ML20236P085

Informs of Results of Evaluations Performed to Determine Susceptibility of Facility Steam Generator Tubes to High Cycle Fatigue Similar to Failure at North Anna Unit 1 on 870715.Facility Procedures Adequate for Event

ML20236P085
Person / Time
Site:	Point Beach, North Anna, 05000000
Issue date:	11/11/1987
From:	Fay C WISCONSIN ELECTRIC POWER CO.
To:	NRC OFFICE OF ADMINISTRATION & RESOURCES MANAGEMENT (ARM)
References
CON-NRC-87-114 VPNPD-87-494, NUDOCS 8711170139
Download: ML20236P085 (6)

Text

{{#Wiki_filter:__ 'h e Wisconsin Electnc eom cown i !1 4 . 23i W MICHIGAN.P.o BOX 2046, MILWAUKEE,W153201 (414)277 2345 -{ VPNPD-87-494 i NRC-87-11'4 I November (11,. i 1987. y 1 i U.S. NUCLEAR REGULATORY COMMISSION Document Control Desk i Washington, D.C. 20555' Gentlemen: DOCKET 50-301~ SUSCEPTIBILITY'TO HIGH CYCLE FATIGUE FAILURE OF. STEAM GENERATOR TUBES: POINT BEACH NUCLEAR PLANT, UNIT 2 l } The' purpose of this letter is to inform you of the results of-evaluations performed.to determine the susceptibility of Point l Beach' Nuclear.' Plant Unit'2 steam generator tubes to high. cycle i ' fatigue: failure similar to that which is believed to have caused'the steam generator tube. failure at North Anna Unit 1 on ] July 15, 1987. The details of our evaluations and the associated conclusions.were presented at a meeting attended'by representa- { tives'from the Nuclear Regulatory Commission staff, Westinghouse i Electric Corporation, and Wisconsin Electric on November 4, 1987. As discussed in our October 7, letter, the. evaluations included increasing the ' scope of the eddy current testing _ program for i the current Unit 2 refueling and maintenance outage to identify tubes in rows 8 through 13.which exhibit denting at the top tube support plate and to identify the antivibration bar (AVB) insertion depths for each column in rows 8'through 13. These eddy current examination results provided input to the thermal-hydraulic, fatigue, vibration and local flow analyses for the Point Beach Unit 2 steam generators. l Based on our evaluations, no tubes in the Point Beach Unit 2 steam generators are susceptible to tube fatigue failure similar to that which occurred at North Anna Unit 1. Therefore, modifications, preventive tube. plugging, or other preventive 4 actions to preclude such an ent at Point Beach are not required. The failure mechanism for the North Anna Unit 1, Steam Generator C, i 001 8711170139 871111 PDR ADOCK 05000301 i nQ P PDR

.4 6, NRC Document Control Desk November 11, 1987 Page 2 Row 9 Column 51 (R9C51) tube has been determined to be limited displacement fluidelastic excitation due to local flow conditions in. conjunction with denting of the tube at the top tube support plate. The denting provided an additional source of mean stress and reduced damping during tube vibration. Fluidelastic_ excitation providedithe necessary displacement and alternating stress which, when combined with the mean stress effects on fatigue, caused failure of the R9C51 tube. It has been further concluded that the fatigue of the R9C51 tube was long term and had been develop-ing since denting occurred during the first fuel cycle. Normal operating transients were not a significant factor. Westinghouse has developed conservative acceptance criteria to determine susceptibility to high cycle fatigue failure of other North Anna steam generator tubes and of steam generator tubes at other facilities. The general approach is to specify that tube vibration displacements and total stresses are sufficiently small such that fatigue failure will not occur. Specifically, the acceptance criteria for a. tube which is dented at the-top tube support plate are 1) AVB support or a stability ratio which is less than or equal to 90 percent of the stability ratio of North Anna R9C51 and, 2) a stress ratio relative to North Anna R9C51 which, after applying the ten percent reduction in the stability ratio, is less than or equal to 1.0. The ten percent reduction in stability ratio reflects tube vibration stresses of less than 4 KSI based on the maximum calculated stresses for North Anna R9C51. At 4 KSI the fatigue usage is less than 0.021 per year. Even assuming a tube developed a through-wall crack up to 125 mils long, such a tube would not experience crack growth at these stress levels. The tube evaluation for Point Beach Unit 2 steam generators was performed by Westinghouse Electric Corporation and included j the following interrelated analyses: l 1 I (1) The three dimensional flow analysis code (ATHOS) was used to calculate flow conditions in regions of interest assuming uniform flow conditions. The ATHOS model was set up with j geometry parameters specific to Point Beach Unit 2 steam generators. Detailed secondary side velocity, density and void fraction distributions were calculated for tube rows 8 through 12. The data from the ATHOS analysis provided input to the vibration analysis codes. l (2) Tube stiffness, frequency, and fluidelastic stability ratios o were obtained by dynamic analysis for tubes in rows 8 through j 12 using the FLOVIB code. The inputs for this evaluation were the flow field characteristics of velocity, density, and void fraction obtained from the ATHOS analysis. The 1 t l [ l l.

a 6 NRC Document Control Desk [ November 11, 1987 i Page 3 dynamic analysis results were then used to calculate tube stress ratios to identify tubes which were potentially l susceptible to high cycle fatigue failure similar to North i Anna R9C51. (3) Axisymmetric finite element analysis was used to determine. the mean stress of tubes at the tube intersection with the 1 2op tube support plate during 100% power conditions. (4) Concurrent with the evaluations in 1 through 3 above, wind tunnel tests were performed to determine the effects on fluidelastic instability of columnwise variations in'AVB i insertion depths. Simulated AVBs were inserted at different depths in adjacent tube columns to evaluate various con-l figurations of tubes and AVBs. The instability threshold l characteristics were determined for specific tubes of interest. These characteristics were then compared to the local instability threshold characteristics of North Anna R9C51 to determine tube specific relative flow peaking factors for Point Beach Unit 2. (5).Also concurrent with evaluations in 1 through 3 above, all unplugged tubes in rows 8 through 13 in both steam generators were eddy current tested over.the length of the ' tube above the top tube support plate to identify dented tubes and AVB locations. 1 The data obtained from the various analyses were combined as described below to provide disposition of the tubes within the potentially susceptible region of rows 8 through 12. The eddy current test data showed that virtually all tubes are dented at the sixth support plate, as expected. The eddy current test data also showed that no tubes have wall thinning indica-tions at the AVB locations. Therefore, it is unlikely that those tubes have been unstable. Additionally, the eddy current data were used to define AVB insertion depths and the extent of tube support for each column of tubes within the region. Virtually all row 11, row 12 and row 13 tubes are supported except for a few peripheral columns in rows 11 and 12. Most row 10 and row 9 tubes are supported. Some row 8 tubes are supported. The analyses identified 16 dented and unsupported tubes which either had relative stability ratios approximately greater than 0.90 or stress ratios greater than 1.0 (assuming local flow conditions to be the same as at North Anna R9C51). These were as follows: _ _ _ _ _ _ _ _ _ _ = -

d -. (; -NRC. Document. Control Desk November 11, 1987 Page 4' f.- Steam Generator A Steam Generator B Row-Column Row Column lI l 12 2 12 2 12 91 12 91 11 2 11 2 11 3 11 91 11 4 10 4 11 91 10 5 10 3 10 90 10 4 10 5 The local flow peaking due to actual AVB insertion depths was then factored into the evaluation for each of the above tubes. This resulted in the relative stability.and stress ratios shown below. All ratios are in comparison to North Anna R9C51. Steam Generator A Steam Generator B Stability Stress Stability Stress Tube Ratio Ratio Tube Ratio Ratio R12C2 0.84 0.48 R12C2 0.84 0.48 R12C91 0.84 0.48 R12C91 0.84 0.48 R11C2 0.61 0.09 R11C2 0.61 0.09 R11C3 0.86 0.65 R11C91 0.61 0.09 R11C4 0.88 0.76 R10C4 0.70 0.21 R11C91 0.61 0.09 R10C5 0.82 0.57 -R10C3-0.71 0.24 R10C90 0.71 0.24 R10C4 0.70 0.21 R10C5 0.69 0.20 The current fatigue usage was calculated for the tube with the highest stress ratio, i.e., the tube at location R11C4 in steam generator A. This calculation assumed denting has been present since initial operation, even' though denting was not detected in Unit 2 until 1975. The conservative cumulative fatigue usage -for the-13 fuel cycles for this worst case tube is only 0.081. Thus, even the worst case tube would not experience fatigue failure during the remaining term of the operating license. Based on the evaluations we conclude that Point Beach Unit 2 steam generator tubes are not susceptible to fatigue failure at the top support plate similar to that which occurred in North Anna Unit 1. Therefore modifications, preventive tube plugging or other measures to preclude such an event are not required.

c l

I

a 6 NRC Document Control Desk November 11, 1987 Page 5 As discussed in our October 7 letter, we have also reviewed our capability to detect and assess leakage and'take appropriate action during the progression of a tube failure similar to that which occurred at North Anna Unit 1. Based on our evaluation, we have determined that enhancements in existing procedures for trending of data, intermediate action levels and time intervals between sample analyses are appropriate. Our primary-to-secondary leakage detection capabilities and responses include a combination of continuous on-line monitoring, l: administrative procedures, technical specifications and emergency ( operating procedures. Continuous monitoring of each unit for primary-to-secondary leakage is provided in the control room by two unit specific radioactivity monitors and one shared monitor. Each unit has an air ejector monitor and a steam generator blowdown monitor. Downstream of the unit specific air ejector monitors, the air ejector discharges are combined and are monitored by a combined air ejector monitor. In essence, the unit air ejector monitor and combined air ejector monitor are redundant monitors in series for each unit. All of these monitors provide alarms and continuous data which are stored for the purpose of trending by the plant process computer system in the control room. All of these monitors use scintil}ation degectors and have sensitivity levels of begween 10-and 10 pCi/cc. Based on a sensitivity level of 10-pCi/cc (minimum sensitivity guidelines prescribed in Regulatory Guide 1.97 for the instrumentation) and our latest beginning of cycle primary coolant chemistry, the unit air ejector monitors can detect and trend radioactivity from leak rates in the order of ten gallons per day. Administrative procedures will prescribe leakage measurement methods, time intervals between measurements and intermediate actions based on leak rates or monitoring instrumentation data. Based on the planned revisions to existing procedures, our primary-to-secondary leakage detection and assessment capabilities will ensure that time is available to take appropriate action during the progression of a steam generator tube leak with time dependent leakage characteristics similar to those of the North Anna tube failure. We plan to modify our procedures prior to return to power from the current refueling outage. During the November 4, 1987 meeting, we were also asked if emer-gency procedures address multiple simultaneous tube ruptures. Emergency operating procedures for Point Beach are based upon Westinghouse Owners Group Emergency Response Guidelines which consider plant response for up to ten simultaneous tube ruptures.

( 7c -- --

s

(;. g a, ~ NRC Document Contgol: Desk. ? i L November 11, 1987 l

Page 6 Thus,: Point 1 Beach procedures' provide 1 appropriate response for.

t'this-extremely;unlikely event. Two emergency-operating procedures, LEOP-0 and EOP-3,would be used in the event.of< single or' multiple . tube ruptures...The7firstLoperator response to a steam generator otube rupture is to use'EOP-0, Reactor Trip or~ Safety Injection. ,The1 operator. verifies that automatic actions have taken place,to ensure core' protection. The operator would then'be directed by-EOP-0 to'EOP-3, Steam Generator Tube Rupture. The. objective of, ~ EOP-3 is-to establish steamLgenerator conditions oto minimize environmental' releases'and to allow a controlled cooldown of the plant. JAs1withlall of'our emergency operating procedures, EOP-0. and'EOP-3 have been-verified and validated, and the Point Beach-Nuclear Plant operators'have been' trained on the-procedures as ~ part of Lthe -licensed operator training program. Please' contact.us.if you have_any! questions regarding this -information. Very trulyDyours, -ih (?tc/jy' J y C. W. Fay. -iVice President-Nuclear Power Copy.to;NRC Regional Administrator, Region III NRCLResident-Inspector _ -}}