ML20035F142

Safety Evaluation Re SEP Topic III-5.A, Effects of Pipe Breaks Inside Containment

ML20035F142
Person / Time
Site:	Haddam Neck File:Connecticut Yankee Atomic Power Co icon.png
Issue date:	04/14/1993
From:	Office of Nuclear Reactor Regulation
To:
Shared Package
ML20035F141	List: ML20035F140 ML20035F142
References
TASK-03-05.A, TASK-3-5.A, TASK-RR NUDOCS 9304200325
Download: ML20035F142 (8)

Text

__.

j J

/'gom "Ic,

c l

'e UNITED STATES

1. ,'n NUCLEAR REGULATORY COMMISSION 2

{

.,E WASHINGTON, D. C. 20555 t,

a

%, '...../

{

l t

1 SAFETY EVALUATION BY THE OFFICE OF NUCLEAR REACTOR REGULATION SYSTEMATIC EVALUATION PROGRAM TOPIC III-5.A.

l

" EFFECTS OF PIPE BREAKS INSIDE CONTAINMENT" f

CONNECTICUT YANKEE ATOMIC POWER COMPANY l

1 HADDAM NECK PLANT DOCKET NO. 50-213 j

i 1

1.0 INTRODUCTION

i An evaluation performed by the staff as part of the Systematic Evaluation i

Program (SEP) under SEP Topic III-5.A, " Effects of Pipe Breaks Inside

(

Containment" was forwarded to Connecticut Yankee Atomic Power Company l

(CYAPC0/the licensee) by letter dated October 12, 1982.- In that letter the i

NRC staff concluded that (1) the definition of high energy fluid systems, (2) the determination of high energy pipe break locations and break types and (3) the pipe whip and jet impingement assumptions are generally consistent with currently accepted standards. The staff review, however, did identify the t

following 10 issues concerning pipe breaks inside containment which were not i

adequately addressed:

I

(

1.

cascading effect sequences, l

2.

jet impingement from circumferential breaks, 3.

strain level functionality criteria, 4.

containment integrity criteria, 5.

jet impingement effects on target-piping, 6.

effects on instrumentation, s

7.

main coolant loop breaks, i '

9.

main steam / main feedwater interactions, and 8.

plant shutdown Method 3, j

10.

core deluge piping break effects.

1 By letters dated December 17, 1984, and January 29,.1993, CYAPC0 responded to the above issues. An evaluation of these responses is provided below.

j 2.0 EVALUATION l

2.1 CASCADING EFFECT SEOUENCES 4

By letter dated December 17, 1984, CYAPC0 provided the staff with their review l

of the effects of cascading sequences.

The licensee stated that cascading is i

controlled by separation of piping or by physical barriers. For insttnce, cascading effects due to a pipe break in'a loop area are limited to that particular loop or breaks in the steam supply system are limited to that j

b i

9304200325 930414 PDR ADOCK 05000213 i

P PDR

~.

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

....... f

l.

l l

' particular train. Regarding multiple blowdown effects, the licensee states t

the primary system is still limited by the double ended guillotine break of the reactor coolant system (RCS) loop piping, and the secondary side is l

limited by a main steam line break together with the associated feedwater line break which were previously analyzed in their September 17, 1982 submittal.

The licensee states that cascading effects including multiple blowdown transients were addressed and there are no scenarios more limiting than those previously identified.

Based on the licensee's response, the staff concludes that it is conservative to assume that the particular safety system associated l

with or in the area with the postulated pipe break becomes inoperative.

Therefore, the staff agrees that the most limiting scenarios have been addressed and the effects of cascading sequences is controlled by separation of piping or by physical barriers.

2.2 JET IMPINGEMENT FROM CIRCUMFERENTIAL BREAKS By letter dated October 12, 1982, the staff stated that CYAPC0 needed to evaluate circumferential breaks.

Furthermore, in the case of circumferential breaks, jets in conjunction with pipe whip should be considered to sweep the arc travelled during the whip. CYAPC0 has stated that the high energy line break (HELB) analysis considered circumferential breaks and that the jet was assumed to travel the arc defined by the whipping pipe.

Based on the licensee response, the staff considers this issue resolved.

2.3 STRAIN LEVEL FUNCTIONALITY CRITERIA By letter dated October 12, 1982, the staff requested CYAPC0 to clarify the allowable strain level utilized in its HELB analysis.

By letter dated December 17, 1984, CYAPC0 stated the HELB analysis did not take any credit for this type of detailed analysis on impacted equipment. Based on this response the staff concludes that this concern is not applicable to the Haddam Neck Plant HELB analysis and considers this issue reselved.

2.4 CONTAINMENT INTEGRITY CRITERIA By letter dated October 12, 1982, the staff requested CYAPC0 to provide additional information to justify their conclusion that no pipe break could fail the material or degrade in any manner the containment liner. By letter dated December 17,1984, CYAPC0 stated that the crane wall protects the containment liner from most pipe breaks. The containment building wall consists of a 4-1/2 feet thick right circular cylindrical concrete wall lined with a thick steel plate. The liner is 1/4 inch thick on the bottom, 3/8 inch thick on the cylindrical walls and 1/2 inch thick on the dome. The liner plate is flush with the concrete containment wall and attached by Nelson studs 24" on center.

CYAPC0 has determined that the limiting break from a load j

standpoint is a 12" feedwater line break. The feedwater pipe is perpendicular to the containment wall and the only type of interaction which could possibly affect the liner plate integrity would be a ripping action from the jagged edge of a broken pipe.

CYAPC0 states that there does not exist sufficient loading or interaction of the type which could possibly affect the integrity of the liner plate for the most limiting pipe whip or jet impingement.

i l

i In addition, by letter dated January 29, 1993, the licensee stated that as part of their current HELB reverification program, a detailed HELB evaluation was performed on the main steam and feedwater lines inside containment to evaluate the effects of a pipe break on the containment structure. The i

calculations developed main steam and feedwater jet impingement loads and reviewed the potential for pipe interaction with the containment structure.

-l Based on these calculations and evaluations, which considered the piping geometry and location, the support scheme, and the location of pipe breaks, i

the licensee concluded that:

l 1.

Both the main steam and feedwater piping would not impat.t the containment wall following a pipe break, and 2.

the containment structure is capable of withstanding the resultant jet loads without degrading the integrity of the liner plate.

Based on the above the staff concludes that there is no HELB inside containment that could cause damage to the containment liner to degrade the liner's function as an environmental barrier. The staff considers this issue resolved.

2.5 JET IMPINGEMENT EFFECTS ON TARGET PIPING By letter dated October 12, 1982, the staff requested CYAPC0 to clarify how they used the size differential criteria in the jet impingement effects evaluation. By letter dated December 17, 1984, CYAPC0 stated an evaluation of jet impingement of piping targets was performed. CYAPC0 stated that due primarily to the physical separation of required safety systems, it is not possible to lose all three of the available shutdown methods. Therefore, CYAPC0 concludes that the plant can achieve a safe shutdown when considering the effects of jet impingement on piping regardless of ratio of pipe sizes.

Based on this response the staff concludes this issue resolved.

2.6 EFFECTS ON INSTRUMENTATION An evaluation of the minimum instrumentation required for safe shutdown, assuming a worst case incident where nonphysically separated instrumentation is rendered inoperable by a single HELB, determined the following minimum sets of instrumentation for each shutdown method-

\\

SHUTDOWN METHOD RE001 RED INSTRUMENTS

1. Main Feed a.

Pressurizer Level b.

Pressurizer Pressure c.

Steam Generator Level d.

Loop T-hot or Core Exit Thermocouples L

..-.-.---.J

.. )

2. Auxiliary Feedwater a.

Pressurizer Level b.

Pressurizer Pressure c.

Steam Generator Level d.

Loop T-hot or Core Exit Thermocouples e.

Demineralized Water Storage Tank Level

3. Feed-and-Bleed a.

Pressurizer Level b.

Pressurizer Pressure

~

c.

Loop T-hot or Core Exit Thermocouples d.

Refueling Water Storage Tank Level, Volume Control Tank Level e.

Pressurizer Relief Valve Monitors, Containment Water l

Level, and Containment High Range Radiation Detectors f.

Containment Pressure g.

RCS Wide-range pressure A discussion of the above instrumentation follows:

a.

Pressurizer Level - CYAPC0 replaced the transmitters for the main control l

board mounted equipment for the three loops of pressurizer level during Cycle 14. While this upgrade provided separation of safety and non-safety l

systems, it did not reduce the degree of independence, separation, and isolation provided in the original design. Therefore, in the case of the loss of pressurizer level instrumentation due to a HELB, the plant could shutdown using feed-and-bleed which does not need pressurizer level or pressurizer pressure indication. The licensee has noted that feed-and-bleed is the least c

desirable of the shutdown methods and would only be used for a HELB where the reactor coolant system remains intact and the steam generators or systems l

servicing them are not available.

Plant emergency procedures would not l

require the operator to use feed-and-bleed for the loss of pressurizer level due to a HELB.

Even if there is a loss of pressurizer level or pressure indication, CYAPCO's emergency operating procedures (E0Ps)-would continue to use main feed or auxiliary feedwater and rely on other indications to verify primary system heat removal. As noted earlier CYAPC0 considers feed-and-bleed j

to be the option of last choice when all other options have been exhausted.

l The E0Ps support this philosophy. However, losing pressurizer level indication will inhibit safety injection termination. The termination criteria for safety injection is based on part in regaining pressurizer level.

i This means in effect, a partial feed-and-bleed (i.e., the operator will inject i

but will not open the power operated relief valves) is used in conjunction with Plant Shutdown Methods 1 and 2.

Based on the above, the staff agrees that even with a loss of pressurizer level there is sufficient means to safely cooldown the plant and that the licensee has minimized reliance on feed-and-bleed to cool down the plant.

l

.. b.

Pressurizer Pressure - CYAPC0 states that in the case where pressurizer pressure instrumentation is lost due to a HELB, RCS pressure instruments would be used instead. The RCS pressure instruments are physically separated from the pressurizer pressure instrumentation. The staff agrees that the RCS pressure instrumentation could be used instead of pressurizer pressure instrumentation for use in Plant Shutdown Methods 1, 2, or 3.

i c.

Steam Generator Level - CYAPC0 states that modifications have been made to the wide-range level indication instrumentation.

Each steam generator now has two redundant Class IE wide-range level detectors.

In addition, even though one channel of steam generator level instrumentation for all four steam generators is routed in one conduit, the redundant channels have been routed taking into account HELBs. As part of the Steam Generator Level Instrumentation Upgrade, the licensee will modify the steam generator narrow-range instrumentation such that it is redundant and physically separated.

Based on the above, the staff concludes that the steam generator level i

instrumentation is redundant and physically separated and this issue is resolved.

l l

l d.

Loop T or Core Exit Thermocouples - The licensee states that the l

cabling of"these instruments inside containment was upgraded and is now l

physically separated and redundant. The staff considers this issue resolved.

l e.

Refueling Water Storage Tank Level, Volume Control Tar.k Level, and l

Demineralized Water Storage Tank Level - The licensee states that all these instruments are outside the containment and are not affected by pipe breaks inside containment. The staff considers this issue resolved.

f.

Pressurizer Relief Valve Monitor, Containment Water Level, Containment High Range Radiation Detectors - Containment Water Level and Containment High Range Radiation Detectors were installed in response to NUREG-0737 and are all physically separated and redundant.

Pressurizer Relief Valve Monitor, which was also installed in response to NUREG-0737, utilizes a single channel.

l However, other indications of relief valve position are available such as

~

temperature elements in the common power-operated relief valve (PORV) header and in the line to the pressurizer relief tank, and relief tank level, pressure, or temperature.

Based on_the above, the staff concurs that sufficient instrumentation exists to determine Pressurizer Relief Valve position and considers this issue resolved.

2.7 MAIN COOLANT LOOP BREAKS By letter dated October-12, 1986, the staff requested CYAPC0 to verify that the seismic loads assumed in Unresolved Safety Issue (USI) A-2 are compatible with the seismic loads assumed in the SEP and that the reactor coolant pressure boundary (RCPB) leakage detection systems, the type and number of the systems provided be commensurate with the guidelines of Regulatory Guide 1.45.

By letters dated December 17, 1984 and June 16, 1989, CYAPC0 confirmed that SEP seismic analysis performed for the Haddam Neck Plant demonstrated that the maximum bending movements do not exceed the allowable limit specified in 1

l'-

l

.. l Generic Letter (GL) 84-04.

This seismic analysis identified certain RCS modifications necessary to ensure that loads on the RCS piping will be within i

those assumed in USI A-2.

All but one of the subject RCS modifications were i

I completed during the 1987-1988 refueling outage. The only remaining modification was the replacement of steam generator hold-down bolts. The modifications to the steam generator hold-down bolts were not necessary as the i

safety factors on the bolts were determined to be within allowable limits. By License Amendment No. 116, dated May 31, 1989, the staff approved technical specifications that govern the operability of leakage detection systems that are capable of detecting an RCS leak of I gpm for 4 hours4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br />. By letter dated July 11, 1989, the staff concluded that CYAPC0 met the conditions specified in GL 84-04 so that the asymmetric blowdown loads resulting from double ended pipe breaks in main coolant loop piping need not be considered as a design basis for the Haddam Neck Plant. Based on the above, the staff considers this issue to be resolved.

1 2.8 PLANT SHUTDOWN METHOD 3 By letter dated October 12, 1982, the staff requested CYAPC0 to identify under what circumstances plant shutdown method 3 (feed and bleed) is intended j

to be used and how the energy is removed from the primary system.

By letter l

dated December 17, 1984, CYAPC0 stated that Method 3 is to be used when the reactor coolant pressure boundary remains intact and the steam generators or systems servicing them are not available for normal heat removal. Reliance on feed-and-bleed is minimal. Therefore, feed-and-bleed is needed when:

I 1.

a pipe break outside containment were to compromise capability to inject either main or auxiliary feedwater into the steam generators, 2.

a pipe break inside containment were to compromise capability i

to utilize the steam generators to remove core decay heat, and 3.

some other failure resulting in the loss of steam generator decay heat removal.

Feed-and-bleed entails injecting water into the RCS with a charging pump (or HPSI pump) and allowing water to discharge from the pressurizer power-operated relief valves (PORVs) into the pressurizer relief tank and then into the containment sump. The water would be drawn from the sump, cooled by the.

4 residual heat removal heat exchangers and returned to the suction of the charging pumps or high-pressure safety injection (HPSI) pumps. CYAPC0 has qualified the low temperature overpressure protection system (pipes and valves) so that it is qualified for the temperature, pressure, and flow conditions that would exist.

Feed and bleed is the least desirable shutdown method. However, for a limited set of pipe break locations in portions of the main steam, feedwater and auxiliary steam piping feed-and-bleed is the primary cooling method.

The licensee has committed to several modifications to decrease the reliance on Plant Shutdown Method 3, feed-and-bleed. The licensee proposed the following modifications to be completed during the next two outages:

m

i l

i l

i '

1.

implement a dedicated erosion / corrosion program for the piping in the Terry turbine building to reduce the probability of pipe break in the Terry turbine room, 2.

install a motor driven auxiliary feedwater pump, powered by l

emergency onsite power source, outside of the Terry turbine room, 3.

provide direct injection flow path to the steam generators, 4.

dedicate the demineralized water storage tank to the auxiliary feedwater system, and 5.

house the electric auxiliary feedwater pump, the automatic initiation support skids and direct injection piping and valves in a new seismically designed enclosure.

The licensee is also considering physical separation of the Terry turbines and pumps which would further reduce the reliance on feed-and-bleed.

By License Amendment dated February 24, 1992, the staff agreed that the above i

modifications and erosion / corrosion program would reduce the probability of a HELB in the Terry turbine and the reliance on feed-and-bleed to an acceptable level. The licensee has adequately responded to our request for information regarding feed-and-bleed.

2.9 MAIN STEAM /FEEDWATER INTERACTIONS By letter dated October 12, 1982, the staff requested CYAPC0 to clarify the apparent inconsistency between the matrix and interaction evaluation of the main steam line breaks. By letter dated December 17, 1984, CYAPC0 stated that the scenario given in the interaction evaluation is correct and that for a given main steam line break, the only potential feedwater line interaction occurs with the feedwater line corresponding to the steam generator. The matrix is in error but the evaluation is correctly stated.

Based on the above the staff considers this issue resolved.

2.10 CORE DELUGE PIPING BREAKS By letter dated October 12, 1982, the staff questioned CYAPC0, coald a single active failure in the emergency power system be a more limiting case than the loss of the motor operated valve in the unaffected train for o core deluge line break.

By letter dated January 29,1993, CYAPC0 stated ns result of the LOCA reanalyses and the analyses performed in support of perranent modifications to resolve various single failure vulnerabilities associated with the ECCS, CYAPC0 did determine that the limiting single failure after a i

core deluge line break is the failure of a diesel. However, CYAPC0 performed-analyses which demonstrate that for a break of this size, adequate results are obtained for the injection phase without any credit from the low-pressure injection system (LPSI). A postulated core deluge line break would behave differently in the recirculation phase than the same size break occurring i

elsewhere in the RCS. This is due to the higher LPSI flowrate associated with 4

I i

.. j a core deluge line break which causes an earlier time to recirculation. The long-term modifications described in CYAPCO's April 1, 1987, letter have been.

analyzed and show acceptable results for all postulated single failures.

The staff concurs with the licensee's analysis and considers this issue resolved.

i

3.0 CONCLUSION

t l

The staff has completed its review of SEP Topic III-5.A and concludes that if a pipe were to break inside containment, the plant can safely shutdown without-loss of containment integrity. The licensee has provided three shutdown methods (Main Feed, Auxiliary.Feedwater, and Feed and Bleed). There is no-HELB !aside containment which could incapacitate all three shutdown methods.

l In addition, piping integrity is maintained by two inspection programs:

inservice inspection and erosion / corrosion. While these-programs cannot preclude the possibility of a high energy line break inside containment they do improve the probability that a high energy line break will not occur.

Based on the above, the staff considers SEP Topic III-5.A to be complete.

Principal Contributor:

A. Wang Date: April 14, 1993 i

I t

l l

}

4

-