ML20214R172

From kanterella
Jump to navigation Jump to search

Safety Evaluation Supporting Amends 50 & 42 to Licenses DPR-21 & DPR-65,respectively
ML20214R172
Person / Time
Site: Millstone  Dominion icon.png
Issue date: 06/19/1978
From:
Office of Nuclear Reactor Regulation
To:
Shared Package
ML20214R167 List:
References
TAC-7355, NUDOCS 8706050385
Download: ML20214R172 (9)


Text

. .

./ 'o UNIT E D STAT ES y / ,, NUCLEAR REGULATORY COMMISSION

{  :

WASHINGTON, D. C. 20555 1

k ....+ /

SAFETY EVALUATION BY THE OFFICE OF NUCLEAR REACTOR REGULATION SUPPORTING AMENDMENT N0. 50 TO LICENSE NO. DPR-21, AND AMENDMENT NO. 42 TO LICENSE NO. DPR-65 l l

NORTHEAST NUCLEAR ENERGY COMPANY j MILLSTONE NUCLEAR POWER STATION UNIT NOS.1 AND 2 j DOCKET NOS. 50-245 AND 50-336 Introduction On February 13, 1978, Northeast Nuclear Energy Company (NNECO) submitted a report entitled, " Steam Dilution Off-Gas Recombiner/ Augmented Off-Gas l System" which provides a system description together with a safety analysis to establish basis for operation criteria. By the same submittal, NNEC0 proposed to amend its operating licenses DPR-21 and DPR-65 by changes l in the Environmental Technical Specifications for Millstone Nuclear Power Station, Units 1 and 2. These changes will limit the off-gas in-process activity rate to assure that the off-site doses resulting from potential accidents associated with operation of the Steam Dilution Augmented Off-gas System (SDA0GS) would not exceed established criteria.

This safety evaluation considers the installed SDA0GS which NNEC0 has proposed to operate. Although the Environmental Technical Specifications for Millstone Units 1 and 2 originally proposed by NNEC0 had a hiaher off-gas activity limit, the revised limit we have specified was discussed with and accepted with reservations by representatives of NNECO. NNEC0's reservations are based on their inability to justify or confirm the validity of.our accident model for the release of the radionuclide ,

inventory from the charcoal beds to the atmosphere.

Discussion ,

The augmented radioactive off-gas treatment system (A0GS) which NNECO installed to meet the radioactive effluent limits of the Nuclear Regulatory Commission is described in previous NNECO reports dated July 1973 and August 1975. The gaseous waste was to be treated sequentially by (1) 3 a hydrogen recombiner system and (2) a xenon-krypton treatment system.

During testing in the last quarter of 1975, an inherent-deficiency, referred to as the catalyst migration problem, was discovered which raised questions concerning the future operability of the recombiner portion of the off-gas system. This problem related to the air. recycle 1 concept employed in the Millstone Unit No.1 off-ges recombiner system. -]

a POR8706050385 780619 ADOCK 05000245 P DR.

1 i

l This air recycle feature made' the entire system susceptible to contami-nation with small particles of catalyst, a substance used to initiate the recombination of the hydrogen and oxygen gases in the recombiner.

During preoperational testing, it' was found that fine particles of this catalyst material had contaminated parts of the recombiner system which j would normally contain explosive mixtures of hydrogen and oxygen during l reactor operation, thus creating the potential for hydrogen explosions. q NNECO therefore modified their A0GS to a steam dilution recombiner system to eliminate the problem related to air recycle and catalyst migration.

I Design of the SDA0GS The proposed SDA0GS is a modification of the A0GS to utilize steam dilution instead of recycle air dilution of the off-gas stream. The second stage ejector' of the steam jet air ejector (SJAE) is modified to bypass the aftercondensers and discharge the motive steam and gas to the process pipe. The process stream, containing a gas / steam mixture, with hydrogen concentration diluted to below 4.0 volume percent, is transported to the recombiner system.

The recombiner system consists of two full capacity redundant trains each containing a preheater, a catalytic recombiner, an off-gas condenser, a jet compressor, an after-cooler. condenser and an associated instrumentation and control system. The preheater utilizes. plant auxiliary steam to preheat the gas / steam off-gas mixture from 250'F to 320 F. The superheated steam-diluted mixture enters the recombiner where free hydrogen and oxygen react in the presence of precious metal-coated metal base grid catalyst bed to form water. The gas exits the recombiner at approximately 730*F and enters the off-gas condenser where it is cooled to 130 F. The condensed water is drained to a.subcooler, cooled to 110 F and returned to the main condenser. A jet compressor provides the motive force for the offgas leaving the off-gas condenser.

The gas exits the jet compressor at 340 F and enters the after-cooler condenser before being transported to the xenon-krypton treatment system (XKS). The jet compressor is capable of discharging 50 SCFM at 22.7 psia. A minimum flow of 25 SCFM is required.by the XKS. l Makeup air from the plant station air system is injected automatically into the gas stream at the preheater to maintain system flow at 25 SCFM if condenser air inleakage falls below 25 SCFM.

The XKS is a low temperature (-20*F) charcoal adsorption system. The system consists of two sections: pretreatment and charcoal adsorption.

The pretreatment utilizes glycol cooler units which are designed to  ?

cool the off-gas to -20 F and dryers to dehumidify the steam to a dew-  !

point of -90 F. Two charcoal beds operate in series', each containing 4

J

l q

l l

1 11,000 pounds of activated charcoal. There are three thermocouples 1 in the first bed and one in the second bed. 'Each of the thermocouples j has temperature indication and high temperature alarm in the~ control 3 d

room. The high temperature alarm is set at 20*F above the operating temperature of -20 F. After decay in the charcoal beds, the offgas flows to HEPA filters prior to being released to the environs from the .

375 foot stack.

The Xe-Kr Building which houses the XKS is a seismic Category I j structure. In addition, the charcoal beds and associated process stream piping and valving in the Xe-Kr Building and the plant stack:

are designed to seismic Category I criteria.

Evaluation-

' At present, the unrecombined off-gas is transported to the stack via a buried delay pipe, which provides approximately 50 minutes of delay.

Routing the off-gas through the SDA0GS will provide _ additional delay of the noble gases and removal of the iodine isotopes by adsorption on the charcoal contained in the charcoal beds. Our evaluation of the expected performance under normal and abnormal conditions follows.

Evaluation of Normal Operation When the system is in operation, the charcoal beds are expected to provide delay times of 1.3 days for krypton and 50 days for xenon, while removing essentially all radiciodine isotopes. The ventilation system of the Xe-Kr Building ventilates'the air.in the building and-any small system outleakage to the elevated stack release point. The SDA0GS will be helium leak tested prior to operation to detect and (thus) minimize system leakage. 1 i

In the event of recombiner system malfunction, as indicated by instru-mentation alarms such as low preheater outlet temperature or high trans- 9 port-pipe hydrogen concentration,.the SDA0GS will be bypassed and the off-gas routed to the original delay pipe providing a minimum delay of-30 minutes prior to being exhausted through HEPA filters and the plant stack. The recombiner system utilizes main plant condensate, auxiliary steam, service air, instrument air and station A-C electric power.

Partial or total loss of these support services will be directly alarmed in the control room, or indirectly alarmed as a result of creating an upset condition in the gas stream, and may result in the bypass of the SDA0GS by the operator.

_4 Evaluation of Hydrogen Reaction )

System components, piping and valves are designed to withstand the peak pressure of a hydrogen explosion within the SDA0GS. The steam dilution in the system minimizes the probability of hydrogen ignitions prior to being recombined at the catalytic recombiner. The off-gas / steam mixture from the second stage ejector of the SJAE bypasses the after-condenser and discharges to the preheater of the recombiner system. The presence of steam dilution keeps the hydrogen concentration below the 4% volume detonable level . The minimum flow of 25 SCFM required by the XKS is provided by makeup air from the plant station air system and injected j automatically into the gas stream at the preheater when the main 1 condenser air inleakage is low. l I

Rupture discs in the system have been blanked off. It has been determined l j

that the actuation of the rupture discs is too slow for pressure relief in the event of a hydrogen detonation but could result in a subsequent off-gas leakage path. The treatment of liquid drains takes on added importance as the drain seals (e.g. loop seals) could be blown by a hydrogen detonation pressure transient. The liquid drains of the condensate from the SDA0GS are piped to the main condenser to minimize the probability of off-gas outleakage.

We have concluded that the SDA0GS will maintain system integrity under hydrogen reactions. The probability of a hydrogen detonation is minimized I by steam dilution and the problem of catalyst migration is eliminated by the absence of air recycle. We have also concluded that the probability of outleakage of offgas resultina from pressure transients is minimized by the elimination of rupture discs and the piping of condensate liquid drains to the main condenser.

Evaluation of Charcoal Fires There is a possibility that hydrogen reactions in the SDA0GS may initiate a fire in the charcoal delay beds. If the reaction is of the detonation type, the detonation front would move through the charcoal bed so rapidly (approximately 8500 fps) that it would be unlikely to initiate a fire.

If the reaction is a deflagration (fire) type, with a slow burning front of approximately 10 to 20 fps, the charcoal may be ignited, since the temperature of the hydrogen-oxygen reaction is about 4700 F. The charcoal used in the SDA0GS will have been previously exposed to temperatures of 1800 F to burn off entrained organic material which would be ignited at lower temperatures than the charcoal itself. Each charcoal tank is expected to be filled with 5.5 tons of activated charcoal, leaving a void fraction of approximately 0.41. In a hydrogen reaction, the

radiolytic oxygen would preferentially combine with the hydrogen. The only oxygen that would be available to sustain a charcoal fire would be that associated with the air inleakage and oxygen which is adsorbed on the charcoal. We have estimated that if the oxvoen in the tank partially oxidizes the char:oal (conversion to carbon monoxide), there would only be sufficient oxygen present in the void spaces and adsorbed on charcoal of a charcoal tank to consume a small fraction of the charcoal. However, with a flowing system, the burning will continue until the offgas flow to the tank is shut off. We estimate that under the expected conditions, if the deflagration has already passed through the tank, there will be sufficient air to oxidize 0.2 pounds of charcoal per minute. The first charcoal tank in each train will have three temperature indicators in the bed with alarms to alert operators in the control room when bed temperatures reach >20 F above the normal condition.

Operator action would isolate the SDA0GS Tn a timely fashion. The ]j second charcoal tank in each train will also be equipped with a tempera-ture element in each bed. The peak pressure associated with a charcoal (

fire is less than that associated with a hydrogen detonation. As  !

discussed previously, system integrity could be maintained during repeated hydrogen reactions. j A charcoal fire, which is assumed to occur locally near the inlet of the first charcoal tank in each stream, would result in local liberation j of noble gases. These noble gases would be reabsorbed on the downstream l portion of the first bed or on the second bed. Under the worst case, j it is expected that only a small portion of the charcoal radionuclide 4 inventory would be released in any postulated charcoal fire. However,  !

for conservatism, we analyzed the complete release of radionuclide 1 inventory in the charcoal beds as a result of explosion or fire. This evaluation follows.

Evaluation of Potential Accidents The Xe-Kr Building which houses the off gas processing system charcoal beds is a Seismic Category I structure. In addition, the charcoal beds, l the charcoal bed piping and isolation valves, the pretreatment equip- i I

ment and the plant stack are designed to the Seismic Category I criteria.

The SJAE and the recombiner system are located within the non-Seismic l Category I turbine building, i While our evaluation indicates that the SDA0GS integrity will be main-tained, under hydrogen explosion or charcoal fire, we have considered '

the failure of the SDA0GS at the Xe-Kr Building and the simultaneous failure of the off-gas piping in the turbine building. We considered the release from non-Category I piping to occur at the turbine building.

The failure of the XKS in the Xe-Kr Building coupled with the failure of building ventilation system could result in the charcoal bed inventory ground level release during the first hour after the incident. Table 1 presents the basic assumptions used in our analysis. The source term

released is the average inventory'in process and transport piping, recombiner system, an equilibrium loading on the charcoal beds and -

an hour's release with a delay of 5 minutes from the=SJAE to the point of failure (this assumes that there is no isolation of the SJAE for-an hour following system failure).

Table 2 presents the estimated dose consequences at the exclusion area boundary from the three contributing sources, i.e., the one-hour release 4 I

from the SJAE, the release from inventory in process, transport piping, and recombiner system and the release of Xe and Kr from the Xe-Kr Building.

The dose consequences are well within the guidelines 'of 10 CFR Part 100 and meet the whole body _ dose criteria in Standard Review Plan (SRP) a 15.7.1 and are, therefore, acceptable.

Evaluation of Technical Specifications Technical Specification 2.4.2.2.K has been revised to include a lower off-gas release rate of the SJAE to limit dose consequence of the failure of the entire 50A0GS to 5 rem or less at the exclusion area boundary.

Based on the accident analysis discussed above, we have determined that SDA0GS operation at an off-gas release rate at'the air ejector of no '

more than 1.47 Ci/sec (<0.52 Ci/sec at 30 minutes) will limit the-potential consequences of the total failure of the off-gas system, including continued operation of the air ejector for one hour, to less than 1 rem thyroid dose and 5 rem total body dose over a period of two hours at the site exclusion boundary. These dose values are also within the guidelines of 10 CFR Part 100 and SRP 15.7.1. The noble gas inprocess activity limit at 1.47 Ci/sec also provides a degree of assurance that offgas system operation will not continue with excessive fuel failures.

ETS 2.4.2.2G has been revised to eliminate reference to an interim offgas system. The need for such a system will be eliminated when the SDA0GS is placed in operation.

Environmental Considerations We have determined that the amendments do not authorize a change in effluent types or a significant increase in the total amounts of-effluents nor an increase in power level and will not result in any significant environmental impact. Having made this determination,'we have further concluded that the amendments involve ~an action which is insignificant from the standpoint of environmental impact and pursuant to 10 CFR 651.5(d)(4), that an environmental impact statement or-negative declaration and an environmental impact appraisal need not be prepared in connection with the issuance of these amendments.

l Conclusion We have concluded, based on the considerations discussed above, that:

(1) because the amendments do not involve a significant increase in the probability or consequences of accidents previously considered and do not involve a significant decrease in a safety margin, the amendments do not involve a significant hazards consideration, (2) there is reasonable assurance that the health and safety of the public will not be endangered by operation in the proposed manner, and (3) such  !

activities will be conducted in compliance with the Commission's.

regulations and the issuance of these amendments will not be inimical to the common defense and security or to the health and safety of the public. ,

Attached: .

Tables 1 and 2 l l

1 Date: June 19,1978 l

l l

T ABl.E 1 j i

I,SSUWT10:,5 USED TO ESTI!'ATL CONSEQUlt!CES Of A j ft,1 LURE 1!! 11?L l'.illSTO:a 1 STEh:1 DILUTIOil AUCliEliTED Off-GAS SYSTEl1 350,000 uC1/sec Noble gas release rate 6 3011inutes 4 f

1971 GE liix e Distriliation of !!uclides l 1

l l

1 Ilesidence 11ma on Charcoal Beci' - )

1 1

1186 hours0.0137 days <br />0.329 hours <br />0.00196 weeks <br />4.51273e-4 months <br />

):enon 31 hours3.587963e-4 days <br />0.00861 hours <br />5.125661e-5 weeks <br />1.17955e-5 months <br /> Krypton i

1 i

i

s:1;ua, Tim for Sterr.: Jet !.ir Eject or 1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br /> j l

yj r} ci E>:clusion to . L undary 3 i

-3

):e-Kr Bu i l d n _, (t rcu .uf l evel ) 2.6 x 10 sec/rf  ;'

-4 3 T ert>ine 1:ui idin.! 'y au:,.i IeveI) 6.4 x 10 sec/ci e

4 e

R

,. . wx,k ,n

% 41s P ,

- y. ,

-- . ~ . .,

+g-

  • e ameimmenee,emie*, -, p.nesne.eec e ,y,..m

TACl.E 2 Estimted Com uquences of a Fallure' of the e

14illstone 15tecm !)ilution Aurpented Off gas System Total e dy Gamma Dose O Exclusion Area Boundary Souroe of ik1eese (,$J,0Q, (_Ilep) 1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br />. release from SJAL 0. 'y f>

Inventory from piping and recorbiner r.y:,1va 0.00 invento:y frora X13 2_. 95_

Total 3.5 Ibic: Off-gs reh a'- rua of 1.4 / :i/si t at ' air cjector monitor would corresp'.nd 1.e totci loi.' pa , dem of 5 I: era.

e 9

0

  1. e e

q g.' [

,.p i

= u .

.Am-