DEFECT CHARACTERISATION ON ZIRCONIUM ALLOY BAR … · DEFECT CHARACTERISATION ON ZIRCONIUM ALLOY...
1
R.K. Chaube, S. Acharya, S.Krishna Reddy, A.Lakshminarayana, B.Prahlad, N. Saibaba Nuclear Fuel Complex, Hyderabad-500062, India DEFECT CHARACTERISATION ON ZIRCONIUM ALLOY BAR MATERIAL INTRODUCTION: Nuclear Fuel complex (NFC), Hyderabad manufactures zirconium alloy structural and components for Nuclear Reactors. End plug, being used for encapsulation of Pressurized Heavy Water Reactor (PHWR) cladding tubes, is considered as important core component. These are machined from Zirconium alloy bar material. The challenge in manufacturing of end plug material was to avoid micro flaws and porosities, which affect the integrity of fuel element. The bar material undergoes non-destructive examination (NDE) such as ultrasonic (UT) as well as eddy current testing (ET) at finished stage with stringent defect standard to avoid presence of these defect on end plugs, which otherwise may affect the fuel element integrity. Finally the fuel elements are being subjected to helium leak testing after assembly. To ascertain the root cause, investigations were carried out on defective bar material detected during UT and ET examination. Multi-NDE techniques as well as characterization tools such as scanning electron microscope (SEM) with energy dispersive spectroscopy (EDS), optical microscope, X-ray residual stress measuring system etc. were used for the characterization. Detailed studies carried out showed presence of chain like porosities, micro-piping and inclusions in axial direction of the bar material. Some of the findings of these investigations were given clue for required modification of quality assurance plan (QAP) at all stages starting from ingots to finished bars. The paper describes the findings of defect characterization using several NDE techniques and as well as characterization tools. Fig.1 Process flow sheet •UT & Visual • Chemical Analysis •UT & Visual • ECT • Chemical Analysis •Tensile test • Metallography • Corrosion test Zirconium alloy ingot Extrusion to billets Extrusion to bars Swaging Annealing Straightening & grinding Finished Bar QUALITY ASSURANCE AT INGOT STAGE The ingots are UT tested by contact method using a 2-MHz probe. A reference defect standard consists of flat bottom hole (FBH) of 3.2 mm Ø and 12.5-mm length is used. Visual examination is carried out on the machined surface for defects such as porosities, blowholes, cavities. etc. QUALITY ASSURANCE AT BILLET STAGE Billets are visually inspected for surface defects such as open cracks, porosities etc. Each billet is ultrasonically tested using a 4-MHz probe. The reference defect standard consists of 1.6 mm Ø FBH and 12 mm deep in axial direction. Specimens corresponding to top, middle and bottom of every ingot are chemically analyzed for alloying elements and about 20 impurities. Specimens from every batch subsequent to beta quenching are examined to check uniformity of grains. QUALITY ASSURANCE AT FINISHED STAGE UT is done with tightened naturaldefect standard in immersion method. Probe frequencies used are 10-MHz and 15-MHz. ECT is performed with a differential OD encircling coil using an automatic driving system. Frequency of 30-KHz is used to ensure depth of penetration upto 3-mm. Fig. 5a Fig. 5b Fig. 5c Fig. 5d Fig. 6 Fig. 5 (a-c) Compares ECT signals from different defect standard with a typical defect indication in bar material [Fig. 7d]. The same defect was revealed in UT-B scan at a depth of about 3 mm [Fig. 6]. 0.4-mmø radial hole 0.3-mmø radial hole Natural defect std. Defective portions were also investigated by X-ray radiography which revealed high-density material in the defective region. Examination under SEM revealed the presence of a foreign particle entrapment along with adjacent void region [Fig. 7a]. The EDS confirmed it as Tantalum [Fig. 7b]. Fig. 7a Fig. 7b Fig.9c Fig. 8a Fig. 8b Fig. 8c Fig.9a Fig. 9b Fig.8d Fig. 8e Fig. 9d Fig. 9e Several defective portions were examined under optical microscope & SEM. Chain-like group porosities [Fig. 8(a-e)] as well as individual porosities & Zr- based inclusions (sizes 30-m to 100-m) were revealed [Fig. 9(a-e)]. Fig. 10a Fig. 10b Defects detected by conventional UT were confirmed using advance NDE like phased array UT. Fig. 10a & 10b represents results of defect free & defective bar respectively. Innovative approaches made in QA procedures and modifications incorporated with respect to testing and defect standards improved the end plug quality. Very fine defects detected by conventional NDE methods were confirmed using advance NDE techniques. Microscopic studies established that fine porosities and flaws of the order of 30-μm can be successfully detected by the present test systems. CONCLUSIONS Several modifications in the quality assurance procedures as well as in the manufacturing process were incorporated based on above characterizations. Adoption of triple melting instead of double & allotment of middle part of heat for bar Avoiding addition of ZrO 2 powder to exclude presence of inclusions Tightening of UT defect standard in billet stage using a 0.8-mm Ø flat bottom hole replacing 1.6-mm Ø and introduction of UT at intermediate stage Introduction of natural defect standard to tighten defect standard in U T and E T of finished bar and introduction of ECT to improve sub-surface defect detection. Process modification by introducing barrel straightening instead of 3X3 cross roller straightening to reduce the amount of residual stress being introduced. MODIFICATION IN QA PLAN Fig.2 UT system for ingots & billets Fig.3 ECT system for bars Fig.4 Driving system for ECT Specimens from every lot are evaluated for metallurgical and mechanical properties. Hydrogen, nitrogen and oxygen content are also analyzed to have better control on the in-service material performance. Typical natural defect DEFECT CHARACTERISATION Fig. 11a Fig. 11b Fig. 12a Fig. 12b The bar material shown continuous indication in ECT [Fig. 11a] were characterized in optical microscopy which revealed heavy deformation twinning at the peripherial region (0.05-mm) of bar cross-section [Fig. 12a]. On the other hand ECT accepted bar material revealed no such twinning. Finally it was confirmed that residual stress introduced by 3X3 cross roller straightener was responsible for such indication in ECT.
Transcript of DEFECT CHARACTERISATION ON ZIRCONIUM ALLOY BAR … · DEFECT CHARACTERISATION ON ZIRCONIUM ALLOY...
R.K. Chaube, S. Acharya, S.Krishna Reddy, A.Lakshminarayana, B.Prahlad, N. Saibaba Nuclear Fuel Complex, Hyderabad-500062, India
DEFECT CHARACTERISATION ON ZIRCONIUM ALLOY BAR MATERIAL
INTRODUCTION: Nuclear Fuel complex (NFC), Hyderabad manufactures zirconium alloy structural and components for Nuclear Reactors. End plug, being used forencapsulation of Pressurized Heavy Water Reactor (PHWR) cladding tubes, is considered as important core component. These are machined from Zirconium alloy barmaterial. The challenge in manufacturing of end plug material was to avoid micro flaws and porosities, which affect the integrity of fuel element. The bar material undergoesnon-destructive examination (NDE) such as ultrasonic (UT) as well as eddy current testing (ET) at finished stage with stringent defect standard to avoid presence of thesedefect on end plugs, which otherwise may affect the fuel element integrity. Finally the fuel elements are being subjected to helium leak testing after assembly. To ascertain theroot cause, investigations were carried out on defective bar material detected during UT and ET examination. Multi-NDE techniques as well as characterization tools such asscanning electron microscope (SEM) with energy dispersive spectroscopy (EDS), optical microscope, X-ray residual stress measuring system etc. were used for thecharacterization. Detailed studies carried out showed presence of chain like porosities, micro-piping and inclusions in axial direction of the bar material. Some of the findings ofthese investigations were given clue for required modification of quality assurance plan (QAP) at all stages starting from ingots to finished bars. The paper describes thefindings of defect characterization using several NDE techniques and as well as characterization tools.
Fig.1 Process flow sheet
•UT & Visual• Chemical Analysis
•UT & Visual• ECT• Chemical Analysis•Tensile test• Metallography• Corrosion test
Zirconium alloy ingot
Extrusion to billets
Extrusion to bars
Swaging
Annealing
Straightening & grinding
Finished Bar
QUALITY ASSURANCE AT INGOT STAGEThe ingots are UT tested by contact method using a 2-MHz probe. A referencedefect standard consists of flat bottom hole (FBH) of 3.2 mm Ø and 12.5-mm lengthis used.
Visual examination is carried out on the machined surface for defects such asporosities, blowholes, cavities. etc.
QUALITY ASSURANCE AT BILLET STAGEBillets are visually inspected for surface defects such as open cracks, porositiesetc.
Each billet is ultrasonically tested using a 4-MHz probe. The reference defectstandard consists of 1.6 mm Ø FBH and 12 mm deep in axial direction.
Specimens corresponding to top, middle and bottom of every ingot are chemicallyanalyzed for alloying elements and about 20 impurities.
Specimens from every batch subsequent to beta quenching are examined to checkuniformity of grains.
QUALITY ASSURANCE AT FINISHED STAGEUT is done with tightened naturaldefect standard in immersion method. Probe frequencies used are 10-MHz and 15-MHz.
ECT is performed with a differential OD encircling coil using an automatic driving system. Frequency of 30-KHz is used to ensure depth of penetration upto 3-mm.
Fig. 5a Fig. 5b Fig. 5c Fig. 5d Fig. 6
Fig. 5 (a-c) Compares ECT signals from different defect standard with a typical defect indication in bar material [Fig. 7d].
The same defect was revealed in UT-B scan at a depth of about 3 mm [Fig. 6].
0.4-mmø radial hole 0.3-mmø radial hole Natural defect std.
Defective portions were also investigated by X-ray radiographywhich revealed high-density material in the defective region.
Examination under SEM revealed the presence of a foreignparticle entrapment along with adjacent void region [Fig. 7a]. TheEDS confirmed it as Tantalum [Fig. 7b].
Fig. 7a
Fig. 7b
Fig.9c
Fig. 8a Fig. 8b Fig. 8c
Fig.9a Fig. 9b
Fig.8d Fig. 8e
Fig. 9d Fig. 9eSeveral defective portions were examined under optical microscope & SEM.
Chain-like group porosities [Fig. 8(a-e)] as well as individual porosities & Zr- basedinclusions (sizes 30-m to 100-m) were revealed [Fig. 9(a-e)].
Fig. 10a Fig. 10b
Defects detected by conventional UT wereconfirmed using advance NDE like phasedarray UT. Fig. 10a & 10b represents results ofdefect free & defective bar respectively.
Innovative approaches made in QA procedures and modifications incorporated withrespect to testing and defect standards improved the end plug quality.
Very fine defects detected by conventional NDE methods were confirmed usingadvance NDE techniques.
Microscopic studies established that fine porosities and flaws of the order of 30-µm canbe successfully detected by the present test systems.
CONCLUSIONS
Several modifications in the quality assurance procedures as well as in the manufacturingprocess were incorporated based on above characterizations.
Adoption of triple melting instead of double & allotment of middle part of heat for bar Avoiding addition of ZrO2 powder to exclude presence of inclusions Tightening of UT defect standard in billet stage using a 0.8-mm Ø flat bottom holereplacing 1.6-mm Ø and introduction of UT at intermediate stage Introduction of natural defect standard to tighten defect standard in U T and E T of finishedbar and introduction of ECT to improve sub-surface defect detection. Process modification by introducing barrel straightening instead of 3X3 cross rollerstraightening to reduce the amount of residual stress being introduced.
MODIFICATION IN QA PLAN
Fig.2 UT system for ingots & billets
Fig.3 ECT system for bars
Fig.4 Driving system for ECT
Specimens from every lot are evaluated for metallurgical and mechanical properties.
Hydrogen, nitrogen and oxygen content are also analyzed to have better control onthe in-service material performance.
Typical natural defect
DEFECT CHARACTERISATION
Fig. 11a Fig. 11b
Fig. 12a Fig. 12b
The bar material shown continuous indication inECT [Fig. 11a] were characterized in opticalmicroscopy which revealed heavy deformationtwinning at the peripherial region (0.05-mm) of barcross-section [Fig. 12a]. On the other hand ECTaccepted bar material revealed no such twinning.
Finally it was confirmed that residual stressintroduced by 3X3 cross roller straightener wasresponsible for such indication in ECT.
