Semantic search
Jump to navigation
Jump to search
| Start date | Reporting criterion | Title | Event description | System | LER | |
|---|---|---|---|---|---|---|
| ENS 57167 | 7 June 2024 04:00:00 | 10 CFR 21.21(d)(3)(i), Failure to Comply or Defect | Part 21 Report - Check Valve Fails Bench Test | The following is a synopsis of information received via email: Flowserve Corporation identified a defect with their 0.5 inch model 1878 piston check valve during bench testing. The piston check valve with adjustable cracking pressure received by Duke Energy Catawba Nuclear Station failed bench test performance for designed cracking pressure with an allowable range of 180 - 220 psig. The valve is used to provide overpressure protection of containment penetration piping due to thermal expansion of fluid trapped between the containment isolation valves when the valves are closed. The valves are required to crack open before the maximum 220 psig setpoint to perform this function. Corrective actions include revising Flowserves assembly and test procedures for the 0.5 inch model 1878 piston check valve. Affected known plant: Catawba Nuclear Station The name and address of the individual reporting this information is: Aaron G. Goodbar Nuclear Quality Assurance Manager Flowserve Corporation 1900 S. Saunders St. Raleigh, NC 27603 1-888-200-5389 | ||
| ENS 56219 | 11 September 2022 04:00:00 | 10 CFR 21.21(d)(3)(i), Failure to Comply or Defect | Part 21 - Flowserve Solenoid Valve Deficiency Report | The following is a synopsis of information provided by Flowserve via fax: Flowserve Corporation hereby submits the enclosed written notification of the identification of a defect, in accordance with the requirements of 10 CFR 21.21(d)(3)(ii). This notification pertains to the failure of a solenoid valve (model 38878-8) at Catawba Nuclear Station. List of plants / customers potentially impacted: Duke - Catawba Luminant - Comanche Peak Exelon - Byron Exelon - Braidwood NextEra Energy - Seabrook The solenoid valve received by Duke Energy failed to shift position when the coil was initially energized. This solenoid valve is part of the feedwater isolation valve assembly and failure to shift will prevent the feedwater isolation valve from closing on demand. Flowserve was initially notified on September 11, 2022. Flowserve provided response to Duke on September 15, 2022 to confirm that Flowserve would perform the evaluation within 45 days upon receipt of the solenoid valve. Flowserve received the solenoid valve for evaluation on October 5, 2022. During Flowserve's investigation into the root cause of the solenoid valve failure, Flowserve was unable to replicate the failure. Based on examination of the solenoid valve components, excess thread lock compound had been applied to the set screw and nut on the coil cap. The excess thread lock compound then ran down the coil cap and wicked between the coil cap outside diameter and the end cap inside diameter. Once this excess thread lock compound cured, the coil cap was locked in the de-energized position, preventing the coil cap from shifting. After the customer left the solenoid valve energized for 10 minutes, the coil cap broke free from de-energized position and was able to shift freely. In the procedure for assembly and testing of this solenoid valve, the solenoid valve is not energized after application of the thread lock compound. Flowserve will (1) revise the assembly and test procedure for the Model 38878 solenoid valve to add a final test after the thread lock component has been applied and been allowed to cure to verify the solenoid valve shifts when energized, and (2) provide training to assembly and test personnel on the importance of ensuring that excess thread lock compound has not been applied. These actions will be completed by December 15, 2022. Nuclear power plants with model 38878 solenoid valves that have not undergone acceptance testing should verify that these suspect solenoid valves will shift on demand when energized. For additional information, please contact Kayn Dills, Flowserve Corp Quality Manager (800-225-6989) | Feedwater | |
| ENS 55974 | 2 February 2022 04:00:00 | 10 CFR 21.21(d)(3)(i), Failure to Comply or Defect | Part 21 REPORT- Failure of Flowserve Solenoid Coil | The following is a synopsis of information received via facsimile: A Model 38878-8 solenoid valve failed a routine coil resistance test at Catawba Nuclear Station while installed on a feed water isolation valve actuator. The solenoid valve was returned to Flowserve where the low resistance was confirmed. The solenoid coil was then sent to the Original Equipment Manufacturer (OEM) for further evaluation. The OEM (Ohmega) completed their analysis and found the reason for failure to be associated with the magnet wire, but the exact point of failure could not be located due to the construction of the coil. Additionally, Flowserve compiled shipment data for the subject coil and found there to be at least 273 instances where the part was shipped to customers. Of those 273+, this case is the only known instance of a failure associated with the coil. Sites that Flowserve shipped the Model 38878-8 Solenoid Valve to: Comanche Peak, Catawba, Braidwood, Byron, Beaver Valley, Seabrook Due to the rigorous functional testing and the historical reliability of the coil in the field, Flowserve does not believe this incident is indicative of an issue with the manufacturing or testing of the coil and concludes that this issue does not affect other coils currently in service. Ohmega suggests a possible manufacturing improvement of winding the coil with a varnish to provide extra insulation of the magnet wire. Flowserve suggests that plant operators using these solenoid coils measure the resistance of the coil periodically, especially after the coil has been energized for testing or service. | ||
| ENS 55505 | 4 August 2021 04:00:00 | 10 CFR 21.21(d)(3)(i), Failure to Comply or Defect | Part 21 - Actuator Failure | The following is a summary of the report provided by Flowserve: Flowserve - Limitorque was notified by Framatome Nuclear Parts Center that TVA Browns Ferry Nuclear (BFN) plant reported that a Limitorque SMB actuator had failed to operate electrically. It was noted by TVA that the actuator failure was discovered following a recent steam leak in the vicinity of the actuator. Investigation at BFN identified the failed component as the DC electric motor. The motor was removed from the actuator and disassembled by BFN personnel. Subsequently the motor was returned to Flowserve - Limitorque for evaluation. The subject motor is designated as a safety related basic component. Flowserve' s investigation, in conjunction with the motor original equipment manufacturer (OEM), has concluded that during manufacture, the subject motor sustained mechanical damage to the stator assembly field coil which led to the failure. As the dedicating entity, Flowserve is reporting this deviation in the assembly of the motor as a defect per the requirements of 10 CFR 21. Flowserve's investigation has concluded that this manufacturing defect is an isolated occurrence. No other instances of similar mechanical damage to the stator field coil insulation of Peerless DC motors have been observed or reported. The defect is specific to a quantity of one motor originally supplied to Framatome NPC on Limitorque Order Number 164879.001. This motor was subsequently supplied to the TVA Browns Ferry Nuclear Plant. Identification of Component: Limitorque part # P-140-851-18F0 Description: Peerless 7.5 ft-lb, 250 volt DC motor Serial Number: E18-99091-2 manufactured in August 2018 Corrective Actions to Date: The motor OEM is implementing additional assembly controls and quality control verification concerning the installation and final appearance of the coil support rod that damaged the stator field coil. This action, to be completed by October 15, 2021 will supplement the existing inspection checks on fully assembled motors, which includes insulation resistance tests and high potential tests which verify the integrity of the motor insulation system. Technical questions concerning this notification can be directed to Kyle Ramsey, Engineering Specialist, Flowserve - Limitorque Actuation Systems: kramsey@flowserve.com | ||
| ENS 53891 | 20 February 2019 05:00:00 | 10 CFR 21.21(d)(3)(i), Failure to Comply or Defect | Part 21 Report - Flowserve Valve Assist Springs May Prevent Full Closure | The following report was received via fax: Description: GE BWXT, acting on behalf of Ontario Power sent five valves back to the Flowserve facility for evaluation of the spring which could spread apart when the flapper rotates open and wedge between the valve body and flapper, creating the potential for the valve flapper to stick open. Evaluation: Shop inspections of the returned valves confirmed the potential for the spring legs to rotate outwards and potentially wedge between the flapper and seat retainer. Of the five valves evaluated, two of the valves had the spring legs rotate outward and after repeated cycles got caught between the flapper and disc; on one of the instances the spring legs prevented full closure of the flapper. If the valves have a safety-related function to isolate flow and must transfer closed, then the springs could prevent the valve from performing its safety related function. The evaluation determined the cause is an inadequate spring design which allows the spring to deflect and permit the legs to move outward. Extent of Condition: In discussion with the customer, it is believed the problem did not exist with valves originally supplied from the Worcester, Scarborough Canada facility; although this cannot be confirmed by Flowserve. It is believed the extent of condition resides with valves and replacement springs supplied from the Flowserve Raleigh, NC facility when the product line was transferred and the spring vendors changed; even though it has been confirmed that the springs supplied out of the Raleigh, NC facility comply with the product drawing requirements. The springs were first sold in valve assemblies or as replacement parts from the Raleigh facility starting in 2008. Valve Scope: The scope of impacted valves is the Worcester series 44 swing check valves. (Please note Worcester also has a series 44 three piece ball valve, which is not in the scope.) The Raleigh facility has supplied parts or valves for four Worcester series 44 swing check valves assemblies which use these springs. The drawings numbers for these valve assemblies are listed below: Drawing Valve Size Customer KN44-0590 1.5 X 1 X 1.5 Bruce Power, GE BWXT & Ontario Power KN44-0630 2 Comanche Peak 14-107362-001 1.5 X 1 X 1.5 GE BWXT 16-118733-001 1.5 X 1 X 1.5 GE BWXT Corrective Actions: 1) Owner's may remove the springs. The valves will check-off and seal under low pressure conditions without the assistance of the spring; this was demonstrated in shop tests using tap water pressure (approximately 60 psig) and will initiate closure when full open without the assistance of the spring. 2) Flowserve is evaluating alternate spring designs along with the possibility of adding guides to prevent the undesired spring movement. Any new design will be proof tested. Respectfully submitted, Mark Rain, PE Product/Design Engineering Specialist Flowserve Corporation Flow Control Division 1900 S. Saunders St. Raleigh, NC 27603 | ||
| ENS 52212 | 31 August 2016 04:00:00 | 10 CFR 21.21(d)(3)(i), Failure to Comply or Defect | Part 21 Notification - Flowserve Swing Check Valves | The following is an excerpt of the part-21 notification: Susquehanna Nuclear reported two size 3 class 900 Anchor Darling swing check valves were not passing LLRT (local leak rate tests). After disassembly there appeared to be wear between the hinge arm and seat ring of the valve body apparent on the hinge. The interference was not severe enough to stop hinge arm motion of the disk, but did affect ability of the valve to seal during LLRT. Affected Serial Numbers: BQ752, BQ753, and BO809 Affected Sites: Ginna, Diablo Canyon, Brunswick, Davis Besse, Kewaunee, Conn Yankee, Monticello, Susquehanna, Framatome, Chin Shan, and Kuosheng Engineering Evaluation 10CFR21 No. 91 POC: Wade Shephard: 919-832-0525 | ||
| ENS 48157 | 3 August 2012 04:00:00 | 10 CFR 21.21(d)(3)(i), Failure to Comply or Defect | Part 21 - Defective Valve Seal Cap | The following was provided via fax: Background: On July 17, 2012, Duke Power - McGuire NPP notified Flowserve Corporation that they had attempted to install a Seal Cap supplied by Flowserve on a Kerotest 1-1500 Y-Type Globe Valve. They found that the Seal Cap could not be installed on the valve due to interference with the Valve Stem. Discussion: The Seal Cap is used to contain leakage past the Valve Diaphragm. The Seal Cap fits over the Valve Stem after Handle removal and is screwed onto the Yoke until it contacts the top of the Body. It is then seal welded to the Body to create a leak tight cap. At this point the Seal Cap is considered pressure retaining. Inspection of the Seal Cap revealed an inside flat bottom in lieu of the conical bottom shown on the drawing. The conical drill point allows clearance for the valve stem when the Seal Cap is installed on the valve. Review of a layout shows the Seal Cap with a Flat bottom cannot be completely installed whether the valve is open or closed. Conclusion: The Seal Cap with a flat bottom cannot be properly installed on the Kerotest Series R9900 size 1/2 to 2 Valves for which it was designed. The Valve Stem will prevent complete installation and the ability to seal weld the Seal Cap to the Body. The Seal Caps supplied would not be available at the Nuclear Power Plant site for an urgent repair if required. The following Nuclear Power Plant Utilities were provide these Seal Caps without the conical shaped area. TVA - Purchase Order# 00020077-00050 and 00020077-00018 First Energy - Purchase Order# 45108999, and 45254808 Alabama Power - Purchase Order# QP040387, QP040872 and QP060575 Duke Power - Purchase Order # NM18494 001, NM25245, 00108212 and 00124677. Based on the above, the Nuclear Utilities need to be notified concerning this deviation so that an evaluation may be performed to determine if this constitutes a defect that could create a substantial safety hazard. | ||
| ENS 48004 | 25 May 2012 04:00:00 | 10 CFR 21.21(d)(3)(i), Failure to Comply or Defect | Part 21 Report - Published Flow Coefficients for Pilot Operated Relief Valves Greater than Actual | This is to notify the US Nuclear Regulatory Commission, in accordance with the provisions of 10CFR-Part 21 of a potential deviation identified by Flowserve Corporation. On May 25, 2012, Flowserve Corporation notified Exelon - Byron and Braidwood Nuclear Power Stations of the results of a Steam flow test performed by Wyle Labs to confirm steam flow capacity against that specified for WKM PORVs. A refurbished size 6 class 900 WKM model 70-19-9 angle pattern control valve with 4" linear trim had been tested to determine its maximum steam flow capacity. The maximum steam flow was approximately 74% of the anticipated flow rate based on the original published WKM rated Cv for the valve. Background and Discussion: As part of their power up rate project, Exelon Braidwood Station procured larger 4" linear valve trim from Flowserve, Raleigh for their steam generator PORVs to increase the valve's steam flow capacity. The original valves were supplied by ACF Industries, WKM Valve Division in the mid to late 1970's with 3" linear trim. The designs for the WKM PORV valves were subsequently acquired by Flowserve. The larger trim was installed in a site spare PORV and tested for steam flow capacity. CFD analysis performed by Flowserve on the tested valve with the 4" trim, determined a Cv value that would yield steam flows similar to the test results. Analysis of the valve with the original 3" trim produced similarly reduced Cv values. Based on this testing and subsequent CFD analysis it appears that the originally published WKM rated Cv values for WKM angle control valve model 70-19-9 for sizes greater than size 2 are higher than actual values. The actual Cv's are believed to be 65% to 75% of the original WKM ratings. Conclusion: Based on the above, the Nuclear Industry needs to be notified concerning this deviation so that an evaluation may be performed to determine if this constitutes a defect that could create a substantial safety hazard. Although Flowserve subsequently acquired the rights to the WKM PORV designs, it does not have the historic sales records from ACF industries (WKM Valve Division). The total number of WKM valves potentially affected and their installed locations are not known. | Steam Generator |