ML18086A711
| ML18086A711 | |
| Person / Time | |
|---|---|
| Site: | Salem  |
| Issue date: | 06/16/1981 |
| From: | Reiter L, Rippe R, Thomas Taylor Public Service Enterprise Group |
| To: | |
| Shared Package | |
| ML18086A710 | List: |
| References | |
| S-1-R200-MSE-09, S-1-R200-MSE-090, S-1-R200-MSE-9, S-1-R200-MSE-90, NUDOCS 8106230648 | |
| Download: ML18086A711 (9) | |
Text
- ..
SGS/~-*
S-l-R200-MSE-090 OPS~G
'AGE 1 CF REVISION NO.
CATE.:
9 0
6/16/81 The Energy People AT TIT~E:NO. 1 UNIT.
SALEM NUCLEAR GENERATING STATION
1.0 PURPOSE
This evaluation assesses the safety impact on continued reactor operation of a TV Camera Lens Housing lost in the reactor coolant system on oi about June 9, 1981.
The evaluation includes consideration of other previously identified debris in the system, and addresses the non-existence of an Unreviewed Safety Question (10CFR50.59)
2.0 REFERENCES
- 1.
List of items missing from TV inspec£ion boat between 12:30 a.m., June 9 to 11:00 a.m. June 10, by Buck Joiner,
~estinghouse Nuclear Service Division, dated June 10, 1981.
- 2.
Letter dated June 14, 1981, F. Noon (W) to H. J. Midura, Westinghouse Safety Evaluation, Television Camera Lens Housing, Salem Unit I.
- 3.
Westinghouse letter PEB-79-051 (Proprietary Class II) dated June 1, 1979, E. A. Bassler (~) to M. G. Arlotti
(~).
- 4.
Mechanical Safety Evaluation S-l-R200-MSE-086-Rev. 0 Metal Impact Monitoring System Indication -
Salem Unit No. 1.
3.0 DISCUSSION
On June 9, 1981, No. 11 steam generator inlet channel head was entered for the purpose of investigating the cause of a previous Loose Parts Monitoring System indication (Ref. 4).
Not finding an apparent cause of the indication in the inlet channel head, examination of the No. 11 hot leg piping was begun by floating a TV camera (Elmira 1250) down the hot leg (approximately 1/2 full of water) on a block of styrofoam.
Apparent interference with the RTD instrumentation and turbulence caused by the 14-inch RHR suction line broke the camera head, a length of nylon lanyard, and a piece of tape, loose from the camera assembly (Ref. 1).
The former items are missing in the piping.
PB139/0l 1 MDF 10.5-la
f'EVISION:
0 PAGE 2
OF DATE:
6/16/81 In order to evaluate the impact on plant operational safety of these parts remaining in the plant systems, four possible scena~ios were postulated, and reasonably probable consequences evaluated.
The scenarios are:
- 1.
The camera remains in the hot leg.
- 2.
The camera entered the RHR system, and resides anywhere in the system, possibly having passed through the 11 RHR pump and corning to rest on the 11 RHR heat exchanger tube sheet.
- 3.
The cam~ra entered the RHR system, passed through the pump, and was broken into pieces sufficiently small to pass through the 5/B=in. RHR heat exchanger tubes and on toward the RCS cold legs and reactor vessel bottom.
- 4.
The camera entered the RHR system, passed through the pump deformed but essentially intact, and passed on to the reactor vessel bottom via the RHR heat exchanger by-pass.
Consequences to each of the above scenarios were evaluated in following paragraphs A through D, while more general concerns are address!d in paragraphs E and F:
A.
Camera in Hot Leg P8139/10 2 The material will probably move upon resumption of RCS flow.
Starting 11 RCP first will cause normal flow in the leg, pushing the material into the 11 steam generator inlet channel head.
Neither the tube sheet nor the tube-to-tube welds will be damaged by such material (Ref. 2, Par. 1).
Subsequent breakup of the camera would result in its transport through the steam generator tubes to the reactor vessel (RPV) bottoms.
Consequences of this will be evaluated for a later scenario.
Starting an RCP other than No. 11 first will result in backflow, possibly sweeping the camera into the RPV upper internals.
Residence here will not be long, as normal flow will sweep the parts back into the loops--no damage to the upper internals is expected.
There is a very small possibility that a control rod could become jammed by a camera part, however the normal control rod exercise program will verify the freedom of all rods during operation.
Even if one control rod should become jammed or inoperable, the plant has been analyzed for one "stuck" rod, and therefore this should present no safety concern.
(Ref. 1, Par. 1)
MDF 10. 5-lb
.. EVISION:
Q 3
o;:
9 DATE:
6/16/81 Consideration of only one jammed rod, rather than a multiple"common-mode" type occurrance, is justified on the basis of ~he extreme improbability of the occurrance.
Diametral clearance between a guide tube and a control rod is only 0.067 -
inches (approximately 1/32-in. annular clearance).
One would have to postulate existence of a particle with a size range to fit this clearance, yet a taper to then subsequently* jam upon rod movement.
Very few camera parts fall into this size range prior to deformation.
Also, flow in the control rod/guide tube interfa~e region is generally upward, making it extremely unlikely for a particle to fall out of the flow in just the right spot to cause such a problem.
(Similarly, it is extremely unlikely for particles to be carried from the vessel bottom, as the inlet to the guide tube bottom is essentially a stagnant flow zone).
B.
Camera in RHR Suction Piping Or Heat Exchanger There is no present safety concern regarding the camera residing in the RHR system.
The RHR system will be isolated during power operation from the RCS.
Recommendations for RHR valve operability checks, RHR heat exchanger inspection and excessive pump vibration assessment (a good indicant of impeller damage or distortion) will give adequate assurance of future system operability.
A more detailed analysis of possible effects on active RHR system or interacing safety system components is given below.
No damage to the RHR Heat Exchanger is expected.
(Ref. 2, Par. 2)
C.
Camera Broken Apart -
Passes Through RHR Heat Exchanger Given that the camera passed through the RHR pump and was deformed sufficiently that some pieces were small enough to pass through the RHR heat exchanger tubes, they would pass to the bottom of the reactor vessel via the cold leg connections.
Previous experience by Westinghbuse has shown that pieces of this size do not pose safety problems in the vessel (Ref. 2, Par. 3: Ref. 3).
Metallic fretting of fuel pins, control rod jamming, (see A. above) or fuel channel flow blockage have been evaluated and are not safety concerns.
There is currently 13 square inches of fuel grid strap in the RCS (Ref. 3).
A fuel channel flow nozzle has an inlet area of about 70 square inches (0.383-inch diameter holes).
Thus the accumulated grid straps plus a conservative estimate of the maximum blockage which could be caused by the camera,* even if assumed to accumulated at one fuel channel inlet nozzle, would block no more than half of the flow area.
Per reference 3, current Salem FSAR analyses P8139/10 3 MDF 10. ~lb
fllEVISION:
0 PAGE 4
o;:
DATE:
6/16/81 (Par. 3.4.3.10) show that even with 100% blockage of a fuel assembly, flow recJvery occurs about 30 inches downstream of the blockage.
For nozzle blockage, flow recovery would thus be achieved in the lower portions of the core where DNB and LOCA are not limiting concerns.
D.
Camera Travels Essentially Intact (via Heat Exchanger By-Pass to bottom of Reactor Vessel)
In this case, the RHR Heat Exchanger would not limit pieces of the -camera to those which could pass through the tubes.
Rather, the camera could pass intact, or in several smaller pieces, directly to the bottom of the Reactor Vessel.
Previous testing and analyses by Westinghouse (ref 2. par.
- 4) has shown that no structural or functional problems will result in the vessel, internals, or fuel from such pieces.
Flow induced rise of the pieces into the core plate and fuel nozzles, even coupled with the 13-square inches of grid strap still unaccounted-for will not cause problems with either the hydraulic or thermal performance of the fuel.
E.
Concerns Regarding Non-Metallic Components of the lost material.
P8139/0l 4 Several components of the lost camera head are non-metallic in composition (ref. 1), as were the nylon lanyard (30-inch long, 0.125 diameter) and black tape.
In all, these non-metallic components comprise about 4 ounces of material, which represents about 0.3 parts per million by weight of the RCS fluid inventory.
Westinghouse Engineering Chemistry has evaluated these materials and quantities, and has concluded that no safety concern exists from a chemistry standpoint (ref 2. page 3).
The non-metallics in the camera head generally fall into two categories:
- 1.
epoxies and phenolics -
characterized by relatively high melting points, and
- 2.
polycarbonates and polyesters -
characterized by lower melting points.
In an RCS environment, the epoxy-phenolic class would be expected to soften, but retain shape, and to deteriorate over the long term.
Though retaining shape, significant softening would preclude any appreciable tensile or shear strength, and hence likewise preclude ability to mechanically bind operating components.
MDF 10. 5-lb
lllEVISION:
DATE:
6/16/81 The polycarbonate-polyester class would be expected to totally melt and travel around the system as globs or stringers (pricipally depenaent upon local flow regimes).
Depending on specific manufacturing properties, the nylon lanyard could fall into either of the above classes.
It could be postulated that the lanyard might remain intact, and lodge in an inverted hairpin configuration, through two fuel assembly nozzle holes.
There is no detrimental effect of this occurrence, adequate coolant flow, no abrasion of fuel, no adverse chemistry effects, and no fuel pin adherarrce can be expected.
There is no reason to believe that the materials will have any propensity to stick to metallic components in a flow stream, (hence no concern with plate-out on fuel pins) but rather they will tend to accumulate in stagnant zones throughout the system (e.g. upper vessel head, stagnant pressurizer surge line, etc.) and reside there with no consequence.
No mechanism can be found whereby any of these materials could adversely affect system instrumentation.
F.
Assessment of Effects on RHR system and adjacent Safety
. systems PB139/01 5 In addition to the above analyses of possible effect~ on RHR and RCS components, an assessment of effects, consequences, and safety implications of the lost material on RHR and adjacent system safety functions was made.
Starting at the 14-inch RHR suction line connection to the RCS No. 11 hot leg, and assuming that the camera head was drawn into RHR, the flow path(s), components, systems and interconnections along the flow path were evaluated:
Valves RH-1, RH-2 These 14-inch gate valves are located inside containment and function as the boundary between the low pressure RHR system and RCS.
Lost material could feasibly be lodged in the valve sealing surfaces.
Should this occur, it would be noticed via short valve stroking, leak testing, relief valve operation, and excessive RCS leakage during startup, at which time corrective action could easily be taken.
There is no safety consequence to this eventuality, as the valves remain shut during power operation and during performance of system safety function.
MDF 10. 5-lb
P8139/10 6 "EVISION:
Q CF 9
DATE:
6/16/81 Feedline From Refueling Water Storage Tank (RWST)
The next major system interface is lhe junction with the 12-inch RWST line.
Evaluation of the piping arrangement here shows a possibility for material hideout in the short dead leg bounded by check valve SJ-70.
Radiographic examination should be performed at this location i1 accessible.
Should material be residing here, no adverse consequences beyond those previously evaluated will result, i.e., the material will probably be swept into the RHR flow stream upon withdrawal of water from the RWST to RHR.
12-RH4 and 12-RHR Pump Suction Throughout this event, only the 11 RHR loop has been in service, thus no flow has been seen in the 12 RHR loop.
The 12 loop suction (inlet to valve 12 RH4) however, is a bottom connection from the common suction, and could have received material falling out of the flow stream.
The location has been examined by radiography (using a pre-qualified procedure which demonstrated clear ability to identify camera parts within the system piping) with negative result.
Impact of material hideout in the 12 RHR loop upon 12 RH2 loop startup would be identical to the 11 loop evaluation and thus acceptable.
Valve 11RH8 Continuing in the llR HR loop, material would pass through the pump (acceptability of which is to be assessed via a vibration check) and through check valve 11RH8.
Material lodged in the seat of RHB could only be of consequence if, in the accident mode, a single failure of llRHR pump were assumed.
A leaking RHB would permit recirculation of No. 12 Pump flow, thus robbing RCS of injection (or recirc) flow.
It is felt that this is a low probability event, but if it did occur, full RHR function is not needed until well into an accident (after significant RCS depressurization), thus allowing sufficient time for detection of reverse flow through the 11 loop and subsequent manual isolation of the loop.
Valve llRHlO, et al Valve llRHlO (and many other system valves) are essentially locked open maintenance valves, and material lodging in their seats, through improbablei is of no safety consequence.
MDF 10. ~lb
- -~
P8139/10 7
.. EVISION.
Q CATE:
6/16/81 Valve 11RH29 and min-flow Recirculation Line Although the recirculation line takes suction from the bottom of the 14-inch header, and this could reasonably be expected to draw in debris from the header, the connection is in such a location that all fluid passing through the min-flow must have gone through the heat exchanger, thereby limiting its size to one which will easily pass through the line and not occlude flow.
No detrimental effect on RHR pump minimum flow requirements is expected.
Valve liRH12, heat exchanger bypass, and control valve RH-20 Valve RH12 is open only during shutdown cooling, so that, in conjunction with flow control via RH20, temperature control can be established.
It is normally closed during reactor operation and system safety function operation.
If material were lodged in the seat of RH12, it would be identified during the system visual check recommended prior to system use.
Neither RH12 nor RH20 have safety functions requiring their operability.
8-inch line to containment spray headers and Safety Injection Pump Suction The piping arrangement was examined in the vicinity of the 8-inch branch feeding the SI pump (or charging/SI from the 12-RHrt loop), with the purpose of identifying any possibility of detrimental debris effect on the function of these two interfacing safety systems.
The arrangement is such that only flow which has been "filtered" by the RHR heat exchanger can approach this connection thus precluding possibility of injestion of large debris into these pumps.
The piping arrangment is such that normal flow from the RHR heat exchanger toward the cold legs is flowing upward in a vertical pipe run where these norma~ly dead legs take off as the branch of a Tee.
It is extremely unlikely that any small debris would enter these dead legs for subsequent pump injestion, however should several small pieces take this route, total effect on the pumps would be minimal.
Valves SJ43 -
Check valve from RHR into accumulator discharge Small debris lodging by chance in the sealing surface of these valves would be easily identified on startup via inability to hold accumulator pressure upon opening the accumulator isolation valves.
Correction at this time would preclude any future safety consequence.
MDF 10.~lb
.-ev1s10N.
0 PAGE 8
OF 9
DATE:
6/16/81 Valves SJ56 -
Accumulator discharge check valves Small debris lodging by chance in the sealing surface of these valves would be identified and corrected upon initial pressurization of RCS, with no subsequent safety impact.
Valves SJ49-Cold Leg Injection Stop Valves Upon termination of injection, alternation of hot and cold leg recirculation would require closure of the SJ49 valves.
As only debris "filtered" by the RHR heat exchanger could approach these valves and by chance hang-up in their seats, flow leakage would be minimal and have no identifiable effect on the overall effectiveness of the recirculation cooling process.
4.0 CONCLUSION
The following conclusions have been reached by the Engineering Department regarding the subject event:
- 1.
Though the event was unfortunate, there is no identifiable mechanism which could reasonably be expected to preclude continued safe operation of the unit prior to material recovery.
- 2.
Reasonable efforts should continue to locate the missing material.
- 3.
No reasonable mechanism has been identified which will either
- a.
increase the probability of occurrence or consequences of an accident or malfunction of equipment important to safety previously evaluated in the Safety Analysis Report.
- b.
create the possibility for an accident or malfunction of a different type then any evaluated previously in the Safety Analysis Report, or
- c.
decrease the margin of Safety as defined in the bases of the Technical Specifications.
Thus, we believe that continued operation of the plant prior to material recovery does not constitute on Unreviewed Safety Question, as defined in 10CFRS0.59.
P8139/10 8 MDF 10. 5-lb
AEVISION:
0 9
PAGE 9
OF DATE:
6/16/81
......Jlj
.lo 5.0 RECOMMENDATIONS It is recommended by the Engineering Department that:
- 1.
Whenever possible, reverse flow in the No. 11 Reactor Coolant Leg be avoided, by starting 11 RCP first and securing it last.
- 2.
The Reactor Vessel lower head and No. 11 Steam Generator be searched during the next refueling outage and foreign debris removed.
- 3.
A boroscopic examination of the 11 RHR heat exchanger inlet tube sheet and head be conducted at the earlie~t possible time.
- 4.
A general visual examination of the RHR system be conducted to ensure freedom from excessive pump vibration, and correct function of valves prior to use.
~~
T. N. Taylor Principal Engineer TNT:dlh/az P8139/10 9
~~~
L\\1411
]~R~ngcj;fr
~1 L. A. Reiter R. D. Rippe Asst. Chief Mech. Eng.
MDF 10. ~lb