INSPECTION TECHNIQUE FOR BWR CORE SPRAY THERMAL SLEEVE WELD

Transcription

1 More Info at Open Access Database INSPECTION TECHNIQUE FOR BWR CORE SPRAY THERMAL SLEEVE WELD ABSTRACT J.L. Fisher, G. Light, Jim Crane, Albert Parvin, Southwest Research Institute, USA Yoshimasa Sugawara, Tokyo Electric Power Company, Japan The core spray piping system in typical boiling water reactor (BWR) pressure vessels includes nozzles with thermal sleeves. Each sleeve consists of a carbon steel pipe welded to a stainless steel pipe [a dissimilar metal (DM) weld], which in turn is welded to a connecting head that feeds coolant to the remainder of the core spray system in the vessel. These welds are typically made in the fabrication shop, where they are machined flush to the pipe surfaces. The carbon steel pipe is welded at the field site to the carbon steel vessel nozzle. Usually, the outer surface of this carbon steel field weld is left in the as-welded condition. Stress corrosion cracking (SCC) has been found in other lightwater reactor locations with DM welds and also with stainless steel-to-stainless steel welds. Access to the outer surface of the thermal sleeve for inspection could be achieved from the vessel interior, by inserting a probe into the annulus between the thermal sleeve and the nozzle. However, due largely to the limited access conditions, these welds are not generally inspected. In a series of projects with Tokyo Electric Power Company (TEPCO), SwRI developed a laboratory prototype system for this inspection problem. The system includes a thin, flexible probe delivery mechanism to insert probes into the annulus, a thin (7 mm thick) ultrasonic probe, and a video inspection system that could also be inserted into the annulus. The system was successfully demonstrated on a mockup with implanted SCC flaws. INTRODUCTION The core spray piping system used in boiling water reactor (BWR) pressure vessels is a key component of the plant operating and safety equipment. This system includes a thermal sleeve, which consists of a carbon steel pipe welded to a stainless steel pipe, which is in turn welded to a connecting head that feeds coolant to the sparger feeder pipes of the core spray system. These welds are typically made in the fabrication shop, where they are machined smooth to the pipe surfaces. The carbon steel pipe is welded at the site to the vessel nozzle. Usually the outer surface of this carbon steel field weld is left in the as-welded condition. Stress corrosion cracking (SCC) has been found in other BWR and pressurized water reactor (PWR) locations with dissimilar metal (DM) welds and also with stainless steel-to-stainless steel welds. However, due to very limited access conditions for the thermal sleeve DM and stainless-to-stainless metal welds, there is no known inspection system for these welds. An investigation was conducted to develop a method that could be used to examine the DM and stainless-to-stainless steel welds for cracks extending 50 percent through wall and oriented either parallel or transverse to the weld. GEOMETRY DETERMINATION AND ACCESS STUDY A key step of the investigation was to define the range of geometries that effect the inspection and the accessibility. There are many different core spray thermal sleeve configurations, including different gaps between the nozzle bore and the thermal sleeve piping, different thermal sleeve pipe wall thicknesses, and different axial locations of the welds with respect to the inside surface of the BWR nozzle. Access to the metal welds in the core spray system was determined to be possible by reaching into the narrow annular gap formed by the bore of the core spray nozzle and the outside of the core spray thermal sleeve, or by scanning from the outside surface of the nozzle. Initial testing from the outside surface of the core spray sleeve was deemed not to be adequate. Therefore, the internal access as illustrated in Figure1 was investigated. 165

2 MOCKUP DESIGN An annulus gap mockup was fabricated to simulate the dissimilar metal butt weld, carbon steel butt weld, and nozzle annulus for the purpose of assessing the NDE delivery methods and evaluating the viability of visually inspecting the nozzle annulus gap prior to ultrasonic testing (UT) probe inspection. The annulus gap mockup (illustrated in Figure 2) was constructed from 316 stainless steel and A508 forged carbon steel to match actual plant construction. The mockup was designed and constructed to have interchangeable annulus sleeves to allow for simulation of multiple annulus gap sizes and offsets. A total of three annulus sleeves representing annulus gaps of 8 mm, 15 mm, and 42 mm were constructed. These three sleeve geometries represent 81% (37 of 46) of all thermal sleeve gaps of TEPCO plants. The remaining nozzles are categorized as follows: 13% (6 of 46) with gaps below 7 mm (extremely challenging) and 7% (3 of 46) with gaps greater than 42 mm (not as challenging but requiring a different scanner approach). A series of internal and external EDM notches and external circumferentially-oriented implanted cracks to simulate SCC were placed in the DM weld and heat-affected zone. Ultrasonic Sensor B Wall Thickness A Similar Metal Weld Dissimilar Metal Weld Weld Location Figure 1. Illustration of access to the dissimilar and similar metal welds in the core spray system by (a) reaching into the narrow annular gap formed by the bore of the core spray nozzle and the outside of the core spray thermal sleeve. The inspection plan is to first inspect the annulus with an articulating videoprobe to evaluate the condition of the annulus gap and also to detect signs of SCC that might be visible on the OD surface of the thermal sleeve. Visual inspection prior to UT probe insertion would also reduce the potential for damaging the probe delivery device. To test the videoprobe device, three visual indication marks, each approximately 1.5 mm x 5 mm, were placed on the outer diameter of the dissimilar metal weld. These marks represent through-wall cracks. Carbon Steel Base 8in NPS Pipe Sch. 60 Sections (Stainless Steel and Carbon Steel) OD = 8.625in (21.9cm) Wall Thickness = 10.3mm (.41in) Dissimilar Metal Buttweld Carbon Steel Metal Buttweld Water Water Interior Adjustable Gap 12in NPS Pipe Sch. 40 OD = 12.8in (32.4cm) ID = 12.0in (30.5cm) Adjustable Gap Sheet Metal Sleeve Figure 2. Thermal sleeve geometry mockup. The mockup can be oriented vertically and filled with water to simulate inspection conditions. 166

3 SYSTEM CONCEPT This inspection system requires access to the interior of the vessel and use of the refueling crane for delivery. The inspection system concept design places a low profile conventional bulk wave transducer inside the annular gap of the core spray nozzle, as illustrated in Figure 3. The transducer is designed to scan the surface close to the weld of interest, which increases its ability to detect and characterize a defect. In addition, the interior scanning unit would include visual capabilities to allow the operator to see the physical condition of the annular gap and DM weld prior to inspection with the ultrasonic sensor. Figure 3. Illustration of a low profile bulk wave transducer inside the annular gap of the nozzle The scanning concept consists of 5 major subsystems. The subsystems are described below. It should be noted that in these descriptions, the designations axial, radial, and circumferential are with respect to the core spray nozzle. (1) Probe subsystem. There are 2 different probe systems: (1) UT probe subsystem and (2) Video testing (VT) probe subsystem. The UT probe subsystem consists of a flexible support member to which the UT sensor is attached. The flexible support member is used to transport the UT sensor into the annular gap and provide good contact between the UT transducer and thermal sleeve. The VT probe system is designed to support and transport a video capture device into the annular gap. The UT probe subsystem is shown in Figure 4. It includes a 45-degree, in diameter transducer. The VT probe subsystem model is shown in Figure 5. It has a CCD imager camera with a resolution of 500 x 582 pixels. Lighting is provided by 2 LEDs mounted on either side of the camera. Figure 4. UT probe subsystem, with a breadboard track. The probe and track are only approximately 5 mm tall, to fit in a 7 mm gap 167

4 LEDs Camera VT Probe Head Figure 5. VT Probe subsystem, showing camera and LEDs for lighting. The manufactured subsystem could also fit into a 7 mm gap. (2) Sensor Delivery Transport (SDT) subsystem. The SDT subsystem is designed to deliver the probe subsystem into the annular gap. It includes axial and circumferential drive motions. The design concept uses 2 SDT subsystems, oriented 180 apart circumferentially. A semirigid single bend retractable prototype device was developed and integrated with the rest of the system. It consists of a thin concave sheet of material (main body) placed in a guide track to assist in directing and making the probe main body rigid. The guide track has a drive assembly attached, which extends and retracts the main body into and out of the annulus. The main body and guide track are shown in Figure 6. The drive assembly consists of a belt drive that is driven by micro servo motors. An optical encoder was used to obtain the position of the probe main body. A rigid interchangeable pre-shaped guide was used to direct the probe s main body into and out of the annular gap of the annular gap mockup. (3) Circumferential Scanning (CS) subsystem. The CS subsystem is designed to adjust the position of the SDT to match the diameter of the core spray nozzle thermal sleeve. The CS subsystem will be used to revolve the SDT subsystem 180 around the circumference of the nozzle during inspection, since each SDT subsystem can only inspect approximately 180 without this motion. In addition, as described later, it will provide radial motion of the SDT subsystem that will be used as part of the SDT insertion process. The CS subsystem is shown in Figure 6. Figure 6. Circumferential Scanning subsystem with Sensor Delivery Transport subsystem attached. (4) Positioning subsystem (PS). The positioning subsystem will align the CS subsystem with the core spray nozzle thermal sleeve, in order to deliver the sensor into the annular gap. The PS subsystem would attach to the sparger feeder pipe with a specialized clamping mechanism to provide registration of the entire system with respect to the vertical location of the nozzle once in place, suction cups also located on the PS will deploy against the wall to secure the device and minimize any possible load on the feeder pipes. Also, a centering, clamping mechanism will concentrically locate the system with the core spray nozzle, by clamping on the nozzle header. Once the PS subsystem is engaged, the CS subsystem will be engaged to deliver the SDT probe. A solid model of the PS is shown in Figure

5 Figure 7. PS subsystem shown with clamp mechanism to provide registration to the feeder pipe, suction cups to maintain position on the vessel wall. The CS and SDT subsystems are also shown. (5) Transport subsystem. The transport subsystem is designed to deliver the overall unit to the core spray thermal sleeve, and will utilize the refueling crane. This subsystem was not prototyped. For a better understanding of the interior scanner system, Figure 8 identifies the five major interior scanner subsystems. Transport Subsystem Positioning Subsystem Circumferential Scanning (CS) Subsystem Sensor Delivery Transport (SDT) Subsystem Probe Subsystem Figure 8. Interior scanning system concept showing the 5 major subsystems The completed portions of the system that were manufactured (probe subsystem, SDT, CS) are shown in Figure 9. Figure 9. Assembled system components, mounted on the thermal sleeve mockup 169

6 UT INSPECTION RESULTS Table 1 provides a summary of the flaws in the thermal sleeve mockup, as measured. All of these flaws were located in the outer surface of the mockup. Table 1. Summary of the flaws in the mockup Flaw Type Orientati on Flaw Location Lengt h (mm) Depth (mm) Width (mm) Distance From Weld C/L Degree Location mm Location S1 SCC* CIRC WELD C/L N/A N/A S2 SCC CIRC WELD C/L N/A N/A S3 SCC CIRC S/S N/A S4 SCC CIRC C/S N/A C1 EDM NOTCH CIRC C/S C2 EDM WELD CIRC NOTCH C/L N/A C3 EDM NOTCH CIRC S/S As expected, the SCC crack response was much lower (by approximately 6 db) than the EDM notch response, and detection through the stainless steel side of the weld was more difficult than detection through the carbon steel side. For the nozzles investigated, it should be possible to inspect from both sides of the weld. All EDM notches were easily detected. An example of a crack response is shown in Figure 10. Figure 10. Left - Crack S1 detected from carbon steel side of the weld. Right - Crack S1 detected from stainless steel side of the weld. VT INSPECTION RESULTS The VT probe subsystem was scanned over a portion of the exterior mockup, in the region of flaw C3. Figure 11 and Figure 12 show images of this region. In Figure 11, the flaw is visible; in Figure 12 the label C3 is clearly visible. 170

7 Flaw C3 Figure 11. VT probe in the region of Flaw C3, which is visible in the image Figure 12. VT probe moved slightly away from Flaw C3. The label C3 is clearly visible in the image CONCLUSIONS Based upon the testing conducted, the following conclusions were reached. (1) A new prototype interior scanner consisting of the UT probe subsystem, VT probe subsystem, Sensor Delivery Transport subsystem, and Circumferential Scanner subsystem, was designed and manufactured. The experimental results indicate that the concepts developed in these projects can be applied to inspection of the dissimilar and similar metal welds in typical BWR core spray thermal sleeves. (2) The UT and VT probe subsystems successfully handle annulus gaps of approximately 7 mm to 22 mm. (3) All implanted crack defects were detected. In order to ensure high reliability of crack detection, the dissimilar metal weld should be inspected from both the carbon steel side and the stainless steel side ACKNOWLEDGEMENTS This work was sponsored by the Tokyo Electric Power Company. 171