ML060230395
| ML060230395 | |
| Person / Time | |
|---|---|
| Site: | North Anna  |
| Issue date: | 01/18/2006 |
| From: | Crawford S, Stephen Cumblidge, Doctor S, Harris R, Schuster G Battelle Memorial Institute, Pacific Northwest National Laboratory |
| To: | Office of Nuclear Reactor Regulation |
| References | |
| PNNL-SA-48066 | |
| Download: ML060230395 (53) | |
Text
NDE Measurements on CRDMs Removed from Service Steven R. Doctor Stephen Cumblidge George Schuster Susan Crawford Robert Harris Presented at NRC Program Review Rockville MD January 18, 2006
Acknowledgements Work funded by the Nuclear Regulatory Commission (NRC) Office of Nuclear Reactor Research.
NRC Project Manager Wallace Norris on JCN-Y6867 and Carol Moyer for JCN-Y6534 PNNL programs support the cooperative work between the NRC and the Electric Power Research Institute (EPRI)
CRDMs from North Anna 2 supplied by EPRI 2
Outline NDE Goals NDE Techniques used Calibration Standards and Results NDE Results for Nozzle #59 NDE Results for Nozzle #31 Proposed Cut Plan Conclusions 3
NDE Goals Use removed-from-service CRDMs (cut from the head of North Anna 2) containing PWSCC to assess the extent of degradation and effectiveness of NDE techniques.
Find all degradation and other flaws in the J-groove weld and buttering area and within the penetration tube wall under laboratory conditions.
Verify and expand on previous NDE work during ISI and by ISI teams in inspections conducted at PNNL in 2004.
Provide extremely precise and detailed information on the flaws to facilitate and guide developing a destructive test plan.
4
NDE Performed on CRDM Nozzles The interiors of both penetration tubes were inspected using eddy current (ET) and time of flight diffraction (TOFD). The penetration tube of Nozzle #59 was also inspected using visual testing (VT) via Microset.
The J-groove weld metal of both nozzles was volumetrically inspected using zero-degree immersion ultrasound.
The J-groove weld area of both nozzles was examined using visual testing via Microset.
5
Calibration Pieces The ET and TOFD techniques were calibrated using an alloy 600 tube with a series of EDM notches cut into the surfaces.
Notch Designation Surface Orientation Depth A ID Axial 2 mm (0.08 in.)
B ID Axial 4 mm (0.16 in.)
C ID Axial 8 mm (0.32 in.)
D ID Circumferential 2 mm (0.08 in.)
E ID Circumferential 4 mm (0.16 in.)
F ID Circumferential 8 mm (0.32 in.)
H OD Circumferential 2 mm (0.08 in.)
I OD Circumferential 4 mm (0.16 in.)
J OD Circumferential 8 mm (0.32 in.)
K OD Axial 2 mm (0.08 in.)
L OD Axial 4 mm (0.16 in.)
M OD Axial 8 mm (0.32 in.)
6
Calibration Pieces Continued A CRDM nozzle cut from the Midland pressure vessel head.
The Midland pressure vessel never saw service, so no service-induced flaws are present in the nozzle.
Contains a number of fabrication flaws.
7
Calibration Continued The VT equipment was tested using Microset replicas made from stainless-steel samples with a variety of artificial implanted cracks Crack opening dimensions (COD)
<10 microns (<0.0005 in.) to 150 microns (0.006 in.) wide.
The samples have been machined smooth, but some contain grinding marks.
8
NDE Methods - Eddy Current Testing Eddy Current testing is sensitive to the presence of surface-breaking cracks.
ET is effective in detecting non-surface breaking cracks if crack is near to the surface.
ET provides crack length, but not depth, information.
ET does not require coupling media but must be in close contact with the surface for best test sensitivity and at a constant distance to provide consistent test results.
Two frequencies, 150 and 350 kHz, and two gains, 15 dB and 30 dB, were used.
9
and 8 mm Notches and Scribe Marks 150 kHz, 15 dB Gain 350 kHz, 15 dB Gain 350 kHz, 15 dB Gain 10
TOFD Technique TOFD uses two ultrasonic transducers to detect features between them.
TOFD is extremely sensitive to cracking.
TOFD is used by industry for detection and depth sizing of cracks.
Inspection Angle Frequency Notes
(#) (degrees) (MHz) 1 60º 7.5 MHz Very sensitive, detects many innocuous fabrication flaws 2 60º 5.0 MHz Industry Standard 11
TOFD Calibration Lateral wave Notch A-scan View Side View C-scan View End View Flaw Orientation Flaw Location Depth Response (dB)
Circumferential Outer Diameter 2 mm (0.08 in.) 1.7 4 mm (0.16 in.) 4.2 Axial Outer Diameter 2 mm (0.08 in.) -4.5 4 mm (0.16 in.) -6.4 8 mm (0.32 in.) -6.8 Axial Inner Diameter 2 mm (0.08 in.) Not Detected 4 mm (0.16 in.) -0.4 8 mm (0.32 in.) -3.8 12
NDE Methods - Immersion Ultrasound Immersion ultrasound at zero degrees was used to examine the J-groove weld.
UT performance degrades as the ultrasound propagates through the J-groove weld and buttering because of the coarse-grained anisotropic weld metal.
Four frequencies were used, 5 MHz, 2.25 MHz, 1 MHz, and 500 kHz. The 5 MHz ultrasound is very sensitive to flaws in the fusion zone between the J-groove weld and the penetration tube, while the lower frequencies provide increasing sensitivity to flaws successively deeper into the weld metal 13
Immersion UT Calibration alibrated by examining the fabrication aws in the Midland CRDM using the Fabrication mersion zero-degree UT system. Flaws e found good sensitivity for Weld brication flaws at the fusion zone Passes nd were able to image individual eld passes deeper in the weld metal.
his scan was taken at 2.25 MHz.
14
Visual Testing via Microset Enhanced VT can be a useful tool in detecting and characterizing component surface features.
A Canon 1Ds Mk 2 camera with a 180 mm 1:1 macro lens was used.
Because the CRDMs are radioactive, contaminated, and have a complex geometry, the visual testing of the J-groove weld was performed using a Microset replica.
Microset is an epoxy-like polymer that is applied on a surface as a liquid and then it hardens, making an extremely high resolution replica of the surface that can capture details as small as 0.1 microns in size.
When photographing the Microset replicas, the camera was mounted on a graduated slide bar and a graduated tripod.
15
VT Calibration Results Cracks as small as 25 microns wide on the stainless steel samples were imaged using the Microset and the digital camera. The visual testing is much more sensitive to cracks that cut across machine marks and scratches and less sensitive to cracks that follow such surface features.
16
NDE Examination of Nozzle #59 The penetration tube interior surface was first examined with ET as it requires no coupling fluid.
The penetration tube interior surface was then examined using TOFD.
After the TOFD was completed the penetration tube was plugged at the bottom and filled with water to conduct the immersion UT testing of the J-groove weld and buttering.
Finally, the J-groove weld was covered in Microset and the Microset was removed and set aside for later visual examination.
This procedure was repeated for Nozzle #31.
17
Eddy Current Results for Nozzle #59 Indications consistent with shallow cracks have been detected on the ID of the Nozzle #59 penetration tube.
15 dB Gain 350 kHz 150 kHz 350 kHz TOFD Compatible Scan 18
Eddy Current Results for Nozzle #59 Large dynamic range in the ET scans allowed the 15 dB gain examinations to pick up scratches and other subtle fabrication conditions such as the conductivity change caused by the J-groove weld. 30 dB gain scans greatly magnified these innocuous indications, had saturated signals, and were less useful than the 15 dB gain examinations.
350 kHz, 15 dB Gain Display set to show small indications 19
Supplemental Visual Testing on Nozzle
- 59 Penetration Tube Based on the Eddy Current results, a Microset replica of the penetration tube interior was located and photographed.
It was determined that the interior was not smoothly machined, showed several deep axial scratches, pit-like indications, and contained a large rough patch coincident with the round eddy current indication.
Axial Scratches Circumferential Machining Marks 10 mm 20
Nozzle #59 Penetration Tube:
VT Indications Region of Pit-Like Indications Round Rough Indication 5 mm 5 mm 21
Nozzle #59 Penetration Tube:
VT CrackCrack-Like -Like Indication Crack-like indication located at 315° clockwise (45° CCW) rotation and 140 mm axially in the penetration tube.
10 mm Pit-Like Indications Crack-Like Indication 22
TOFD Results on Nozzle #59 Many Indications were found. The only ones with time-of-flight shapes were located in the weld metal. Other strong indications were detected and recorded.
TOFD Results, 5MHz, 60L, Nozzle 59 TOFD Results, 7.5MHz, 60L, Nozzle 59 250 200 Axial Position (mm)
WeldRepair Weld Drop Through Intrusion 150 WeldRepair Weld Drop Through Intrusion Indications Indications Region of Indications Region of Indications Region of Indications 100 Region of Indications 50 0
0 50 100 150 200 250 300 0 50 100 150 200 250 300 Circumferential Position (mm) Circumferential Position (mm) 23
Nozzle #59 Penetration Tube Summary Data fusion shows very little agreement.
24
Nozzle #59 JJ-Groove
-Groove Weld NDE The J-groove weld metal of Nozzle #59 was volumetrically examined using immersion zero degree UT and the surface of the weld was examined using VT via a Microset replica.
25
Immersion Ultrasound at Zero Degrees Results for Nozzle #59 #59-- 5 MHz Indications inside the penetration tube Indications inside the fusion zone Circ.
Circ.
360 deg.
deg.
Axial Axial 26
2.25 MHz Immersion UT Results for Nozzle #59 Indications inside the fusion zone Circ. Circ.
360 360 deg.
deg.
Axial Axial 27
1 MHz Immersion UT Results for Nozzle #59 Indications inside the fusion zone Circ.
360 deg.
Axial 28
500 kHz Immersion UT Results for Nozzle #59 Circ.
360 deg.
Axial 29
VT Results for Nozzle #59 JJ-Groove
-Groove Weld Only small crack-like indications were detected, possibly PWSCC or hot cracks 3 mm Possible Possible Micro Crack Micro Crack 3 mm 30
VT Results for Nozzle #59 JJ-Groove-Groove Weld One crack-like indication of roughly 1cm long was found.
5 mm 31
Nozzle #59 Summary Many indications were found in the penetration tube using TOFD and ET, but the TOFD and ET results are inconsistent as to where flaws are located.
VT found many axial scratches and pit-like indications that may have accounted for much of the ET results. One crack-like indication was discovered, although this was not solidly corroborated by the ET results.
The examination of the J-groove weld using VT and UT showed some crack-like indications in the weld.
UT found strong reflection from the interference fit. One region of ultrasonic transmission was seen but is likely the result of the construction technique used to make the interference fit.
32
Eddy Current Results for Nozzle #31 350 kHz We again found very high 15 dB Gain sensitivity with the 15 dB gain scan. We were able to detect the J-groove weld, the zero-degree scribe mark, and scratches possibly accidentally introduced by a mock-thermal sleeve during round-robin testing on this nozzle in 2004 No crack-like indications were detected.
33
TOFD Results on Nozzle #31 Several TOF shaped indications were detected. None were corroborated by the ET testing. Also, none of the indications showed a break in the lateral wave.
TOFD Results, 5MHz, 60L, Nozzle 31 TOFD Results, 7.5MHz, 60L, Nozzle 31 250 200 Axial Position (mm) 150 WeldIntrusion Weld Drop Through WeldIntrusion Weld Drop Through Indications Indications TOF Shape TOF Shape 100 Region of Indications 50 0
0 50 100 150 200 250 300 0 50 100 150 200 250 300 Circumferential Position (mm) Circumferential Position (mm) 34
Example TOFD Indications in Nozzle #31 Both at 7.5 MHz 35
Nozzle #31 Penetration Tube Summary No surface-breaking ID flaws detected in the penetration tube.
No ET indications, many TOFD indications in the penetration tube wall.
No breaks in the lateral wave were detected in the TOFD Data.
No VT was performed because there were no areas of strong interest to investigate.
36
Nozzle #31 JJ-Groove
-Groove Weld NDE The J-groove weld metal of Nozzle #31 was volumetrically examined using immersion zero degree UT and the surface of the weld was examined using VT via a Microset replica.
37
5 MHz Immersion UT Results for Nozzle #31 Circ.
360 deg.
Axial 38
2.25 MHz Immersion UT Results for Nozzle #31 Circ 360 deg.
Axial 39
1 MHz Immersion UT Results for Nozzle #31 Circ.
360 deg.
Axial Axial 40
500 kHz Immersion UT Results for Nozzle #31 Circ.
360 deg.
Axial 41
Visual Testing Results for Nozzle #31 10-millimeter long crack-like indication located at 100° CCW 5 mm 42
Visual Testing Results for Nozzle #31 5-10 millimeter long crack-like indication located at 90° CCW 10 mm 43
Visual Testing Results for Nozzle #31 Crack-like indication at 270° CCW 10 mm 44
Nozzle #31 Summary The penetration tube shows no strong ET indications and no breaks in the lateral wave using TOFD.
Immersion UT detected a very large indication in the J-groove weld metal that starts at 0° and continues to 50°.
The J-groove weld metal shows cracking at 90°-
100° via VT and an indication at 90° was detected by immersion UT.
VT detected a crack-like indication at 270° that was not corroborated by the immersion UT.
45
Nozzle #31 Summary In many places in the interference fit, the UT was well-coupled and produced a mottled pattern in the UT results.
This is in strong contrast to the results seen in Nozzle #59 and the calibration Midland CRDM, where the interference fit allowed very little coupling between the penetration tube and the carbon steel.
Nozzle #59 2.25 MHz Nozzle #31 2.25 MHz 46
Data Fusion With Previous Examinations Nozzle #59
- 59-- JJ-Groove
-Groove Weld Weld Surface Weld Interior ISI Cracking found using ET Small cracks found by PNNL VT Weld Repair Intrusion PNNL UT Strings of PNNL UT Indications Composite Diagram 47
Nozzle #59
- 59-- Penetration Tube 48
Data Fusion With Previous Examinations Nozzle #31 #31-- JJ-Groove
-Groove Weld Weld Surface Weld Interior ISI Cracking found using ET Small cracks found by PNNL VT Strings of PNNL UT Indications Note- Difficult Topography for ET Weld raphy Composite Diagram 49
Data Fusion With Previous Examinations Nozzle #31
- 31-- Penetration Tube 50
Proposed Cut Plan For Nozzle #59 The J-groove weld will be cut out and cut into quarters.
The penetration tube will be cut lengthwise along the 90°-270° plane and the regions that caused the ET and VT indications will be closely examined using direct-metal VT and UT.
51
Proposed Cut Plan For Nozzle #31 The J-groove weld will be removed and cut into quarters along the 50°-
230° plane and the 135°-315° plane.
The penetration tube will be cut in half along the 45° and 225° plane and investigated with UT and direct metal VT to determine if any cracks were missed.
52
Conclusions The PNNL NDE inspections found crack-like indications in both Nozzles #59 and #31 that along with the ISI and round robin indications need to be validated by DE.
z The penetration tube of Nozzle #59 showed crack-like indications z The J-groove weld and buttering of Nozzles #59 and #31 showed crack-like indications Supplemental NDE should be performed on cut pieces to guide the DE analysis and to assure that the areas to be examined are optimized before polishing and sectioning.
The J-groove weld of both Nozzles #59 and #31 showed evidence of embedded fabrication flaws such as lack-of fusion.
53