ML20235T451

Safety Evaluation Re Facility Core Support Flow Vent Sys. Continued Operation of Facility W/Current Core Support Flow Sys Configuration Acceptable

ML20235T451
Person / Time
Site:	Fort Saint Vrain
Issue date:	02/24/1989
From:	Office of Nuclear Reactor Regulation
To:
Shared Package
ML20235T441	List: ML20235T440 ML20235T451
References
NUDOCS 8903080290
Download: ML20235T451 (4)

Text

E, = . . ,

L.

and; i E o ' UNITED STATES g

8 g NUCLEAR REGULATORY COMMISSION

, g <; WASHINGTON, D. C. 20655

%...../

SAFETY EVALUATION _.BY THE OFFICE OF NUCLEAR REACTOR REGULATION RELATED TO CORE SUPPORT FLOOR. VENT SYSTEM PUBLIC SERVICE. COMPANY OF COLORADO FORT ST. VRAIN NUCLEAR GENERATING STATION DOCKET NO. 50-267

1.0 INTRODUCTION AND BACKGROUND

The licensee has reported leakage of the Fort St. Vrain (FSV) reactor through the core support floor (CSF) as early as 1982. By letter dated May 7,1982, the licensee reported thet there was leakage from the reactor primary coolant system to the CSF vent system. There was also leakage from the liner cooling system (LCS) tubes in the CSF. The licensee supplied an-evaluatio, of the acceptability of this condition. There is no record of the staff reviewing this issue.

More recently, two events involving the CSF vent system were reported.

These events occurred on April 4 and 7, 1988. In both events, the reactor was scrammed from a high power level. Primary coolant was released into the CSF vent system in excess of the system's capacity. The system's relief valve (V-6389) lifted and an uncontrolled release of primary coolant took place. Both releases were below 15 percent of the applicable limits in the plant Technical Specifications (TS). The offsite doses associated with these releases are fractions of a millirem, because the releases were processed by the Reactor Building's filter system.

The events led to a concern about the operation of the CSF vent system and its potential role in releasing radioactivity to the environment. The staff elected to review a recent evaluation of the CSF vent system submitted by the licensee dated March 18, 1988.

Subsequent to the start of the staff's review, on July 6, 1988 the licensee discovered that a single valve (V-111063) which is in the CSF vent system flow path was nearly blocked shut and had become inoperable. This discovery led to an additional concern that a single failure could defeat the function of the CSF vent system. Potential concerns involved the possibility of the CSF structure deforming and blocking the flow of primary coolant needed to remove reactor decay heat.

2.0 EVALUATION 2.1 Evaluation of Low Level Leakage via the CSF Vent System l The CSF vent system could be a path for a steady release of low level radio-activity escaping to the environment. The following measures are in place to prevent this occurrence.

8903080290 890224 PDR ADOCK 05000267 ll P PDC

i First, the level of activity in the reactor's primary coolant sy* tem is sampled weekly under TS SR 5.2.11. This assures that the current low levels of activity are maintained well below TS limits. The flow through the CSF vent system is continuously monitored by a flow element (FT-6375). The leakage limits through this flow path is not directly limited by the TS, but the general leakage limi,ts of LC0 4.2.9 which apply to the reactor vessel closures would provide a basis for judging acceptable limits. In reality, the small size of the CSF vent system piping (about one inch) precludes very large flows through the system. The overall release of radioactivity from the system is also governed by the overall requirements of the TS in Section 8, which are specifically concerned with radioactive effluents. All release paths from the CSF vent system are monitored by the reactor building stack monitors and all releases pass through the reactor building filter system.

The licensee has calculated that the current releases experienced by the plant are about two percent of the Maximum Permissible Concentrations allowed by 10 CFR Tiart 20. Significant increases in either the CSF leak rate or the primary coolant activity levels would be required before CSF routine leakage would be a problem.

Additionally, the licensee has taken certain corrective actions to assure the optimal functioning of the CSF vent system. This has included servicing of the waste gas compressors and the CSF system filters to assure the system operates at full capacity. The setpoint on the system relief valve (V-6389) has been increased to 10 psig to reduce its tendency to lift on minor system transients. The licensee has also evaluated operating the CSF vent system with a higher setpoint, ie. at 100 psig.

In view of the above, the staff finds that there are adequate protective measures in place to assure that releases from the CSF vent system are adequately monitored. The system's performance is adequately monitored and it is highly unlikely that significant uncontrolled releases can take place.

The staff concluded that this aspect of the CSF vent system is acceptable.

2.2 CSF Vent System Accident Considerations Accident considerations for the CSF vent system involve two possible failure modes of the system. The first failure mode is that the vent system line or lines is sheared and the vent system is open to the reactor building environment.

The second case is when the single vent line is blocked and the CSF internal pressure rises to the normal reactor coolant primary system pressure.

Shearing failure of the CSF vent line potentially leads to very slow depressurization of the reactor vessel. Shearing of other lines, such as lines in the helium purification system piping has already been analyzed by the licensee and reviewed by the staff as part of the original license review. The CSF vent line is smaller than the cases already analyzed, and so the consequences of that accident are much less severe, and would result in millirem doses.

The currently estimated leak rate would be about 20 lbs/ hour. The licensee's instrumentation has measured flows of up to 10 lbs/hr during shutdown transients following a reactor scram. The exact mechanism for determining the leak rate cannot be modeled. However, the leak rate is limited by the small pipe diameter and the resistance to flow through that pipe. These parameters are well established. Therefore, the staff accepts the licensee's basis that the leak rate would be less than 20 lbs/hr. A leak rate of 140 lbs/ hour would be needed to reach 10 CFR Part 20 levels.

The radiological impact of such releases can be calculated from established system characteristics. The release into the Reactor Building can be determined from the flow rate calculated above, and the known concentration of radionuclides in the reactor primary coolant. All releases within the Reactor Building are processed by the Reactor Building Ventilation Systems filtered release path. The filtering and dispersion characteristics of this engineered safeguard are well established. Hence, the staff accepts the licensee's conclusions concerning these releases.

Reverse pressurization of the CSF is a more complex issue. This can occur if the pressure control system fails, or a single failure causes the valve V-111063 to fail closed. Then the CSF pressure will gradually come into equilibrium with the reactor coolant system primary pressure of about 600 to 700 psig. This potentially becomes a problem only if a rapid depressurization accident occurs. In that case, the CSF floor is now internally pressurized, while the primary system pressure drops rapidly to atmospheric. The CSF liner will tend to bulge out under an internal pressure differential of more than about 200 to 220 psid. The deformation of the CSF liner could block the annular flow path needed to cool the core after the rapid depressur.ization accident.

The staff has previously examined the probability of the rapid depressurization accident in a Safety Evaluation and a Technical Evaluation Report issued on July 21,1988. The staff concluded that the probability of this accident was about 3x10E-5 per year. Likelihood of the single valve V-111063 failing independently is about 3x10E-2 per year. Thus, the overall probability of the CSF failing in this mode is about 1x10E-6 per year.

The significance of an event with a probability of 1x10E-6 per year can be considered as follows: Generic Letter No. 88-20 dated November 23, 1988 concerned individual plant examination for severe accident vulnerabilities.

Appendix 2 suggests the criteria for a significant functional sequence is a contribution of 1x10E-6 or greater per (reactor) year to core damage. Thus, this sequence could be considered as just at the limit of significant vulnerability and require further consideration.

The licensee has initiated tighter controls over valve V-111063. This includes sealing the valve open and more frequent surveillance of its position. Additionally, the licensee has installed a pressure indicator upstream of this valve, which effectively monitors the performance of the CSF vent system controls and warns of potential overpressurization of the CSF internal space.

p

\.

I

e. j j

I-L The resident inspector has noted that the licensee has also installed an l additional bypass line from the CSF vent directly to the gaseous waste vacuum tank. This provides an additional vent path should V-111063 fail closed.

We believe these additional measures sufficiently reduce the vulnerability to this functional sequence, so that the CSF vent system is acceptable in this regard.

The licensee has also examined the role of the CSF vent system in other internal and external events. These include: design basis earthquake and tornado, high energy line break, Appendix R fires, and, permanent loss of forced circulation cooling. In each case, malfunctions of the CSF vent system had no significant consequences on the results of the already analyzed accident scenarios or the consequences of these accidents.

3.0 CONCLUSION

S In conclusion, the staff has reviewed two concerns about the CSF vent system. The staff examined the possibility of significant routine leakage through this path from the reactor primary coolant system to the environment.

The steff also examined the system's impact on potential accident situations.

In the first case, the staff found that adequate controls are in place to assure that routine operation of the system is monitored and clear limits are set on what are. acceptable releases from that routine operation. In the second case, the staff concludes that the CSF vent system operating in its current condition does not significantly increase the overall risk from all analyzed reactor accident sequences at FSV. Thus the staff concludes that continued operation of FSV with the current CSF vent system configuration is acceptable.

Dated: February 24, 1989 Principal Contributor: Kenneth L. Heitner, PD-IV

____.m___.______-__..____..__u .__--_w -------- wJ