set Point U

[ e Nominal 2400 m.

2350 0 0.4 0.8 1.2 1.6 2.0 Trip Delay Tic:e, sec 0.001 363 PEAK PRESSURE VERIUS TRIP DELAY TIME START UP ACCIDENT USING ALL RODS AT 5.

(6K/K)/SEC. FROM 10 ' RATED POWER NM FIGURE 14 9 THREE MILE ISLAND NUCLEAR STATION

,

'

l &l?{y /////

,

1. $(I4 k IMAGE EVALUATION TEST TARGET (MT-3) kCk NNNN 1

l l

l l

l 1.0 .S't121 o 22 g"2 2 I ,: 3t l,l -

"

k a

1.25 IA l.6

_

'

l

,

,

h MICROCOPY RESOLUTION TEST CH ART

$ '

%

A % 3,, / 4%

fi[>ff>p ,

'{sQ)fsb 4 <;p o

g;,s

-

l  ; .

.

.

. ,

-

l ~ . Lt ._ .,

_ _ _ _ .____ __ _ _ _ _ _ __

i O l l

i

.

i 2550 l '

'

2525 [ Safety set Valve point

,

2500 l f- 2475 l

'

Hominst Y I

2450 -

E ~ N

,  %

g 2425 l

2400 )

0 2 4 6 8 10 i Tripped Rod Worth, % ok/k l

l t

I 0002 001 l PEAK PRF55URE VERSUS TRIPPED ROD WOR 1 A ST. T UP ACCIDENT USING ALL RODS 5.8 x 10*' (OK/K)/SEC. FROM 10*' RATED PC

, NM FIGURE 141-THREE MILE ISLAND NUCLEAR STATIOt

O

.

l

!

I 1

l l

1 l

l l

!

2550 , g g  !

1 High Flux ,

_ High Presstee 2525 Invel Trip Ievel Trip i e t

'

,

\ ( Safety ialve

2500 Set Point a l

\

2475

'

.

= n~ 4 ,, . ,

xs l

'

2450 1

2 2425 2d 0 -0.4 -0.6 -1.2 -1.6 - 2. 0 5

Doppler Ccefficient, (ak/k)/F x 10 0002 002 PEAK PRESSURE VERSUS DOPPLER COEFFI FOR A START UP ACCIDENT USING ALL RO 5.8 x 10 4 (OK/K)/SEC. FROM 10 RATED Pi NEP/F FIGURE 141 THREE MILE ISLAND NUCLEAR STATIO

.

y +-- e

.

.

O

~-

2500 2490 3

1 248o -

E Nominal #

2470 l s#

s #

h 2h60 #

h /

2450

-4 0 +4 +8 +12 +16 Moderator Coefficient, (ak/k)/F x 10 5

,

~ 0002 003 PEAK PRESSURE YERSUS MODERATOR COEFFI FOR A START UP ACCIDENT USING ALL ROD 5.8 x 10 4 (AK/K)/SEC. FROM 10 ' RATED PO'

.

MM FIGURE 141:

THREE MILE 15 LAND NUCLEAR STATIOb

_ _ _ _ __ ._ _. __

, , , , , ,

O l

. ,

l

' ' '

, , , ,

'

, a f

,

,

,

1

, , , i

' ' ' i i l 12 0 '

NEUTRON 80

, l  ; ,

, 9ill, POWER , */. * ' ' ' ' > ' ' 'I i '

' ' ' ' ' ' ' ' ' ' ' ' ' '

40 , , , , , x, ,

'

, i , , i s

. , , , , , , , i e i l i i i l  ! I f I I

, ! , , i I , 1 i ' ,

THERMAL 12C i i - *

• i , , , , , i I ' ' ' i i '

i

! ' , i i

, i , ,

POWER , % , ,,. , , , , , , , ,N ,

80 ,i,,

ii , i i

, , ,

, i i

i , , , g,

-

, , , , , , , ,

,; , , , , , , , ,,

, , , , , , , , , , , , ,,i, i a , , ! iii,i i e i l l

' ' ' ' ' ' ' ' ' ' ' ' ' ' ' '

200 i i i i i i i t I t i i I i t i FUEL i , i ., wi .

U ' ' ' ' ' ' ' ' ' N '

TEMPERATURE i i i i ii ii l i iN i CHANGE , F -200 , , , , , ,, , ,

i N

' ' ' ' ' ' ' ' ' ' '

-400 l I , ' i! i i ii I i ! l i iiii,,ii i i,iii i i i i i i e i i t i i i ie i i l

AVERAGE ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' '

5 ie i , , , i i i i

,' , , ,

CORE MODERATOR i i i i i ' i - , i i i iN i l TEMPERATURE '''''''''''N

-5 ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' A )

CHANGE,F i ! 6 i i i i i i i i i i i i i i i , , , . , i i , i , , , , , ,

1

_,g i , , i , , , , . , , i 1

, i , ,

!

i

.-5 SEC- l 2500 i l l ' l , , , l ' l ' , l l '; l l

, ,

' ' ' ' ' ' ' ' ' '

REACTOR 2350 i , , , , i i i i i v' ' ' ', , , x' ,' ,

SYSTEM ' ' '

' ' ' ' ' '

2200 l PRESSURE 2050 ll'lllll l'll,N

,,,,;

,

l PSIA , , , , , , , , ,x l

' '

1900 lll l!l l'l'l l 1

1 t

'

0002 004 ROD WITHDRAWAL ACCIDENT FROM I O POWER USING A 1.2r.SK/K ROD GRO 5.8 x 10 5 (OK/K)/SEC; HIGH PRESS

l REACTOR TRIP 15 ACTUATED l l M :iP!7/ 1 FIGURE l l

l THREE MILE ISLAND HUCLEAR STA l l

l 1

. _ _ _ _ _ _ _ _ _ _ _ _ _

O O O i

I

!

j 2450 l I I l l 1 l l l l l l l l l l l l l l l 3

-

High Pressure High Flur --

Imvel Trip 4

• Invel Trip 2400 '

~

-

e Nominal 2n.

- 2350 5

e -

_

, 2300 e

y. ,N Q -

-

n O

$ h

"

r $ >E 2250 F

'$N

$0<

g rsg -

-

t CA8 0 mO, 2200 l i l l l l l l l l l l l l l l l l l l l l c *O k 6 8 10" ~3

10.

• 10 r- r*

7[ ,m g ] Rod Withdrawal Rate, (ak/k)/sec e 5 OO C ym Eg N

>

-!

n F- 'i>t~ O O

LM

-. - _ - _ - . . _ - _____

__ ___ _ _ _________ - ___________________ _ _ _ _ _ _

O

.

2500 2475 I e

2' 2h50 2 O f h

'

2k25 2 /

m /

2400 Nominal 2375 2350 1 0 0.4 0.8 1.2 1.6

Trip Delay Time, sec 0002 006 PEAK PRESSURE VERSUS TRIP DELAY TIME l ROD WITHDRAWAL ACCIDENT FROM RATED F l U$lHG A 1.2'.' OK/K ROD GROUP; HIGH PRE!

REACTOR TRIP 15 ACTUATED N_EP"F FIGU R E 1.t- 15 THREE MILE ISLAND NUCLEAR STAT 106 l

-_. __ _ _ _ - - _ _ _ _ _ _ - _ - - - -

l l

O .

l 1

l l

l I

i l

2425

.

4

)

High Flux = 1 High Pressure Trip l Trip

= 2400 o i

• 2375  ;

/ v Nominal l

$ f

$ 2350 5 /

2325 2300 0 -0.4 -0.8 -1.2 -1.6 -2.0 Doppler Coefficient, (ak/k)/F x 10 5 0002 007 PEAK PRESSURE VERSUS DOPPLER COEFFl(

FOR A ROD WITHDRAWAL ACCIDENT FROM R POWER USING A i.2T.OK/K ROD GROUP NM FIGURE 141 THREE MILE ISLAND HUCLEAR STATIO

.

O

-

100

\b 80 IN I = 70,000 lb-ft 2

• \ l , ,

i H

N 60 0 1 N b l E 40 M

.

I' 20 O

O 4 8 u g Time, sec g 0002 008 PERCENT REACTOR COOLANT FLOW AS A FUN

'

0F TIME AFTER LOSS OF PUMP POWER N EF"F FIGURE 14-17 THREE MILE ISLAND NUCLEAR STA fl0N l

.- __ __ _ _ _ _ _ _ _ _ _ __

.

O 17 l

Trip Set Point 511 = =

(107,5%) l MWF N I u 7 Mini:num ONE Ratio

- i in Hot Channel at 114%

}

a

\

N \ Power Stead;r State (1.38) 6e  % w l l l \

N

. N /~ I = 70,000 lb-*

'

$

Rated Power with \

Overshoot and Desdband l C g3 1.4 (102%) '

N N j a)  !

.%

Io

'

O -r gz l

l i 5 13 l l

I 1.2 _  !

100 102 10k 1C6 106 110 112 114 Overpower at Which Coastdown Begins, %

0002 009 i

MINIMUM DNBR WHICH OCCURS DURING TH COASTDOWN FOR VARIOUS INITIAL POWER LE' REY: 7-21-67 NM FIGURE 14 18 CHANGED OVERSHCOT FROM THREE MILE 15 LAND NUCLEAR STATION 103 TO 102'. AMEND.1 (7 2167)

.

O o.06

-

0.05 k -

Ns 0.04 h

l

\

\

0.03 \

o N O.02 m --

N N.

o.o1 0

0 loo 200 300 400 &

500 Time after Break, sec 0002 010 REACTOR SYSTEM COOLING RATE 4 IN.2 STEAM LINE BREAK N E7"F FIGURE THREE MILE 15 LAND NUCLEAR ST.

40 l

BOL Parameters O s = -1.1u x 10-5c ,3 3a 35 g = 6.0 x 10-5 gg)g T = 0 3 see Delay I = 5.47 x 10-5 ,,,

'

30 ,

EOL Parameters EOL a = -1 36 x 10-5 (g g )g g g = Assume Zero

" t = 0 5 see Delay 25 t

= 2 75 x 10-5 see l[

5 2

20 i

.

/

O 15 10 g BOL Case

_

0 1 0.1 0.2 03 0 . 16 05 0.6 0.i C=tro1 aa vora, mA 0002 011 l

PERCENT CORE EXPERIENCING ONS AS A FUNCT l

OF UECTED CONTROL RCD WORTH AT ULTIMATE F EM l

FIGURE 14-20

(

THREE MILE ISLAND NUCLEAR STATION

- . _ _

__ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

+.

O .

\

25 BOL Parameters EOL a = -1.14 x 10-5 (ak/k)/F D

2.0 - g = 6.0 x 10'5 (ak/k)F T = 0.3 sec. Delay a k* = 5.47 x 10 sec f

8 EOL Parameters 15 -a D = -136 x 10-5 (ak/k)/?

o g = Assume Zero y T = 0.3 sec Delay O *^ ' - " 75 * #=

1.0

/

/g 05 ' / i BOL

- Nominal Case 0

0.1 0.2 03 0.4 03 0.6 0 d Control Rod Worth, M /k i

I Ol2 2R H10 REACTION AS A FUNCTION OF EJ CONTROL RCD WORTH AT ULTIMATE P(

NM FIGURE 14 THREE MILE 15 LAND HUCLEAR STAT

. _ _

__

6

_ r.or. rc - tan I

/

g _ g = -1 36 x 10-3 (d /k)/F /

a = Assume Zero /

2

= o.3 see Delay 2+ = 2 75 x 10-5 ,,e f

[

Ultimate Power BOL Parameters EoL 1

- g = -1.14 x lo-> (d/k)/F

-

, ,

og =6.0x10-5(gfg)jp p f f p

6 _ T = 0 3 see Delay / /

-

t* = 5.47 x 10~5 see / '

'

4 1

-

/

2

'

Y '

Ulti

.

L f

3 . /

/

E 6 Y

/l I g / l

/

, /l 10-9 Ultimate "O"

Power 10 /

,- . .

- , ,

6 / I

/ I 4 / I l I

2 0 l 10 0.1 0.2 03 o.k o.5 0.6 07 c-1 Re um, Sa/*

0002 013 O REACTOR NEUTRON POWER VARI, WITH EJECTED CONTROL ROD WC

_A5Catsyry FIGU RE l

THREE MILE 15 LAND NUCLEAR ST l

. ._ .-___ ___ _ ____

O 140 -

'J1timagPower

'

/

120 / ~

I t

UltigtePowe 100 '

nom nni. ' -

Case 1

80 BOL Parameters O s = -11' x 1o-' ca/x)/r

= 6.0 x 10-5 (a/x)/r

,

!

= 0 3 see Delay '

2+ = 5.h7 x 10-5 ,,c l

EOL Parameters i

= -1 36 x 10-5 (d /k)/F

= Assume Zero 10-9 Ultimate

'

BOL

-

T = 0 3 see DeJay J+ = 2 75 x lo-> see

- 1 20 s

g - Nomtnnl l 0 - W ,

Case 0.0 0.1 0.2 03 0.h 0.5 0.6 Control Rod '4 orth, %2/k

!

0002 014 REACTOR THERMAL POWER AS A FUN OF EJECTED CONTROL ROD WORT

-

NM FIGURE '

THREE MILE ISLAND NUCLEAR STA*

- _ _. . . . _- . . - . __

O 60 , , , , ,

BOL Parameters -,

g = - 1.14 x 10-5 (g/g)fy g = 6.0 x 10~3 (ak/k)/F Ultimate Power 50 - T = 0 3 see Delay l l

'

P = 5.47 x 10~3 sec  ;

- EOL Parameters / l

go _ g = -1 36 x 10~3 (6k/k)/F

' I 3 g = Assume Zero l

• • * "#

_. T = 0 3 see D L l' = 2 75 x 10~ gaysee 30 '

'

a y 8 #

• i

-9 o <

ase j Ultimate 8 20 Power B0 3 /

n f

+ - Nomin,1 Case 3 10 '

b

/

/

O p #

g 0.1 0.2 03 0.4 0.4 03 0.6 Control Rod Wor *2 , M /x 0002 015 l

lO ENTHALPY INCREASE TO THE HOTTEST FUE YERSUS EJECTED CONTROL ROD WORTi M FIGURE 14-24 THREE MILE ISLAND NUCLEAR STATION 1

l

__ _ _

.

O

.:

BOL Parameters g = 6.0 x 10-E (d /k)/F a/k = 0 5%

• t = 0 3 see Delay C 60

\ 2+ = 5.47 x 10 -3 sec l l l Oi @

s 1.0

!

N \

N

%

a

$ Nominal Case g

%

E N E

20

-0 5 -0.7 -0 9 -1.1 -1 3 -1 5 -1 7 5

Doppler Coefficient, (ak/k)/F x 10 0002 016 THE EFFECT ON REACTOR NEUTRON POW VARYING THE DOPPLER COEFFICIEN ROD EJECTION AT 10 ULTIMATE POW NM FIGURE 14-THREE MILE ISLAND NUCLEAR STATK

- -.

1 O

--

Ak i

. e i e i i l BOL Parameters

-

g = -1.14 x 10~5 (d/k)/F w s /b 0 55 C 40 -

T = 0 3 see De P = 5.47 x lo" jaysec cl.

g Nominal Case o a se o

t 8

c:

32 i i e i i 0 3 6 9 12 15 Moderator Coef ficier t, (.1k/k)/F x 10 5

.

0002 Ol7 THE EFFECT ON REACTOR NEUTRON POW VARYING THE MODERATOR COEFFICIE ROD EJECTION AT 10 ULTIMATE POW MM FIGURE 14-THREE MILE ISLAND NUCLEAR STATK

_.

. . . - _ _ _ - _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

.

O 24 BOL Parameters g = 6.0 x 10-5 (&/k)/F

• 20

\ A a /x = o.5%

T = 0 3 see Delay N -5 sec s.. 2+ = 5.47 x 10 f i x

l l l

]16 nom nal Case N.% ,

e  % j o

u ,

'

E j

$ 12

'

-0 5 -0 7 -09 -1.1 -1 3 -1,3 -1 7 Doppler Coefficient, (ak/k)/F x 10 l l

1

,

1 l

l l

0002 018 THE EFFECT ON REACTOR THERMAL POk VARYING THE DOPPLER COEFFICIES ROD EJECTION AT 104 ULilMATE pow eMM FIGURE 14-THREE MILE ISLAND NUCLEAR STATIC

._ __ _ _ - - - _ _ - - - _ _ - -

l I

O

.

%"

l i . . .

i i i BOL Parameters w

g _ g = - 1.14 x 109(s/k)/r C & /k = 0 5% [

f _t = 0 3 see Delay Oe e t- = 5.u7 x 1o -5 sec

/

7

$U fg Nomtnmi Case 11 -

/

s 15 i t , , , ,

0 3 6 9 12 15 18 5

Moderator Coefficient, (ak/k)/F x 10 0002 019 THE EFFECT ON REACTOR THERMAL POW ,

VARYING THE MODERATOR COEFFICIE  ;

'

RCD EJECTION AT 10*' ULTIMATE POW NM FIGU R E 14-THREE MILE ISLAND NUCLEAR STATit I l

..

- _ _ _ _ _ -_ -_ ___ _____.

.

O

.

112

, , g i i BOL Parameters nl -

g = -1.14 x lo-' (&/k)/F

/

g = 6.0 x 10-3 (3 /k)/F /

/ ~-

j no -

s/k=0.3% -

f* = 5.47 x 10 ' see EOL Paramtars

-

109 r g = -136 x 10-5 (&/k)/F 108 -

g = Assum Zero s/k=0.3%

u' 107

-

2* = 2 75 x 10 see

/

i Ultimate Power, BOL

[ 105 Y 8

j 104

-

Case

/ /

E 103

/ /

102

'

j Ultimate Power, EOL 101 g

/

100 0 0.1 0.2 03 0.4 03 0.6 07 0.8 09 Trip Delay, sec 0002 020 REACTOR THERMAL POWER YER$US TRll DELAY TIME ROD EJECTION AT ULTIMATE F

'

NM FIGURE 1415 l THREE MILE ISLAND NUCLEAR STATION

-

BOL Parameters g = -1.14 x 10-3 (ok/k)/F 42 - g = 6.0 x 10-5 (ag/g)/7 ,

Ak/k=05%(10 -9 Ultimate Power)

-ok/k = 0.3% (Ultimate Power) p = 3.47 x 10-5 3c

.

38

_. EOL Parameters -'

g = -1.36 x 10-5 (g/k)/F f. Ultimate

~b " * / OL and ECL

$

% ok/k = 0 3% (Ultimate Power) /

l i

1 J+ = 2 75 x 10

-5 /

O 34

-

sec ,

l

.

3

=

'

a q 30

$  !

e /

e O 8

%

M r o

E E I o

!

,

2 22 ) ,

1 e /

$

• Nominal Case c

2 l 18 4 _9 gytim,t, ,

y Power BOL s l i

1. )

'

14 l l 0 0.2 0.4 0.6 0.8 1.0 Trip Delay (:), sec 0002 021 O ENTHALPY INCREASE TO THE HOTTEST FUE:

VER505 TRIP DELAY TIME ROD EJECTli ME3&F FIGURE 14-3C THREE MILE ISLAND NUCLEAR STATich l

-

- - .

l

.

_i i i i . . . _8-

-

_ _

-

O E

- -

-

l

/

/ /

/ /

/ -

-

- - ,/  :

.

- - / _

- 8*

i -

t

-

4 -

_ u 2 Eu p

-

I I

1 I

~

- -

_

-

_ -

_

-

_-

-

-

.

/ -

t

=

1

.

<  :

-

+

~

_

E ) -

E

-

/

/ .

_

!

, s' _

/

/

-

-

_=

< _

-

, _

-

/ _

-

_

- _

~

- . , , , o 8 e g e

< 8 8

.

8

~ 8 .

~ ~ -

3Tsd 'ssnses2d

.

O 0002 022 l

l LOFT SEMISCALE BLOWDOWN TEST N  !

!

VESSEL PRES $URE VERSUS TIME

)

NM FIGU RE THREE MILE ISLAND NUCLEAR STA*

l

_ _ _ _ _ _ . _ . - __.

.

O l l

!

l

.

' ~

l i I

100 l

80 l

'

N 5 \

%

a S 60

(

i 40 h

22% Measured 2o )

n 0

0 10 20 30 40 50 60 Time, sec 0002 023 O

PREDICTED PERCENT MASS REMAINING VERSUS LOFT TEST HO. 546 Mg FIGURE 14 32 THREE MILE 15 LAND HUCLEAR STATION

.

_ - - - - - ' _ - -

-

O '

.

I 70 l t

.

I 60 50

?o ^

• s 'o

}f ,

-

f 3

~ i ,

=

\ l O e i I g 30 0

20

)

%

10 ~ '.

.A l N 0

0 1 2 l .% m 3 4 5 6 7 8 9 Time, see

.

i -

.0002 024 CORE FLOW VER5U5 TIME FOR A ID, DOUBLE ENDED PIPE RUPTU

,

pg FIGUR E THREE MILE 15 LAND HUCLEAR ST l

1500 I 1400 A

.$ I n' DNB at j

'

d 1300 0.25 '

see i

}"

- 1200 c Calculaeed by Quinn's I

{,100

-.

[ Modified Sieder-Tate Equation s

-

'

O s/ '

, 1000 ig 900 A y s

\

N j 800 3 Slump Model /

O 700 _ Simulation \

w N 600 c N

-

E 500 y

w 400 b i

N '

300 200 \,\,

100 L \-

0 0 1 2 3 4 5 6 7 8 9 10 Time, sec 0002 025 HOT CHANNEL CLAD SURFACE HEAT TRA COEFFICIENT AFTER DNB VERSUS Til l FOR A 36 IN. ID, DOUBLE ENDED PIPE RUI NEFfF FIGU R E 1.t.

THREE MILE ISLAND NUCLEAR STATI

,

<

'

I i )

O l i

I 24

,

1 l

Core Top l

__.- - i

\ l s I 14" Pipe-50: N2 y  ;  ;  ; - -

• 14" Pipe-33% N2 16

] e i 12", Pipe-33% N2 ,

i  ;

m

,

• 12" Pipe-50% N2

'

i / /

h12  ! /

'

t 5 A

- '{ f l / Core Bottom 58 i

'

f

!

3 u

//

4

//r 0~5 0 10 15 20 25 30 35 40 45 50 55 60 Time, sec 0002 026 REACTOR VESSEL WATER VOLUME VERSUS TIME 36 IN. ID, DOUBLE ENDED PIPE RUPTURE FOR 600 CORE FLOODING TANK OPERATING PRE 55URI NM FIGURE 1.t 35 THREE MILE 15 LAND NUCLEAR STATION AMEND. 6 (18-68)

-

.

O 2500

"

l 2400

,

Hot Sp 2300

/

2200 600 lb-12"-50% N

/ ' 1, 1

/

• t 2100

<

/

'

600 lb-14"-33% N

[ / y )

600 lb-14"- 4 2000 -

50% N I

[

1900 \ ' '

-

.800

'

, ,# '

1700 / l t I 400 lb-14"-33% N 600 lb-12"-33% N _

400 lb-14"-50% N

/

1600

/ '

[ [ \ 1000 lb-12"-33% N 1500 e

'

1000 lb-12"-50% N

' ' '

10 12 14 16 18 20 22 2s 26 28 30 32 34 Quench Time, sec 0002 027 CORE FLOODING TANK ANALYSIS; MAXIMUM TEMPERATURE VER$US TIME TO QUENCH F(

l 36 IN. ID, DOUBLE ENDED PIPE RUPTUR l

!

,

ME FIGURE 14 3*

i THREE MILE ISLAND HUCLEAR STATich

.

O 2800 2600

.

l n.

!

- 2400 8

3 1 2 2200 '

!

I a

2000 '

1 \

c u 1800 Nominal N *

'1 Design N Point

]a 1600 g

_.

!a 1400 E

1200 0 200 400 600 800 1000 1200 1400 1600 1800 2000 2200 F.ax1=um Heat, Btu /hr-ft _y 2

.

0002 028 MAXIMUM HOT SPOT CLAD TEMPERATURE VE:

MAXIMUM HEAT TRANSFER COEFFICIENT AFTE FOR A 36 IN. ID, DOUBLE ENDED PIPE RUPTL M EFF FIGURE 1.t 35 THREE MILE ISLAND NUCLEAR STATION

_- _- .

___ - _ _ _ _ _ - - _ _ _ _ _

i lO

'

6 s s s s s 1

2800

\ h=0

• y 9 2600 2400 2200 ' I h = 15 Stu/hr-ft 2 _y 2000

= 20

[l f1800

?

%,1600 I e / \

1400 - \

u S.1200

}

~ 1000 t\

800

/

r N l\

h = 100 I

l 600 400 200

' '  ! ' ' '

0 O 10 20 30 40 50 60 Ti:ne , see 0002 029 HOT SPOT CLAD TEMPERATURE YERSU FOR %IN. ID, DOUBLE ENDED PIPE RUF AND VARIABLE QUENCH COEFFICIE N E7"F FIGURE 1 THREE MILE ISLAND NUCLEAR STAT

O .

2000 g l

'

l g 500 F 1800 / I 280 F u .

. \

3 1600 , 200 F

'

k 100 F E.1400 0!

5 200 3

l}

$

g 1000

$

"

800 7 600 0 10 20 30 40 50 60 Time, see C002 030 HOT SPOT CLAD TEMPERATURE VER$U:

FOR 36 IN. ID, DOUBLE. ENDED PIPE RUP l AND VARIABLE SINK TEMPERATUR NM FIGURE 1 THREE MILE 15 LAND NUCLEAR STAT

- -

.

' i66sisii i1 .... iii.iii.. .t...i. .. .i, . ..is . .. ..... .., .

~ g

-

-

_-

-

O

-

-

.

,

-

3 -

c.

-

1 -

-

.!.

- -

". '

E I N

, m

,

o ( f "e-

-.

.

~

-

-

~

-

_ ) -

-

-

e .

-

n' .

-

~, -

-

'

-

• .

- -

_

M

~

-

_

.

e u

I L

"

-

' 'N A .\ -

-

N N  :

=

.?.

-

\  :

~

o -

-

a 2"

55

"

.

'

N -

-

o -

~

• • • 5 -

i:

x3 \  ;

O

-

-

N T

-

_

-

-

-

L

-

-

-

-

ei,iii. i ..... .., .,..,,,,, ......,,,

7o

..,,,,i,, ,,,,,,,,,, ...,,..,i e a e ~ ~ o g,01 x 9T *ssvg 0002 031 O m ss ,L,,,,1o ,. CTo ,o,L, FOR THE SPECTRUM 0F HOT LEG RUP' NN FIGURE 1 THREE MILE ISLAND NUCLEAR STA1

.

_. . . . . . _

_

_

_

. -

O  :  :

.

.i -

-

,

-

-

-

<,

3 o i t/ o.~. /: -

-

Y ~VT

/ /:::

-

/

-

-

_

// Y=

( -

~

=

_

_

/ fi _

-

,

-

-

~

- "

- .

"

n - - E

' '

-

f - C 4

o a

k O

J  : s

_

_

_

. .

.

-

.

c. -

_.

.- .

_ g" -

;

/r o.a.,

,

a  :

-

s-

-

-- l

s:  :

.

-

_

-

-

/ -

.

-

N 9 t t t 1 M

• "

= 0 *

,.m . .,,,........, .....

0002 032 O REACTOR COOLANT AVERAGE PRESS FOR THE SPECTRUM 0F HOT LEG RUP' NEFT FIGURE 14 THREE MILE ISLAND NUCLEAR STATit

.

ydie46 ill is4444, 4i 4 is344e i4 ildeje 4is a64 f46J 1 3 46431 4 ilk

- _

O

- _

_ _

_ _

_ _

_ _

_ _

/ -

E

_

/ 5

_

l

_ _

_ _

_

o. _ \

U l

)

i

! O

~

- 3,

.

!  !

_ _

.

l

=

=.

,

. _

g  ;

-

-

t

" '

_ 8

_

"

1

-  :  : '

a -

.

2" I

( 3 -

5 5

• 5

- 4

.

_ _

_ _

_ _

o 3

_

\ 3

_

~

. _

_ _

_ _

_ _

,,,,,,,,, ,,,,,,,, ..,,,,,,, ,,,,,,,,, ,,,,,,,,, ,,,,,,,,, ~o 3 S S

• 2 2 -

di sd 'o2nessag Suipting 2ossesy 0002 033 REACTOR BUILDlHG PRESSURE VERSUS 36 IN. ID, DOUBLE ENDED RUPTUR NEPT FIGURE 14-REV: 7-21-67 THREE MILE ISLAND HUCLEAR STATI REVISED FROM '4 TO 3 EMERGENCY COOLING UNITS AMEND.1 (7 2167)

,

l

"

- . . i . . .

'o

-

j _

2 E

O  : t

-

I.

. I

-

1 aC 32O

..

33 [d 2:

gjl- " '

l

,

- 3"I 12' I /- j

211= $5 /: I

- si , - ~.

'

n_g=8

-

e4 ee /  : i

-

/ _

{

2 m

o N

-

-

i l

-

-

• : ,

-

- _  ; l

= l

-

_ t

.

1 - 8

-

-

.

9 c

-

O -

-

-

-

-

-

-

-

-

-

o

. .

-

4

• :

-

-

==

-

.

, , , , o,

-

S S S S R 2 5!sd sansssJd Suist!ng soisesy REACTOR BUILDING PRESSURE VERSU:

O;

\v FOR A 36 IN. ID, DOUBLE ENDED RUP WITH AND WITHOUT COOLING OF THE RSClRCULATED SPRAY WA*

NEFfF FIGURE 1.5 .

THREE MILE ISLAND NUCLEAR STA1 wgsc. i 7 zi 6--

. _ - .

. . . . 4 i , 4

. . "'o _

i ll l  :

)

-

l Sj -

-

813 h/  ;

-

-: /

3& bi f

/

2. ~8 3 :=

/ , .

. r -

_o

.Ex :: 1 I , -

l#3 g 15 i /  :

. 2.1* 22 /

n- si ' .

_

/ -

_

hggaee f .

/ ,

/ \

l / l n 1

/  :

o J

l

-

l l-i

.

-

.- .

'

E -

!

-

./ 3" 3 l

_

/ 1

.- "

? I 5

)

-

hl -

\ t *

.\ 4 2 . "o  ;;

%  ;

.

.

2\

--

3 3

--

.

-

_

-

.I =

.

g _- l

.  :' O

-

I ~

_

-

, -

j I -

.

\ j  :

w

-

.

.

\ _

_'

~

\

x

-

g _

-

\

N i

-

N  !

, , i \ r i ,  !

i o

-o S S

• tt. @ ~ $ S- S-3 eanteaea.et i 0002 035 l

f%

'

REACTOR BUILDING ATMOSPHERE A

$ UMP COOLANT TEMPERATUF'.E5 FOLLC A 36 IN.10 DOUBLE ENDED RUPTUF NM FIGURE 1.t 4:

THREE MILE ISLAND NUCLEAR STATI i AMEND.1 (7 2! 67) l

,

...i.i. ii. is ... . . . . . iii .is e ns e i s is a i iw -o O

.

_

_

_

.

_

_

_

_

_

~

/[

-

.

.

_

/ "o

.

- -

_

_

_ _

_

_

_

_

_

_

. . _

_

\

) a S

.

,  :

-

_-

_- n

-

-

J 8 -

. u _

-

_ j 0, _

J f a  :

-

5" "5

-

5

( .

.

i

.

c.

-

- \-s -

-o C

' _

_

_

_

_

_

_

_

_

_

_

_

_

_

.

s e,

- -

-

_

_-

_

,

__ _

_

_

_

_

_

_

I. . .. .l.,, ., , .. - .. . . '

,

S S 9 R R 2 -

3;sd 'an.ssa.:d 82Z -*.E 23M* *M 0002 036 l

N 1

l REACTOR SUILDING PRESSURE VER$U!

AFTER RUPTURE 3.5 FT2 RUPTUr NM FIGURE 14

. ,. ..

u .t : ....-:

. THREE MILE ISLAND NUCLEAR STAT l

1 tEr S E Ft:N . TO 3 9E*iE*0Y :L:9G 'J.i t *! hENO. I ~ 27 C

1 f

.

, .. 4 .. . ..i464 .4. i4 ...... 6 4 . 4 ..... ..... ... ....i.., o

...l.4 O

.

. 1

. . 1

. .

. .

. .

. .

. .

"

. . .

o

,

. <

_

_

.

l _-

- c ,

-

.

,

N.

"

$ -

i 1

-

-

"o ~

u

-

I 1[  :

.

.

. g

_

g

. .

3

_ _

- -

_

_ N .

5 e

.-

i O

_

I  : -

_ g .

-

.

.

g .

_ v -

m.

_ g .

3.

-

t. -

J!

n S

_ _ e

.

_ l

,

_

_

_

l .

- ~

_

_

_ _

f

,,,,.,,,. s . . .,,,,. , . . . ,,i,. , i c3

• * *-

3 S 2 2 l 5t ed 'aanssead Sulpitag ao):veg 0002 037 o

'

' d REACTOR BUILLING PRESSURE VERSU AFTER RUPTURE 3.0 FT2 RUPTU REV: 72167 REVISED FROM 4 TO 3 EMERGENCY THREE MILE ISLAND NUCLEAR STAT COOLING UNITS AMENO.1 (7 2167)

.

.

.. i..... ii.e.iisi ..i464. ai 466 ..a6 .. 46.4. t i se .ei... 61 "o -

O I G

,

.

-

"o J. .

e  %

'

5 i F h E ~

0

/

0 *5 .

E *

?

I w i as g

?'.

5 I

< -

.-

c O

_

• o

-

,1 ,.i, ii., 1 ...,, ,,, ,,t. ..., ,,re vi,', I 3 S R 2 2 Sisd 'aanssaJd Su!pitng sol: ay 0002 038 REACTOR BUILDING PRESSURE VER$US AFTER RUPTURE 2.0 FT2 RUPTUF N_EP/F FIGURE 14-REV: 7- 21-67 THREE MILE ISLAND NUCLEAR STATI REVISED FROM 4 TO 3 EMERGENCY COOLING UNITS AMENC. I (7-2167)

.

' ria64466 f 44 asia 4 2 . 64664. . . . 44 4.,44 . 6 .. 66. . s....i.,6

• o

,% - -

_-

.

.

.

.

.,

-

_

_

_

J m

.

_ O

.

-

_

-.

a

_

_

_ -.

.

. _

.

i -

_

.,

e.

3 v

_

.

_ .

_ m

-

- -

- * - *

-

~

3  % ~

• _

\ _

~

Y

,

-

_  ;

!

E O _

_

.

.

_

_

_

O

-

_

-

-

-

.2 -

_ =a _

-

Y

-

8: -

2

- G O

_ -

_

_

-

.

_

-

-

-

_

_

_

t t ? 1 1 e 1 ? ! '

?? i t *1 '  ? ? ! t t t if ' I t ? ' t f

!8 t f ff fe e et e ?t f e g 2 R S S 31sd 'sanssaJd Su3pting sossess 0002 039

) REACTOR BUILDING PRESSURE VERSU!

AFTER RUPTURE - 1.0 FT2 RUPTUF REV: 7 21-47 NEF"F - FIGURE 14 REVISED FROM '4 TO 3 ENERGENCY THREE MILE ISLAND NUCLEAR STAT C001. LNG UM t TS AMEND.1 U-2167)

gia666664 4.i14 : 44 4 46 .646: i.44saisi 4 . 64 ae is iiseii4 1 6 -e

.

.

-

-

% .

g .

.

_

-

_

-

.

.

_

m

. o

. -

. .

. .

. -

.

.

- -

.

-

A a

o E

.

N 5

~

!

.  ;

.

.  ;

.

.  ;

-

.

.

.

'

.

g

\ _

o O

. ~

.

_

- S -

.

, _

.

3 _

.

g _

.

v

_

U  !

.

.

I

"

o I d

-

-

-

-

1

=

-

-

-

-

_.

.

-

gi.,,,, i i,,,,,,it ,.i.vi,,, ,,,,,r.,, vii.,..,i .,,.,,.,,'e 3 3 3 3 S 3

• Sied 'ianssead Suipting zo23esy 0002 040 l

REACTOR BUILDlHG PRESSURE VERSU!

AFTER RUPTURE 0.4 FT2 RUPTUF

.EV:

. ,,_21.o.., MM FIGURE 14

?EVISED FRCM 4 TO 3 ENERGENCY THREE MILE 15 LAND NUCLEAR STATI l ccouMG UNirs wEsc. H7 2167) 1

.

l l

O O O

.

UM%

" $ .?

Eb{

G l' T

-

-, 2 ?, 3 10 . . . . .

, _ i . . . .iii. , , . . . . . i i ii.. . . ..p

-,  ; -

' I

~

s

_ 3 Emergency Cooling Units Total p' I g -

,~ _

y .

- . _ _ _ _ _ .

11 ' /

o

-<

.

/ 4 f Steam-Air Hixture_.J 2 l

.

l0

! ,

y -- - - - -

Liquid

--

'o

'

-

.

/ \ [ -

-

-

j

/

, Decay -

" /

-

Keat -

0

,

-

/ Structures Removal Cooler { -

t j l j 10

-

_

~

N nN m om -

3 Emergency a

n> O

-

_

'm b m -

tooling Units y .o

_

Z" e _

_

i:

> "_ ,6E -

-

r m

u. r-z >

f* O o-O C 0 o z cI 10 t i_t i i n i

'. oI i i i i n L1 i i i s ul a i i i s ui i e i iniin i i as o

-

_ O -1 0

s. my O 10 10 10 1

10 2

10 3

10 il

'

h

• m .

's h5 oo Time after Rupture, sec

> o~ mM O

=c oy 4 Um

> c. ,M

-

--'*

4

O O O

.

SN5

@ $ .T 55?M c ,,

i E s.

-4 x ~s 43 8

w 3 10 e i i ie iisie e i iimii . . . .

iisei[-

i i

_

. .

. . . i sp

, ..

i,i -

3 Emergency Cooling Units ' I

-

5 [ Total -

1. "

"'

p

_

'_

/ -

/ Steam-Air e4isture

/

10

.

/

-

• Liquid -

-

"

-

_

/

/ Decay

~

f / Heat -

I Structures R** oval 5 go l  ! Cooler

_

m y

_

-

s m 3 Emergency

-

2 > .

m n

,_g

.

Cooling Units _

M O ..

.

~

r- 'o p u

, m ms o ,o0 . . . ..... . . . . . . . . . . . ..... . . . . . . . . . . . . . . . .

$Z O 10 '

~

10 10' 10

> CO 10 to

.o n o Time after Rupture, sec wm Q Z mZ A C n Cm Q

-

5 f* am go 5 5 m 3 C 4 N

> o m-2 C mZ <

an a m

-4 m w Z t.

_ _ _ _ _ _ _ . _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _____ ___ _

ll i

.

O .

.

.

.

_ ~ _ _ - _: - _~ - ~ - ~  ; :

m1 0

i

.

i

.

i

.

e

.

i N

.

.

e n 10 n

. r (

. u i t i

.

a e i

.

r r e u e p t

.

m a e r i T e

.

'

r o

p a

m T p

m e

i g

n

• V p l

o c m I l o e

. g u C 0 s

.

.

- n S y

i e

1

.

e i

n

. d g c r

.

l n n i u

i u

i e i t

. d g p B l r e u

. r i

o e s R o B mt i E i r t

c r n e

.

a 3 U t e t o i f

R c a a e e m

.

R 0 i n

. 1 T O

i

.

e

.

i

.

i

.

i

.

e

.

i 0

.

. i 0

l a

.

s

.

i

.

i

.

i i

g -

i i

- - _ - ~- [ _  : :

0 0 0 o

M 3 2 5 5 1

0 1

0 5

ol

-

_

_

_

m yEt _ i

)

QCgh. C4u "m$ao= mC : gn4%on@om

.

_

_

.

_

O 4- S= E*i-oC,r?mzSo ,Qg m >w p MCI e ,$ ua z JT

_

o' 4gm - >n m" gm Nc"U-4cmm

, II

%2 ~g TO n Oc a m ._

-

52"G DY = gQ#g" 41mm"

- E,m

-

Gr zD g n Q b* e4> -

O? Eo EGm ,rm gS$-

.l l

_ _ _ _

O u O, O"'

=

? 'a2 .4 s

aGy .

• '

e ,e

=

a a.

gx s a

w 300_ . . i r iii . . ... . . . . .. . . im 7 i.is i i i cri

3 Emergency Cooling Units M

m 250 _ ',

Q -

.

Reactor Building Vapor Temperature 200 ;- -- - -

,

_

I Reactor Building $ ump Temperature

$ ~ -

% 150

@

L

-

/ '

w -

_

gg __

_ __ _ . _ _ _ _ _

' -4 ~

m >mKmm  :

m *1>

m -4 tre n 50 - -

- - - - - - -

,b ma N>y E o r- m -4 m -

m -

C C _

>

I u

_ m mW _

_a m C 0~ '

' ' ' ' ' 41 * *

• 88 ' ' ' ' ' ' ' '

Q{

o z C " r-N ,> O 10

-t 10 0 1 2

' ' ' 81**'-

3 A " *- A ' "

9 10 10 to 10 l' 0z $#

'

Z 0

Time Af ter Rupture, sec r e h OC>

,1, Q

~-$ m ~m -4 O O a y o~ "d M Q N C mO> N w m Cz1 z

-4 m oO

> a Y m en O O

A

.

_ _ _ . _ _ _ _ _ _ _ _ _ _ _ _ . _ _ _ . _ _ . _ _ _ _ _ _ _ _ _ _ _ _

_ - -_ -

o 4 . .. .. .. . . . .. ,, ..... .... . ..  %

U> -

.

_

- -

.

.

-

.

.

.

.

.

.

/ -

.

.

7  :

- .

.

.

.

.

.

-

~. .

-

, .

.

. i .

.?.

\

_ f

.

"o -

u

.

- 3 .

- c -

J

- l e ~

.

_ 2

-

.

.t

- .

_

. =.

s O .

.

n

.

,

-

..

-

.

.

. .

.

g .

. t

.

.

.

.

.

.

.

O

-

E

.

.

.

.

.

M

.

.

_

M

.

.

.

. . .

,

S S S R R

.... .. ... . .... .,, .,=..'

0002 045

' m,J CRITERION 17 CASE FOR 36 IN.

DOUBLE ENDED RUPTURE REY: ,...

,,).e.7 M E7"F- FIGURE

-

tEvts D .:tcM 2 70 3 B Et1E3cy THREE MILE ISLAND NUCLEAR ST

ctiss Ju is w E.c. : 72167.

.

hV

,

l 3: 3 Emerger.:y -

"

0:olir.g *.' nits + /

2 3; rays 3

1 70 /

3 Emerger.y Onli g waits

c' * '

f i

.. ..

..

'

'

/ s

~$

[-

' JC '

/ /

2'- l l

.

,

• 2: Y

<

/

Stru:1.res J . I

20: 40: 60: 3:0 *. :: '2:0

. '400

. ;60 1300 2 00 22: 2k:0 2600 2S:0 30:c T're  : :=;;ete 3es:ti:n, se:

,

0002 046 f 's l j REACTOR BUILDING ZR REACTION CAPAB 1 FOR 55 PSIG DESIGN PRESSURE 4EY: 7-21 67 NM - FIGURE 1.t.

(

t REVISED F9CM 4 TO 3 BE.1GENCY THREE MILE ISLAND HUCLEAR STATI(

1 l 000L:NG UN4T3 AME.NC.1 :7 21 o7:

l l

[

10 .

, s . > . <>> . . . < .>. . < s 8 -

Exclusion Distance

-

6

-

4 -

-

Boundary of Low Population 2 - +

Zone

"

30-Day Dose 5

, 1 24-Hour Dose / N

-

I

"

6 -

"- .

2

\

-

5 1 -

%

'

cn - 2-Hour Dose 2 ,

c l

10 0 l

- ~

-

,)

(O 8

-

-

6

-

4 -

g

-

2 -

-

-l , . . , , ,,, , , . , , , , , .

10 6 10 10 Downwind Distance, ft.

0002 047 THYRolD DOSE FROM LOSS-OF COOLANT ACCI 2-HOUR, 24 HOUR,& 30 DAY DOSES REY: 7-21-67 REVISED TO INCLUDE MORE NN FIGURE 14-55 CONSERVATIVE FlSSION PRODUCT THREE MILE ISLAND HUCLEAR STATION RELEASE. AMENo.1(7 2167)

1 r

l 10 , i 3 3 , 3

_ 3 3 3 3 ,3 33 3 3 , ,

,

~

. \

~

,

30-Day Dese 4 - 2L-Ho w Dese 1

-- -

10 CFa 100 Limic

!

2 -

!

- 2 10

\

~

_

-

_ *  ;

o

6 -

l

.

g

_  % 1

> l

• 1 4 -

r i .

l 0

2 -

Exclusion

[ Distance

.

%

gg i

- '

_

_

6 -

_ 2-Hour Dese

/

4 -

-

2 _

Botudary of Loc Eopulation Zone 1[ ' ' ' ' ' I t ' ' I I I i '  !  ! !

2 2 4

g 6 10 2 4 6 4 2 4 e 10 Dev=vind Distance, f t 0002 048 O MAXIMUM HYPOTHETICAL ACCIDENT THYRolt ASSUMlHG 100 PERCENT CORE MELTDOW NM FIGURE 14 56 THREE MILE ISLAND HUCLEAR STATION

e -

10 pJ 3444

=

iiii ##4, iiii ##ii i,4, , 66# iiii iiii sie. i i i . i, i i t

=

.

_ _-

_

_

1 . _

. .

_ _

_ Exclusice _

O Distence 10 5 )- j

e j

_ _

,

_  ; _

~

l l

!- 10~1 ,

k I l

i  :

.  :

o l

-

[ 30-Day Dese

-

-

o c -

8-Hour Dese -

c - ~

o

. 10

_

-

l [ 2-Ecur Dese

_

0 y 1

=

5

_

,

5

_

u _

_

u l

. _ _

!

lo" 3 & l e

._

i e

_

-

-

i

-

-

, .

!

i i 1 _A E I I' E

-

1 I  ;

-

!

--

!

_

l \ >

_

10-5 ,,,, ,,,, ,,,, ,,,, ,,,, ,,,,:,,,, ,,,, ,,,, ,,, ,, , ,,

0 1000 2000 3000 4000 5000 6000 Distance Frc= Edge of Reactor Building, it 1

'

0002 049 l

b INTEGRATED DIRECT DOSE FOLLOWING MHA 31/2 FOOT REACTOR BUILDING WALL THICI NM FIGURE 14 57 )

THREE MILE ISLAND HUCLEAR STAT 10h

!

l