ML20205F047

Calculation of DBA-1 Pcrv Penetration Leak Rate

ML20205F047
Person / Time
Site:	Fort Saint Vrain
Issue date:	08/08/1986
From:	PUBLIC SERVICE CO. OF COLORADO
To:
Shared Package
ML20205F044	List: ML20205F038 ML20205F047
References
P-86497, TAC-56565, NUDOCS 8608190076
Download: ML20205F047 (11)

Text

'

Attachment 1 to P-86497 i (Sheet 1 of 7) r ATTACHMENT I TO P-86497 i

CALCULATION OF DBA-1 PCRV PENETRATION LEAK RATE l

6 e

61 Tf7 %

Prepared By: d6 Date:

V Reviewed By: Tk' k Date: f/f/f*g

/

I 8608190076 PDR 860013ADOCKPDR 05000267 4 P

I

, Attachment 1 to P-86497 (Shee2 2 of 7)

ATTACH 1ENT I E.Al.CULATION OF DBA-1 PCRV PENETRATlQN LEAK RATE i

NomenclatyCe;

, f4 = equivalent diameter = 4Rg f = friction factor = 16p/GDe g - acceleration of gravity = 32.2 ft/sec2 G = mass flow rate per unit area, Ib/hr ft2 L = length of the leak path p = pressure, lbs/in 2 Ap = differential pressure between inlet and outlet of the leak path R = gas constant RH = hydraulic radius RN = Reynold's Number = GDe/p p - absolute (dynamic) viscosity e = weight density, Ib/ft3. p/RT I

. Attachment I ts P-86497 (Sheet 3 ef 7)

Assumotions:

1. The total PCRV helium inventory at normal operating pressure is, (per FSAR Section 4.2.1):

Primary Coo 1 ant System........... ...... .... . ..... .... .... .. 7370 1bs PCR V Pene tra t i ons..................................................... ...... 810 l bs Total PCR V i nventory................................. 8180 lbs Purification, Storage, and Aux 111ary Systems...... 710 lbs To t a1 I nve nto ry............. ........ . ... .. ............................ 8 890 1 b s

2. The maximum design leak rate from all components including the PCRV and helium storage system Is 100% of the total helium inventory per year (FSAR Section 5.2.3).

This leakage would, in actuality, be almost entirely purified helium from small valves external to the PCRV. However, for purposes of DBA-1 leak rate analysis, it is conservatively assumed that the PCRV penetration leakage is 100% of the primary coolant system and PCRV penetration inventory per year and that all the leakage occurs via the PCRV penetration closures.

3. a) Design primary coolant pressure is 700 psia at the circulator discharge (FSAR Section 1.3 and 4.2.1) so the normal differential pressure between primary coolant and atmosphere is approximately 688 psi.

b) The PCRV pressure is reduced to S psig following a Permanent Loss of Forced Circulation Accident so the post-accident differential pressure between primary coolant and atmosphere is S psi (FSAR Section D. I.3.4.2) 2

. Attachment 1 to P-86497 (Sh n t 4 of 7)

ASSUMPTIONS (continuedh

4. The PCRV Inventory following depressurization to 5 psig following a permanent LOFC accident is:

Location Temperature Mass Top Plenum i186*F 22 lbs Core 2137'F 5 lbs Bottom Plenum 900*F 15 lbs Steam Generators 674*F 34 lbs Lower Cavity 453'F 122 lbs Core Barrel 633'F 27lbs TOTAL 225 lbs PCRV penetration interspace inventory is conservatively ignored (approximately 20 lbs at 5 psig) and the interspace pressurization system is assumed no longer functioning.

5. Leakage flow is laminar.

i 3

f , Attachment 1 to P-86497 (Sheet 5 of 7)

Calculation:

1. For faminar flow:

4fG 2 L3 Ap =

2gt De .)l 16p and; f= for RN < 1500 GD e

2. By substitution:

32pG I L Ap = l ge (D e2

3. Solving for G:

e G=

32pl

4. Then the ratio between the flow rate at 17 psia (5 psid) and the flow rate at 700 psia (688 psid) is:

G 37 AP ,7 3 9 37

=

l (All other terms concel)

G 7oo @79o) e 73o Ap term density term

5. The Ap term, relating DBA-1 conditions to normal design conditions is:

Ap 5 psid i7 ,

i Ap 688 psid 138 7oo

6. By ideal gas laws:

=

e RT P

I7 = 17 (Penetration temperature assumed the some t E for both cases) 700 700 4

. Attachment 1 to P-86497 (Sh:et 6 of 7) l Calculation (continued):

6. (continued) Using the average pressure in the leak path to reflect the average density, the density term relating DBA-1 conditions to normal design conditions is:

e 37 h(17 + lZ  ;

' 700 25 f(700 + 12)

7. Therefore:

G ,7 i  ; j G 7oo 138 25 3450 That is, the PCRV penetration leak rate at DBA .1 conditions is approximately a factor of 3450 less than that at normal design pressure (1.e., a factor of aboutl38 less due to differential pressure and a factor of about 25 less due to density).

8. For the Revision 4, FSAR Section D.1.3.4.2 analysis of the PCRV penetration leak rate, the density effect was conservatively ignored and the leak rate at 5 psid 1s:

yea

• 100%

• 365 days

• 138

=

0.1628 lb/ day Converting to percent of inventory per day, The calculated PCRV penetration leak rate at DBA-1 pressure is:

0.1628 lb/ day x = 0.072X/ day at 5 psid 25 s it should be noted that, if the density effect were included, the calculated PCRV penetration leak rate would be:

I 0.072 %/ day x = 0.003 %/ day 25 5

Attachment 1 to P-86497 (Sheet 7 of 7)

Calculation (continued):

9. Technical SpectfIcation LCO 4.2.9 limits total secondary penetration closure leakage from all PCRV penetrations combined to 400 lbs/ day at normal design pressure.

The LCO allowable leak rate corresponds to the following e DBA-1 leak rate:

400 lb 1 1 100%

day

• 138

• 25

• 225 lb

=

0.05 X/ day at 5 psid This is less than the 0.072 %/ day attributed to PCRV penetrations in Revision 4, FSAR Section D.l.3.4.2 and both values are well within the historic DBA-1 overall leak rate assumption of 0.2 %/ day.

The historic DBA-1 overall PCRV leak rate assumption for accident consequences analysis is:l O.2 %/ day l at 5 psid r

Results:  :

The PCRV penetration leak rate during DBA-1, with the PCRV depressurized to 5 psig, is conservatively estimated to be 0.072 %/ day.

This value has been used in Revision 4, FSAR Section D.l.3.4.2 as the contribution of PCRV penetration leakage to overall PCRV leakage.

Technical Specification LCO 4.2.9 limits leakage during normal operation to a value that corresponds to 0.05 %/ day with the PCRV depressurized to 5 psig. '

Conclusion-The above results, together with the corrected concrete permeability calculation in Reference 1, confirm the conservatism of the historic DBA-1 overall leak rate assumption of 0.2 %/ day.

6

Attachment 2 to P-86497 UPDATED FSAR (Sheet 1 of 4) Revision 3 0.1-53 The core activity fractions remaining gas borne in the PCRV and available for leakage from the PCRV are compared in Table 0.1-11 with the arbitrary and more conservative gas borne fractions that were used, based on the assumptions in TID-14844 for a pressurized water reactor. The arbitrary TID-14844 release fractions were used as the PCRV escape fracticns in comparative dose calculations, i.e.,

the calculated core releases of noble gas, iodine and particulate activity were multiplied by 1., 0.25 and 0.01 respectively, to define the activity available for leakage from the PCRV. The doses resulting from this accident were calculated for purposes of comparison only for both sets of the release fractions listed in Table 0.1-11, since the TID-14844 values are not considered to be applicable to the HTGR system. The TID-14844 assumptions were, however, used by the AEC-DRL in their safety evaluation of the Fort St. Vrain reactor during the construction permit period. The purpose of the release comparison is to show the relative safety of the HTGR in relation to the arbitrary TID-14844 assumptions.

0.1.3.4.2. pCRV Leakage and Reactor Building Effects PCRV Fressure Effects. The PCRV pressure will be reduced to 5 psig after the onset of the accident, and heating of the coolant gas in and around the reactor core combined with the 99% release of noble gases from the fuel elements will not result in a further rise of the PCRV internal pressure. Calculation shows that the helium contraction due to cooling in the steam generator c vity is more than sufficient to counterbalance the gas expansion in and around the Core.

PCRV Leakage Assumptions. The outer (secondary) penetration closures will have a maximum design basis total leak rate of 100% of the PCRV volume per year at design pressure. This is about 3.65 ft'/hr at 700 psig. This maximum leakage is based on the economics of helium loss. If an internal PCRV pressure of 5 psig is assumed, the maximum leakage rate from the penetrations would be about 0.3 ft /hr, conservatively assuming that the secondary closures have 8

reached their limiting total normal operating leakage of 3.65 ft'/hr at 700 psig as stated above. This takes no credit for the inner, i primary closure. This is a fractional leak rate of less than IE-l 05/hr (0.024%/ day) based on a PCRV gas volume of 32,000 f t8 .

An arbitrarily conservative and non-mechanistic estimate of PCRV leakage is obtained by assuming that the liner has failed completely (or does not exist) and only the concrete permeability controls the leakage. The internal 5 psi pressure differential gives a PCRV leak rate of about 8.33E-05 fraction per hour (0.2%/ day),

r.early ten times larger than that controlled by the penetrations.

The calculation of the permeation rate for the PCRV concrete for these conditions were reported in Question 0.2 of Amendment No. 9 to the Fort St. Vrain construction permit application.

Reactor Building Ventilation and Filtration System. Fission product release from the PCRV into the reactor building is calculated simply as the product of the fractional activity available for leakage (see Table 0.1-10) frem the PCRV and specified PCRV leak

UPDATED FSAR Attachment 2 to P-86497 Revision 3 (Sh'.et 2 of 4) 0.1-54 rate. A very significant fraction of the non-noble gas fission product activity leaking from the PCRV would be expected to plateout

]

in transit along the tortuous leakage paths through the PCRV concrete structure. The enormous surface area, the low temperatures, the low leakage velocities and the caustic chemistry of the concrete would all combine to enhance this plateout. However, depletion of the leaking fission product inventory by plateout, both in the PCRV concrete and in the reactor building, was conservatively neglected in this analysis due to the lack of applicable experimental data. Thus, no credit for fission product plateout or settling in the reactor building has been assumed although these effects would be expected to reduce the activity released from the plant.

All leakage from the PCRV during this accident will be processed through the reactor plant ventilation system which will continue to operate normally. Particulate and iodine activity will be filtered with 95% and 90% efficiencies, respectively, as conservatively assumed for accident conditions described in Section 14.12. Iodine filter efficiencies of 90% are considered conservative (Ref. 17) and have been used in the release calculations. The fodine filters will be treated for removal of methyl iodine. The filters are tested in place to assure there is no bypass leakage. The constant fractional ventilation rate of the PCRV area of the reactor building during the accident would be the normal rate of about 1.5/hr with two of the three redundant filter train systems operating.

The filtered effluent from the plant is discharged at high l velocity from the roof top vent. Figure D.1-31 shows the integrated I activity release from the plant, using the experimentally determined PCRV fission product release fractions from Table D.1-7. Also shown l in Figure 0.1-31 is the integrated release calculated by applying the TID-14844 release fractions to the core release source term in the PCRV.

D.1.3.5. Offsite Doses The doses at the 16,000 meter low population zone outer boundary over the duration of the accident were calculated using both TID-14844 and the Applicants methods. The dose calculation methods are further described in Section 14.12. The detailed results and controlling parameters of these calculations are given in Table D.1-

, 12. Table 0.1-12 compares the doses shown above, which were I calculated using the methods of fission product transport developed l in Section 0.1.2.2., with those calculated using the TID-14844 l fission product release fractions given in Section D.1.3.4.1 (100%

l noble gas, 25% iodine, 1% solids). Also listed in Table 0.1-12 are the net fission product release fractions from the PCRV and reactor plant. The atmospheric dilution factors and breathing rates used for l the two release cases are also given for comparison. Table 0.1-12 i

points out the relative differences between the detailed calculational results predicted by the applicant and those based on TID-14844 fission product release fractions and more conservative meteorological assumptions. Parametric analyses have shown that the calculated doses for thi,s accident are not sensitive to assumptions

UPDATED FSAR Revision 4 Attachment 2 to P-86497

- D.1-48 (Shent 3 of 4) primary coolant system is simply the product of the total quantity of f any nuclide released from the core at any time and the escape (

fraction for the particular nuclide as listed in Table D.1-7. The calculated core release fractions, the experimentally determined escape fractions from Table 0.1-7, and the product of the two are listed in Table D.1-10. For example, the calculated fractional release of Cs from the core during the accident is 0.75, and the escape fraction from Table D.1-7 is 0.057(94.3% is absorbed or condensed within the primary system), giving a net gas borne fraction of 0.75 x 0.057 = 0.043 of the total Cs inventory available for leakage from the system.

The gas borne fission product activity that is available for leakage from the PCRV during the accident is shown vs. time in Figure D.1-34. These releases of noble gcs, iodine and all other activity are based on the calculational methods discussed in Section 0.1.2.2.

and in Section 0.3.4. of Appendix D.3.

The core activity fractions remaining gas borne in the PCRV and available for leakage from the PCRV are compared in Table D.1-11 with the arbitrary and more conservative gas borne fractions that were used, based on the assumptions in TID-14844 for a pressurized water reactor. The arbitrary TID-14844 release fractions were used as the PCRV escape fractions in comparative dose calculations, i.e.,

the calculated core releases of noble gas, iodine and particulate activity were multiplied by 1., 0.25 and 0.01 respectively, to define the activity available for leakage from the PCRV. The doses ,-

resulting from this accident were calculated for purposes of ( ~

comparison only for both sets of the release fractions listed in Table 0.1-11, since the TID-14844 values are not considered to be applicable to the HTGR system. The TID-14844 assumptions were, however, used by the AEC-DRL in their safety evaluation of the Fort St. Vrain reactor during the construction permit period. The purpose of the release comparison is to show the relative safety of the HTGR in relation to the arbitrary TID-14844 assumptions.

I l

0.1.3.4.2. PCRV Leakage and Reactor Building Effects PCRV Pressure Effects. The PCRV pressure will be reduced to 5 psig after the onset of the accident, and heating of the coolant gas in and around the reactor core combined with the 99*; release of noble gases from the fuel elements will not result in a further rise of the PCRV internal pressure. Calculation shows that the helium contraction due to cooling in the steam generator cavity is more than l sufficient to counterbalance the gas expansion in and around the Core.

l PCRV. Leakage Assumptions. For calculation of offsite doses

! l an integrated leak rate of 0.2%/ day is assumed throughout the l duration of the accident.

l l Economic operation of the plant requires that the rate of l helium leakage be low. For design purposes, the uncontrolled leakage l of purified helium from all components including the PCRV and the l helium storage system was estimated to be 100% of the total helium

UPDATED FSAR

• Revision 4 Attachment 2 to P-86497 D.1-49 (Sh::et 4 of 4) l inventory per year (Section 5.2.3.). For this accident analysis it l is conservatively assumed that the PCRV leakage is 100% of the l primary coolant system and the PCRV penetration inventory per year,

! and is attributed entirely to the outer (:ccondary) penetration l closures. This is about 22.4 lb/ day at normal full power operating l inventory and pressure. At the reduced PCRV pressure of 5 psig that l would be attained following the accident, the maximum leakage from i the penetrations is extrapolated to be about 0.16 lb/ day. This is l equivalent to 0.072%/ day, based on a calculated PCRV inventory of l about 225 lbs. of helium at 5 psig. The extrapolation is based on l laminar flow behavior through the leak. Conservatively, no credit is l taken for density variation with pressure, which should result in a i significant additionti reduction in leak rate at 5 psig.

l Furthermore, no credit is taken for the primary closure l leaktightness.

l In Appendix D.2.1.2., " Possibility of Liner Failure," it was l concluded that this loss of forced circulation accident would not l result in liner failure and the liner would remain intact and leak l tight throughout the accident. This was supported by extensive l analysis. Despite the above, the AEC directed (Ref.33) that an l arbitrarily conservative and non mechanistic estimate of PCRV leakage l be obtained by assuming that the liner has failed completely (or does l not exist) and only the concrete permeability controls the leakage.

l An ' internal PCRV pressure of 5 psig was assumed, which was calculated l to give a PCRV leak rate of about 8.33 E-05 fraction per hour l (0.2%/ day). Subsequently, the historic 0.2%/ day PCRV permeation leak l rate used i'or fission product escape from the PCRV been found to be l overly conservative by a factor of 200. The corrected PCRV concrete l permeaticn leak rate at 5 psig is calculated to be 0.001%/ day in l Reference 34.

l Nevertheless, for calculation of offsite doses, the historic l 0.2%/ day is assumed to exist as an uppper limit of all potential l l contaminated primary coolant leakage including permeability through I the PCRV concrete and leakage from penetrations. This is also l cnsistent with LWR licensing assumptions. In actuality, leakage of l radioactive helium could not occur (even if the primary closures were l not leaktight) during the several hour period of controlled PCRV l depressurization while the penetrations are pressurized with purified l helium. However, for conservatism this PCRV leak rate of 0.2%/ day is l applied uniformly throughout the duration of the accident.

I Reactor Building Ventilation and Filtration Syst m. Fission product release from the PCRV into the reactor building is calculated simply as the product of the fractional activity available for leakage (see T&ble D.1-10) from the PCRV and specified PCRV leak rate. A very significant fraction of the non-noble gas fission l product activity leaking from the PCRV would be expected to plateout l in transit along the tortuous leakage paths through the PCRV concrete l

structure. The enormous surface area, the low temperatures, the low l leakage velocities and the caustic chemistry of the concrete would all combine to enhance this plateout. However, depletion of the l

t ._ -