ML19309H501
| ML19309H501 | |
| Person / Time | |
|---|---|
| Site: | La Crosse File:Dairyland Power Cooperative icon.png |
| Issue date: | 05/08/1980 |
| From: | NUCLEAR ENERGY SERVICES, INC. |
| To: | |
| Shared Package | |
| ML19309H500 | List: |
| References | |
| 81A003, 81A3, NUDOCS 8005130407 | |
| Download: ML19309H501 (15) | |
Text
{{#Wiki_filter:. 8 0 0 5 i s 0 '/47 O ATTACHMENT 2 TO DPC LETTER LAC-6905, MAY 8, 1980 l NES Document No. 81A0033, Rev, 1 LACBWR ROD DROP PROBABILITY STUDY Prepared for DAIRYLAND POWER COOPERATIVE LA CROSSE, WISCONSIN By NUCLEAR ENERGY SERVICES, INC. Danbury, Connecticut 06810 i i I-
(/ DOCUMENT NO. AlannT1 REV. 1 7 NUCLEAR ENERGY SERVICES, INC, PAGE 1 nF 14 LACBWR ROD DROP PROBABILITY STUDY Prepared for DAIRYLAND POWER COOPERATIVE LA CROSSE, WISCONSIN By NUCLEAR ENERGY SERVICES, INC. Danbury, Connecticut 06810 !I'Ys f9 h 0 9 7 /. O ',} de
- f y,j;e 4al' (q!
li! i d dr :1 u i u : 5 ', I \\ i 'A 8 (]} if'e Syf i d;u _ _. _ - _ _ _ _ _. _ ~ Project Application Prepared By Date 5101 AM 8N 3-14-79 APP RdV A LS TITLE / DEPT. SIGN ATU R E DATE Proj. Mgr. 8-28-77 V. P. Eng. 3 ~JdO 7i 8'~EI' N O. A. Mgr. A s l l
DOCUMENT NO. 81A0033 NUCLEAR ENERGY SERVICES, lNC. PAGE 2 op 14 TABLE OF CONTENTS PAGE 1.
SUMMARY
3 2. INTRODUCTION 3 3. CRITERIA 3 4. RDA EVENT DESCRIPTION 4 4.1 Nuclear Considerations 4 4.2 RDA Sequence of Events 6 5. RDA EVENT PROBABILITIES 7 .5.1 D'isconnect 8 5.2 Stuck Rod. 9 5.3 High Rod Worth. 9 5.4 Non-Detection. 10 5.5 Drop Timing. 11 6. RESULTS AND CONCLUSIONS 11 7. REFERENCES 13
NUCLEAR ENERGY SERVICES, INC. DOCUMENT NO. 81A0033 PAGE 1 OF 14 1.
SUMMARY
A study has been conducted for the Lacrosse Boiling Water Reactor (LACBWR) to determine the probability of a rod drop accident (RDA) exceeding the 280 cal /gm fuel enthalpy limit (Ref. 1). The LACBWR RDA was evaluated against an occurrence probability acceptance criterion of 10-7 per reactor year. It has been concluded that the expected probability of the RDA occurrence at LACBWR with energy desposition greater than 280 cal /gm is 8.2x10-14 and is adequately low when applied against the stated criterion. Even with exceedingly conservative assumptions, the maximum probability of an unacceptable RDA occurrence has been calculated to be
- 4. 8 x 10-9 Therefore, no further design or operational considerations are necessary to assure the continued safe operation of the Lacrosse Reactor.
2. INTRODUCTION The study performed analyzes the events and probabili-ties of a rod drop, accident necessary to exceed fuel enthalpy deposition of 280 cal /gm at LACBWR. Each event pertinent to the RDA is identified and dis-cussed in detail and the probabilities associ ted with the events developed. The probabilities af the individual events are combined to form over-all occurrence probabilities. The results of the study are evaluated against the accepted criterion as a basis for evaluation. It is recognized that some of the parameters in any RDA analysis are not amenable to precise interpretations. An effort.has been made, therefore, to establish a range of probabilities. The lower end of the range. (lower probability) is de - termined by assuming realistic / expected values in the context of current NRC positions and precedents, while the upper end (highest probability) is defined by assuming very conservative values. 3. CRITERIA The results of this study must be evaluated against accepted criteria as a suitable basis for judgement. Such criteria are provided in WASH-1270 (Ref. 2) which established the safety objective for the con-sequences of an ATWS event.
DOCUMENT NO. 81A0033 NUCLEAR ENERGY SERVICES, INC. PAGE 3 OF la An occurrence probability of less than 10-7 per reactor year was adopted as the desirable safety level for very unlikely postulated accidents. Specific application of this criterion in a RDA analysis is found in the NRC staff report, " Generic Item II A-2 Control Rod Drop Accidents (BWR) ", dated June 1, 1976. (Ref. 3). The NRC Report con-cluded that an occurrence probability of 10-7 is a suitable safety objective for the rod drop accident. That report, while not strictly applicable to LACBWR, contains very similar assumptions and comparable event probabilities. It further acknowledges that the analysis would be similar for LACBWR and that quantitative results aro expected to be of the same order. Therefore, sufficient precedent is concluded to have been established for the application of this criterion as the evaluation basis for the LACBWR RDA probability study. 4. RDA EVENT DESCRIPTION 4.1 Nuclear Considerations There are 29 cruciform-shaped control rods in the LACBWR core, a central rod with others distributed symmetrically about it. Beginning from a fully shutdown reactor, startup and power escalation pro-coed by first withdrawing all 29 rods p'artially followed by full withdrawal of the C bank (comprised of 20 rods). The central rod ( #1) and the 8 rods in the central region (banks A and B) are withdrawn for power escalation and to compensate for fuel burnup. Figure 1 illustrates the LACBWR control rod layout. A detailed three dimensional space-time kinetics rod drop analysis has been performed for LACBWR using BWKIN code (Ref. 4). Anticipated rod patterns for cycle 6 startup and power operations have also been reviewed. The resultant core parameters which relate to rod worth are presented below: The minimum rod worth necessary to exceed 280 cal /gm energy deposition in a rod drop accident has been calculated to be 1.4% Ak/k. This rod worth may only be reached under certain conditions. Those conditions are:
DOCUMENT NO. 81A0033 NUCLEAR ENERGY SERVICES, INC. PAGE 5 OF 14 I 26 l I I4 I I /S 7 I /6' IOi 17 i -l 1 - 7'~7hI---$ I 6 I I I ihO f 2N I I 2 I I p' 1 - F i ii l I 24Q l t' IOr .i i i 9 0 e i lhL_5 l 24 I I 4 I I to I i-(l soi o [ Or i: l E I I ?P 1 l g IQ pn l + i I Pf I IN CORE FLUX MONITORS Q .v PLANT NORTH Bank Rod #'s A 7, 9, 11, 13 I B 6, 8, 10, 12 C 2-5, 14-29 [ Figure 1 - LACBWR Control Rod Map
DOCUMENT NO. 81A0033 NUCLEAR ENERGY SERVICES, INC. PAGE 6 OF 14 (1) Reactor Power The reactor must be at power in the range of 45% to 100% power. None of the control rods were found to have a potential rod worth exceeding 1.4% Ak/k at either cold startup or during power escalation. (2) Drop Height The disconnected rod must have a potential drop height of 35" or more. A drop height of less than this distance will not cause any of the affected rods to reach a potential worth exceeding l.4% Ak/k. (The required separation length to exceed 1.4% Ak/k for Cycle 6 is actually longer than the 35 inches assumed in this study.) (3) Core Position The tip of the stuck rod must be situated in the top half of the core. Specifically, once the rod tip is withdrawn past the core midplane, potential rod worth is reduced to below 1.4%A k/k. (4) Number of High Worth Rods Of the 29 control rods in the Lacrosse Reactor, 16 have been found to be potentially high worth (greater than 1.4 % Ak/k) rods during some period of operation. 4.2 RDA Sequence of Events There are definite events necessary in a rod drop accident to exceed 280 cal /gm. These events must meet certain conditions and follow a specific sequence. Those rod drop events are identified in this section. (1) Rod Disconnection The rod separates from the control rod drive mechanism. The rod was either improperly coupled or fails in a fashion that creates the disconnect. This disconnect must occur while the rod tip is in the upper half of the core, under conditions described above.
NUCLEAR ENERGY SERVICES, INC. DOCUMENT NO. 81A0033 7 14 PAGE OF (2) Rod Sticking The disconnected rod must stick and remain in place as the drive is moved away. The rod must stick in the upper half of the core and the drive mechanism must be lowered at least 35 inches below the stuck rod. (3) Non-Detection The disconnect must go undetected, despite. indications on nuclear instrumentation. The affected rod drive must be withdrawn repeatedly in to 1" increments throughout power opera-tion as burnup proceeds (35 to 70 times) with-out corrective action until the separation between the bottom of the stuck rod and the top of the drive mechanism is 35 inches or-greater. (4) Rod Worth The affected rod must have a high potential worth. While it varies somewhat with the rod pattern and time in cycle, a RDA must have a reactivity worth greater than 1.4% ok/k to exceed 280 cal /gm. Many of the rods in LACBWR (13 out of 29 rods during Cycle 6) will never attain a potential worth of this value. (5) Rod Drop Timing The stuck rod must drop. The timing during which the rod must drop is important..A window exists in the reactor power status during which an RDA has the potential to exceed 280 cal /gm. This window exists at operation above 45% power. 5. ROD DROP ACCIDENT EVENT PROBABILITIES Each event pertinent to the rod drop is analyzed in detail, and associated probabilities provided and justified. Relevant aspects of plant design, opera-tional procedures as well as the statistical bases are discussed.
81A0033 NUCLEAR ENERGY SERVICES, INC. DOCUMENT NO. O 14 PAGE OF 5.1 Rod Disconnection In the NRC Staff Report (Ref. 3), it is stated that GE plants have experienced eight disconnects in 200,000 withdrawals. These disconnects occurred in 10 GE plants with at least two modes of uncoupling: sticking of the upper locking handle and a moved strainer. LACBWR has a more positive latching mechanism and has not yet experi-enced any disconnects. In the 12 years of LACBWR operation, it is estimated that the control rods have experienced about 25,000 complete withdrawals (including developmental and Tech. Spec. testing, operator training, power operation, etc.) This data base would pro bability of 1.2x10 yide a maximum rod disconnect pro-The probability for disegnnect based on the GE experience of 8 disconnects in 2x10 withdrawals is 4x10-5 This value should be conservative for LACBWR because of the more positive latching mechanism and will be used as the expected rod disconnect probability. Neither the expected nor the maximum rod disconnect probability takes credit for the event having to occur with the rod worth in the RDA range.
DOCWE NUCLEAR ENERGY SERVlCES, INC. PAGE 9 OF 14 5.2 Rod Sticking No precise data exist on stuck control rods. Accordingly, the rod sticking probability value of 10-3 per with-drawal established in Reference 3 will be used in this study. At LACBWR, this implies one stuck rod every two years, assuming approximately 450 rod withdrawals annually (LACBWR has been averaging 15 startups per year). An exceedingly conservative value of 10-2, corresponding to close to 5 stuck rods per year, will be used as the maximum boundary in this analysis. In LACBWR, zircaloy and stainless steel fuel shrouds form the control rod channels. A program is currently implemented to replace irradiated zircaloy shrouds on the basis of a conservative irradiation exposure limit to minimize shroud creep and deformation. No such phenomenon has been observed with stainless steel shrouds. The limit ensures that the control rod channel is at least 0.11 inches wider than the control rod sheath during operation. No credit is taken' for the sticking event having to occur under the other conditions necessary to create a potential rod worth in the RDA range., 5.3 Non-Detection LACBWR routinely verifies that the control rods are following their drives by performing a control rod following exercise for each rod as part of the startup procedure. This is performed by moving each rod in and then returning it to the original position and observing the decrease and increase of the power level and corresponding change of reactor period while the reactor is just critical at very low power level. During reactor heatup and powerescalations, the reactor period, the power level and heatup rate are monitored as the control rods are moved in multiple small in-crements. 1
NUCLEAR ENERGY SERVICES, INC. PAGE 10 OF 14 During power operation, especially at levels higher than the 45% power level with the rods of concern in the upper half of the core (the RDA range established in Secti;n 4), rod withdrawal for fuel burnup compensation is accomplished by incremental movements of to 1 inch at a time. The effect of the rod movements is readily discernable and regularly monitored on the reactor power level, reactor period and steam flow instrumentation. This requirement for frequent small withdrawals provides many opportunities for control rod dis-connection detection. Since a 35 inch separation is prerequisite for a rod worth potential in the RDA range, a minimum of 35 withdrawal increments will have been performed subsequent to a rod sticking event. As proposed in the Staff Report (Ref. 3)a value of 10-1 is assigned to the failure to detect nuclear instrumentation response every time the drive is actuated. Assuming conservati'.ely only 8 detection opportunities (approximately one out of every possible four), the total probabiligy on rod disconnect non-(10-1) = 10-8, detection would be Even if one assumes a.25 probability of non-detection of non-movement of the rod on the nuclear instrumen-tation and again only 8 detection opportunities, the low value for non-detection probability ~is calculated to be 1.5 x 10-5, 5.4 Rod Worth It is evident that a number of control rods will not have sufficient reactivity worth to exceed the 1.4 % Ak/k value required to exceed 280 cal /gm. As discussed in Section 4, the LACBWR core characteristics are such that 16 of the 29 control rods have potential worth in excess of 1.4% Ak/k at some time during Cycle 6. Therefore the probability of the disconnected rod being a high worth rod would be 16/29 or 5.5x10-1 To account for different rod patterns in future cores and to allow for uncertainties, it is assumed that conservatively 21 rods might have worths in of7.2x10-{heRDArangewitharesultantprobability for the determination of Maximum Probability.
DOCWEM W. NUCLEAR ENERGY SERVICES, INC. PAGF 11 OF ~4 5.5 Rod Drop Timing It is conservatively assumed that rod worths greater than 1.4% are possible whenever the reactor is operating above 45% of rated power. This condition exists during approximately 10 months of a calendar year and therefore the_progability_of.theRodDropTimingis10/12or 8.3x10-A No basis exists for probability development of the stuck rod dropping. Therefore, the drop probability for a stuck rod is assumed to be 1, especially since the stuck rod is more likely to drop due to the-period-of-increased vibration at power operation. 5.6 Total Rod Withdrawals As previously discussed, LACBWR has experienced an average of 15 startups per year, giving approximately 450 withdrawa]s per. year. ,1 6. RESULTS AND CONCLUSIONS The attached Tarle 1 is a summary of event probabilities developed in this study. ) ihe total calculated expected probability for the rod drop accident resulting in fuel enthalpy deposition in excess of 280 cal /gm limit, is 8.2 x 10-14 As a result of the comparison of this probability with the acceptance criteria as discussed in WASH 1270, (probability of occurrence less than 10-7 per reactor year), it is concluded that the probability of a rod drop accident resulting in excess of 280 cal /gm occurring at LACBWR is adequatcly low when applied against stated criteria. Even when exceedir. gly conservative values are combined, the overall probability for an unacceptable RDA is calculated to be 4.8 x 10-9 which is still lower than the 10-7 criterion.
DOCUMEM NUCLEAR ENERGY SERVICES, INC. PAGE 12 OF 14 TALLE 1 LACBWR ROD DROP PROBABILITIES Expected Maximum Event Probability Probability Rod Disconnection
- 4. 0 x 10-5 1.2 x 10-4 Rod Sticking 10-3 10-2
-5 Non-Detection 10-8 1.5 x 10 High Rod Worth 5.5 x 10-1 7.2x 10 1 Rod Drop Timing
- 8. 3< 10-1 8.3x 10-1 2
Total Ro'd Withdrawals 4.5 x 10 4.5 x 102 Probability / year of 8.2x 10-14 4.8 x 107 9 Rod Drop Resulting in I >280 cal /gm l i
DOCWENT NO. NW f. NUCLEAR ENERGY SERVICES, INC. M'E 13 PAGE op 14 7. REFE RENCES 1. USNRC Standard Review Plan, Section 15.4.9, " Spectrum of Red Drop Accidents (BWR) ", NU REG-75/ 087. 2. WASH-1270, " Anticipated Transient Without Scram for Water-Cooled Power Reactors", Sept. 1973. 3. NRC Staff Report, " Generic Item IIA-2 Control Drop Accidents (BWR) ", June 1, 1976. 4. BWKIN Code, B & W Version of MEKIN, " Nuclear Reactor Core Kinetics Code", NPGD-TM-425, Nov. 1977. 5. WASH-1318, " Technical Report on Analysis of Pressure Vessel Statistics From Fossil-Fueled Power Plant Service and Assessment of Reaetor Vessel Reliability in Nuclear Power Plants", May 1974.
" ^ NUCLEAR ENERGY SERVICES, INC. REVISION LOG "^ " "E ",QE D E S C RIPTIO N APPROV AL g, DATE 1 5/5/80 all CRA01323 - extensive revision through-X d e out report: no revision number required. _}}