ML20212B794

Final Rept for Maximum Allowable Primary to Secondary Leak Rate Using

ML20212B794
Person / Time
Site:	Prairie Island
Issue date:	10/20/1997
From:	NORTHERN STATES POWER CO.
To:
Shared Package
ML20212A923	List: ML20212A920 ML20212B794
References
NUDOCS 9710280208
Download: ML20212B794 (16)

Text

_ _ -

l ATTACHMENTB FINAL REPORT FOR MAXIMUM ALLOWABLE PRIMARY TO SECONDARY LEAK RATE USING 1 Ci/gm D.E.1-131 I

i Steamline Break Radiological Consequences Introduction For this analysis the complete severance of a main steamline outside containment is assumed to occur.

The atTected SG will rapidly depressurize and release radiciodines initially contained in the secondary coolant and primary coolant activity, transferred via SG tube leaks, directly to the outside atmosphere.

A portion of the iodine activity initially contained in the intact SGs and noble Eas activity due to tube leaka;;e is teleased to annosphere through either the atmospheric dump valves (ADV) or the safety valves (MSSVs). This analysis evaluated the maximum pennissible primary to secondary leak rate which could exist in the faulted SG without exceeding the allowable dose rates at the site boundary, the low population zone or the control room. This section describes the assumptions and analyses performed to determine the amount of radioactivity released and the offsite and control room doses resulting from this release.

Input Parameters and Assumptions The analysis of the steam line break (SLB) radiological consequences uses the analytical methods and assumptions outlined in the Standard Review plan (Reference 1). For the pre accident iodine spike it is usumed that a reactor transient has occurred prior to the SLB and has mised the RCS lodine concentration to 60 pCi/gm of dose equivalent (DE) 1131. For the accident initiated iodine spike the reactor trip associated with the SLB creates an iodine spike in the RCS which inercases the iodine release rate from the fuel to $e RCS to a value 500 times greater than the release rate corresponding to the maximum equilibrium RCS Technical Specification concentration of 1.0 pCilgm of DE l 131.

The accident initiated lodine spike is calculated assuming the letdown flowrate of 40 gpm and 100%

etliciency ofiodine in the demineralizer beds. The durution of the accident initiated iodine spike is 8.0 hours0 days <br />0 hours <br />0 weeks <br />0 months <br />.

The noble gas activity concentration in the RCS at the time the accident occurs is based on a fuel defect level of 1.0%. This is approximately equal to the Technical Specification value of 100/E bar pCilgm for gross radioactivity. The iodine activity concentration of the secondary coolant at the time the SLB occurs is assumed to be equivalent to the Technical Specification limit of 0.1 pCi/gm of DE l 131. .

The amount of primary to secondary SG tube leakage in the intact SG is assumed to be equal to the Technical Specification limit of 150 gpd The allowable leak rate for the faulted SG was determined to be 1.22 gpm based on a density of 62.4 lbm/ff.

No credit for iodine removal is taken for any steam released to the condenser prior to reactor trip and concurrent loss of offsite power.

I

e' The 50 connected to the broken steamline is assumed to boil dry within the initial 15 minutes following the St.B. The entire liquid inventory of this SO is assumed to be steamed off and all of the iodine initially in this S0 is released to the environment. Also, iodine carried over to the faulted SO by So tube leaks is assumed to be released directly to the environment with no credit taken for iodine retention in the SG.

An iodine panition factor in the intact SO of 0.01 (curies I /gm steam)1 curies 1/gm water) is used (Reference 1).

All noble gas activity carried over to the secondary through so tube leakage is assumed to be immediately released to the outside atmosphere.

Eight hours after the accident, the RHR System is assumed to be placed into service for heat removal, and there are no funhet steam releases to atmosphere from the secondary system.

The thyroid dose conversion facton, breathing rates, and atmospheric dispersion factors used in the dose calculations are given la Table 1. The parameters associated with the control room HVAC modes are summarized in Table 2. The remaining maior assumptions and parameters used specifically in this analysis are itemized in Table 3.

Control Room Model The prairie Island control room HVAC system operates in one of two modes. Mode I is the nonnal HVAC mode, in which 1.835 cfm of air flow is outside air and 10,000 cfm is recirculated air all of which is unfiltered. Mode 2, which consists of 100% recirculated air within the control room a ponion of which is filtered. In each mode there is 165 cr m of unfiltered in leakage. The parumeters associated with the control room HVAC modes are summarized in Table 2.

For the steam line break accident it is assumed that the HVAC system begins in Mode 1. Following the steam line break it is assumed that the system is shifted to Mode 2 within 2 minutes, Description of Analyses Performed The analysis of the steam line break (St.B) radiological consequences uses the analytical methods and assumptions outlined in the Standard Review Plan (Reference 1). Both the pre accident iodine spike and accident initiated iodine spike models are analyzed for these release paths.

2

4. '-

Acceptance Criteria i-The offsite dose limits for a St.B with a pre accident iodine spike are the guideline values of-10CFR100. These guideline values are 300 rem thyroid and 25 rem y body. For a SW tith an cecident initiated iodine spike the acceptance criteria are a "small fraction of' the 10CFR100 guideline values; or 30 rem thyroid and 2.5 rem y body. The criteria defined in SRP Section 6.4 (Reference 2) will be used for the control room dose limits: 30 rem thyroid. 5 rem whole boay and 30 rem beta skin.

Results l

The offsite and control room thyrold, y body, and beta skin doses due to the SLB are given in Table 4.

Conclusions it was determined that for a primary to secondary leak rate in the faulted SG of less than or equal to 4620 grams / minute or 1.22 gpm based on a density of 62.4 ft'. the offsite thyroid and whole body doses are within the current NRC acceptance criteria for y steamline break accident based on the low population zone dose for the accident initiated spike case. The control room thyroid, whole body and t

beta skin doses are also within the current NRC acceptance criteria for the control room.

References 1.

NUREG-0800, Standard Review I'lan 15.1.5, Appendix, A, " Radiological Consequences of Main Steam Line Failures Outside Containment of a pWR, Rev. 2, July 1981, 2.

NRC SRP Section 6.4," Control Room Habitability System", Rev 2 July 1981, NUREG 0800,-

3 l

. o

TABLEI DOSE CONVERSION FACTORS, BREATillNG RATES AND ATMOSPilERIC DISPERSION FACTORS Thyroid Dose Conversion Factors "'

isotope (rem /curle) 1131 1.07 E6 1 132 6.29 E3 1133 1.81 E$

l 134 1.07 E3 1 135 3.14 E4 Time Period Breathing Rate "'

(hr) (m'/sec) 08 3.47 E 4 Atmospheric Dispersion Factors "'

(sec/m')

Site Boundary 0 2 hr 6.49E-4 Low Population Zone 0 8 hr 1.77E 4 Control Room 0 8 hr 5.58E 3 E 'P Publica: ion 30

• alatory Guide 1.4

• d't.saR Appendix H 4

r _ _ _ _, , . . . _ , . . - _ _ _ . _

O TABLE 2 CONTROL ROOM PARAMETERS Volume 165,000 ft' Un61tered inleakage 165.0 cfm Total Flow Rate 12000 cfm Un61tered Makeup'in leakage Mode 1 2000 cfm Mode 2 165.0 cfm Filtered Makeup Mode 1 0 cfm Mode 2 0 cfm Filtered Recirculation Mode 1 0 cfm Mode 2 3600 cfm Filter Efuciency Elemental 90 %

Organic 90 %

Particulate 95 %

Occupancy Factors .

0-8 hours 1.0 5

e TABLE 3 ASSUMPTIONS USED FOR SLB DOSE ANALYSIS Power 1721 MWt Reactor Coolant Noble Gas Activity Prior to Accident 1.0% Fuel Defect Level Reactor Coolant lodine Activity Prior to Accident Pre Accident Spike 60 pCi'gm of DE l 131 Accident initiated Spike 1.0 pCi/gm of DE l 131 i Reactor Coolant lodine Activity increase Due 500 times equilibrium to Accident initiated Spike release rate from fuel for 8.0 hours0 days <br />0 hours <br />0 weeks <br />0 months <br /> after SLB Secondary Coolant Activity Prior to Accident 1 0.1 pCl'gm of DE l 131 SG Tube Leak Rate During Accident (Based on a Density of 62,4 ft))

latact SG 150 spd '

Faulted SG 1.22 gpm lodine Partition Factor Faulted SG

- Intact SG 1.0 (SG assumed to steam dry) 0.01 Duration of Activity Release Secondary System 8 hours9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br /> Offsite Power Lost Steam Release from Intact SG 254,400 lb (0-2 hr) 486,000 lb (2 8 hr) 6

-___j

, o

.g TABLE 4 SLB OFFSITE & CONTROL ROOM DOSES Site Boundary (0-2 hr) Dose Acceptance Criteria Thyroid: Accident initiated Spike 6.60 rem 30 rem Thyroid: Pre Accident Spike 8.02 rem 300 rem y-body 0.027 rem 2.5 rem j

• ow Population Zone (0-8 hr)

Thyroid: Accident initiated Spike 23.34 rem 30 rem Thyroid: Pre-Accident Spike 7.62 rem 300 rem y-body 0.070 rem 2.5 rem 4 Control Room (0-8 hr)

Thyroid: Accident initiated Spike 29.78 rem 30 rem Thyroid: Pre Accident Spike 11,67 rem 30 rem y-body 0.0042 rem 5 rem Beta skin - 0.05 rem 30 rem 7

e Enclosure 3 4

Section 2.2 of Design Basis Accident Radiological Study for the Prairie Island Control Room Transmitted to the NRC by Letter Dated July 20,1981 4

4 4

d

(

s 4

i

(

4

'?

it

_ Q 1

, 2. 0 = Control Room Air'Jorne Doses -

2.1 General' Licensing- Consideration ,

I The requirements to show acceptable post LOCA doses in the ControlLRoom (CR), (NRC's letter of 5/7/80) result in the-need to evaluate the DBA-LoCA and the subsequent pathways for release of_ radioactivity. ,.

The dose calculations _were performed to show compliance of the Control' Room (CR) with GDC 19.- l 2.2' Me thodology The guidelines given in _ SRP 6.4_ and- R.G. _1.3 were used with an exception of the X/Qs for CR and TSC. Atmospheric dis-persion factors are based on the Halitsky Methodology from Meteorology and Atomic Energy 1968, as discussed in-Section 3.2.2.

2.2.1 Assumptions and. Bases Regulatory Guide 1.4 was used to determine activity levels in the containment following?a DBA-LoCA. Activity releases are based on a containment leakage rate of 0.254 per day -

for the first- day and 0.1254 per day thereaf ter._ . Table 1 lists the assumptions and parameters _used in the analysis.

The majority of. the containment leakage: will be - collected in the shield building and exhausted to the atmosphere through the.954 efficient SBVS filters as an elevated

. release from the main stack. However,- there _ exist - certain release pathways from the containment which will bypass the~SBVS filters. The bypass _ leakage was quantified by as-suming that it of the primary containment leakage bypasses both the SBVS and the KBSVS systems .directly to the atmos-

_-phere.

2.2.2 Atmospheric Dispersion Factor (Xfg)

The following discussion is yn./Qexplanation methodology of and the a reasons valus of for the .use of the Halitsky Ke =2.5 instead of .the Murphy. methodology -(Ref. 2) which SRP-6.4 suggests asi an interim position.

Historically, the preliminary work on building wake X/Qs

'was' based on a series of wind tunnel tests by James Halitsky et' al . :Halitsky summarized these results in Meteorology and Atomic Energy in 1968 (Ref.1) . In 1974 e( their paper based K. Murpny and K.Campeof-NRCpublisg/Qmethodologywhich on a survey of existirg data. This M-26/9

+

, +

presented equations without derivation or justification, was adopted as the interim methodology in SRP 6.4Xin 1975.

Since that time a series of actual building wake fg measurements have been conducted at Rancho Seco (Ref. 3) and several other papers have been published documenting the results of additional wind tunnel tests.

Reviews of the Murphy Eq. 6 and discussions with the author overtheyearshavedeterminedthatthy*guildingwakecor-rection factor, (K+2)/A, and K=3/($/d) were derived from the Halitsky data in Figure 37 of Ref. 2 from Murphy's paper. The Halitsky data was from wind tunnel tests on a model of the EBR-II rounded (FWR Type) containment and the validity of the data was limited to .5 <s/d <3 (Ref.1,

. Sect. 5.5.5.2). The origin and reason for the +2 in K+2 is not known. All other formulations use K only, and for sit-an 1 the use of K+2 imposes an uationswhereKislesstg/Q.

unrealistic limit on the For the Prairie Island plant, the building complex is com-posed of low, square edged buildings and two cylindrical shield buildings. For the HVAC intake on the Auxiliary Building roof, the intake will be subject to a building wake caused by the portion of the shield building above the roof of the Turbine Building-Auxiliary Building complex.

Since the Murphy methodology is overly conservative , a sur- -

vey of the literature was undertaken. It was found that the Halitsky wind tunnel test data (Ref. 1, Section 5.5.5) conservatively overestimated K values "by factors of up to possibly 10". Given this conservatism, it was felt that a the use of a reasonable K value from the Halitsky data should be acceptable. A review of Figures 5.29c from M& AE A value

( Re f . 1) resulted in K values in_the 2 to 3 ragge. Informa-t of K=2.5'was chosen to get a /Q of 5.33 x 10~ .

tion from other sources, as indicated below, has also shown that this should be a conservative value.

< In a paper by Walker (Ref. 4), control room X /Q's were experimentally determined for floating power plants in wind tunnel tests. Different intake and exhaust combina-tions were considered. Using the data for intake 6, and

( in Re f . 4 ) /Q values of 2.95 x 10-3 and stackAexgsust,were 3.73 x 10~ found after adjusting the wind speed from 1.5 m/see to 0.6 m/sec. These values are approximately two orders of magnitude lower than the conservatively calculated value for Prairie Island.

M-26/9

, . .e -

4< ,

4 a

i In a- wind . tunnel test by Hatcher (Ref. 5), a model' ind us- - ,

trial complex was used ; to test dispersions due to the wake.

Data obtained from their tests show that K has a value less than 1, and decreases as the test points are moved closer  :

to the_ structure.- Meroney and Yang (Ref. 6) in a study to determine optimuu stack heights, show that for short stacks (6/5 of building height), K reaches a value of approximately ,

0.2 and: decreases closer to the building. They concluded that the Halitsky methodology _ was " overly conservative" .

perimental tests show that K = 2.5- used to These recent determine the eg/Q for Prairie Island is a conservative estimate by, at least, a factor of 2'and possibly by 10 or j more.

f Field 3),

' and u~ /Q gests were were made on obtained. the from LData Rancho Seco round facility topped (Ref.

contain-

' ment releases and square edged auxiliary building releases wege used to simulate the Prairie Isgand Measured

~2case.

u~

1 x/Q v glues ranging from 8.07 x 10~

10~

m to apppoximately-w 4

in the 10"gre m' gound . - Although range for those-most values- of- u~ /Q were casesLapproximating the i Prairie Island conf value of 8.07 x 10~gguration, the worst at Pasquill G andRancho1.8 m/secSeco with case a building area of-2050 m 2 is used.for comparison-purposes.

Whun adjusted to the Prairie Island: cengitigns with a wind speed of 0.6 m/sec and an area of 782 m a- /Q of 3.

4 which is 1.5 times smaller than the value 5.33 x 10~g3 x -10~3,-

calcu-lated for Prairie Island using the Halitsky wini- tunnel data. ,

It was concluded that sufficient data and field tests exist to give-a reasonable assurance that the chosen'X/0 is a: con-servative one, over and above the conservatism implied by using the 5th percentile wind speed and wind direction.

factors.

. -2.2.3 Resul ts The radiological exposures in the CR are included in Table 2.

The doses f all within the GDC 19 guidelines values.

(

l i M-26/9

s c ,

[ g. ,

4 2.2.4 . References

1. D. H. Slade, ed., Meteorology and Atomic Energy - TID 3 24190 (1968).

2. K. G. Murphy and K. M. Compe, " Nuclear Power Plant Control

! Room ventilation System ')esign for Meeting General Cri-terion 19", 13th AEC-Air Cleaning Conference.

3. Start, G. E., J. H. Cato, C. R. Dickson, N. R. Ricks,

• G. H. Ackerman, and J. F. Sagendorf, " Rancho Seco Building

" Wake Effects on Atmospheric Diffusion, NOAA Technical Memo-randum, ERL ARL-69, (1977).

! 4. Walker, D. H., R. N. Nassano, M. A. Capo, 1976: " Control Room ventilation Intake Selection for the Floating Nuclear Power Plant", 14th ERDA Air-Cleaning Conference.

5. Hatcher, R. N. , R. N. Meroney, J. A. Peterka, K. Kothari,

1978

" Dispersion in the Wake of a Model Industrial Complex",

NUREG-0373.

4 6. Moroney, R.N., and B. T. Yang, 1971: ? Wind Tunnel Study on

! Gaseous Mixing due to Various Stack Heights and Injection Rates Above an Isolated Structure", FDDL Report CER 71-72 RNM-BTY16, Colorado State Univ.

l.

4 4F

• 4 e

i i

f

s:

.g'd

- TABLE 1, LOSS-OF-COOLANT ACCIDENT

PARAMETERS TABULATED FOR POSTULATED ACCIDENT ANALYSES DESIGN BASIS l ASSUMPTIONS l_ I.- Data and Assumptions Used to l- Estimate Radioactive Sources i from Postulated Accidents A. Power Level. (MWt) . 1721.4 l

• i B. Burnup NA C. Fission Products Released 100%

from Fuel (fuel damaged)

D. Iodine Fractions 0.04 1 (1) Organic (2) Elemental 0.91 (3) Particulate 0.05 l

II. Data and Assumptions Used to Estimate Activity Released A. Primary Containment Leak 0.25 (0-1 day)

Rate (t/ day) 0.125 (1-180 days)

B. No mixing is assumed to occur in the shield building prior to release to the atmosphere C. Bypass leakage (t of primary 1

- ccatainment leak rate)

D. SBVS Adsorption and Filtration Efficiencies (4)

(1) Organic iodinen 951

-(2) Elemental iodine 95 95 (3) Particulate iodine (4) Particulate fission products 95 III. Dispersion.(sec/m 3): .

R - Building Wake f

A.g/Q.forTimeIntervalsof (1) 0-8 hrs 5.33 x 10~3 (2) 8-24 hrs 3.14 x 10~3 (3) 1-4 days 2.00 x 10-3 (4) 4-30 days 8.79 x 10-4 (5)30-180 days 4.40 x 10~4

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

9:

TABLE 1 (Continued)

DESIGN BASIS l ASSUMPTIONS IV. . Data for CRs A. Volume of CR (f t3) 116,840

- B. -Recirculation Rato through 3,000 Charcoal Filters C. Efficiency of Charcoal (t) 95 Adsorber 4

• D. Unfiltered Inleakage Rate (h+~1) 0.06 E. Occupancy Factors:

0-1 day 1,o

• l-4 days 0.6 4-30 days o,4 30-180 days o,4 i

4 5

e b

A-33

' c' : .

4*'

TABLE 2 AIRBORNE ACTIVITY INSIDE THE- CONTROL ROOM DOSES FROM A DBA LOCA (0-180) DAYS Thyroid Whole Body Skin Doses (REM) 15.4 1.0 26.9 GDC 19 Dose Guidelines (REM) 30 5 30 75*

• If protective clothing is worn O

e M-26 4

_w-.-- - e ,, - -