Loading...
HomeMy WebLinkAbout851167.tiff OEfkP NEON, UNITED STATES W NUCLEAR REGULATORY COMMISSION _ REGION IV t °r 011 RYAN PLAZA DRIVE,SUITE 1000 i " O ARLINGTON,TEXAS 70011 JUL 121985 In Reply Refer To: Docket: 50-267 WELD COMITY rii1, ";c 3; NERD JULi 61985 ' Public Service Company of Colorado � �iJ' ATTN: 0. R. Lee, Vice President , Electric Production SRMILET. (O. P. 0. Box 840 Denver, Colorado 80201 Dear Mr. Lee: Our consultants at EG&G Idaho, Inc. have reviewed your June 13, 1985 (P-85201) submittal related to the control rod drive and orifice assembly (CRDOA) bearings. A copy of their evaluation report is enclosed for your information and comment. Based on our review of the information you have provided and the enclosed evaluation, we have concluded that the installed bearings are acceptable replacements for the original design. In order to ensure continued acceptable bearing performance, we request that the test programs discussed in P-85201 be completed and that bearing performance be factored into your improved CRDOA surveillance and preventative maintenance program. We have also completed our review of your June 7, 1985 (P-85195) submittal related to the epoxy used to attach temperature sensors in the CRDOAs. We agree that this material appears to be adequate for the attachment of the sensors. It is our understanding that the epoxy will be included in the upcoming environmental qualification program. We recommend that the epoxy be inspected for deterioration as part of the CRDOA preventative maintenance program. If you have any questions on this subject, please contact us. Sincerely, E. H. Johnson, Chief Reactor Project Branch Enclosure: Evaluation of CRDOA Bearings cc w/Enclosure: (cont. on next page) 851167 Li ,, 10 °l I„IRs Public Service Company of Colorado -2- cc: Mr. D. W. Warembourg, Manager Nuclear Engineering Division Public Service Company of Colorado P. 0. Box 840 Denver, Colorado 80201 Mr. David Alberstein, 14/159A GA Technologies, Inc. P. 0. Box 85608 San Diego, California 92138 Kelley, Stansfield & O'Donnell Public Service Company Building 550 15th Street, Room 900 Denver, Colorado 80202 Chairman, Board of County Comm. of Weld County, Colorado Greeley, Colorado 80631 Regional Representative Radiation Programs Environmental Protection Agency 1860 Lincoln Street Denver, Colorado 80203 Mr. H. L. Brey, Manager Nuclear Licensing/Fuels Div. Public Service Company of Colorado P. 0. Box 840 Denver, Colorado 80201 J. W. Gahm, Manager, Nuclear Production Division Fort St. Vrain Nuclear Station 16805 WCR 19} Platteville, Colorado 80651 L. Singleton, Manager, Quality Assurance Division (same address) Colorado Radiation Control Program Director EVALUATION CRDOA BEARING REPORT FROM PUBLIC SERVICE COMPANY OF COLORADO Bert L. Barnes July 1 , 1985 EVALUATION OF CRDOA BEARING REPORT FROM PUBLIC SERVICE COMPANY OF COLORADO Summary An independent review of the Public Service Company of Colorado letter report indicates the conclusions reached within the report to be generally valid and that the replacement bearings are statistically and operationally as good as the original CRDOA bearings. These conclusions are based upon an independent review of technical literature on ball bearing failure, dry molybdenum disulfide lubrication, ball bearing materials, bearing clearances and tolerances, life testing, life calculations, contact stresses, duty cycles and loadings. Where enough data was provided. calculations described within the Reference 1 report were independently spot checked and found to be correct. The Public Service Company of Colorado CRDOA Report provides positive support for the NRC Regulatory approval for all proposed changes to the CRDOA hardware described by this report. Object The object of this report was to review the Public Service Company of Colorado report (Reference 1 ) to determine if the Fort Saint Vrain control rod device and orifice assembly replacement ball bearings were equivalent to or at least as good as the original bearings used prior to January 1985. Introduction This letter report was prepared under the provisions of the NRC Form 189 for FIN A6701, Task Order R. to provide technical assistance to the NRC Region IV office. The report (Reference 1 ) reviewed herein summarily describes actions taken by Public Service Company of Colorado and their subcontractors (GA Technologies, Industrial Tectonics, Inc. , and SKF Industries) to justify the use of replacement ball bearings differing only slightly from the original ball bearings for the Fort Saint Vrain control rod drives and orifice assemblies. The items discussed below follow the format and chronological order of the Reference 1 report. Page and paragraph numbers cited refer to this report. Commentary Review All the bearing raceways and some of the balls are 440C stainless steel (pg. 2, paragraph 3). This material is recognized as an excellent ball bearing material (Reference 2), though it is not as commonly used as E52100 type steels. The 440C materials have an upper limit on hardness of Rockwell C60 (200°F temper), while the E52100 steels have a Rockwell C64 hardness upper limit (200°F temper) . Reference 2 (pg. 433-438) generally indicates that bearing load capacity and fatigue life are both very 1 sensitive to hardness. A few points of increase to hardness can easily double the load capacity of a given material and maximum fatigue life is also obtained with the materials of highest hardness. However, in the presence of moisture and oxidizing contaminants, 440C bearing materials may be superior (Reference 2, pg. 324) to the E52100 steels because of corrosion and oxidation resistance. In either case. 440C is an excellent material for this application. The tungsten carbide balls, 17-4 PH and nitralloy 135M cages are also generally recognized (Reference 2) as being good materials for these applications. Though Reference 1 was generally lacking in detail , the sintered bronze lubricant reservoirs are expected to be equivalent to the "Bearite" material used earlier in the original designs. The choice of dry molybdenum-disufide powder, (pg. 2, paragraph 3) as a lubricant is also excellent for this application, given that the high radiation level and operating temperature make the use of common oil and grease lubricants impractical . The applicable temperature range for this lubricant exceeds 600°F (Ref. 2, pg. 212 and pg. 234, Ref. 9, pg. 26. Ref. 3, pg. 145-146, Ref. 4, pg. 163-164) . Molybdenum disulfide appears to be the best choice in dry lubricants for temperatures around 200°F to 250°F; for higher temperatures, Pb0 and CaF2 exhibit superior performance. It should be noted that molybdenum disulfide lubricants are not all the same; depending upon the binder material and other additives, the performance of this lubricant can be variable (Reference 2, pg. 212) . The Reference 1 report gives no details regarding the specific molybdenum disulfide lubricant used and how it compares with the original lubricant. However, the testing program does at least partially verify that the lubricant used was of high grade and applicable for this service. The dimensional differences between original and replacement bearings. including reductions in the number of balls, do not appear to contribute to altering the reliability of the control rod drive mechanisms (pg. 3. paragraph 1 ). Though the dimensional data given in Appendix A of Reference 1 are incomplete, the tolerances and clearances for corresponding parts in the original and replacement bearings are very similar . A comparison of ISO Tolerance Tables (starting on page 228 of Ref. 5) indicates that there are no significant differences in the original and replacement bearings. This conclusion is also supported by data from Reference 6 on bearing tolerances, though, as might be expected, the tolerances for instrument bearings listed in Reference 6 are in general tighter than those used for these dry lubricated ball bearings. Though Appendix A (Reference 1 ) does contain limited data on comparative surface finishes for the retaining rings and cages , no such data is listed for the balls and raceways. Reference 2, pg. 436-438 indicates that bearing life (fatigue) is very much a function of surface finish. Better surface finishes in general exhibit significantly longer fatigue life. Ball bearing rolling resistance, and in particular the "breakaway torque" for dry lubricated ball bearings, are very much a function of ball and raceway surface finish. Not all surfaces on a ball bearing are critical ; the roughest surfaces have about a 32 finish while ball and raceway finishes are generally much better than this . A 2 comparison of replacement and original ball bearings with regard to this important factor is not possible without this data. Also, a comparison of the actual critical surface finishes for original and replacement balls and cages is more important than a comparison of the specified values. In this regard, Reference 1 is deficient. The finish specifications given for the retaining ring (Part Number SLR-01210-222, Appendix A, pg. 2) are at best very very rough. The 63 and 32 outside diameter surface finishes for the cage (Appendix A. pg. 3) are also relatively rough for critical parts such as these. However, the cage outside diameter surface finish is probably relatively unimportant compared to that of the ball pockets. All parts which touch the ball bearings themselves and the inner and outer raceway critical surfaces should have RMS surface finishes of 4.0 or better. In particular, the cage ball pockets, the sintered bronze lubricant reservoirs, and other surfaces of the cage which rub or roll against the inner or outer races should have RMS surface finishes of 4.0 or better, but certainly no rougher than RMS 10 at the very worst. The retaining ring specified surface finish (63 RMS original, 125 RMS replacement) is probably not critical . The actual surface finish is probably much better than this. However, lacking the lack of surface finish data for other critical parts is an error of omission for a design parameter which strongly influences bearing life and 'breakaway torque", the primary failure mechanism of this entire control rod device. A 125 RMS surface finish is typically what results from a rough lathe turning or roughing cut on a milling machine with no surface polishing or sanding. In order to facilitate comparison of original and replacement bearings, all critical surfaces should have a listing of both specified and actual as-measured surface finishes. While it may be that the replacement bearings have critical surface finishes equal to or better than the original bearings, Reference 1 provides no evidence of this. Reference 1 (pg. 3, paragraph 3) indicates the design parameters considered critical in assessing bearing performance to be, applied loads, internal geometry, materials, and lubrication. In this application in particular, contamination should also be considered as a critical parameter. The failure of the control rod drives has been suspected of being associated with or linked to the presence of water and water vapor in and around these bearings. Reference 2, pg. 215-216 indicates that contaminants are generally undesirable. Some contaminants can significantly increase the wear rate of the bearing though no specific reference was found in the literature search to water or water vapor causing premature failures. In this respect, both original and replacement bearings must be considered equally resistant to contamination. The 80/20 figure cited (pg. 4, paragraph 2) for assumed duplex bearing loading distribution seems like an excellent choice. The actual load distribution is probably much better than this. However, lacking more specific data, this assumption is probably conservative. 3 The radial load data (pg. 4, paragraph 2) is impossible to evaluate without more data. However, the loads listed seem reasonable by superficially considering the geometry of the parts, the loads applied to the cable by the control rods. and x h the re ct ond at coft st fnt duiratio each ach stage esof the, pg.gear box. If the diagram (Appendix these loads could be verified. The accuracy of the fatigue life calculations is directly related to the accuracy of the bearing loadings. In similar fashion, the validity of the testing program is dependent on accurately modeling the actual bearing loads. The bearing operational life cycle data (pg. 5, paragraphs 1 and 2) is presented in sufficient detail that the numbers can be verified by calculation. Verification calculations indicate all the numbers calculated within these paragraphs to be both reasonable and correct. The general worth and applicability of the fatigue life calculations described in Reference 1 (pg. 5, paragraph 3 to pg. 8, paragraph 4) are deserving of comment. 1 . The fatigue life of a bearing is related to a subsurface material failure and as such is not the surface failure mechanism observed for these bearings. This statement is supported by the observation that the bearings with the lowest predicted life (pg. 8 tabular L-10 lives) have never failed in service. In fact, the only bearings that have failed (due to increases in surface roughness) have fatigue lives of between three and four orders of magnitude greater than the bearings that are predicted to be the shortest lived. Fatigue life calculations are worth while to facilitate comparison of original and replacement bearings, but fatigue failures apparently have little or nothing to do with the failures observed for the control rod drive bearings . The fatigue life of original and replacement bearings, while not exactly equal, are roughly equivalent (pg. 8. tabular data). 2. The results of the fatigue life calculations are based upon the commonly used 10% failure criterion and as such are statistically highly variable. To quote from Reference 2, pg. 167: "If a number of similar bearings are tested to fatigue at a specific load, there is a wide dispersion of life among the various bearings. For a group of 30 or more bearings, the ratio of the longest to the shortest life may be of the order of 20 or more. A curve of life as a function of the percent of bearings that failed can be drawn for any group of bearings. For a group of 30 or more bearings, the longest life would be of the order of four or five times the average life. The term life, as used in bearing catalogs, usually means the life that is exceeded by 90 percent of the bearings. This is the so-called 8-10 or 10-percent life. The 10-percent life is one-fifth the average of 50-percent life for a normal life-dispersion curve". 4 Considering that the 37 control rod drives each employ 14 ball bearings, three of which are acknowledged to be critical, the probability of an early fatigue, life failure significantly differing from the mean predicted life is very high. If the predicted fatigue lives were not so long (51 years minimum continuous operation), this would justify concern. Even if the fatigue life were reduced by a factor of 100 for the 111 (37 units x 3 resulting bearings/unit) shortestfa critical motor bearings, the tiguelife would exceed 1000 years. 3. The statements made (pg. 8, paragraph 2) regarding the applicability of fatigue life calculations for oil lubricated bearings not being directly applicable o dry film lubricated dies bearings are true. Reference 2, (pg. 383, paragraph b) this statement by saying: "All experimental evidence obtained to date indicates that the inverse cubic relation between load and life. which was found to exist for point contact with conventional bearing materials with mineral-oil lubrication, is approximately true for other materials and lubricants and for bench-type fatigue testers used for studying the effects of different variables on rolling-element-bearing fatigue". 4. The Hertzian contact stress criterion of 368 KSI being one-third the Brinnel hardness number was not verified by a review of NASA-SP38 (Reference 2). This is a large reference; perhaps the author missed finding the words to verify this. Also, since no data was given for bearing material hardness, it was not possible to verify that 368 KSI was one-third the Brinnel hardness number. The 368 KSI contact stress cited seemed reasonable when compared to Hertzian stresses described in Reference 2 (pg. 383 to 384). It should be noted that calculated Hertzian contact stresses are not particularly sensitive to applied radial loads. Minor errors in the calculation of the applied radial loads (pg. 4, paragraph 2) will not result in significant errors for calculated Hertzian contact stresses because the contact stress varies as the cube root of the normal force (radial load). The predicted fatigue life, on the other hand, is very sensitive to errors in predicting radial loads. The predicted fatigue life of a ball bearing varies inversely as the cube tabulardata) onthe the load. Because the lowest predicted fatigue life (pg. order of 51 years of continuous operation,rati onl ) sd rror n, y gross ses in the radial load prediction (pg. 4. reason for alarm. However, as stated earlier, the radial loads were not verifiable from the data provided in Reference 1 . If for any reason the actual loads are later found to be higher than the predicted loads, these calculations should be repeated. The physical testing programs described (pgs. 9 to 11 ) seem reasonable. However, the conclusions drawn from a small sample size can be 5 very misleading and very inaccurate. The quote cited earlier from Reference 2, pg. 161 provides ample evidence that the life of seemingly identical ball bearings is statistically highly variable. To test two or three bearings of a given type and then to predict or even hint that 111 3 per drive unit and 37 drive units) of these bearings will last as long is not prudent. This is probably the biggest single factor that makes the test results inconclusive. The 30 oz-in. bearing torque criterion (pg. 9, paragraph 2) described seems to be a questionable choice for a failure criterion. The shim motors each employ three bearings; a cumulative (3 bearings) frictional bearing torque exceeding 15 in.-oz constitutes a failure of the control rod drive mechanism. For this reason, a more realistic failure criterion might be based upon the total frictional torque from any three bearings exceeding 15 in.-oz. Any conclusions drawn on the basis of tests with a two bearing 30 in.-oz failure criterion are inconclusive. The point in time when the cumulative bearing torque from two bearings exceeded 2/3 x 15 in.-oz = 10 in.-oz might also be meaningful . The reasoning behind the choice of 65 lb and 15.3 lb radial loads was not clear (pg. 10, paragraphs 3 and 4) . The table on page 4 lists loads of 50.8 lb and 12.2 lb as being closest to the test loads for the shim motor. Though the failures experienced in shim motor bearings are not directly related to fatigue life failures, the fact that fatigue life is known to vary inversely with the cube of the radial load should provide motivation to very carefully select the test loads. Perhaps the differences in these figures is an indication of additional static loading not included in the load table on page 10. The Reference 1 report provides no explanation for these differences. Some tests did use a helium environment and select other operational duty cycle test parameters which were conservative (more demanding) than the actual duty cycles. However, no tests were done at elevated temperatures duplicating that of the real operational environment of 200° to 250°F. The lubrication of the bearings has been acknowledged as being critical to bearing life. Though molybdenum disulfide lubrication is capable of much higher temperatures (exceeding 600°F for some applications), certainly life tests run at room temperature are not necessarily applicable to operational temperatures of 200° to 250°F. Also, prolonged temperatures of 250°F will reduce the hardness of the 440C bearing races if a temperature of less than 250°F was used to temper these parts when originally heat treated. As stated earlier, ball bearing life has been shown to be directly related to material hardness. A few points reduction in hardness can significantly reduce ball bearing life. In this report, both original and replacement bearings are expected to be equal . The lack of including both radiation and pressure effects in the test series seems justified because the author has found no evidence to indicate that bearing life is sensitive to these parameters. 6 Conclusions This author is in general agreement with all the conclusions from physical testing (pg. 11, paragraphs 2 to 4). In particular, this author believes the replacement bearings are roughly equivalent to the original bearings. Though the test results are inconclusive, and some of the test parameters and criteria are questionable, the author believes the replacement bearings to be equivalent to the original and suitable for use in the control rod drives. 7 REFERENCES 1 . D. W. Warembourg, Public Service Company. Colorado, letter report to E. H. Johnson, USNRC, Region IV, Evaluation of CRDOA Bearings. June 13, 1985. 2. E. E. Bisson and W. J. Anderson. Advanced Bearing Technology, Cleveland, Ohio, NSA SP-38, Lewis Research Center. 1964. 3. P. Freeman, Lubrication and Friction, London. Whitefriares Press Ltd. , 1962. 4. F. P. Bowden and D. Tabor, The Friction and Lubrication of Solids, Clarendon, Oxford University Press, 1958. 5. A. Palmgren, Ball and Roller Bearing Engineering, Philadelphia, third edition, S. H. Burbank and Co. Inc.. 1959. 6. Anti-Friction Bearing Manufacturers Association, Inc. , "AFBMA Standards for Instrument Ball Bearings.' Section 12, Revision 2, AFBMA, New York. October 1966. 7. Plant Engineering Training Systems. Unit 4, Bearings Lubrication, Edited by Sayre, Clifford R. , Technical Publishing Company, Barrington, Ill . , 1970. 8. Plant Engineering Training Systems, Unit 2, Lubrication, Edited by Sayre, Clifford R. , Technical Publishing Company, Barrington, Ill . , 1970. 9. FMC Corporation, Link-Belt Bearing Division, Bearing Technical Journal , Indianapolis, Indiana, First Edition, 1970. 8 ,.P no ?Jost �4r UNITED STATES W" it NUCLEAR REGULATORY COMMISSION I i ; REGION IV •• -g e? 611 RYAN PLAZA DRIVE,SUITE 1000 e° ARLINGTON,TEXAS 78011 JUL 1 0 1985 ,11119 COUNTY COMMISSIONERS Docket 50-267 D Cif-PP��R \ JUL 161985 Mr. 0. R. Lee, Vice President �e � r• Electric Production Public Service Company of Colorado P.O. Box 840 Denver, Colorado 80201 Dear Mr. Lee: As a result of the problems encountered during the testing of the emergency diesel generator sets (EDGs) on December 18, 1984, the NRC has reevaluated the Fort St. Vrain (FSV) emergency electrical systems. The results of our review are contained in the enclosed Safety Evaluation (SE). We have concluded that, even with the potential EDG single failure and independence problems identified in the SE, FSV can be operated safely for an interim period while corrective measures are being pursued. Your commitment to provide an evaluation of the above problems, within 90 days of your June 14, 1985 (P-85208) submittal , will be confirmed in the listing of various commitments that will be issued with the authorization to restart FSV. In addition to the safety questions addressed in the enclosed SE, the question of plant operation in possible nonconformance with the established licensing basis (FSAR) must be examined from a regulatory compliance viewpoint. We have discussed this matter with members of your staff and are awaiting additional information. If your have any questions on this matter, please contact the NRC Project Manager, P. Wagner, at (817)860-8127. Since the reporting requirement relates solely to FSV, OMB clearance is not required under P.L.96-511. Sincerely, ti: ea. o E. H. Johnson, Chief Reactor Projects Branch Enclosure: SE on EDGs cc: (see next page) I 1 _ 1_ 4 I lY I4 C cc: Mr. D. W. Warembourg, Manager Nuclear Engineering Division Public Service Company of Colorado P. 0. Box 840 Denver, Colorado 80201 Mr. David Alberstein, 14/159A GA Technologies, Inc. P. 0. Box 85608 San Diego, California 92138 Kelley, Stansfield & O'Donnell Public Service Company Building 550 15th Street, Room 900 Denver, Colorado 80202 Chairman, Board of County Comm. of Weld County, Colorado Greeley, Colorado 80631 Regional Representative Radiation Programs Environmental Protection Agency 1860 Lincoln Street Denver, Colorado 80203 Mr. H. L. Brey, Manager Nuclear Licensing/Fuels Div. Public Service Company of Colorado P. 0. Box 840 Denver, Colorado 80201 J. W. Gahm, Manager, Nuclear Production Division Fort St. Vrain Nuclear Station 16805 WCR 19} Platteville, Colorado 80651 L. Singleton, Manager, Quality Assurance Division (same address) Colorado Radiation Control Program Director 0fl. mewfr UNITED STATES JUL 0 5 1985 O * NUCLEAR REGULATORY COMMISSION ft • i REGION IV 011 RYAN PLAZA DRIVE,SUITE 1000 fib ARLINGTON,TEXAS 76011 SAFETY EVALUATION BY THE OFFICE OF NUCLEAR REACTOR REGULATION PUBLIC SERVICE COMPANY OF COLORADO FORT ST. VRAIN (FSV) NUCLEAR GENERATING STATION DOCKET NO. 50-267 EMERGENCY ELECTRICAL POWER SYSTEM INTRODUCTION By task interface agreement (TIA) No. 85-02, Rev. 1, dated February 5, 1985 the NRR staff was requested to review the subject design with reference to the problems encountered during the testing of the emergency diesel generator sets (EDGS) on December 18, 1984. The staff has reviewed FSV's latest revision of the FSAR and electrical schematic drawings of the emergency power systems. The review focused primarily on the compliance of the design with the redundancy, independence and single failure criterion established in the FSAR. BACKGROUND Emergency electrical distribution system at FSV is a 3 bus (two redundant and one swing bus) system with two 100% load capacity EDGs. Each EDG has two tandem engines each rated to 1/2 of the generator output capacity. If only one of the two engines operates in one EDG system, the other redundant EDG must also be operative to supply the shutdown load with at least one of its engines operating. The intended logic at FSV is to start both EDGs simultaneously and let the first EDG with rated voltage, frequency and 100% output (both engines operating) be connected to its assigned 480 volt bus together with the swing bus. The first generator on line assumes sequence "A" loading which is sufficient for an orderly shutdown and continued maintenance of the plant in a safe shutdown condition. The second generator, if available with rated voltage and frequency, will assume sequence "B" loading. On December 18, 1984, with the reactor shutdown and the PCRV depressurized, the loss of offsite power and turbine trip semiannual surveillance test was initiated by blocking one EDG (EDG-A) to intentionally make the other EDG (EDG-B) first on line and assume sequence A load. Due to the nonavailability of one of the two engines with EDG-B, this logic could not be completed and breaker did not close. The logic should have made the intentionally blocked EDG-A as the second generator in line and should have closed its supply breaker to initiate sequence B loading on EDG-A. The EDG-A breaker also did not close, thus causing loss of both redundant emergency power supplies to the essential buses. EDG-A failure to supply power was attributed to the inadvertent trip of exhaust temperature switches on both engines of EDG-A due to the loss of instrument power to these switches. This event necessitated a review of the FSV emergency electric system to establish the following: (1) Independence and redundancy of the onsite AC power supply distribution system and the safety loads to perform their safety function. JUL 0 5 198: - 2 - (2) Reasons for EDG-B's inability to get connected to the bus when only one of its two engines failed and the other was available to supply 1/2 of the designed capacity of EDG-B. EVALUATION The FSAR maintains in Section 8.2.5.1 that the AC and DC power systems in FSV design are each redundant systems; the onsite power supplies are completely independent and meet the single failure criterion. Our review of FSV's onsite electric system drawings (EDG breaker and bus tie breaker schematic diagrams and auxiliary tripping relays control diagrams), on sample basis revealed the following information. 1. Automatic closure of one redundant EDG breaker is dependent on the operation of components associated with the other redundant EDG. Interlocks from one division providing permissive in the breaker close circuitry of the other could potentially prevent the required operation of both circuits and render both emergency power supplies incapable of performing their safety function. 2. Each redundant EDG should be capable of supplying 100% of its rated power when both engines operate and 50% of its rated power when only one is operative. The EDG breaker should close for rated voltage and frequency irrespective of whether one engine is operating or both. FSV design (EDG breaker schematic) indicates that the breaker will not close if one of the two engines is inoperative. Both identified discrepancies, were explained to the licensee in detail in a meeting held on May 16, 1985. Based on the available information, it is the staff's understanding that the automatic operation of the redundant EDG circuit breakers is dependent on each other which is contrary to the FSAR requirement. This discrepancy could potentially render both emergency power supplies incapable of automatically performing their safety function. However, in case the EDG breaker fails to close automatically, manually operated electrical control breaker closing circuitry is available in FSV design to initiate closure of the breaker immediately after identifying the failure of the automatic circuitry. The licensee confirmed that the manual circuitry is not affected by the failure of the automatic breaker closure circuitry. Our review of the FSAR indicates that besides the EDGs, the FSV design includes an alternate means of providing electric power for cooling the reactor, in case both offsite power and EDGs are not available. This power source is capable of operation, independent of disruptive faults or events, such as a major fire in the congested cable areas. This system is named "Alternate Cooling Method (ACM)" and manually started to restore 480V electric power at the control terminals of the required safety equipment within two hours (1 to 2 hours). The ACM power source is a non Class 1E, 4160 volt, 60 Hz AC diesel generator, rated at 2500 kW (equal to the combined rating of both EDGs) and is located JUL 05 198E - 3 - away from the existing plant structure. The ACM is designed to accomplish the following functions by means of local manual initiation: (a) To maintain the reactor subcritical using Reserve Shutdown System. (b) To resume PCRV liner cooling, thereby cooling the core and the PCRV. (c) To allow depressurization of the PCRV through the helium purification system. (d) To establish operation of the Reactor Building Exhaust System and radiation monitoring of the exhaust effluent to the atmosphere. Additionally, the ACM can power the plant lighting, fire pumps, service water pumps and plant ventilation system. The staff reviewed Section 8.2.8.5 of the FSAR and noted that in the unlikely event when both EDGs are not available coincident with the loss of offsite power, and the ACM power is restored to the emergency equipment by manual means within two hours, then adequate core cooling and depressurization of primary coolant system can be achieved maintaining integrity of the core and the PCRV. CONCLUSION The staff's evaluation of the FSV EDG electric system has identified some discrepancies in EDG breaker control logic regarding independence of the redundant EDG system. However, the inherent capability of the PCRV and core with an alternate non-Class lE power source (ACM) provides an added assurance of safe shutdown capability. In the interim, until the licensee proposes any necessary modification in the EDG breaker automatic control circuitry, manually operated switches are available to override the automatic control circuit failure and close the breakers to provide power to operate equipment necessary for safe shutdown of the plant. This can readily be accomplished well within the time frame available to prevent damage to the reactor. It is the staff's conclusion, as it relates to the emergency power system prob- lems identified, that the plant operation may resume without undue risk to the health and safety of the public. However, the licensee should establish a sche- dule, without undue delay, for the review and resolution of the potential single failure and independence problems for EDGs identified in this report. Date: JUL 05 1985 Principal Contributor: I. Ahmed, DSI y SSINS No. : 6835 IN 85-50 UNITED STATES WEIR COUNTY COMMISSIFERS NUCLEAR REGULATORY COMMISSION OFFICE OF INSPECTION AND ENFORCEMEN 0 WASHINGTON, D.C. 20555 JUL'1 8 July 8, 1985 GREELEY. COLO. IE INFORMATION NOTICE NO. 85-50: COMPLETE LOSS OF MAIN AND AUXILIARY FEEDWATER AT A PWR DESIGNED BY BABCOCK & WILCOX ADDRESSEES: All nuclear power facilities holding an operating license (OL) or construction permit (CP). Purpose: This information notice is being provided to inform licensees of a significant reactor operating event involving the loss of main and auxiliary feedwater at a pressurized water reactor. Information in this notice is preliminary and was obtained from the special NRC fact finding team which is investigating the event. A complete report of findings will form the basis for further communi- cations or actions related to this event. The NRC expects that recipients will review this notice for applicability to their facilities. Suggestions contained in this notice do not constitute NRC requirements; therefore, no specific action or written response is required. Description of Circumstances: On June 9, 1985, the Davis-Besse plant was operating at 90% power with Main Feedwater Pump 2 in manual control because problems in automatic had been experienced. A control problem with Main Feedwater Pump 1 occurred, and it tripped on overspeed. Reactor runback at 50% per minute toward 55% power was automatically initiated. Nevertheless, 30 seconds later, the reactor tripped at 80% power on high pressure in the reactor coolant system. One second after reactor/turbine trip, one channel of the Steam and Feedwater Rupture Control System (SFRCS) was automatically initiated due to a spurious signal indicating low water level in Steam Generator 2. Both Main Steam Isolation Valves (MSIVs) closed. Three seconds after the actuation, the SFRCS automatically reset. Closing of the MSIVs isolated the turbine of the operating main feedwater pump from its source of steam. The pump continued to supply feedwater to the steam generators for a few minutes as it coasted down. Four and a half minutes after reactor trip, water level in the steam generators began to fall from the normal post-trip level which is 35 inches. After MSIV closure, steam release to atmosphere continued to remove decay heat. One minute later, Channel 1 of SFRCS actuated when the water level in Steam Generator 1 actually reached the SFRCS setpoint at 27 inches (See Figure 1). SFRCS started Auxiliary Feedwater Pump 1 and initiated alignment of it to Steam Generator 1. 8507080156 IN 85-50 July 8, 1985 Page 2 of 4 Within seconds after automatic initiation of Channel 1 of SFRCS, the operator actuated both channels of SFRCS; however, he inadvertently actuated both SFRCS channels on low steam pressure instead of low water level . When an SFRCS channel is actuated on low steam pressure, a rupture of the steam line associated with that channel is presumed to have occurred. The SFRCS closes the steam generator isolation valves, including a valve in the auxiliary feedwater line, and aligns the auxiliary feedwater pump to the other steam generator. Because both channels had been manually actuated on low steam pressure, both steam generators were isolated from both auxiliary feedwater pumps. Five seconds after the operator' s inadvertent actuation of both channels on low steam pressure, SFRCS Channel 2 received an actual low water level actuation signal . Because low pressure initiation takes precedence, alignment of the auxiliary feedwater pumps remained unchanged. At six minutes into the event as both auxiliary feedwater pumps were accelerating, they tripped on overspeed. In summary, all main feedwater had been lost, both steam generators were isolated from feedwater and were boiling dry, all auxiliary feedwater pumps were tripped, pressure of the reactor coolant system was rising, and reactor coolant system temperature was increasing. Within one minute after the operator' s inadvertent actuation of the SFRCS on low steam pressure, the mistake had been recognized and the SFRCS had been reset. If equipment had performed in accordance with system design requirements, the operator' s error might not have had a significant impact on the event. The auxiliary feedwater isolation valves should have reopened automatically, but the valves did not reopen. The operator then tried to reopen the valves from the main control panel , but the valves would not reopen. Operators were dispatched to locally start the auxiliary feedwater pumps, open the auxiliary feedwater isolation valves, start the nonsafety-related motor-driven startup feedwater pump, and valve it to the system. Pressure and temperature in the reactor coolant system continued to rise because there was not sufficient water in the steam generators to provide an adequate heat sink. At 13 minutes after reactor trip, reactor coolant system pressure reached 2425 psig, and the Pilot Operated Relief Valve (PORV) opened three times to limit the pressure rise. On the third lift, the valve remained open. The operator closed the PORV block valve and reopened it two minutes later after the PORV had closed. Approximately 16 to 18 minutes after reactor trip, the operators had the startup and auxiliary feedwater pumps running and the valves aligned. Water levels were beginning to rise in the steam generators. Reactor coolant temperature reached a maximum of 594° F and then started to decrease to normal . Refilling of the steam generators caused the reactor coolant system to fall to 1716 psig and about 540°F before returning to normal (See Figure 2). At 30 minutes after reactor trip, plant conditions were essentially stable. IN 85-50 July 8, 1985 Page 3 of 4 Discussion: For several minutes after reactor trip, the steam generators were unable to cool the reactor coolant system adequately. The first problem contributing to this event was the loss of all main feedwater due to closure of the MSIVs. The licensee' s hypothesis, based on information from Babcock & Wilcox, is that turbine trip caused a pressure transient upstream from the turbine stop valves which caused the outputs of the redundant steam generator level instrumentation channels to oscillate widely for several seconds. The licensee believes that this caused a spurious low level actuation of SFRCS which closed the MSIVs. Three additional problems contributed to this event by affecting the availability of both trains of the auxiliary feedwater system. The first occurred when the reactor operator pressed the wrong SFRCS buttons. The second occurred when both auxiliary feedwater pumps tripped on overspeed. The third occurred when both auxiliary feedwater isolation valves did not reopen when SFRCS was reset. Control buttons for the SFRCS are arranged in two vertical columns. Each column of buttons controls one SFRCS channel . The operator should have pressed the fourth button from the top in each column. Instead, the operator pressed the top buttons causing isolation of both steam generators. Both auxiliary feedwater pumps are driven by Terry turbines which tripped on overspeed early in the event. When this occurred, steam was being supplied to the turbines via crossover lines, which are longer than the normal supply lines and include long horizontal runs. The licensee believes that significant condensation may have occurred in the crossover lines. Further, the licensee believes that the quality of steam arriving at the turbines may have been affected significantly by the configuration of the crossover lines and may have caused the overspeed trips. The auxiliary feedwater system isolation valves have Limitorque motor operators. The motor operators have torque switches which prevent overtorquing of the valves by disconnecting power to the motors. When the valves are being opened, additional torque is required to overcome friction while the gates are being unseated and while a significant pressure differential may exist across the gates. During the initial part of the opening stroke, the torque switch in the motor operator is bypassed by a bypass switch so that full motor torque is developed if necessary. The licensee believes that these bypass switches went off bypass too early. The valves did not reopen until an operator unseated them by hand. IN 85-50 July 8, 1985 Page 4 of 4 No specific action or written response is required by this information notice. If you have any questions about this matter, please contact the Regional Administrator of the appropriate NRC regional office or this office. alEdward Jordan, Director Divisio of Emergency Preparedness and Engineering Response Office of Inspection and Enforcement Technical Contact: R. W. Woodruff, IE (301) 492-4507 Attachments: 1. Figure 1 - Steam Generator 1 Level and Pressure 2. Figure 2 - RCS Temperature and Pressure 3. List of Recently Issued IE Information Notices Attachment 1 IN 85-50 ' LB83 SG 1 SU RANGE LVL, 983 (IN) July 8, 1985 o i.S' 50 75' /00 /JS 4-0 /75 200 say .2-SO t I 1 P932 SG 1 OUT SIM PRESS.PT12B2 PSIR 600 ESo Tan 750 goo $Sb 900 9.522 ioao /ctso //co — - 0 T cn •tsi el _. ...........o a ...........e.40 9 5 �1 ! . . ! . • IHT 4 . FIGURE I : STEAM GENERATOR I LEVEL auD PRESSURE Attachment 2 IN 85-50 July 8, 1985 P725 RC LOOP I HLG WR PRES5.SFRS CH 3 • PSIR /We /Soo 1600 M» /S00 /900 .lee NOo faoo K. •4t'O f T709 ' RC RVG NR TEMP ', F 5.10 453 d 5+0 .SS° $6o 37o a a sy0 boo 610 620 C'-44 0 --1 - - - —y- 0%- i ham-. -0 _ ........... ............. 3 AN . - _ : ---I 3 1+ as b FIGURE z: RCS TEMPERATURE: I u r, oor_ CC '0 Attachment 3 IN 85-50 July 8, 1985 LIST OF RECENTLY ISSUED IE INFORMATION NOTICES Information Date of Notice No. Subject Issue Issued to 85-49 Relay Calibration Problem 7/1/85 All power reactor facilities holding an OL or CP 85-48 Respirator Users Notice: 6/19/85 All power reactor Defective Self-Contained facilities holding Breathing Apparatus Air an OL or CP, research, Cylinders and test reactor, fuel cycle and Priority 1 material licensees 85-47 Potential Effect Of Line- 6/18/85 All power reactor Induced Vibration On Certain facilities holding Target Rock Solenoid-Operated an OL or CP Valves 85-46 Clarification Of Several 6/10/85 All power reactor Aspects Of Removable Radio- facilities holding active Surface Contamination an OL Limits For Transport Packages 85-45 Potential Seismic Interaction 6/6/85 All power reactor Involving The Movable In-Core facilities holding Flux Mapping System Used In an OL or CP Westinghouse Designed Plants 85-44 Emergency Communication 5/30/85 All power reactor System Monthly Test facilities holding an OL 85-43 Radiography Events At Power 5/30/85 All power reactor Reactors facilities holding an OL or CP 85-42 Loose Phosphor In Panasonic 5/29/85 All power reactor 800 Series Badge Thermo- facilities holding luminescent Dosimeter (TLD) an OL or CP Elements OL = Operating License CP = Construction Permit MEMORANDUM ��DF cow „„, DIVISION OF DISASTER John P.Byrne FQ _ EMERGENCY SERVICES — - DIRECTOR * �' Gy * Camp George West HELD CONNI�CO COMMISSIONERS ` •- • Golden,Colorado 80401 *7876 (303)273-1624 1 JUL 151985 DATE: 10 July 1985 GREELEY. COLO. TO: (See Distribution) FROM: John P. Byrne, Di ct r, Division of Disaster Emergency Services SUBJECT: Summary Report f 98 rt St. Vrain Exercises 1. Attached is the summary report for the 1985 Fort St. Vrain full scale exercise for your review and action as necessary. 2. The next full scale exercise will be conducted in 1987. In the interim, we will conduct periodic functional drills and exercises to sustain emergency preparedness. 3. While F0SAVEX-85 is still fresh in our minds, please forward any changes or revisions to the Radiological Emergency Response Plan to Bruce Smith as soon as possible. 1 attachment F0SAVEX-85 Exercise Report • /js cs COLORADO • DEPARTMENT OF PUBLIC SAFETY Distribution Office of the Governor Department of Administration Department of Public Safety Department of Health Department of Agriculture Department of Highways Department of Social Services Department of Law Department of Military Affairs Department of Local Affairs Department of Natural Resources Division of Disaster Emergency Services Division of Colorado State Patrol Weld County Commissioners Weld County Sheriff Weld County Emergency Management Director Weld County Health Department Weld County Agricultural Extension Service National Weather Service Federal Emergency Management Agency Environmental Protection Agency Food and Drug Administration Department of Energy American Red Cross Salvation Army Public Service Company of Colorado OVCD%� MEMORANDUM - --v t -DIVISION OFDISASTER __ _ John P.Byrne �$ EMERGENCY SERVICES DIRECTOR N J� * �� 44., *) Camp George West + • • *- Golden,Colorado 80401 1876# (303)273-1624 DATE: 1 July 1985 TO: John P. Byrne, Director, DODES • FROM: Bruce N. Smith, Exercise Directoz] 21. Summary Report for 1985 Fort St. Vrain Exercise SUBJECT: 1. The 1985 full scale exercise to test major off—site functional elements of the Colorado State Radiological Emergency Response Plan (RERP) for the Fort St. Vrain Nuclear Generating Station was conducted on 18 June 1985. The exercise short title was FOSAVEX-85 and was a joint participation exercise with the Public Service Company (PSC) of Colorado. The objectives of the off-site exercise were to exercise and evaluate the following: a. Notification procedures and public warning dissemination systems, to include activation of NOAA Weather Radio Network and passage of an exercise warning message to the Emergency Broadcast System (EBS) lead station. b. Deployment of response elements and their direction and control, to include activation of the State elements of the Forward Command Post (FCP) and establishment of a controlled area and traffic control points. c. Radiological monitoring, assessment and reporting procedures from field deployed Health Department monitoring teams through the FCP to the State Emergency Operations Center (EOC) . d. Operational communications and associated direction and control procedures at all levels. e. Food and agricultural control procedures. f. Dissemination of official, coordinated, pertinent and accurate information to the news media and general public. c s COLORADO DEPARTMENT OF PUBLIC SAFETY 2 This report is based upon the comments and observations of the official state exercise observation_and evaluation team, review of exercise logs and message traffic files, and discussions held during the post- exercise critique. 2. Overall, the exercise response was excellent and was indicative of sustained, on-going emergency preparedness training programs at all levels. In particular, the level of seriousness and professionalism displayed by exercise participants at all locations was exemplary--as one observer commented, "If you didn't have prior knowledge of an exercise, you would think an actual emergency existed". No major deficiencies were noted and the capacity to protect the health and safety of the public in the vicinity of the plant was clearly demonstrated. Contrary to what is often encountered, communications at all levels during FOSAVEX-85 were exceptional. 3. As in every successful exercise, areas were identified where improvements are possible. These "lessons learned" are the most valuable results of the exercise because, with proper follow up and training, they lead to improved emergency response capabilities. The following observations and comments are provided to assist in such improvements: a. State Emergency Operations Center (EOC) (1) The initial notification sequence was timely and efficient, due primarily to the rapid and correct reaction of the Weld County Communications Center. (2) Access processing at the EOC was degraded because of outdated organizational access rosters which required excessive delays and frequent escorts. (3) The exercise scenario was forced to be excessively artificial because of existing regulations. Scenario time compression forced a too rapid transition to a General Emergency category. PSC and DODES will once again request a determination from the Nuclear Regulatory Commission (NRC) and the Federal Emergency Management Agency (FEMA) as to whether a General Emergency category could ever be reached at Fort St. Vrain, given the radioactive inventory and reactor design. (4) EOC activation was smooth and efficient but some agencies required by the REAP to report to the EOC were notified but did not report. (5) Television and radio broadcasts were not monitored to keep track of public information being disseminated to assure accuracy (6) There is a need for more effective and detailed information exchange between PSC and State agencies when emergency classes change. (7) If NRC is going to interact with State agencies concerning 3 off-site activities, this needs to be formally recognized and incorporated into the RERP. b. Forward Command Post (FCP) (1) Rapid and efficient access was also a problem at the FCP because of outdated access rosters and the frequent need for individual, personal validation by the FCP Director. A positive aspect was the superb job of access control done by Weld County Deputy Sheriff Park -- if you weren't cleared, you didn't enter, period. (2) The FCP message center function needs improvement in the area of physical movement of message traffic between the FCP and the communications van. (3) A larger scale map of the plant vicinity, with clearly marked road identifications, would improve the efficiency of plotting and reading. (4) Non-essential vehicle traffic was allowed to pass through the alley outside the FCP and parking was not fully controlled. Both factors created unnecssary congestion and impeded efficient personnel movement. (5) The establishment and control of traffic control points by the Weld County Sheriff's Department was especially effective. (6) The physical layout of the FCP needs to be changed to take full advantage of video monitor equipment and to eliminate duplicate posting of the same data. Key staff elements, particularly the Department of Health and DODES, need to be more effectively located to enhance interaction and reduce distractions from peripheral activity. (7) It was significant and appropriate that the Department of Health senior representative continued to request field monitor readings at specific locations in the controlled area even after being informed that the on-site leak was contained and emergency class downgraded. c. Field Monitor Activity (1) After initial problems with frequency selection, the field monitor direction, control and reporting radio network functioned extremely well. To my knowledge, this is the first time that this capability has been effectively demonstrated and represents a significant step forward. (2) Field monitor team vehicles need to be identified in some manner so as to be readily recognized as emergency vehicles. (3) Even though marked improvement was evident, more practice in proper radio communication procedures would be beneficial. (4) Some method (vests, hats, etc.) should be used to identify field monitor team members to expedite assembly, briefing and dispatch. Improvement is also needed in issuing controlled area access badges to team members. Signs should be used to identify team assembly area to minimize confusion.- - (5) Better maps are needed for field monitor teams. (6) More actual demonstration of the use of monitoring equipment and sample collection procedures is needed as well as collection of air sampling equipment filters and aggregation and transport of samples to the laboratory for analysis. (7) Most monitoring equipment issued was overdue for calibration and personal dosimeters were not readily available for individual issue to monitor team members. (8) Greater attention is needed relative to emergency worker/monitor exposure control procedures and status briefings to same on the radiological emergency. /js J``¢M neau4J UNITED STATES NUCLEAR REGULATORY COMMISSION REGION IV • c e 611 RYAN PLAZA DRIVE,SUITE 1000 Y�'44 � .coq ARLINGTON,TEXAS 78011 JUL 12 1985 Docket: 50-267 !VD CBDDT3 t rz SSrR RS Public Service Company of Colorado r ;L ATTN: O. R. Lee, Vice President a1- 16935 � I Electric Production P. O. Box 840 Denver, Colorado 80201 clRca,car.czranm, Dear Mr. Lee: We have reviewed the various responses you have submitted related to our October 16, 1984, "Preliminary Report Related to the Restart and Continued Operation of the Fort St. Vrain Nuclear Generating Station" and the subsequent meetings. Our review of the germane submittals is presented in the enclosed Safety Evaluation (SE). Our review indicates that numerous issues have been resolved. There are, however, additional issues which must be resolved prior to plant restart and some longer term issues for which commitments for resolution must be acceptable prior to restart. Some of these issues are addressed by the enclosed SE; others will be the subject of separate correspondence. If you have any questions on this matter, please contact the NRC Project Manager - P. Wagner at (817) 860-8127. Sincerely, E. H. Johnson, Chief Reactor Project Branch Enclosure: Safety Evaluation cc: Mr. D. W. Warembourg, Manager Nuclear Engineering Division Public Service Company of Colorado P. 0. Box 840 Denver, Colorado 80201 (cont. on next page) �,� me Ho (8S Public Service Company of Colorado -2- Mr. David Alberstein, 14/159A GA Technologies, Inc. P. 0. Box 85608 San Diego, California 92138 Kelley, Stansfield & O'Donnell Public Service Company Building 550 15th Street, Room 900 Denver, Colorado 80202 Chairman, Board of County Comm. of Weld County, Colorado Greeley, Colorado 80631 Regional Representative Radiation Programs Environmental Protection Agency 1860 Lincoln Street Denver, Colorado 80203 Mr. H. L. Brey, Manager Nuclear Licensing/Fuels Div. Public Service Company of Colorado P. 0. Box 840 Denver, Colorado 80201 J. W. Gahm, Manager, Nuclear Production Division Fort St. Vrain Nuclear Station 16805 WCR 194 Platteville, Colorado 80651 L. Singleton, Manager, Quality Assurance Division (same address) Colorado Radiation Control Program Director ENCLOSURE tRR REG& UNITED STATES W° °� NUCLEAR REGULATORY COMMISSION L REGION IV w 611 RYAN PLAZA DRIVE,SUITE 1000 �'►� po ARLINGTON.TEXAS 70011 SAFETY EVALUATION BY THE NUCLEAR REGULATORY COMMISSION FORT ST. VRAIN NUCLEAR GENERATING STATION PUBLIC SERVICE COMPANY OF COLORADO DOCKET NO. 50-267 Introduction As a result of the failure of six Control Rod Drive Mechanisms (CRDMs) to automatically insert their associated neutron absorber material (2 control rod strings per CRDM) following a reactor scram on June 23, 1984, the NRC issued a confirmatory action letter (CAL). The CAL, dated June 26, 1984, required Public Service Company of Colorado (PSC or the licensee) to take various actions and to maintain the Fort St. Vrain (FSV) reactor in the shutdown condition until the NRC authorized restart. This failure to insert problem, in conjunction with other areas of concern, prompted the Director of the Office Nuclear Reactor Regulation, NRC, to direct an overall conduct of operations evaluation of FSV. A team of NRC personnel and NRC consultants from the Los Alamos National Laboratory conducted audits of the FSV facility on July 9 through 11, 1984 and August 1 through 3, 1984. The results of these audits are contained in a "Preliminary Report Related to the Restart and Continued Operation of the Fort St. Vrain Nuclear Generating Station" (Assessment Report) which was transmitted to PSC by letter dated October 16, 1984. This safety evaluation presents the NRC findings related to the various PSC submittals and meetings in response to the Assessment Report. Other areas of concern which require resolution prior to plant restart (e.g. , the concrete reactor vessel prestressing tendon system problems and the emergency electrical systems operation) are discussed in separate correspondence. Evaluation The Assessment Report required the completion of a number of items prior to the restart of FSV and additional long term items following restart. PSC responded to the Assessment Report requirements by letter dated January 4, 1985, (P-85003) with commitments to resolve the NRC concerns. Numerous meetings and submittals expanded and clarified those commitments. Our evaluation of the PSC commitments and corrective actions is contained below and follows the numbering sequence used in the Assessment Report. A. Items Required to be Completed Prior to Restart 1. Actions Required for Control Rod Problems a. "Ensure that future scram signals will result in all rods automatically being inserted into the core. The licensee must identify the failure mechanism and take corrective action for -2- the rods that did not scram; or if the cause cannot be positively identified through examination or analysis of the drive mechanisms and the circumstances of the failure, other compensatory measures must be taken to provide assurance of reliability of control rods. These measures could reasonably include refurbishing all drive mechanisms. Regardless, of any other measures taken to remedy the failure to scram problem prior to reactor restart, PSC must outline and commit to periodic inspection/preventive maintenance and surveillance programs for control rod drives and associated position instrumentation. b. "Implement procedures to prevent overdriving the control rods past the rod-in limit. c. "One 20-weight percent and one 40-weight percent reserve shutdown hopper should be functionally tested to assure that the reserve shutdown capability is fully available. d. "Until the long term corrective actions are completed, the licensee should develop a procedure that will require a reactor shutdown under conditions where purge flow is lost or when high levels of moisture exists in the coolant. e. "Implement a procedure for recording representative samples of CRDM temperatures at all operating conditions until continuous recordings capability is available." By letter dated January 31, 1985, PSC provided the plans to resolve the Control Rod problems. Our review of this, and supplemental information, is described in Attachment 1, Safety Evaluation of Control Rod Drive Mechanisms and Reserve Shutdown Systems. Attachment 2 is a copy of the Technical Evaluation performed by our consultant at the Los Alamos National Laboratory. There are, however, a number of items which require additional resolution or which have occurred since the issuance of the Assessment Report. As discussed in the Safety Evaluation (SE) , Attachment 1. The licensee must provide a commitment to operate the plant within the CRDM temperature limits accepted by the NRC. The temperature limits cannot be changed without NRC approval of new temperature limits or alternative methods of assuring CRDM operability; and -3- The licensee must provide a commitment to submit an improved - CRDM surveillance and preventative maintenance program within six months of plant restart. In addition, the following issues must be resolved: The acceptability of the replacement ball bearings used in the - CRDM refurbishment; and The acceptability of the epoxy used to attach the CRDM - temperatures sensors. By letter dated June 14, 1985 (P-85199) PSC stated that the interim Technical specifications (TS) contain a requirement that the CRDM motor temperatures are monitored to ensure the temperatures are within acceptable limits. (See PSC letter P-85180, dated June 7, 1985, and the discussion in Item 3 below. ) In addition, P-85199 committed to submit an improved CRDM surveillance and preventive maintenance program within six months of plant restart. By letters dated June 7, 1985 (P-85195) and June 13, 1985 (P-85201) PSC provided information on the epoxy and the bearings, respectively. Our review of this information is in progress and will be the subject of a subsequent SE. Additional information concerning the CRDM position instrumentation and procedures to prevent overdriving is contained in Item 3 below, "Actions Required for the upgrade of TS" , and in Attachment 3. 2. "Actions Required to Correct Weaknesses Noted in the Area of Overall Conduct of Operations "In the area of overall conduct of operations, the staff confirmed the deficiencies noted in various Region IV inspection reports and in the last two SALP reports. The staff has concluded that PSC must develop a comprehensive program for identifying the underlying causes for the deficiencies and for applying corrective measures. This program should be conducted by a third party consulting organization and should be aimed at reviewing the PSC management structure and practices relative to the operation of FSV with emphasis on correcting deficiencies noted in the various Region IV inspection reports, the last two SALPs and programmatic weaknesses identified in Section 4 of this report. PSC should submit the scope and schedule for this program prior to reactor restart." The scope of this management review program was further discussed with the licensee in a meeting on November 14, 1985. (The summary of this meeting is contained in IE Inspection Report 50-267/84-32 dated May 22, 1985. Based on the information contained in the -4- assessment report and the further understanding gained from this meeting, the licensee commissioned the NUS Operating Services Corporation to perform this review. By letter dated February 28, 1985 (P-85066), PSC provided a copy of the NUS Report, "An Analysis and Evaluation of the Management of Nuclear-Related Activities of the Public Service Company of Colorado" together with their response to that report. PSC provided additional information in their March 29, 1985 (P-85107) submittal. The staff reviewed these submittals and requested additional details regarding PSC's proposed actions for Sections 4.2.5 "Conduct of Operations" and 4.2.6 " Maintenance Practices" of the Assessment Report. This additional information is contained in the licensee's letter dated May 22, 1985 (P-85178). Based on our review of these submittals, we find that PSC has carried out a management review that meets the requirements of the Assessment Report. To provide the means of implementing the corrective actions needed to address the recommendations made in this management review, PSC has developed a Nuclear Performance Enhancement Program. This program is described in the March 29, 1985, letter referenced above. Additional details were provided to the NRC on May 31, 1985, during the SALP management meeting held with PSC at the Fort St. Vrain site. The status of implementation of this program was described to the staff in a meeting on June 17, 1985. The Nuclear Performance Enhancement Program is a proactive management scheduling and followup device that includes all of the significant findings of the NRC Assessment Report, the NUS management audit, NRC inspection findings and company generated findings. Each of the items is assigned a project manager and a schedule for completion is established in accordance with the licensee's overall priority scheme. Progress against this schedule is determined bi-weekly and management is kept advised in order to ensure that any schedule changes are agreed to by management. The staff has determined that the Nuclear Performance Enhancement Program has the structure and capability of carrying out the corrective actions that are necessary. If properly supported by PSC management, many of the deficiencies noted by the NRC in licensee performance should be adequately addressed. 3. Actions Required for the Upgrade of Technical Specifications a. "A high priority effort should be undertaken to review and propose revisions to the existing Technical Specifications to reduce the likelihood of operator error and/or misinterpretation and correct omissions. The staff has determined that a schedule should be -5- developed by PSC which will reflect completion of the review, revision, and submittal of the proposed Technical Specifications by April 1, 1985. b. "To improve control rod and reserve shutdown reliability, the licensee shall propose the following changes to Technical Specifications, and implement interim procedures until the Specifications are approved: (1) A Weekly control rod exercise surveillance program for all partially or fully withdrawn control rods; (2) A Limiting Condition for Operations defining control rod operability and the minimum requirements for rod position indication; and (3) A Limiting Condition for Operations and a corresponding surveillance test to define and confirm reserve shutdown system operability." By letter dated April 1, 1985 (P-85098) PSC provided a submittal titled "Upgraded Technical Specifications for Acceptance Review" in fulfillment of item a. above. Our evaluation of this submittal will be the subject of a future SE. By letter dated March 15, 1985 (P-85089) PSC submitted "Draft Technical Specifications to Improve Control Rod Reliability." Our review of this submittal determined that numerous problems existed and a meeting was held on May 3, 1985 to discuss those problems. The results of this meeting are documented in our letter dated May 28, 1985. By letter dated June 7, 1985 (P-85180) PSC resubmitted the TS, together with a commitment to implement those requirements through the use of interim procedures prior to restart. Our evaluation of this resubmittal indicated that some improvement and clarification was necessary to ensure resolution of our concerns. PSC has agreed to submit information to satisfy our concerns; we will evaluate this information prior to restart. As discussed in Attachment 3 - "Safety Evaluation Related to Control Rod Position Instrumentation", additional operability and surveillance requirements are needed to ensure safe operation. Specifically, the following issues should be incorporated: Additional TS and procedures for determining control rod - full-in position. (See Section 3.4 of Attachment 3. ) Additional surveillance tests on rod position indications. — (See Section 3.4 of Attachment 3. ) -6- Initiation of the backup shutdown system if rod full-in position indication cannot be verified within 1 hour. (See Section 3.4 of Attachment 3.) _ Additional procedural control to prevent inward overtravel of control rods. (See Section 3.2 of Attachment 3.) The proposed TS (P-85180) discussed above provide addition controls and limitations on the operation and testing of the control rods and their indication. In addition, plant operating procedures were revised to prevent inward overtravel of the control rods. (SOP 12-01, Issue 15 which was submitted in letter P-85040 dated January 31, 1985). Some problems have been identified which PSC has agreed to resolve in the submittal discussed above. 4. Actions on Continued Water Ingress "In the area of continued water ingress the staff has determined that PSC must develop a plan to carry out any of those modifications recommended by the PSC "Moisture Ingress Committee" that are determined by PSC to have a high potential to significantly reduce the frequency and severity of upsets involving injection of circulator bearing water into the helium coolant. Any significant reduction would clearly reduce the frequency of plant transients; improve the reliability of overall plant operations; and might, if has an effect, improve the performance of control rod drives. This plan should include a status report to the NRC as part of the annual report on the progress in implementing modifications." By letter dated January 24, 1985 (P-85022) PSC outlined some of the modifications which have been completed, are presently under way, and are planned to be implemented to control the water ingress problem. Since this submittal did not contain sufficient detail to resolve our concerns, PSC agreed, during a meeting at FSV on February 21, 1985, to provide additional information. Additional detail was provided in PSC letter, P-85082, dated March 12, 1985. Our review of this submittal and other germane information will be included in a future SE. B. Actions Required Following Restart The Assessment Report, in addition to requiring that certain items be completed as a condition for plant restart, required some longer term actions. Specifically, the Assessment Report stated: "In addition to the above items required for restart, the staff noted several weaknesses that should be corrected on a longer term basis. The licensee must submit schedules within 60 days of restart for completing these items." The January 4, 1985 (P-85003) PSC letter provided the required schedules, NRC review of the schedules resulted in a meeting on January 15, 1985, to -7- discuss differences. The results of the meeting are contained in our January 17, 1985, letter to PSC. The items listed in the Assessment Report for which action is required following plant restart are: a. "Provide continuous recording of a representative sample of CRDM temperatures at all operating conditions to provide part of the data necessary for the longer term program noted below (Section 2). b. "Determine whether compensating design and/or operational modifications are needed to minimize moisture ingress to the CRDM cavities and minimize temperatures in the vicinity of the rod drives. In the event that temperatures recorded during plant operation prove to be higher than those for which the assembly was initially qualified, take immediate steps to perform environmental requalification testing of a CRDM assembly or hold temperatures to that for which the CRDM has been qualified (Section 2). c. "The present Watt-meter testing of the shim motor during drive-in and drive-out is not a reliable method to verify full insertion or withdrawal of control rods. This test should be refined or an alternative, reliable test for control rod position verification, must be developed (Section 3). d. "Investigate a design change to provide a positive stop on the CRDM position indicator potentiometer shaft to prevent overtravel (Section 3) and provide the results to the NRC. e. "Conduct an integrated systems study to resolve rod position indication, maintenance and operability questions (Section 3). f. "Establish procedures for verification and sign-off by the Maintenance Quality Control (MQC) of key steps in Technical Specifications surveillance procedures (Section 5). g. "Establish a procedure for review and concurrence by the QA organi- zation of safety-related procedures and changes thereto (Section 5). h. "At the time of the audit, the MQC group was reviewing each completed surveillance procedure. The staff concluded that this practice should continue. i. "A review by the QA organization of the content and adequacy of the Technical Specification procedures is important, and the staff has determined that this should be implemented." These items were all discussed in the January 4, 1985 (P-85003) submittal and some were clarified in our January 17, 1985, letter. In addition, -8- some of the items are further discussed as follows: a. Continuous recording of CRDM temperatures as, discussed in PSC letters P-85032 dated January 30, 1985, and P-85199 dated June 14, 1985, has been implemented through the use of multipoint recording devices which will provide frequent monitoring. b. A discussion of the modifications planned to control moisture and purge flow to the CRDMs and plans to qualify a CRDM to higher temperatures is contained in PSC letter P-85032 dated January 30, 1985. c. A discussion of the wattmeter test is contained in PSC letter P-85040 dated January 31, 1985. (The discussion of the wattmeter test in Attachment 3 to this SE indicates that further improvement is necessary. ) d. A discussion of the positive stop to prevent overdriving is contained in P-85032 dated January 30, 1985 and was discussed during our February 21, 1985, meeting. The final resolution of these long term issues will be the subject of a future SE. Summary All of the issues presented in the NRC's Assessment Report have been addressed by PSC as discussed above. Most of the near term issues have been satisfactorily resolved as have some of the long term issues. The remaining issues and additional issues requiring resolution prior to restart will be (or have been) the subject of separate SEs. Since there have been numerous commitments by PSC to provide various documents and/or implement various programs or procedures, we will confirm these commitments in writing in connection with authorization of plant restart. Attachments: 1. SE on CRDM and RSS 2. TER on CRDMs, RSS, and Tendons 3. SE on Control Rod Position Indication. rJ400NEou4r Attachment 1 0 UNITED STATES NUCLEAR REGULATORY COMMISSION w ' WASHINGTON,D.C.20555 r4' .0 SAFETY EVALUATION BY THE OFFICE OF NUCLEAR REACTOR REGULATION FORT ST. VRAIN NUCLEAR GENERATING STATION PUBLIC SERVICE COMPANY OF COLORADO AND RESERVE SHUTDOWN SYSIEM Docket No. 50-267 1.0 Introduction and Background On June 23, 1984, a failure occurred at the Fort St. Vrain Station (FSV) in which 6 of 37 control rod pairs failed to insert on the receipt of a scram signal . As a result of this incident and other problems at FSV, the Director of the Office of Nuclear Reactor Regulation asked his staff to assess several aspects of operations at FSV. The resulting assessment (Reference 1) defined actions required of the licensee before the plant could be restarted. It also defined longer term issues for the licensee to address. Subsequent to the June 23, 1984 event, on November 5, 1984, a reserve shutdown hopper failed to fully discharge during a routine Surveillance Test. Further meetings were held with the licensee on November 28 and 29, 1984 and on January 15, 1985. During the January 15 meeting, additional commitments made by the licensee and requirements for restart were clarified (References 2 and 3). Key commitments made by the licensee included refurbishment of all control rod drive mechanisms (CRDMs) and replacement of all reserve shutdown system (RSS) material . In addition, the licensee provided a number of submittals in response to the restart issues addressed in the assessment report. These submittals are referenced in the Technical Evaluation Report (TER) (Reference 4), which is enclosed with this evaluation. The TER was prepared by Los Alamos National Laboratory under contract to the NRC. This evaluation has been performed by the Office of Nuclear Reactor Regulation in partial fulfillment of Item 1 of Technical Interface Agreement (TIA) 85-01 dated January 2, 1985. This TIA was formulated in response to a request from Region IV (Reference 5). The key issues covered in this evaluation are: - Control Rod Drive Mechanism Failures - CRDM Refurbishment Program - CRDM Temperature Recording and Requalification - CRDM Surveillance and Preventative Maintenance 2.0 Evaluation of Individual Issues Restart of the Fort St. Vrain Nuclear Generating Station is dependent on satisfactory evaluation of several major issues identified in the assessment report (Reference 1). The following evaluations are based on the enclosed TER (Reference 4). - 2 - 2.1 Control Rod Drive Failure Mechanisms (Reference 4 - Section 2.1) Failure of the control rods to insert on a scram signal has been attributed to two mechanisms. The first failure mechanism is the accumulation of debris in both the CRDM motor bearings and gear mechanism. This debris could eventually bind the mechanical drive and prevent control rod insertion. The second failure mechanism is the potential effect of temperature and moisture on the CRDMs. Loss of CRDM purge flow can result in the CRDM binding due to differential expansion effects as well as a change in properties of the molybdenum disulfide lubricant. Since the failure mechanism(s) has not been definitely identified, the licensee has proposed multiple measures in response to each failure mechanism. First, each of the 37 CRDMs will be completely refurbished prior to restart. Second, the temperature of the CRDMs will be monitored. Third, backup source of purge flow capable of two hours of operation will be installed. Fourth, the licensee will institute a program to qualify the CRDMs for a temperature of 300°F. We find that the measures proposed by the licensee address each of the proposed failure mechanisms. However, we conclude that further work must be done to assure that the CRDMs are operable over the remaining plant life. These long term issues will be discussed in Sections 2.3 and 2.4. We conclude the work to establish the CRDM failure mechanism is adequate and an acceptable basis for plant restart. 2.2 CRDM Refurbishment Program (Reference 4 - Section 2.2) In view of the mechanical deterioration of the CRDM motor bearings and gear trains, the licensee has elected to perform a complete refurbishment of the CRDMs. Additionally, since the licensee also found deterioration in the CRDM cables and the Reserve Shutdown System (RSS) material , he elected to replace these items with new materials. CRDM Refurbishment The licensee's program for refurbishment of the CRDMs involves inspection, testing, refurbishment or replacement of all major components. We find that the refurbishment process is thorough and represents an effort to restore the CRDMs as closely as possible to as-new conditions. Furthermore, the licensee has proposed a testing program to establish the acceptability of the CRDMs following refurbishment. This program involves a multiple step testing program, with eventual tests in the reactor core. Certain aspects of this testing procedure are still in development and cannot be reviewed at this time. We conclude that the licensee's mechanical refurbishment program is adequate and provides an acceptable basis to support plant restart. - 3 - However, we also conclude that an improved testing and surveillance program should be formulated to assure continued CRDM operability during plant operation. We have discussed these long term items in Sections 2.3 and 2.4. Other Refurbishment Items The licensee has elected to replace the control rod cables which were found to be failing from stress corrosion. The licensee has selected different materials for portions of this refurbishment to improve the resistance of these materials to future stress corrosion failures. We have reviewed the replacement materials proposed by the licensee and found them acceptable. We recommend continued investigation by the licensee of the sources of chlorine in the reactor and its potential effects on other reactor components. The licensee has also elected to completely replace the materials contained in the RSS hoppers. The replacement was in response to the failure of a hopper to discharge due to formation of boric acid crystals on the RSS material . The licensee selected a new material with improved properties to reduce the likelihood of future boric acid crystals. The licensee has also taken measures to reduce the possibility of moisture ingress into this system. We have reviewed the licensee's program for replacement of the control rod cables and the RSS materials and found it acceptable. 2.3 CRDM Temperature Recording and Requalification (Reference 4 - Section 2.3) The licensee has proposed to upgrade the CRDM temperature measuring systems to provide continuous records. He intends to monitor weekly the CRDM temperature, in all operating conditions. We have reviewed this monitoring program and find it is acceptable for steady state operation. However, we recommend it be supplemented with a transient monitoring program to provide an increased monitoring frequency during transient events that could lead to high CRDM temperatures. The licensee has proposed that the CRDMs are currently qualified to 272°F, and recommends plant operation with an administrative temperature limit of 250°F. In addition, the licensee is pursuing a program to requalify the CRDMs to temperatures of 300°F. Our evaluation of the CRDM qualification based on test data only supports an average operating temperature of 215°F. The licensee has indicated that the 215°F temperature limit would impact on plant operation. We conclude that pending the receipt of new qualification data, the temperature of the CRDMs be limited to 215°F as an administrative limit. In the event that 215°F is exceeded, continuous monitoring of the affected CRDM must be initiated and the results reported to the NRC on a monthly basis. The CRDMs, until requalified, should not be operated at higher than 250°F. We require that the licensee provide a commitment to operate within these temperature limits until further requalification test data is available or provide another method of assuring CRDM operability. These temperature limits cannot be changed - 4 - without NRC approval of new temperature limits or alternative methods of assuring CRDM operability. 2.4 CRDM Surveillance and Preventative Maintenance (Reference 4 - Section 2.4) The licensee has proposed a program of surveillance and preventative maintenance to assure continued operability of the CRDMs and RSS during plant operation. The surveillance program is an interim program, based mainly on weekly ten inch control rod drops. Our evaluation of this surveillance program is that it may not provide the detailed information needed to predict potential failures of the CRDMs to insert on a scram signal over the long term. The licensee is cognizant of the limitations of the current tests, and is developing more thorough surveillance tests. We conclude that the licensee's proposed interim surveillance procedures provide an acceptable basis for plant restart. However, we require the licensee to submit an improved (long term) surveillance program within six months of plant restart for NRC review. The licensee has also proposed a preventative maintenance program for the CRDMs and RSS. Certain of these components are normally removed from the reactor for refueling. At each refueling, the systems and their com- ponents would be thoroughly inspected and refurbished as needed. Additionally, as more sensitive surveillance tests are developed, surveillance data could be used to determine additional candidates for preventative maintenance activities. We conclude that the licensee's preventative maintenance program is an acceptable basis for plant restart. However, the licensee must provide information on the interaction between improved surveillance program (noted above) and the preventative maintenance activities. This information should be provided with the revised surveillance program. 3.0 Conclusions on Restart Issues We have evaluated the licensee's proposed programs to address the issues covered by this evaluation and related commitments made to the NRC (References 2 and 3). We find that these issues, as discussed in Section 2.0, have been adequately resolved to allow plant restart with the exception of the following: 1. The licensee must provide a commitment to operate the plant within the CRDM temperature limits accepted by the NRC. These limits cannot be changed without NRC approval of new temperature limits or alternative methods of assuring CRDM operability. 2. The licensee must provide a commitment to submit an improved CRDM surveillance and preventative maintenance program within six months of plant restart. - 5 - These open issues should be resolved prior to plant startup. 4.0 Long Term Issues Continued plant operation beyond one refueling cycle should be contingent on resolution of the longer term items identified in the Assessment Report of October 16, 1984 (Reference 1). We request that the licensee commit prior to restart to resolution of these long term items as outlined in the Assessment Report (Page viii). Enclosure: LANL TER Date: May 21, 1985 Principal Contributor: K. Heitner REFERENCES 1. TIA 85-01 dated January 2, 1985 2. Memorandum from J. R. Miller and E. Johnson to F. Miraglia dated February 5, 1985 3. Preliminary Report Relating to the Restart and Continued Operation of Fort St. Vrain Nuclear Generating Station, Docket No. 50-267, October 16, 1984 4. Evaluation of Control Rod Drive Mechanism and Reserve Shutdown System Failures and PCRV Tendon Degradation Issues Prior to Fort. St. Vrain Restart. LANL FIN No. A-7290, March 12, 1985 (Enclosed) Attachment 2 Evaluation of Control Rod Drive Mechanism and Reserve Shutdown System Failures, and PCRV Tendon Degradation Issues Prior to Fort St. Vrain Restart NEC Fin No. A-7290 March 12, 1985 Los Alamos National Laboratory • Deborah R. Bennett, Q-13 Gerald W. Fly, Q-13 L. Erik Fugelso, Q-13 Robert Reiswig, MST-6 Stan W. Moore, Q-13 Responsible NRC Individual and Division J. R. Miller/ORB3 Prepared for the U.S. Nuclear Regulatory Commission Washington, D.C. 20555 DISCLAIMER Tnis report was prepared as an account of work sponsored by an agency of the United States Government. Neither the United States Government nor any agency thereof, or any of their employees, makes any warranty, expressed or implied, or assumes any legal liability or responsibility for any third party's use, of any information, apparatus, product or process disclosed in this report or represents that its use by such third party would not infringe privately owned rights. - i - Table of Contents 1.0 Background 1.1 Assessment Report Restart Issues 1.2 PCRV Tendon Restart Issues 1.3 Purpose of the Technical Evaluation 2.0 Control Rod Drive and Orifice Assemblies 2.1 Failure Mechanisms 2.1.1 Motor Brake Malfunctions 2.1.2 Reduction Gear Mechanism Malfunctions 2.1.3 Motor and Motor Bearing Malfunctions 2.2 Refurbishment Program 2.2.1 CRDOA Refurbishment 2.2.2 Control Rod Cable Replacement 2.2.3 Reserve Shutdown System Material-Related Failure 2.2.4 Purge Flow and Seal Replacement 2.3 CRDM Temperature Recording and CRDM Requalification 2.4 CRDM Preventive/Predictive Maintenance and Surveillance 2.4.1 CRDM Preventive/Predictive Maintenance 2.4.2 CRDM Interim Operational Surveillance 3.0 Moisture Ingress Issues 4.0 PCRV Post-Tensioning Tendon System 4.1 Tendon Accessibility, Extent of Known Degradation and Failure Mechanism 4.2 Tendon Corrosion Corrective Measures 4.3 PCRV Tendon Interim Surveillance 4.4 PCRV Structural Calculations by Los Alamos National Laboratory 5.0 Conclusions 6.0 References - ii - Evaluation of Control Rod Drive Mechanism and Reserve Shutdown System Failures, and PCRV Tendon Degradation Issues Prior to Fort St. Vrain Restart 1.0 Background On June 23, 1984, following a moisture ingress event resulting in a loss of purge flow to the Control Rod Drive Mechanism (CRDM) cavities, 6 of 37 control rod pairs in the Fort St. Vrain (FSV) High Temperature Gas- Cooled reactor failed to insert on a scram signal. Subsequently, all six control rod pairs were successfully driven into the core. In July, 1984, an assessment team consisting of Nuclear Regulatory Commission (NRC) personnel from Headquarters, Region III and Region IV, and their technical consultant, Los Alamos National Laboratory, conducted an on-site review of the Control Rod Drive Mechanism failures, overall conduct of plant operations, adequacy of technical specifications and a review of the continued moisture ingress problem. An additional plant visit in August, 1984, reviewed CRDM instrumentation anomalies. 1.1 Assessment Report Restart Issues The results of both assessments were reported in the "Preliminary Report Related to the Restart and Continued Operation of Fort St. Vrain Nuclear Generating Station"1, in October, 1984. The report concluded that Fort St. Vrain should not oe restarted until modifications and/or other corrective actions had been taken, or until all control rod drive mechanisms had been inspected and refurbished to provide reasonable as- surance that the control rods would insert automatically on receipt of a scram signal. More specifically, and as included in this technical eval- uation, Reference 1 required Public Service Co. of Colorado (PSC) to com- plete the following, prior to restart: a. The licensee must identify the CRDM failure mechanism(s) and take corrective actions, or, if the mechansm(s) cannot be posi- tively identified, take other compensatory measures to provide assurance of control rod reliability, which could reasonably include refurbishment of all CRDMs. - 1 - b. The licensee must outline and commit to periodic inspection, preventive maintenance and surveillance programs for control rod drive mechanisms and associated position instrumentation. A change in the Technical Specifications shall be proposed to implement a weekly control rod exercise surveillance program for all partially or fully withdrawn control rods. A Limiting Condition for Operation should define control rod operability, and the minimum requirements for rod position indication. c. The licensee must functionally test one-20 weight % boron and one-40 weight % boron hopper from the Reserve Shutdown System (RSS), to assure the full availability of the RSS, prior to restart. The licensee must outline and commit to periodic in- spection, preventive maintenance and surveillance programs for Reserve Shutdown System material. A change in the Technical Specifications shall be proposed to implement the RSS surveil- lance program. A Limiting Condition for Operation should define and confirm the operability of the Reserve Shutdown System. d. The licensee should develop a proceaure requiring reactor shut- down when high levels of moisture exist in the primary coolant, or when CRDM purge flow is lost. e. The licensee should implement a procedure for recording repre- sentative samples of CRDM temperatures at all operating condi- tions, until continuous recording capability is available. f. The licensee should implement procedure to prevent overdriving the control rods past the "Rod-In" limit. g. The licensee must develop a plan to implement any modifications recommended by the PSC Moisture Ingress Committee that are determined, by PSC, to have a high potential for significantly ' reducing the severity and frequency of moisture ingress events. 1.2 PCRV Tendon Restart Issues As a result of previously identified tendon degradation in the Pre- stressed Concrete Reactor Vessel (PCRV) post-tensioning system, PSC must complete the following, as comfirmed by Reference 2, prior to restart: - 2 - a. The licensee should submit documentation evaluating the mechan- ism(s) causing corrosion on and failure of the PCRV tendon wires, and corrective measures to eliminate further tendon degradation, thereby assuring the continued structural integ- rity of the PCRV and its post-tensioning system. b. The licensee should propose and implement a tendon surveillance program that determines the extent of current tendon degrada- tion in the PCRV, and that systematically monitors the rate of tendon corrosion. 1.3 Purpose of tne Technical Evaluation This document provides a technical review of tne restart issues identified above, and the corrective measures and/or actions proposed by licensee, based on the licensee's January 31, 1985 submittals (References given as used in this document), and the meeting between the licensee and NRC at the FSV plant site on February 20-22, 19d5, as transcribed in References 3, 4 and 5. 2.0 Control Rod Drive and Orifice Assemblies This section includes a review of CHDM failure mechanisms, Control Rod Drive and Orifice Assemblies (CRDOA) refurbishment, CRDM temperature recording and requalification testing, CRDM preventive/predictive main- tenance and surveillance. 2.1 Failure Mechanisms The failures of control rod pairs to scram, under various operating conditions, has been documented since 1982,6,7 and are as noted in Table 1 by region, CRDOA number and CRDM purge flow subheader (total of 8 purge flow subheaders). - 3 - Table 1. Control Rod Failures Date 2/22/82 6/23/84 1/14/85 Region 7 28 6 7 10 14 25 28 28 31 32 CRDOA # 18 44 29 18 14 25 7 44 36 17 15 CRDM Purge 1 1 6 1 7 2 5 1 1 2 3 Subheader # High moisture content in the primary coolant and loss of purge flow were common modes during the 2/22/82 and 6/23/84 events. Substantial descriptions and operating characteristics of the drive motor, friction brake and dynamic braking, the reduction gear mechanism, the cable drum and cable, and the bearing lubricant are provided in Reference 6. The licensee reviewed those CRDM components that could have caused the fail- ures to scram, and postulated various failure mechanisms that could have interacted on each component, as described below. 2.1.1 Motor Brake Malfunctions During a scram, the motor brake is de-energized and released, thereby freeing the motor rotor shaft and gear train assembly to rotate under the torque applied by the weight of the control rods. In the motor brake assembly, failure of the scram contactor to de-energize dc power to the electromagnet was discounted because the operator had removed the brake fuses following the CRDM failures to insert the control rod pairs. According to the licensee, electromagnetic remanence and reduced spring constant in the brake spring plungers (due to elevated tempera- tures) were eliminated as possible failure mechanisms. Some corrosion and rust was identified on the brake disks of CRDOAs 25, 18 and 29. How- ever, the disks of a CRUM motor brake assembly with "discoloration and whatever surface variations"3' p.149, could not be made to stick in an elevated temperature helium environment with high moisture content (test T-228). The licensee concluded that the motor brake was not instrumental in the failures to scram. Los Alamos agrees with the licensee that the motor brake assembly was probably not related to the CRDM failures. - 4 - 2.1.2 Reduction Gear Mechanism Malfunctions The reduction gear train is driven by the motor rotor shaft, and rotates the cable drum with a gear ratio of 1150 between the motor and drum. The condition of the reduction gear mechanism was postulated by the licensee to potentially contribute to a failure to scram through gear tooth or bearing damage, by the presence of large particulate matter pre- venting gear rotation, and/or the presence of particulate matter in the gears or gear bearings reducing the gear train efficiency--i.e., the torque transmitted from the gear train to the motor rotor shaft might have been insufficient to overcome the friction of the motor bearings. The licensee stated that no major damage has been identified on sev- eral inspected reduction gear mechanisms, even though some wear and debris were observed. The licensee's analyses indicated that particulates with a size of 0.030 inches in diameter or greater, and with a comparable material composition as the reduction gear mechanism (implying comparable hardness), would be required to inhibit gear or gear bearing rotation. Analyses of CRDOA debris8 showed the presence of rust, molybdenum di- sulfide and traces of silicon particles, which are relatively soft mate- rials. The average particle of 0.020 inches was uniform in size, and tended to be smaller than that thought to inhibit rotation, even though rust particles on the order of 0.0625 to 0.125 inches were scraped off the ring gear pinion housing of CRDOA 18. However, the presence of debris in the gears and gear bearings tended to support the licensee's case of reduced gear train efficiency when sensitivity studies indicated that the motor bearings were only three times more sensitive to debris than the first pinion gear mesh of the reduction gear assembly, and 500 times more sensitive to debris than the cable drum bearings. Los Alamos agrees with the licensee that the presence of debris , especially in the first pinion gear mesh and the gear bearings, could reduce the efficiency of the reduction gear train, and thereby contribute to CRDM failures. 2.1.3 Motor and Motor Bearing Malfunctions During a scram, the motor is de-energized and does not directly con- tribute to the scram process, even though it operates as an induction generator. However, because 1b-20 inch-ounces of resisting torque on the - 5 - motor rotor shaft can forestall scram,9 the friction from the motor bearings can be a significant contributor to the failure to scram. Pos- sible contributions to increase the friction include debris in the bear- ing race, wear on the bearing ball or race, and changes in the lubricant properties during adverse conditions. The licensee reported that debris was observed in the bearing races of CRDOAs 7, 18 and 44, "roughness in rolling the bearing balls was noted in virtually all of the unrefurbished bearings examined",6 and minor race wear was identified. Reference 8 verified that the major debris constituents could be attributed to the motor bearing materials (which includes bearing balls, races, and other bearing components), whereas minor constituents were indicative of the motor itself. The analysis provided little evidence to support the theory that debris had been "washed" into the bearing races. The licensee also determined, because of the relatively close bearing tolerances and because rod weight alone might not produce sufficient "crushing force" to deform bearing particu- late, that bearing operation could be reduced with the presence of par- ticulate matter. Tne licensee therefore concluded that internally gener- ated wear byproducts in the CRDM motor bearings contributed significantly to the failures to scram. Los Alamos agrees with the licensee that increased friction in the motor bearings, caused by the presence of internally generated debris, could have been a likely contributor to the failures to scram. Los Alamos also agrees with the licensee that the "wash in" theory of debris into the motor bearing races is not supported. Los Alamos contends that the loss of CRDM purge flow allowed primary ' coolant with high moisture content to enter the CRDM cavity. An indepen- dent literature search indicates that the dry film lubricant, molybdenum disulfide, MoS2 , experiences an increase in its coefficient of fric- tion in the presence of moisture38. Therefore, the increased frictional coefficient of the lubricant on the motor bearings, MoS2, may have also contributed to the CRDM failures by resisting motor rotor shaft rotation. 2.2 Refurbishment Program The cause of the failures to scram could be attributed to several mechanisms such as reduced reduction gear train efficiency, internally - 6 - generated debris in the motor bearings causing increased friction on the motor rotor shaft, and possibly an increased frictional coefficient in the dry film lubricant in the presence of moisture. Because the CRDM failure mechanism cannot be specifically delineated, and because of CRDM cable failures, the licensee has undertaken a refurbishment program, in- volving the CRDM motors and reduction gear mechanisms, on all 37 CRDMs. The licensee reported that the CRDM refurbishment process and a testing program will ensure the ability of the control rods to scram under oper- ating conditions. In addition, the licensee has elected to replace the control rod cabling and other connecting hardware in light of recently identified stress corrosion problems, to replace the Reserve Shutdown System material due to the discovery of material "bridging" during hopper discharge, ane to install seals around certain penetrations into the CRDM cavity to mitigate the effects of primary coolant ingress by natural circulation. 2.2.1 CRDOA Refurbishment The licensee has proposed complete refurbishment of all Control Rod Drive and Orificing Assemblies to ensure that the CRDOAs will perform their intended safety functions, and to avoid potential operability prob- lems that could limit plant availability. As specified in Reference 10, the following major components are to be inspected, tested, refurbished or replaced, as necessary: 1. Control Rod Drive (200) Assembly--shim motor and brake assembly, bearings, reduction gears, limit switches/potentiometers. 2. Orifice Control Mechanism--orifice control motor, bearings, potentiometer, gears, drive shaft and nut, drive shaft housing. 3. Control rod clevis bolts. 4. Reserve Shutdown System--boron balls, rupture disks, DP switch. Design modifications include the replacement of control rod cables, cable end fittings, and cable clevis bolts, the installation of new purge seals into the CRDM cavity, and the installation of RTDs (Resistance Tem- perature Detectors) in all CRDOAs--the impact of tnese design changes will be evaluated later in this report. Each CRDOA will undergo the following series of scram tests in the refurbishment process6: a pre-refurbishment, in-core full scram test; a - 7 - pre-refurbishment full scram test in the Hot Service Facility (HSF); a scram test with refurbished reduction gear mechanism and unrefurbished shim motor, using dummy weights; a full scram test using a "standardized" motor, using dummy weights; a scram test with completely refurbished 200 assembly, using dummy weights; a post-refurbishment, full scram test in the HSF; and finally, a post-refurbishment, full in-core scram test. As designated by the licensee in Reference 6, back-EMF voltage meas- urements from the shim motor will be taken for the series of scram tests conducted before, during and after refurbishment, and should define the CRDM operating characteristics. From the back-EMF voltage measurements, the licensee states that they can generate the following information-- voltage versus time, frequency versus time, voltage versus frequency, acceleration versus time, torque versus time, peak angular velocity, time to peak back-EMF and angular velocity, average torque on motor rotor dur- ing acceleration to peak velocity, maximum torque on motor rotor each 10 second interval, maximum deviation of torque values each 10 second inter- val, and gear train efficiency. The licensee has proposed a CRDOA refurbishment acceptance criterion, taking into account the results of the back-EMF voltage measurements and the resulting calculations of acceleration and torque such that6: 1. Tne minimum calculated average torque during acceleration to peak velocity will be 17.0 inch-ounces; this value corresponds to an average acceleration to peak velocity of 98.83 radians/ second2. 2. The maximum torque calculated during "steady-state" will be 7.0 inch-ounces. According to the licensee, final acceptance of a refurbished CRDOA will be based upon the results of its in-core full scram test. Los Alamos agrees with the mechanical refurbishment of all CRDOAs, as the program is currently being implemented by the licensee. In par- 3, pp. 174-75 titular, the replacement of shim motor bearings is con- sidered essential to the refurbishment process. However, the current program of mechanical refurbishment alone cannot ensure CRDOA operability. - 8 - From the documentation presented by the licensee and reviewed earlier in this section, Los Alamos believes that the proposed back-EMF testing and acceptance criteria have potential in providing a data base from which control rod operability might be determined. But, an element of uncer- tainty, as to CRDOA operability based on back-EMF testing, is introduced because the test method and interpretation of its results are still in the developmental stages, and because in-core full scram testing of re- furbished CRDOAs has not yet taken place. Los Alamos recommends that the back-EMF testing method continue to be developed, that the further collection of back-EMF information be used in preparing a statistical data base for possibly defining CRDOA opera- bility, and that more attention be paid to the initial, start-up scram characteristics of the CRDOA, in developing a better understanding of break-away torque effects. In line with Region IV's increased inspection of the refurbishment process, we suggest a review, by Region IV, of all testing results pertaining to CRDOA refurbishment acceptability, after in-core testing is complete, but prior to startup. As an additional method to ensure CRDOA operability during scram, a procedure requiring control rod run-in is recommended. As a post-startup item, Los Alamos recommends that a final determina- tion be made as to the suitability and acceptability of back-EMF testing in defining CRDOA operability. 2.2.2 Control Rod Cable Replacement In September, 1984, the control rod cable on CRDOA 25 was severed in several places during an investigation of a slack cable indication.11 A subsequent metallurgical examination12 of the austenitic 347 stainless steel cable indicated that the cable surface was pitted and cracked, that the delta-like material cracks were typical of stress corrosion cracks, and that the fracture surfaces were brittle in nature. Further investi- gation revealed that the 347 SS cable material was susceptible to stress corrosion when under the existing stressed conditions, and in the presence of chlorides and moisture. Tne potential sources of the chlorides in the primary coolant con- tributing to the chloride stress corrosion are reviewed in Reference 13. The licensee states that the chlorine occurs as two different species--HC1 - 9 - gas and a salt; the sources of the gas species include the fuel rods , H-327/H-451 graphite, PGX/HLM graphite and the Ti sponge, whereas the sources of the salt species include the ceramic insulation, concrete and water, all to varying degrees. As part of tne overall CRDOA refurbishment program, the licensee elected to replace the control rod cable witn Inconel 625, which is con- sidered resistant to chloride stress corrosion, and has increased strength and fatigue properties over the former 347 SS. Cable components and con- necting hardware that were made from materials susceptible to stress cor- rosion, and are being replaced with materials more resistant to stress corrosion include: Component Material 1. Cable and rod portion Inconel 625--high strength of the ball end and resistance to oxidation 2. Anchor, set screw Martensitic steel-high strength, ability to be nitrided, resistance to oxidation 3. Spring, connecting bolt Inconel X-750--high yield strength, resistance to oxidation. Drawing numbers and material information are available in Reference 12. A safety analysis of tne material changes in tne reactor control rod drive and orificing assembly, which are classified as Class I, Safety Related • and Safe Shutdown components, is included in Reference 14. Los Alamos metallurgical analyses on a sample of the corroded control rod cable15 also indicate pitting on the cable surface, ductile and brittle fracture surfaces, and to a lesser degree than the licensee , cracking indicative of stress corrosion cracking. Qualitative measure- ments confirm the presence of chlorine on fracture surfaces. Therefore, Los Alamos agrees tnat chloride stress corrosion contributed to the de- graded condition of the control rod cable. The Los Alamos analysis also observed that a certain particle removed from between the individual cable strands of the Los Alamos sample had a "shaved" appearance, and was - 10 - identified as a 7000 series aluminum alloy--the licensee noted that the control rod cable drum is constructed of 7075 aluminum alloy4; p.20, and that no excessive drum wear had been noted. Los Alamos agrees that the licensee's recommended material changes tend to improve the overall resistance of the CHDOA cable components and connecting hardware to chloride stress corrosion. However, Los Alamos also recommends a continued analysis into the sources of the chlorine and its effects on other reactor components, especially components potentially subjected to high chlorine concentrations such as the bottom plenum or other areas where water could accumulate. 2.2.3 Reserve Shutdown System Material-Related Failure In November, 1984, during the required testing of a 20 weight $ boron and a 40 weight % boron hopper in the Reserve Shutdown System, only half of the RSS material in CRDOA 21 (40 weight $ boron) was discharged. The licensee's examination of the undischarged material revealed that the 4C boronated graphite balls had "bridged" together through a crystal- line structure on the ball surfaces. Analyses on the crystalline material indicated that it was boric acid.16 The formation of the boric acid crystals was caused by moisture reacting with residual boric oxide in the RSS material. It was concluded that the moisture had entered the RSS hopper through the CRDOA vent/purge line by "breathing", and/or by water contamination in the helium purge line. In Reference 16, the licensee proposed a threefold corrective action to the RSS material problems. First, new RSS material, manufactured by Advanced Refractory Technologies (ART) in late 1984 and early 1985, has an order of magnitude less residual boric oxide in the B4C material, and will be installed in all RSS hoppers as part of the overall CRDOA refurbishment program. No effort will be mane to use ART blended RSS material currently in stores4' p.32 unless NRC is notified. Second, an expanded HSS material surveillance program, which will be incorporated into the Technical Specification, will test one 20 weight % boron hopper and one 40 weight $ boron hopper during each refueling outage, and will include visual examinations for boric acid crystal formations, chemical analyses of RSS material for boron carbide and leachable boron oxide con- tent. Third, efforts will be mace to mitigate or eliminate the ingress - 11 - of moisture into the RSS hoppers by installing a knock-out pot, moisture elements, and a back-up helium source for the main CRDOA purge and Reserve Snutdown System purge lines.17 Each knock-out pot will be equipped with a sight glass and a high level alarm in the Control Room. Los Alamos concurs that the crystalline structures on the surface of the B4C RSS balls is meta-boric acid,18 most probably formed by mois- ture reacting with leachable boric oxide in the B4C material. In light of the new RSS material to be used, the increased surveillance efforts , and measures to mitigate the ingress of moisture in the ASS hoppers, Los Alamos agrees that the refurbished RSS should be able to reliably perform its function. 2.2.4 Purge Flow and Seal Replacement Just prior to the June 23, 1984 event when 6 of 37 control rod pairs failed to insert on a scram signal, a high moisture content in the primary coolant resulted in the loss of purge flow into the CRDM cavities. The loss of purge flow may have allowed the additional ingress of moist pri- mary coolant into the CRDM cavities, resulting in mechanisms that may have contributed to the CRDM failures. Because the exact CRDM failure mechanism has not been determined, and to alleviate the possibility of purge flow loss and/or hign moisture content in the primary coolant con- tributing to future CRDM failures, the licensee has proposed several cor- rective measures19 as part of the overall CRDOA refurbishment program. To provide an accurate measure of the purge flow into the CRDM cavi- ties, the licensee has proposed tnat new flow indicators with a range of 0-20 scfm be installed on each helium purge line, providing local indica- tion, remote indication in the Control Room, and an alarm in the Control Room to indicate low flow conditions.20 A minimum of 8 bypass lines (one line serviced by each of tne 8 purge flow subheaders) will be in- stalled prior to restart. Tne licensee intends to install the flow instrumentation4, pg 3-4 on tnese subheaders when the devices are available. As mentioned in section 2.2.3, to reduce tne possibility of moisture ingress into tne CRDM cavities via the helium purge lines, the licensee will install a knock-out pot, moisture elements and a back-up helium source for the main CRDOA purge and RSS purge lines, prior to criticality - 12 - following the fourth refueling outage4, pg 5. The knock-out pots will be equipped with a sight glass and a high level alarm in the Control Room. The helium trailer, which will act as the back-up source of dry helium for purge, can provide helium at a rate of 7.4 acfm (4.5 lbms/hr per pen- etration at 700 psig) for approximately 2 hours.i7 To mitigate the ingress of primary coolant, which could contain moisture, into the CRDM cavity, seals will be installed on four large flow passages into tne CRDM cavity--the two passages in the reserve shut- down tube holes, and the two passages over the eye bolts that penetrate the floor of tne CRDM cavity.21 Cover plates with integral gaskets will also be installed on the four access openings on the lower CRDM housing. Thermal and mechanical analyses22 have determined that the seal additions will not interfere with the R6S performance under the in- fluence of mechanical , thermal or seismic loadings. Tne flow calculations in Reference 22 conclude tnat addition of tne mechanical seals to the RbS pressure tubes and the lifting eyebolts will reduce naturally convective ingress of primary coolant into the CRDM cavity from a flow rate of 0.68 acfm to less than 0.006 acfm. Additional calculations have confirmed that the seals are able to withstand both a design basis slow depressuri- zation transient and a design basis rapid depressurization transient. The licensee has proposed a procedure in Reference 23 tnat basically requires reactor shutdown in the event CRDM purge flow is lost, or if high moisture content is present in tne primary coolant. Los Alamos agrees with tne efforts of tne licensee in monitoring the flow and moisture content of tne helium purge into the CRDM cavities, in restricting the ingress of moisture into the CRD cavities via the purge • lines, and in providing a back-up source of helium in case of purge flow loss. From the review of the provided documentation, Los Alamos agrees that the addition of seals and coverplates with integral gaskets will indeed mitigate the ingress of primary coolant and moisture into tne CRD cavities through penetrations. In addition, Los Alamos believes tnat the procedure requiring reactor shutdown with loss of purge flow or high moisture levels in the primary coolant fulfills the requirements of the assessment report1. Tne licensee defines "high moisture levels" in Reference 23. - 13 - 2.3 CRDM Temperature Recording and CRDM Requalification Tne lack of direct measurements of CROM temperatures curing tne June 23 event, and during steady state and other transient operating condi- tions, has prompted the installation of Rills to monitor tne CRDM cavity closure plate (ambient) , orifice valve motor plate and control roc drive motor temperatures. Strip chart recorders will continuously record tne tnree temperatures for each CRDM,24 and will provide a CALM operating temperature data base. The old data collection surveillance procedure25 4, p.60 will be modified to collect data on a continuous basis. Tne licensee intends to install the permanent recorders prior to restart.4, p'57 The licensee postulates in Reference 26 that "the maximum temperature rating of tne Drive mecnanism which might innibit the scram function is 272°F", and in monitoring CRDM temperatures "the maximum temperature rating of 272°F should not be exceeded curing power operation". Tne licensee has also proposed a CRDOA requalification testing pro- gram that is designee to establisn a temperature at which the CRDOA is qualified for operation.27 The helium test environment will be operated at 2o0°F, 260°F, 270°F, 2o0°F, 290°F and 300°F with a goal of qualifying all CRDOA components for 300°F operation. Results of tne requalification testing are anticipated by tne end of 1965. Los Alamos agrees tnat the placement of CRDOA tnermocouples, and tne continuous data monitoring at all operating conditions is sufficient to provide a CRDOA temperature data base during steady state and transient operating conditions. In addition, Los Alamos believes tnat the CRDOA is currently only qualified to operate up to 215°F based on the original mechanical CRDOA qualification tests, an NRC recommendation,28 and previous Los Alamos calculations.29 Tne licensee's argument tnat the CRDOA is qualified for 272°F operation based on analytical calculations4, p.49 is not sub- stantiated. Tnerefore, Los Alamos recommends tnat CRDOA operation be limited to 215°F until mechanical requalification supports a higher oper- ating temperature. - 14 - 2.4 CRDM Surveillance and Preventive/Predictive Maintenance The licensee has proposed a set of preventive/predictive mainte- nance tests and surveillance inspection procedures that are intended to monitor the performance of the CHDOAs and to determine the overall opera- bility of the CRDOAs during reactor operation. Initial development of these operating tests are considered part of the CRDOA refuroishment pro- gram, and will utilize the data base and resultant trends formulated dur- ing refurbishment. 2.4.1 CR111 Preventive/Predictive Maintenance Tne licensee's CRDOA preventive/predictive maintenance program is proposed in Reference 30. According to the licensee, the normal preven- tive maintenance (PM) program will be implemented on a refueling basis rotational cycle for CRDOAs that would normally be removed for refueling, unless the predictive maintenance (PDM) program indicates tne need for more frequent maintenance. The PM program would emphasize the mecnanical examination and refurbishment of the shim motor/brake assembly, the drive train, control rod cable, reserve shutdown system, position potentiom- eters, limit switches, orifice drive motor assembly, orifice drive lead screw, assorted seals, ;valves, electrical components, bolts and the ab- sorber string. On the other hand, the predictive maintenance techniques would be used to monitor the most important aspect of CRDOA performance--the "scram capability"--by determining the shim motor/brake and gear train performance. The tests proposed in the PDM program include wattage requirements, back-EMF voltages, delivered torque at tne motors, scram times, rod drop rates and torques to rotate motor/brake assemblies. Cer- tain aspects of the PUM program would be implemented on a weekly basis to determine scram capability and temperature performance during power oper- ation. The licensee has also proposed that testing information be acquired during reactor shutdown for trending purposes. Los Alamos concurs with the proposed preventive maintenance pro- gram as outlined by the licensee, on the assumption that data acquired during reactor operation will show that predictive maintenance tecnniques can be used to detect a reduction in CRDOA performance. Tne PDM testing tecnniques are closely linked to tne techniques that are being used for - 15 - the acceptance criteria in the refurbisnment program, and will tnerefore be dependent on the suitability and acceptability of back-EMF testing for determining CRDOA operability, as discussed in section 2.2.1. 2.4.2 CRDM Interim Operational Surveillance Tne licensee's CRDOA interim surveillance program is proposed in Reference 31. The surveillance tests are scheduled on a weekly basis, using a 10" rod drop method on all withdrawn and partially inserted con- trol rods, except the regulating rod.s' pg 82 The surveillance tests will obtain data for analysis and long term trending, exercise the rod, test selected circuitry, verify FSAR (Final Safety Analysis Report)9 assumes scram times, and confirm control rod operaoility. In addition, CitDOA temperature and purge flow information will be collected. For a fully withdrawn rod, analog and digital position information will be obtained, "Rod-Out" lights will be verified on, "Rod-In" aria "Slack Cable" lights will oe verified off, and the rod will be dropped approximately 10" by de-energizing the brake, while back-EMF data are obtainea for future trending. The "Rod-Out" light indication will be verified off, and analog and digital information will be compared, with an acceptable deviation of 10 inches between position indications. The rod will then be withdrawn to the full out position, so that analog ana digital positions can again be ootained. Control rods that are par- tially or fully inserted will undergo variations of this method. Quarterly surveillance tests are intended to supplement weekly sur- veillance information, and to verify redundancy of selected control roe position limit switches. Refueling snutdown surveillance will acquire the same information as the weekly and quarterly tests, except full stroke insertion tests will be performed. The operaoility acceptance criteria, according to the licensee, will be based on distance and time rod drop data used to calculate a conserva- tive average full lengt❑ scram time. A CRDOA will be considered inoper- able if it does not meet the maximum scram time of 160 seconds as defined in the FSAR9. Such an indication would warrant back-EMF testing in confirming scram operability. Los Alamos agrees that the basic surveillance methodology is suffi- cient to exercise the control rod, verify FSAR scram times, and to test - lo - selected circuitry. However, references to 272°F as the maximum CRDOA operating temperature are still considered inappropriate as discussed in section 2.3, and a 10 inch deviation is not considered acceptable between digital and analog position indications--such a deviation coulo inadver- tently lead to control rod overdrive through a misinterpretation of rod position. Also, the back-EMF testing methods and interpretation of the results are still in the developmental stages, and an engineering deter- mination of tne suitability and acceptability of tnis testing metnod in determining continued CRDOA operability will need to be mace before the licensee can finalize tnis portion of tne surveillance program. 3.0 Moisture Ingress Issues The licensee nas submitted32 a listing of the issues considered, and actions taken, by the FSV Improvement Committee (formerly the FSV Moisture Ingress Committee) in significantly reducing the frequency and severity of moisture ingress events. The issues were divided into four categories: 1 . Issues currently under consideration by tne Fort St. Vrain Im- provement Committee. 2. Circulator Auxiliary System modifications yet to be completed prior to startup. 3. Circulator Auxiliary System modifications to be completed prior to startup, provided material availability and schedule permits. 4. Items identified by tne Moisture Ingress Committee wnicn are installed and operational. Los Alamos believes that a listing of intended and installed modifi- cations does not provide any indication as to what any given modification really is, wny tney contribute to tne reduction in potential for moisture ingress events, nor which improvements will substantially reduce the severity and frequency of moisture ingress events. Tne licensee nas com- mitted to submit a more explanatory version of tne actions to mitigate moisture ingress, prior to restart 4, pg 80 4.0 PCRV Post-Tensioning Tendon System In tne spring of 1984, during scheduled PCRV tendon surveillance, tendons with corroded and broken wires were found. Since that time, the - 17 - licensee has evaluated the corrosion mechanism, has performed lift-off tests on selected tendons to determine their load-carrying capability, and proposed corrective actions and an increased surveillance procedures. 4.1 Tendon Accessibility, Extent of Known Degradation ana Failure Mechanism The licensee, in determining the extent of tendon corrosion in the PCRV, determined what fraction of the tendons were available for visual examination ana lift-off tests. Tne tendon system is subdivided into four major groups: the 90 longitudinal (vertical) tendons have 169 wires per tendon; the 210 circumferential tendons in tne PI:tV sidewail have 1)2 wires per tendon, and the 50 circumferential tendons in botn the top ano bottom heaas have 169 wires per tendon; tne 24 bottom cross-heap tendons, and 24 top cross-nead tendons nave 169 wires per tendon. Of the four groups, the licensee states the following accessibility33: Tendon Group Both Ends Acces. One End Acces. Neither End Acces. Longitudinal Visual 20 69 1 Lift-off 0 74 16 Circumferential Visual 2d1 2 Lift-off 236 62 12 Bottom cross-heau Visual 20 4 0 Lift-off lb 4 4 Top cross-neaa Visual 17 7 0 Lift-off 16 6 2 Tne number of tendons witn Known broken wires as identifies in tue licensee's 1904 surveillance,34 include° 10 longitudinal tendons with 1 to 22 broken wires, 2 circumferential tendons witn 2 and 15 broken wires, 8 bottom cross-Head tendons with 1 to 19 broken wires, and no top cross- head tendons with broken wires. In some cases, tne total number of cor- rodeo, broken wires include wires broken during lift-off tests, or during retensioning. - 18 - The results of 74 longitudinal lift-off tests35 indicated that tendons with identified broken wires generally had a slightly smaller lift-off value than intact tendons. Thirty lift-off tests on circumfer- ential tendons snowed little change in lift-off value. Some of the fif- teen bottom cross-head tendon lift-off tests showed a definite reduction in lift-off value for tendons with multiple wire breaks. The value of tne lift-off test on one top cross-head tendon was nominal. All lift-off test values exceedeu tne minimum limits. The licensee conducted metallurgical investigations into the cause of the corrosion, and determined that microbiological attack on tne tendon NO-OX-IL CM organic grease caused the formation of formic and acetic acids,34,36 Tne acids, in conjunction with moisture in tne tendon tube, vaporized and recondensed on the cooler portions of the tendons--in this case, usually toward the tendon ends. Tne acidic attack resulted in re- duced cross-sectional wire area, stress corrosion cracking, localized tensile overload and wire breakage. Los Alamos believes, based on the documentation presented by the licensee, that microbiological attack of the tendon grease and the resul- tant formation of acetic and formic acids, in the presence of moisture, is a probable cause for tne currently observed tendon corrosion, and has led to the subsequent wire breakage through tensile overload. However, Los Alamos believes that the extent of known tendon corrosion, breakage and previous suryeillance have not been clearly defined by the licensee. Los Alamos therefore recommends that a complete map oe mace that lists each tenoon, its visual examinations and lift-off values, and tne number ano location of corroded and broken wires. An indication of tne degree of wire corrosion would also be desiraole. 4.2 Tendon Corrosion Corrective Measures The licensee evaluated several methods for arresting the corrosion process,34,36 including the use of ozone as a biocide to kill the micro- organisms, the use of an alkaline grease which should not be conducive to microbiological growth, and the use of an inert blanket consisting of nitrogen gas. The licensee's consultants found that the nitrogen atmo- sphere arrested the growth of the microbes in the NO-OX-ID CM organic - 19 - grease34,35, and eliminated the oxygen wnich is necessary for the cor- rosion process to continue. based on these results, and as a snort term action, the licensee has proposed that nitrogen blankets be establisneu on the longitudinal and bottom cross-Head tendons. Long term actions would include further investigations into tne corrosion process and ar- resting techniques, and the possible installation of additional load cells in monitoring the PCRV behavior. Los Alamos believes that the use of a nitrogen blanket to halt tne corrosion process may be suitable, but difficult to implement as proposed. The tendon tubes are not likely to oe leaktight, and maintaining an inert gas atmosphere at a set over-pressure may prove difficult. Consideration might be given to maintaining an intermittent or continuous purge flow through the tendon tubes, as needed, rather than to maintaining a speci- fied overpressure. However, Los Alamos recommends that initially the nitrogen be purged through the individual tendon tubes to remove as much moisture as possible, and tnat gas samples be used to monitor moisture and oxygen reduction. Further investigation into the long term effects of a nitrogen blanket on tendons, tne corrosion process and currently available corrosion acids are also recommended. 4.3 PCRV Tenoon Interim Surveillance Because the total extent of tendon corrosion in tne PCRV is unknown, because the rate of existing corrosion is unknown, and because tne use of a nitrogen blanket as an arrest to tne corrosion process is an unknown, the licensee has proposed an interim surveillance program designed to address eacn of tnese issues.5, pp•164-7• The interim tendon surveil- lance program would include increased visual and lift-off surveillance for three years, or until effective corrosion control Has been estab- lished. Two populations of tendons would be inspected--a population of tendons that have not been previously identifies as being corroded, and a control population with known corrosion. On a six-month frequency, visual surveillance of both tendon ends, when accessible, would include: - 2U - Tendon Group No. of New Tendons No. of Control Tendons Longitudinal 24 6 Circumferential 13 3 Bottom cross-head 6 2 Top cross-head 1 1 • Lift-off tests would be performed on two frequencies--an to month frequency for the population of new tendons, and a b month frequency for the control population. The number of tendons for lift-off will include: Tendon Group No. of New Tendons No. of Control Tendons Longitudinal 12 3 Circumferential 13 3 Bottom cross-head • 3 1 Top cross-head 1 1 As an acceptance criteria, tne licensee proposed that, based on vis- ual examinations, a mandatory engineering evaluation be conducted on any tendon tnat has 20% of its wires broken. For any tendon that has only one accessible end, the mandatory engineering evaluation will be con- ducted when any tendon has ioy of its wires broken. The control tendon population will include those tendons with tne worst known corrosion with ready accessibility. Los Alamos agrees that the increased tendon surveillance program of the nature proposed by the licensee will provide more information on the extent of corrosion in the PCRV by inspecting new tendons each surveil- lance, and at the same time, monitor the rate of corrosion with the con- trol tendon population. Tne increased surveillance should also determine the effectiveness of the nitrogen blanket in arresting corrosion, or any other corrective measure the licensee may propose. Los Alamos recommends that the licensee submit an outline of the intended mandatory engineering evaluation, which should include all lift-off, load cell and relaxation data incorporated into a safety evaluation. The licensee should define the extent of tne visual and lift-off testing procedures, ano could use US/NRC Regulatory Guide 1.3537 for guidance. - 21 - 4.4 PCRV Structural Calculations by Los Alamos National Laboratory The PCRV tendons are intended to apply sufficient compression in tne concrete to balance or exceed the circumferential and vertical tension in the concrete tnat results from the internal pressure. A combined analyt- ical and numerical study39 was undertaken by Los Alamos National Laboratory to evaluate the evolution of these stresses, both to the ini- tial prestressing and to subsequent partial and total rupture of tnese tendons. At the stress levels anticipated in the concrete, and for the anticipated operating life span of tne PCRV, tne concrete benavior was modeled as a linear viscoelastic solid with tne creep strain varying pro- portionally witn tne logarithm of time at constant stress tnroubnout the projected reactor lifetime. A one-dimensional model of a long concrete column of rectangular cross-section, with an embedded prestressing tendon along the length, was used to evaluate the concrete and steel stresses as well as tne nold-down and lift-off forces. Tnese were evaluated for the intact tendons and the degraded tendons. Tne degree of tendon degradation is described tnrough the ratio of tne number of unbroken strands to the original numoer of strands. Initial time of rupture was varied from the time of initial prestressing to 400 days after emplacement. Tne formulation led to an integral equation, which was solved numerically. The hold-down forces decayed approximately with tne logarithm of time and for both the extreme observed degradation (21 broken strands) and for a more extreme case (40 broken strands) , the hold-down force still exceeded the minimum safety design requirements. In addition, several finite element calculations, using the finite element code hDNSAP-C, were made to evaluate complete tendon failure in a 60° sector of the Fort St. Vrain PCRV. This code has an extensive material library of constitutive relations to model the various properties of concrete, together with a specialized element model to simulate pre- stressing tendons. Two rows of vertical and an arc row of circumferential tendons were incorporated in the model as a baseline calculation. Tne tendons were prestressed to 705, of the ultimate and an internal pressure of 775 psi was applied (this pressure is tne internal pressure of tne nelium coolant in the HTGR) and the creep of the concrete and slow decay of tne tenuon stresses were evaluated out to 30,000 days. Then, three - 22 - cases wherein one tendon was removed at one day were evaluated. First the middle vertical tendon in the outer row and in line with the outer buttress was removed. Second, an inner vertical tendon opposite the thinnest portion of the PCRV wall was removed. Finally, an inner layer circumferential tendon at micheight was removed. Stress redistributions at 300 days after ruptures were calculated and snifts of the remaining tendon loads to accommodate the broken tendon were calculated. Regions of local tensile and snear stress in the concrete portion of tne PCRV were identified and related to overall structural integrity. With all tendons present, the mean vertical stress was about -7b0 psi , the radial stress decreased from the applied internal pressure of -705 to about -1200 psi at the ring of circumferential tendons and the tangential stress ranged from -2400 psi at tne inner wall to about -2200 psi at the same place. Removal of a vertical tendon reduced the mean axial stress by about +40 psi, the local tangential stress by -10 psi and did not materially affect the radial stress. Removal of a circumferential tendon reduced the mean tangential stress by +30 psi and the local axial stress by -80 psi. The vertical hold-down force from zero days tnrough 30,000 days decreased linearly and remained above the prescribed safety limit, as did the circumferential hold-down force. Comparison of the analytical solution and a small finite element problem simulating the analytical proolem was made to verify the visco- elastic creep models and the tendon element in the NONSAP-C code. Excel- lent agreement for stresses, strains and nold-down forces was ootained. 5.0 Conclusions Los Alamos concludes that tne licensee, Public Service Co. of Colorado, has made a conscientious effort to address all of the restart issues listed in the assessment report.1 The refurbishment program on all CRLOAs provides confidence in CRDOA operability during reactor opera- tion and the ability to scram, even if the exact "failure to scram" mecn- anism has not been defined. Questions concerning the reliability of the back-EMF testing procedure on the shim motor/brake assembly in determining control rod operational acceptability still exist, but further method development, more experience with result interpretation, and in-core - 23 - testing may alleviate the questions. Until CRDOA operability can defin- itely be ascertained with these methods, we recommend that the licensee have backup measures such as rod run-in following scram. Control rod caole and connecting hardware material replacement, along with replacement of tne Reserve Shutdown System material, serve to rectify the material problems brought on by corrosive mechanisms. ' In light of chloride stress corrosion problems, Los Alamos also recommends that all reactor components exposed to the primary coolant be reviewed for susceptibility to chloride attack, especially the PCRV liner. Review should continue into the source of cnlorine and methods to elimi- nate its generation and presence. The effects of purge flow loss have not been determined to be in- strumental in CRDOA failures to scram, yet the licensee has committed to maintaining purge flow by external means, and to reducing the effects of primary coolant naturally convecting into the CRDOA cavity with extra seal installation. Even though current qualified CRDOA operating temperatures are very much in question, the licensee is in the process of requalifing the mech- anism for temperatures more in line with those anticipated during reactor operation. From a mechanical standpoint, CRDOA preventive/predictive mainte- nance procedures are certainly reasonable, but like the proposed surveil- lance program, they are dependent on back-EMF testing methods wnicn are still in the developmental stages. Evaluation of moisture ingress corrective measures was difficult due to the lacx of information with which to understand the measures taken. The licensee has committed to submit a more explanatory version of the actions to mitigate moisture ingress prior to restart. Tne extent of PCHV tendon degradation is not well known, even if tne licensee may nave determined the cause of the corrosion. Further investi- gation into arresting measures is definitely required, especially because the nitrogen blanket technique may be so difficult to employ. However, the interim surveillance program should provide information on the degree and rate of corrosion, in addition to establishing a tendon wire loss acceptance criteria. The tendon acceptance criteria should ensure PCHV margins to safety. - 24 - b.0 References 1 . "Preliminary Report Related to the Restart and Continue° Operation of Fort St. Vrain Nuclear Generating Station," Docket No. 50-267, Public Service Co. of Colorado, October, 1964. 2. "Review of Dallas Meeting (1/15/85) and Restart Committments", letter from Martin, NRC/Reg IV, to Lee, PSC, 1/17/85. 3. "Fort St. Vrain Meeting, NRC-PSC, February 20, 1985," Volumes I, II and III, recorded and transcribed by Koenig & Patterson, Inc. 4. "Fort St. Vrain Meeting, NHC-PSC, February 21, 1985," Volumes I and II, recorded and transcribed by Koenig & Patterson, Inc. 5. "Fort St. Vrain Meeting, NRC-PSC, February 22, 1985," Volumes I and II, recorder and transcribed by Koenig & Patterson, Inc. 6. "Engineering Report on CRDOA Failures to Scram-Control Rod Drive and Orifice Assemblies," PSC submittal P-85037, 1/31/85. 7. "Failure of Three CRDOAs to SCRAM," PSC submittal P-85029, 1/28/85. 8. "bearing Deoris Analysis," PSC submittal P-85017, 1/18/85. 9. "Fort St. Vrain Nuclear Generating Station, Updated Final Safety Analysis heport," Public Service Co. of Colorado. 10. "CRDOA Refurbishment Program Report," PSC submittal P-85040-2, 1/31/85. 11. "Control Rod Drive Cable Replacement," PSC submittal P-85032-2, 1/20/85. 12. "Control Roo Drive Cable Replacement Report," GA Technologies Document 907622, Attachment 1 to PSC submittal P-85032-2, 1/31/85. 13. "Investigations into Sources of Chloride in FSV Primary Circuit," PSC submittal P-66036, 1/31/85. 14. "Safety Analysis Report--Change in Material of the FSC Control Rod and Orifice Assemblies," Attachment 2 to PSC submittal P-85032-2, 1/31/85. 15. "FSV Control Rod Cable Metallurgical Examinations," draft report from Los Alamos National Laboratory, 3/85. 16. "Report on Reserve Shutdown Absorber Material," PSC submittal P-85027, 1/28/85. 17. "Moisture Control in CRDOA Purge Lines," PSC submittal P-85032-9, 1/20/85. - 25 - 18. "FSV Reserve Snutdown System Material Metallurgical Examinations, " draft report from Los Alamos National Laooratory, 3/05. 19. "CRDOA Moisture/Purge Flow," PSC submittal P-85032-6, 1/20/85. 20. "Modifications to CNDUA Helium Purge Supply," PSC suomittal P-85032-8, 1/20/85. 21. "Control Rod Drive Cavity Seals," PSC submittal P-85032-7, 1/20/85. 22. "FSV CRD Cavity Seals Design Report," GA Tecnnologies Document 907604, Attachment 1 to PSC submittal P-05032-7, 1/20/85. 23. "Operations Order No. 84-17 Describing Operator Actions Upon a Loss of Purge Flow and or Detection of Hign Moisture Levels in Primary Coolant," PSC submittal P-85040-8, 1/31/85. 24. "CND Temperature and Helium Purge Flow Recorders," PSC submittal P-85032-3, 1/20/65. 25. "Current CRD Temperature Data Collection Procedure Wnich Requires Station Manager Notification Upon Discovery of a Measured CRD Temperature in Excess of 25u°F," PSC submittal P-85040-9, 1/31/85. 2b. "Control hod System Operaoility Evaluation Report," PSC suomittal P-85040-1, 1/31/85. 27. "CNDUA Mechanism Temperatures Environmental Requalification," PsC submittal P-85032-1, 1/20/85. 2d. Letter from Robert A. Clark, Cnief, ORu3, to 0. R. Lee, PSCo. , December 2, 1982. 29. Meier, K. , "Fort St. Vrain Reactor Control Rod Drive Mechanism Over- Temperature Problem," Los Alamos National Laboratory, 1982. 30. "CRDOA Proposed Preventive/Predictive Maintenance Program Report, " PSC submittal P-85040-3, 1/31/05. 31. "ChDOA Interim Surveillance Program Report," PSC submittal P-85040-5, 1/31/85. 32. "FSV Improvement Committee Actions," PSC submittal P-85022, 1/24/85. 33. "Tendon Accessibility Report," PSCo. letter from Warembourg, PSC, to Jonnson, NRC/Reg IV, PSC submittal P-84523, 12/14/84. 34. "Lab Report No. 52--Examination of Failed Wires from Fort St. brain Unit No. i," PSC submittal P-o4543-4, 1/24/85. 35. "Liftoff Tests," Attachment 1 to "Engineering Report on Fort St. Vrain Tendons," PSC submittal P-84543, 12/31/84. - 26 - `0,,II,EGV Attachment 3 • . 9! UNITED STATES W • NUCLEAR REGULATORY COMMISSION �� '1 WASHINGTON,D.C.20555 Y� SAFETY EVALUATION BY THE OFFICE OF NUCLEAR REACTOR REGULATION RELATED TO CONIROL ROD POSITION INSTRUMENTATION TO FACILI1Y OPERATING LICENSE NO. DPR-34 PUBLIC SERVICE COMPANY OF COLORADO FORT ST. VRAIN DOCKET NO. 50-267 1. INTRODUCTION This Safety Evaluation concerns control rod position instrumentation (RPI) failures that were recognized subsequent to the partial ATWS event at Fort St. Vrain on June 23, 1984. The NRC issued an Assessment Report [Reference 1], which addressed the ATWS event and the RPI failures, and included recommended actions to be taken at Fort St. Vrain before the plant is restarted. The objective of this evaluation is to review the licensee's response regarding the following issues: the causes of the RPI failures, the methods to prevent further failures, operability and surveillance actions for the RPIs, and the adequacy of the licensee's proposed backup method for verification of rod full-in position. The submittal by the licensee in response to the Assessment Report takes the form of a series of reports dated about January 31, 1985 [References 8-15]. Numerous additional documents were reviewed, including background information about control rod drive design and operation, supplied by General Atomic Company, NRC reports, and Public Service Company of Colorado (PSC) related submittals [1-7, 16-18]. Evaluations and conclusions are provided based on the licensee's response to questions and the information contained in the licensee's submittals. 2. BACKGROUND On June 23, 1984, the Fort St. Vrain plant experienced a partial ATWS event, in that 6 of 37 control rod pairs failed to insert into the reactor core in response to an actual scram signal . Subsequently, on July 31, 1984, PSC reported 11 RPI anomalies to the NRC. The anomalies included improper analog and digital indications of rod position, faulty slack cable, erroneous rod full-in and full-out limit indications. During August 1984, we visited the plant to investigate the RPI failures and related problems. We met with the licensee again at the plant site on November 28, 1984, at which time the licensee indicated the general steps that were being taken regarding the instrumentation problems. On January 31, 1985, the licensee provided documentation of its evaluations and efforts relating to restart. These documents were reviewed by the staff and its consultant, Franklin Research Center. Questions arising from the review were addressed by the licensee during a transcribed meeting at the plant site on February 22, 1985. - 2 - In order to establish the scope of this review, we have revisited the Assessment Report, the Region IV letter of January 16, 1985 to PSC, and the PSC response of January 28, 1985 [1, 13, 17]. The following restart items are within the scope of this review. o Determination of RPI Failure Causes and Corrective Action o Procedures for Prevention of Driving Control Rods Beyond the Full-in Limit o RPI Limiting Conditions for Operation o Position Instrumentation Surveillance Program During this evaluation, we determined that revisions of the plant procedures are necessary to assure that appropriate action is taken upon indication of mispositioned rods following a scram. Additionally, during the February 22, 1985 meeting with the licensee, the Wattmeter test was reclassified as a restart issue. 3. EVALUATION The evaluation is divided into the following sections: Page 3.1 Rod Position Instrumentation (RPI) Failures and Corrective Actions 2 3.2 Prevention of Inward Overtravel of Rods 4 3.3 RPI Minimum Performance Requirements 5 3.4 RPI Surveillance Requirements 3.5 Backup Reactor Shutdown Procedure 9 3.6 Backup Full-in Position Verification Test 9 (Wattmeter) Each section includes the conclusions regarding that topic. 3.1 ROD POSITION INSTRUMENTATION (RPI) FAILURES AND CORRECTIVE ACTIONS Subsequent to the reactor trip of June 23, 1984, the licensee identified 11 control rod position instrumentation (RPI) failures. The failures included simultaneous rod full-in rod and rod full-out indications, full-out switch lights remaining lit, indications of partial rod withdrawal , no position signals, disparity between analog and digital rod position information, and a slack-cable indication. The failures caused conditions in which the operators could not determine the position of certain rods based upon the existing indicators. As described in the October 16, 1984 Assessment Report [1], for one rod pair the analog and digital RPIs indicated it to be at the 40-inch withdrawn position. Simultaneously, the full-in position indicator for this rod was not operable. When asked if the instrumentation should be believed, PSC personnel responded that it did not believe the installed RPIs, but had verified to its satisfaction that the control rod was fully inserted. - 3 - The results of the licensee's investigation into the causes of these failures and his proposed corrective actions are included in Attachment 4 of PSC letter P-85032 [12], dated January 30, 1985. This report suggests that the failure causes for both the switches and the potentiometers were primarily mechanical in nature and were not electrically induced. The licensee confirmed this general conclusion during the February 22, 1985 meeting. Limit Switch Failures The limit switches that indicate full-in or full-out positions are roller plunger-type microswitches. The licensee stated that the 30° slope of the cams that operate the limit switches is too abrupt, causing a high lateral force between the plunger and the shaft tube that eventually results in switch failure. This failure cause may have been compounded by increased friction due to pitting of the plunger and shaft tube from the effects of moisture. Our evaluation of the licensee's description of the failure cause indicates that it should be expected, given the design of the limit switch. Further, the failure mechanism is consistent with our previous experience with such switches. We conclude that the failure cause identified by the licensee is credible. The licensee proposes to change the slope of the limit switch cams to 15° to reduce the lateral force. However, these cams are not expected to be installed at the time of return to service. During the present refurbishment of the CRDs, the operation of the switches will be checked and those switches which are not operational will be replaced in kind. The level of deterioration of the operational switches will not be evaluated. Therefore, potentially degraded but operational switches will be returned to service. Potentiometer Failures There are two separate failure mechanisms associated with the analog and digital position potentiometers: one relates to internal damage of the potentiometers and one relates to external damage of the associated coupling. The internal failure mechanism for the potentiometers is travel beyond the 10-1/2-turn capability. The greatest potential for such overtravel is at the full-in limit, especially if the full-in limit switches have failed. Review of the CRD design indicates that damage may result from an inward overtravel of as little as six inches. There is no mechanical stop that interrupts CRD travel before potentiometer failure might occur. The analog and digital potentiometers share a common drive shaft so that simultaneous failure may occur if both potentiometers exceed their overtravel limits. The licensee proposes to replace the 10-1/2-turn potentiometers with 15-turn potentiometers so that overtravel of the drive will not cause internal damage to the potentiometers. However, the 15-turn potentiometers are not expected to be available for installation at the time of the scheduled plant restart. - 4 - The external failure mechanism associated with the potentiometer is caused or mechanical interference couplingconbthecommon driveen the cams shaft the liformit switches the potentiometers. During an inward overtravel , one of the limit switch cams can strike the coupe nt ce has wn that,thereby breaking occurs, coupling.ithispos a possible fortheeanalognindicatoro if to agree with the digital indicator, and for both indicators to track rod motion, but both indicators can be grossly inaccurate (i.e. , offset) with respect to the actual cations have been licenseeototion of eliminatee rod damageair. No caused by theiinterference betweenotheed by the cam and the coupling. Prior to startup, already-damaged potentiometers will be replaced, as will damaged multi-jaw couplings that connect the potentiometer shaft to the CRD gear train. The replacement potentiometers will be 10-1/2-turn devices, not 15-turn devices. 3.2 PREVENTION OF INWARD OVERTRAVEL OF RODS Since modifications that prevent overtravel or otherwise preclude damage as a result of overtravel will not be in place at the time of restart, procedures must be instituted that preclude intentional inward overtravel of the control rod pairs and reduce the probability of such damage. In Attachment 7 to PSC letter P-85040 [14], the licensee provided a revision of page 17 of 20 of Procedure SOP 12-01, that describes actions to be taken if the "in limit" light is not received when a rod is believed to be fully inserted. The previous practice had been to drive the rod further inward in search of limit switch actuation. In the revised procedure, the operator is directed to first withdraw the rod a few inches, trying to obtain the full-in position indication. The procedure then directs the operator to move the rod switches inward until rod motion automatically stops, or an indication of zero inches is attained, whichever occurs first. Use of the revised procedure will help assure that rod position indications remain available for use. However, no direction or suggestion is given to perform a lamp test or to verify the analog/digital position indication prior to this special rod movement. It is recommended that these steps be added to this procedure. It is also unclear that this particular procedure, SOP 12-01, governs all conditions in which manual rod inward travel in the vicinity of the full-in position may be performed. The licensee should modify all applicable procedures to similarly reduce the likelihood of inward overtravel of the rods near the full-in limit. - 5 - 3.3 RPI MINIMUM PERFORMANCE REQUIREMENTS Control rod full-in limit indications are "important to safety" because they are the primary means of verifying that all the CRDs have fulfilled their reactor scram safety function. Immediately following a plant trip, total reliance on the startup range nuclear instrumentation may be inadequate. The analog position indicators are "important to safety" because they provide the operator with continuous position information for all the rods during operation and following a scram. In contrast, however, the digital position indicators and the full-out position indicators are not as important, but provide information that is useful for accurate position data, fine control of reactor power, and as operational conveniences. They may be used a backup indicators, in the event of a loss of the analog or full-in limit indications. The NRC Assessment Report recommended the Technical Specification Limiting Conditions for Operation be established to define the minimum performance requirements for the RPIs. The licensee has concluded that a specific limiting condition for operation (LCO) governing rod position instrumentation is not necessary. The basis of this conclusion is that other LCOs indirectly require instrumentation to be operable. In Attachment 1 to PSC letter P-85040 [14], the licensee discussed LCOs 4.1.2, Operable Control Rods; 4.1.4, Partially Inserted Rods; and 4.1.8, Reactivity Status that require surveillance tests that in turn require the use of position indicators. The licensee's reasoning is that lack of position indication will preclude performance of a surveillance test and therefore an LCO will not be met. The use of indirect LCOs as a requirement for position indicator operability does not focus attention on the importance of the RPIs, does not provide definitive action statements for partial losses of instrumentation, and does not ensure timely resolution of indication failures. In addition, the use of indirect LCOs is not consistent with the Technical Specifications for other operating plants. In summary, we do not agree with the licensee's conclusions. In light of the possibility of an ATWS event, coupled with the propensity of the position indications to fail , the following concepts related to RPIs must be implemented via procedures prior to restart and via a new Technical Specification LCO soon after restart: At startup and during operation, the analog rod position indicator and full-in rod limit indicator for each control rod pair shall be operable. - 6 - If an analog indicator is inoperable, operation may continue provided that one of the following conditions is met: (1) when the rod is fully inserted, the full-in limit indication is operable or the full-in position has been established by an independent means of verification; (2) when the rod is in a mid-position, rod position is continuously indicated by an operable digital position indicator and the full-in limit indtheifull out and rthe efull-in; or )position when elimit rod iindicators s in the uare operable.11-out on, If the full-in indicator is inoperable, operation may continue provided that one of the following conditions is met: (1) when the rod is fully inserted, the full-in position has been established by an independent means of verification; or (2) when the rod is in a mid or full-out position, both the digital and analog indicators are operable and are known to be accurate at the full-in position, and a digital indicator is continuously indicating the rod's position. In addition to the new LCO described above, a modification to LCO 3.1.2, Operable Control Rods, is necessary. This modification must incorporate the RPI concept that rod all be red newpLCOble areif notts associated met. Without such operability requirements, the operators may tend to lose faith in the position indications to the extent that actual failure of the rods to attain the full-in position may be inappropriately considered to be a faulty indication rather than a valid condition. The LCOs described above are necessary whether or not the proposed modifications are made to the indicator mechanisms. However, the LCO is particularly important for the present restart since the modifications that will reduce the failure rate of the indicators will not be in place. With the required LCO in effect at the time of restart, restraints will be placed on reactor operation to allow safe operation should position instrumentation failure occur. 3.4 RPI SURVEILLANCE REQUIREMENTS The NRC Assessment Report recommends that the licensee establish Technical Specification Surveillance Requirements regarding the RPIs. In Attachment 5 of PSC letter P-85040 [14], the licensee outlines the proposed interim surveillance program for the control rod drives. This program includes partial verification of rod position indication operability at weekly and quarterly intervals during operation and during each refueling outage. For fully withdrawn rods, the proposed weekly surveillance verifies the oerabiit s and betweenithe analog and ity of the ldigital-out mindications rto be no r mores the difference than 10 inches. - 7 However, the program outline does not indicate that the analog and digital readings must correlate to the full-out or full-in position, nor does it state that the indicators must correctly track the rod position when the rod is exercised (i.e., partially inserted and then withdrawn). For partially inserted rods, the weekly surveillance will verify that the analog and digital position indications are within 10 inches of each other. Again, the proposal does not state that the indicators must appropriately track rod motion. For fully inserted rods, there is no proposed weekly surveillance, because other Technical Specification requirements apparently preclude movement of fully inserted rods during power operation. The proposed quarterly surveillance adds one further test to the weekly surveillance. For rods in the full-out position, the operability of each of the redundant full-out limit switches will be verified. No additional quarterly surveillance is proposed for the RPIs associated with partially inserted and full-in rods. The proposed refueling outage surveillance includes verification of the operability of each redundant full-out and full-in limit switch and comparison of the analog and digital position indications during a full travel scram of the analog and digital indications during a full travel scram of the rod. While not directly stated, it is assumed that the accuracy of the analog and digital position indicators will also be confirmed at the full-out and full-in positions during these tests. The RPI surveillance actions proposed by the licensee are worthwhile and appropriate. However, the proposed surveillance program does not encompass all of the necessary aspects of RPI operability verification. To be consistent with the position established in Section 3.3, that the full-in limits and the analog indicators are "important to safety," appropriate surveillance actions are needed for these devices. Further, if the digital indicators are to be used as backup indications, some surveillance actions for these devices are also appropriate. The surveillance tests proposed by the licensee do not address the need to verify operability of the full-in limit switches prior to outward movement from the full-in position. Nor do they verify the accuracy of the digital and analog position indicators at the full-in position prior to the first outward rod motion. Also, the proposed surveillance program description does not provide sufficient detail to determine if the operability of analog and digital position indicators will be evaluated for partially and fully withdrawn rods. - 8 - Since the failure mechanisms identified by the licensee for the full-in switches and potentiometers are associated with overtravel of a rod near the full-in limit, it is appropriate to verify that these position indicators are operable before or at the next outward movement of the rod from the full-in position. This suggests the need for verification test of these RPIs at each reactor start-up. Verification of the operability of the full-in limit switches and the accuracy of the analog RPIs may only be performed when the rods are full-in, as they are at or prior to start-up. Furthermore, simple verification checks at this time would have minimal adverse impact on plant operations. The following concepts should be incorporated into the surveillance program prior to reactor restart. The surveillance requirements that result from these concepts are in addition to the surveillance actions proposed by the licensee and are necessary to assure that instrumentation is operable and reasonably accurate at reactor start-up, during operation, and at the time of a scram. Prior to each reactor start-up, each full-in limit indicator should be verified as operating when the rod is full-in and operable by virtue of the change in the indication when the rod is withdrawn a short distance. Alternatverified t hefirst time dur rod full-in ingioron afteristart-uppthatiairodmis withdrawn be from the full-in position. Prior to each reactor start-up, or during the first outward motion of a rod, the analog position indicator should be shown to be acceptably accurate at the full-in position and be shown to respond appropriately when the rod is withdrawn a short distance. The accuracy requirement should be such that returning the rod to an "0" position as indicated by the analog RPI will not result in overtravel that could cause damage to the potentiometers and the associated coupling. The digital indication should likewise be shown to be operable and accurate at the full-in position. During each weekly surveillance during power operation, the reasonableness of the analog RPI should be verified by comparing the change in analog indi- cation with the direction and time duration of the rod travel . The analog and digital position indications should agree with each other within a predetermined amount. If a larger difference is observed, the licensee should, conservatively, assume that the analog indicator is the inoperable channel , unless it can be proven to be accurate and operable by another means. The combination of the surveillance actions proposed by the licensee and the additional surveillance actions stated herein will provide the necessary assurance that the RPIs are operable and accurate. r - 9 - 3.5 BACKUP REACTOR SHUTDOWN PROCEDURE At the time of a reactor scram, it is essential that the reactor operator have confidence in the control rod position indications and take conservative actions based upon those indications. Therefore, when a reactor scram occurs, the reactor operator must be able to verify that the rods are full-in or must take appropriate alternate shutdown actions. In view of our conclusions regarding the corrective actions to prevent future RPI failures, we have determined that additional action is necessary. Plant procedures should specify that, if after a reasonable but conservative period (e.g. , one hour), more than one rod pair cannot be verified as being at the full-in position, these rods must be assumed not to be fully inserted and the backup shutdown system is to be initiated. Such actions are necessary to assure adequate shutdown. Furthermore, immediately following a plant trip, total reliance on the startup nuclear channels may not be adequate. Therefore, the Fort St. Vrain procedure must be revised, as necessary, to include the following concepts: Following each reactor scram, each rod pair shall be verified to be at the full-in position by one of the following means: 1. the agreement of the analog position indication and the full-in position indications; or 2. the agreement of the analog and digital position indications (that were known both to be operable prior to the scram and to be accurate at the full-in position); or 3. the use of an independent rod position verification method (e.g. , Wattmeter test). If more than one control rod pair cannot be verified to be fully inserted at the end of one hour, the backup reactivity control system must be initiated. Rods that were known to be fully inserted into the reactor prior to the scram may be excluded from the above considerations. 3.6 BACKUP FULL-IN POSITION VERIFICATION TEST The NRC Assessment Report of October 16, 1984 [1], concluded that the installed RPI system may be inadequate to determine rod positions under adverse conditions. The licensee has been using the Wattmeter test as an independent verification of full-in positions. However, as a result of our review, we determined the test was too judgmental to provide convincing verification. Therefore, the licensee was requested to refine its means of verifying that the control rods are fully inserted. - 10 - The licensee's Wattmeter tests analyze the electrical power required by the motor when the control rod pair is moved a short distance in the vicinityletter fthe llinposition . [hese tests were described in AtThe electrical power requirements at the full-in position differ from other positions due to the unwinding the last 10 inches, th cable the e point of rum. As the cable unwinds for approximately connection of the cable to the drum changes from the horizontal to a,vertical s(6 i or'clock)re the positionwherent rthen the momentrum armis isgzero. Because of this change in moment, the power requirements for the motor in both the inward and outward directions near the full-in position are different from other positions of the rod. On an outward pull from the full-in position, the transient power peak is lower and of shorter duration than on outward pulls from other positions, and the power peak is followed by a dip in power below the steady-state level , which also does not occur at other positions. On an inward position isereached.full-inp This ositirise tdoes he onot occur atwer rises iother positions ghtly as the of the rod. The licensee has sufficiently described and demonstrated the phenomena related to the electrical power changes. The uniqueness of these phenomena provides an acceptable concept for verifying the full-in rod position. The key areas of concern continue to be the difficulty of interpretation of the wattmeter charts and the degree to which judgment is required in order to properly interpret the results. The acceptance criteria proposed by the licensee require interpretation of graphical results at or beyond the levels of precision and resolution of good engineering practice. The wattmeter used is a 1000-watt meter coupled to a stripchart recorder. The smallest division on the chart paper is equal to 20 watts. The procedure requires interpretations of differences of 4-6 watts. The procedure also requires interpretation of the decay time of the motor starting transient. The non-standard definitions of decay time stated in the licensee's procedure could allow greatly differing interpretations of the same transient. Under the definitions, one interpretation of the decay time may be as much as twice that of another interpretation for the electrical transient. The licensee proposes to use a wattmeter that has a 1000-watt scale and uses paper with a 500-division scale. The written procedure alternately notsewatts and ally surescale whethervwattssorlmost scaleidivisionseisly, such c that one is correct. - 11 - 4.0 SUMMARY We have evaluated information provided by the licensee regarding: (1) control rod position instrumentation (RPI) failure mechanisms and corrective actions; (2) procedures to prevent inward overtravel ; (3) Technical Specifications relating to RPI operability; (4) RPI surveillance program, (5) backup reactor shutdown procedures; and (6) the backup full-in position verification (wattmeter) test. The conclusions relating to these concerns are summarized below. The licensee has determined that the failure of the limit switches is caused by the steep slope of the limit switch cams, and that two failure causes are associated with the potentiometers. In the first, the potentiometer is forced beyond its 10-1/2 turn overtravel limit and is damaged during inward overtravel . In the second, the shaft coupling for the potentiometers is struck by a limit switch cam during inward overtravel of the rod. The licensee has proposed a modification to the limit switch cam to eliminate the limit switch failure mechanism and has proposed to use a 15-turn potentiometer to eliminate the forcing of the potentiometer beyond its internal travel limit. However, these modifications will not be implemented prior to the scheduled restart, and no modification has been proposed to address the second cause of the potentiometer failures. The licensee has modified Procedure SOP 12-1 to preclude manual inward overtravel of rods following a reactor scram. It is not clear that this procedure covers all conditions in which manual inward motion of a control rod could occur. Therefore, the licensee should confirm that all procedures in which rod travel beyond the full-in limit could occur contain appropriate cautions and controls to reduce the likelihood of damage to potentiometers and their couplings. Contrary to the recommendation in the NRC Assessment Report, the licensee believes that no additional limiting conditions for operation (LCO) for RPIs are necessary and proposed to rely upon indirect LCOs. The use of indirect LCOs does not provide definitive actions to be taken upon the loss of RPIs and is inconsistent with the Technical Specifications of other reactors. Therefore, the LCOs described in Section 3.3 for RPI operability are necessary. The licensee has proposed surveillance actions for the RPIs for use during reactor operation and during refueling outages. These surveillance actions are appropriate and worthwhile. However, they do not cover verification of operability of the RPIs at each startup of the reactor. The failure mechanisms for all RPIs (except the full-out limit RPI) are associated with travel near the full-in limit. Therefore, it is appropriate to verify operability of these RPIs at the next outward motion of the rod (i.e. , at startup). The combination of the licensee's proposed surveillance program and the additional surveillance stated in Section 3.4 combine to provide adequate assurance of RPI operability. - 12 - When a scram occurs, the operator must believe the rod position indicators. If multiple control rod pairs are not indicated as being full-in, the operator must take timely action to initiate the backup reactor shutdown system. The modifications to the operating procedures described in Section 3.5 will assure this action. The wattmeter test proposed by the licensee is an adequate method of verifying the rod full-in position provided that the ease of interpretation of the he inter- polation decreased sthroughttheluseloffa reliance moreappropriate ter- choice of wattmeter range and recorder speed. Date: May 17, 1985 Principal Contributor: J. T. Beard, DL REFERENCES 1. Letter from H. R. Denton, NRC, to R. F. Walker, PSC. Preliminary Report Related to Restart and Continued Operation of Fort St. Vrain Nuclear Generating Station Docket No. 50-267, Public Service of Colorado, October 16, 1984. 2. PSC Presentation and Handout Notes from Meeting at Fort St. Vrain, November 28, 1984. 3. Letter from 0. R. Lee, PSC, to E. H. Johnson, NRC Region IV, Attachments 1 and 2, Observations on Reworked Control Rod Mechanisms, August 28, 1984. 4. PSC History of Observations on 44 CRDOA, Fuel Handling Procedure Work Packet, FHPWP, August 2, 1984. 5. Memorandum from J. T. Beard, NRC, to J. Miller, NRC. Subject: Executive Summary, FSV Assessment Report, September 11, 1984. 6. Rod Control System Equipment I-9303, Operation and Maintenance Manual 93-I-1-327, E-115-265, (Rev. 3), Prepared for PSC, Unit 1, Fort St. Vrain, Colorado, Gulf General Atomic, 1979. 7. Installation, Operation, and Maintenance Manual for the Control and Orificing Assembly for the Fort St. Vrain Reactor 12-D-4-63, FCN-3849, GA-9806 (Rev.) Prepared for PSC of Colorado Fort St. Vrain Nuclear Generating Station, General Atomic, May 1977. 8. Letter from 0. R. Lee, PSC, to E. H. Johnston, NRC Region IV. Subject: Comments on Index No. 3 (January 15, 1985 Meeting Minutes) , January 28, 1985. 9. Letter from H. L. Brey, PSC, to E. H. Johnson, NRC Region IV. Subject: Description of CRDM Failures on January 14 and 16, 1985 (Work Reports) , January 28, 1985, P-85029. 10. Letter from 0. R. Lee, PSC, to E. H. Johnson, NRC Region IV. Subject: Creation of FSV Improvement Committee and Items of Consideration, January 24, 1985, P-85022. 11. Letter from D. W. Warembourg, PSC, to E. H. Johnson, NRC Region IV. Subject: Results of CRDOA Debris Analysis Test Report Attached, January 18, 1985, P-85017. - 2 - 12. Letter from 0. R. Lee, PSC, to E. H. Johnson, NRC Region IV. Subject: Transmittal of Technical Reports, January 30, 1985, P-85032. Attachment 1: CRODA Mechanism Temperatures Environmental Requalification Attachment 4: Control Rod Instrumentation Attachment 5: Cable Anchor Welding Attachment 6: CRDOA Moisture/Purge Flow 13. Letter from R. D. Martin, NRC Region IV to 0. R. Lee, PSC. Subject: Minutes of Meeting of NRC Region IV and PSC on January 15, 1985. Commitments by PSC on Plant Upgrades, January 16, 1985. 14. Letter from PSC (author unknown) to R. Martin, NRC Revion IV. Subject: (See Attachments), January 31, 1985, P-85040. Attachment 1: Control Rod System Operability Evaluation Report Attachment 5: Control Rod Drive and Orificing Assembly Interim Surveillance Program Attachment 6: Wattmeter Use to Determine Inserted Absorber String Position Attachment 7: Page 17 of SOP 12-01, Discussion of No "In Limit" Light on Fully Inserted Rod. 15. Letter from D. W. Warembourg, PSC, to E. H. Johnson, NRC Region IV. Subject: Cover Letter for Attachment, January 31, 1985, P-85937. Attachment: PSC Report No. EE-12-0010, 'Failures to Scram - Control Rod Drive and Orifice Assemblies." 16. Letter from D. W. Warembourg, PSC, to E. H. Johnson, NRC Region IV. Subject: High-Pressure Scram and Ensuring Control Rod Automatic Insertion Failures P-84227, July 23, 1984. 17. Letter from 0. R. Lee, PSC, to E. H. Johnson, NRC Region IV, January 28, 1985. Subject: Response to Reference 13. 18. NRC Memorandum from J. T. Beard, ORAB, to E. Johnson, Region IV. Subject: Fort St. Vrain -- Design Weakness in Reactor Scram System, March 4, 1985. 19. Letter from J. R. Buchanan, Oak Ridge National Laboratory, to Frederick J. Hebdon, NRC, February 14, 1985. Hello