GENES4/ANP2003, Sep. 15-19, 2003, Kyoto, JAPAN Paper 1086 Fuel cladding stress evaluation for the control rods pattern change at the Shimane Nuclear Power Station Unit No1 and No2 Saemi Shimatani 1*, Masayuki Tamura 1, Ikunori Kawanaka 1, Masao Kuwatani 1,Kensuke Tokunaga 2 and Takaaki Mochida 3 1 The Chugoku Electric Power CO.,INC, Hiroshima, 730-8701, Japan 2 Global Nuclear Fuel- Japan CO.,Ltd, Yokosuka,Kanagawa, 239-0836, Japan 3 Hitachi Ltd, Hitachi, Ibaraki, 317-8581, Japan The Chugoku Electric Power Co. Inc. operates two Boiling Water Reactors (BWRs), Shimane Nuclear Power Station Unit No.1 and No.2. In order to ensure fuel integrity, especially to avoid operation-related fuel failures, we have been evaluating fuel cladding stress due to pellet-cladding mechanical interaction. Special care has been exercised in reload core designs, including effects due to control rod pattern changes, since control rod withdrawal could cause a high cladding tensile stress even at a relatively low power level. The cladding stress evaluation procedure is as follows. (1) Calculate fuel cladding stress by a fuel performance code based on irradiation histories obtained from fuel loading and operation plans. (2) Confirm sufficient stress margins by comparison of the calculated stress to the cladding stress criterion, above which cladding failure is statistically predicted from the power ramp test results KEYWORDS: fuel integrity, control blade history(cbh), control rods pattern change, fuel cladding stress evaluation I. Introduction Shimane Nuclear Power Station Unit No.1 has been operated for about 30 years since the initial start-up in 1973. Unit No.2 has been operated since 1988. About 4,500 fuel assemblies (about 280,000 fuel rods) have been so far loaded to the Unit 1 and Unit 2 cores. Fuel assembly types have been updated from the 7x7 design to the latest 9x9 design as shown in Fig. 1. In order to ensure fuel integrity, we have been confirming not only core thermal margins but also cladding stress due to pellet-cladding mechanical interaction in reload core designs. Modern fuel designs employ zirconium liner cladding which is highly resistant to the pellet-cladding interaction failures due to stress corrosion cracking (PCI/SCC). Recent power ramp tests revealed, however, that PCI/SCC is the cause of failures at intermediate burn-up level, but hydride-assisted Fig.1 Changes in Fuel Types at Shimane Nuclear Power Station * Corresponding author, Tel. +81-82-523-6053, Fax. +81-82-523-6065, E-mail: 365588@pnet.energia.co.jp
outside-in cracking was the cause at the local burn-up higher than about 50 GWd/tU 1). We have been evaluating cladding stress in generating reload core designs for many years to avoid PCI failures. This paper describes our methods so far applied for fuel cladding stress evaluation. II. Minimum Control Cell Operations BWR plants are operated with several control rods inserted in the core to control excess reactivity. In current BWR plants in Japan, simple control rod pattern change is performed without large capacity loss or thermal margin loss by control cell core concept 2). One control cell consists of four low reactivity fuel bundles around a control rod inserted during power operation. Together with the axially enrichment zoning 3) and the optimization of axial gadolinia distribution in BWR fuel design, single control rod pattern operation for long term became possible with use of the minimum number of control rod insertion 4). While the long-term control rod insertion, the adjacent control rod distorts the exposure distribution and neutron spectrum of a fuel bundle. Local uranium residual and plutonium production rates remain relatively high adjacent to the inserted cruciform blade, where the blades covers only two faces of the four faces of the square bundles. Cumulative effect of this makes local power peaking high when the control rod is withdrawn. Such an effect is known as the control blade history (CBH). As the thermal power level in control cell is sufficiently low, it will provide ample thermal margin even for the consideration of additional increase in the linear heat generation rate due to the CBH effect. However, control cell fuels have the following characteristics regarding fuel rod cladding stress. (1) Fuel rod power is kept low for long time by the adjacent control rods. (2) When the control rod is withdrawn, fuel rod power is possible to be higher than other fuels of equivalent burn-up. (3) Fuel rod burn-up is relatively high. Control cell fuel rod cladding is in contact with the pellets at a relatively low power level due to the cladding creepdown by the coolant pressure and the pellet swelling. When control rod is withdrawn after the long-term insertion, high cladding tensile stress is possible to be produced by the change in power level due to pellet-cladding mechanical interaction. III. Mechanism of producing fuel cladding stress Figure 2 schematically shows a typical power history and corresponding pellet/cladding dimensional changes of control cell fuel. As shown in Fig. 2, fuel rod local power is suppressed by control rod insertion, and the local power is increased by control rod withdrawal. Cladding stress depends on the local rod power history. (1) (2) (3) (4) (5) (6) T 1 T 2 T 3 T 4 T 5 T 6 P 1 P 2 P 3 P 4 P 5 P 6 T i :Cladding ID T 1 < T 2 T 2 > T 3 T 3 > T 4 T 4 > T 5 T 5 < T 6 Cladding behaviors Thermal expansion Creepdown (Thermal expansion ) Creepdown (Thermal expansion ) Creep * P i :Pellet OD P 1 < P 2 P 2 < P 3 P 3 > P 4 P 4 < P 5 P 5 < P 6 Pellet behaviors Thermal expansion Swelling Thermal expansion Swelling Thermal expansion Relocation Creep * Densification Hot-pressing * : diameter increase, : diameter decrease * effective at slow ramp rate (2) (3) (6) LHGR (1) Exposure (4) (5) Fig.2 Typical Power History and Fuel Rod Behavior of Control Cell Fuel
(1) At fuel fabrication As-fabricated pellet-cladding gap is formed as specified by the fuel rod mechanical designs. (2) Beginning of irradiation Pellet temperature increases by the heat generation in the pellet, and both pellet and cladding expand thermally. Pellet cracks and relocates toward cladding. Pellet thermal expansion is greater than cladding thermal expansion because of higher temperature. Pellet-cladding gap thus decreases at power. In the beginning of irradiation, pellet usually densifies rather than the swell due to irradiation and pellet-cladding gap is open at the power levels typically given in BWR. (3) Burn-up progression Open pellet-cladding gap tends to decrease with the burn-up progression by the cladding creepdown under the coolant pressure and the pellet swelling, and eventually pellet and cladding come into contact. The contact force is moderate at steady power operation because of the cladding stress relaxation and the compressive pellet hot-pressing. (4) Insertion of control rod When the fuel rod power is suppressed by the control rod insertion, pellet heat generation decreases, pellet thermal expansions decreases and pellet diameter decrease. Cladding temperature decreases also, but the degree is small, and thus the pellet-cladding gap re-opens. (5) Burn-up progression under the control rod insertion Pellet-cladding gap decreases by the pellet swelling and the cladding creepdown. After a long-term control rod insertion, pellet-cladding gap is nearly closed at low power. (6) Control rod withdrawal Control rod withdrawal causes a rapid power rise for fuel rod. Pellet temperature increases, pellet-cladding mechanical interaction starts at relatively low power, and the cladding tensile stress is produced. IV. Fuel cladding stress evaluation The cladding stress is evaluated as a part of the reload core designs. Since peripheral fuel rods in the BWR fuel assembly is close to the control blade and it is subject to large CBH effect, we evaluate the cladding stress for the fuel rod having most severe power history. The corner rod is a representative. The power increase due to the control rod withdrawal is highest for the corner rod. Care must be taken in selecting the rod of interest, because the cladding stress depends on the fuel assembly designs and the power histories. The cladding stress evaluation is carried out for two cases, a normal case and a conservative case. Normal case evaluation is based on the initial operation plan, and the conservative case assumes a longer interval between control rod pattern changes, considering possible operational modifications. The fuel cladding stress evaluation procedure is outlined in Fig.3. The fuel cladding stress evaluation procedure is as follows. Core Bundle Fuel Rod Pellet Process Computer 3D BWR Simulator Lattice Calculation Code 3D 3D Distribution of Power & Exposure Local Peaking Factor Geometry Data of Fuel Rod Power History of Each Fuel Rod Input Data of Fuel Performance Code Cladding Stress Analysis Data Evaluation of Power Ramp Test Fuel (1) Cladding stress analysis based on following inputs Manufacture data of power ramp test fuel Base irradiation history in commercial reactor Power ramp history (2) Power ramp test results Criteria of Fuel Integrity Evaluation Determine minimum value of cladding failure stress by statistical evaluation for each fuel type categorized as followings Non Zr-liner fuel Zr-liner fuel Cladding Hoop Stress Fig.3 Outline of Fuel Cladding Stress Evaluation Procedure
(1) Fuel rod power history is evaluated on the basis of three-dimensional core power and burnup distribution data, the control rod pattern change plan, and the local peaking factors. Using the fuel rod power history thus evaluated and the nominal fuel mechanical design variables, cladding stress is calculated by a fuel performance code. (2) Fuel integrity is confirmed by comparing the calculated cladding stress and the stress criterion based on power ramp test results. When the fuel rod power history is evaluated, we adopt latest actual operating data available at the time of a reload core design. In case of control rod withdrawal, we calculate the fuel rod power considering CBH effect on the basis of period of control rod insertion and its elevation. It should be noted that the cladding stress evaluation has a relative importance in the sense that the cladding stress criterion (PCI failure limit) is evaluated by the same procedure and fuel performance code. The cladding stress criterion was determined from cladding tensile stress values as evaluated using actual power history of power ramp tests. On the other hand, when fuel integrity is confirmed in a reload core design, we make a conservative-side simplifying assumption that the power increase is stepwise, ignoring stress relaxation during power rises. V. PCI failure criterion The PCI failure is caused mainly by SCC, and at high burn-up, possibly assisted by radial hydride formation. While PCI/SCC failure is due to corrosive fission products release from fuel and hydride-assisted PCI failure depends on hydrogen pickup from coolant, we found from ramp test experiences that a simple cladding stress criterion can distinguish the failed fuel rods due to PCI. Figure 4 shows the stress evaluation results for zirconium liner fuel rods, which were irradiated in commercial BWRs and ramp tested at test reactors 1,5,6,7). The failure/non-failure boundary is defined by a stress criterion, below which failure probability is regarded as zero. For the old type non-liner fuel, we had to use a very low stress criterion to avoid PCI/SCC failures and had to apply a strict operational restraint in control rod withdrawals to suppress cladding stress. Current fuel designs with zirconium-liner have contributed to increase operational flexibilities. VI. Experience at Shimane Nuclear Power Station The cladding stress evaluation was initially adopted about 20 years ago to reload core designs of Shimane unit No. 1, and we have continued this practice in Unit No.1 and No.2 in these 20 years for five different BWR fuel types, the 8x8 fuel, the new 8x8 fuel, the new 8x8 zirconium-liner fuel (Step I fuel), the high burn-up 8x8 fuel (Step II fuel) and to the 9x9 fuel (Step III fuel). While non-liner fuel types were adopted, care had to be exercised in selecting fuel bundles to load into control cells, because of a low stress criterion of PCI/SCC, sometimes finding alternative options to meet the stress criterion. The cladding stress values of near 90% of the stress criterion were experienced for non-liner fuel types. It should be noted that the stress evaluation results depends on fuel designs and operation plans. Current high burn-up 1.8 1.6 1.4 1.2 Relative Hoop Stress 1.0 0.8 0.6 0.4 0.2 0.0 Ramp Fuel(Sound) Ramp Fuel(Failure) Reload Step1(Nominal) Reload Step2(Nominal) Reload Step3(Nominal) Reload Step1(Conserv.) Reload Step2(Conserv.) Reload Step3(Conserv.) Criterion -0.2 Fig.4 0 10 20 30 40 50 60 70 0 0.5 1 Cumulative Failure Exposure (GWd/t) Probability Cladding Stress Evaluation Results for Ramp Test Fuels and Shimane Reload Fuels
fuel designs with zirconium-liner tend to be subject to higher cladding stress. Zirconium-liner fuel experiences are also shown in Fig. 4. For the high burn-up 8x8 fuel and the 9x9 fuel, the cladding stress tends to increase with extended burn-up. VII. Conclusion The Shimane Nuclear Power Station celebrates for the 30 years anniversary this year, and for zero fuel failure experience since the initial start-up in 1973. Non-failure for 30 years looks like a miracle in Shimane, and is a result of continuous efforts in managing reactors and cores. We believe that core fuel management to avoid PCI failure has contributed to this achievement. Acknowledgment The authors wish to acknowledge the excellent and long efforts by all predecessors. References 1) H. Hayashi, et al., "Outside-in Failure of High Burnup BWR Segment Rods Caused by Power Ramp Tests", TopFuel 2003, Würzburg Germany, Track 1-2 (2003) 2) S. R. Specker, et al., "The BWR Control Cell Improved Design ", Tran. Am. Nucl. Soc., 30, 336 (1978). 3) O. Sugimoto, et al., "BWR Operating Experience at Shimane-1 with WNS-type Initial Fuel", Tran. Am. Nucl. Soc., 39, 912 (1981). 4) A. Nishimura, et al., "Experience with Single Control Rod Pattern Operation in Axially Two-Zoned Fuel", Tran. Am. Nucl. Soc., 50, 567 (1985). 5) K. Inoue, et al., "An Overview of the Joint Development Work on PCI Remedy Fuel", ANS Topical Meeting on LWR Fuel Performance, Orland, Florida, 21-24 (1985) 6) Y. Wakashima, et al., "Power Ramp Test of Zirconium Liner Fuel at High Burnup", Nihon-Genshiryoku- Gakkai (At. Energy Soc. Jpn.), Autumn meeting K35 (1989), [in Japanese]. 7) H. Sakurai, et al., "Irradiation Characteristics of High Burnup BWR Fuels", ANS Topical Meeting on LWR Fuel Performance, Park City, Utah, 128-138 (2000) 8) Y. Matsuo, "Development of PCI-Resistant Fuel for BWR", Karyokugenshiryokuhatuden (The Thermal and Nuclea Power.), 33, No.3 (1982), [in Japanese].