Enrico Lucon

Direct Comparison of Single-Specimen Clamped SE(T) Test Methods on X100 Line Pipe Steel

The clamped single edge-notched tension (SE(T)) specimen has been widely used in a single-specimen testing scheme to generate fracture resistance curves for high strength line-pipe steels. The SE(T) specimen with appropriate notch geometry is a low-constraint specimen designed to reduce conservatism in the measurement of fracture toughness. The crack driving force is taken as either the J-integral or crack tip opening displacement (CTOD); it is generally accepted that the two parameters are interchangeable and equivalent using a simple closed form solution. However, the assumption that they are interchangeable, and to what extent, hasn’t been previously investigated experimentally on the same SE(T) specimen. This paper presents multiple test methods that were simultaneously employed on the same SE(T) specimens. The instrumentation includes: clip-gauges to measure surface crack mouth opening displacements (CMOD) and CTOD by the double-clip-gauge method; strain-gage arrays for direct J-integral measurements; and direct-current potential-drop (DCPD) instrumentation for supplementary crack size measurement. A direct comparison of ductile crack-growth resistance curves generated using J-integral and CTOD is presented here where each represents a different experimental and analytical approach. The two methods are in reasonable agreement over a narrow range of crack growth, differing slightly at initiation and diverging with increasing crack growth. Analysis of the supplementary instrumentation (i.e., strain gages, extensometers and DCPD) will be provided in a future publication.© 2014 ASME

On the Effectiveness of the Dynamic Force Adjustment for Reducing the Scatter of Instrumented Charpy Results

One of the key factors for obtaining reliable instrumented Charpy results is the calibration of the instrumented striker. An interesting alternative to the conventional static calibration recommended by the standards is the dynamic force adjustment (DFA), in which forces and displacements are iteratively adjusted until equality is achieved between absorbed energies calculated under the test record (Wt) and measured by the machine encoder (KV). In this study, the procedure has been applied to the instrumented data obtained by ten international laboratories using notched and precracked Charpy specimens, in the framework of a Coordinated Research Project (CRP8) of the International Atomic Energy Agency (IAEA). DFA is extremely effective in reducing the between-laboratory scatter for both general yield and maximum forces. The effect is less significant for dynamic reference temperatures measured from precracked Charpy specimens using the Master Curve procedure, but a moderate reduction of the standard deviation is, however, observed. It is shown that striker calibration is a prominent contribution to the interlaboratory variability of instrumented impact forces, particularly in the case of maximum forces.

Evaluation of Bias for Two Charpy Impact Machines with the Same Instrumented Striker

Two Charpy machines were used to test NIST verification specimens at three energy levels: low energy (∼15 J at −40°C), high energy (∼100 J at −40°C), and super-high energy (∼240 J at room temperature). The study evaluates the differences observed for the bias between two impact machines and the variation in test data for instrumented versus non-instrumented impact tests. The machines used for testing were of very similar design, and all tests were performed with the same instrumented striker (switched between machines). After testing, the raw force/time data were used for the analyses, without correcting instrumented data by matching absorbed energies measured by the machine encoder (KV) and calculated under the force/deflection test record (Wt). The characteristic forces at general yield (Fgy) and the maximum forces (Fm) were determined in accordance with ASTM E2298-09 from the instrumented impact record that was used to calculate the total impact energy (Wt). The findings show the following: (1) one machine consistently produced higher absorbed energy values than the other machine; (2) the variation in Wt is significantly lower than the variation in absorbed energy measured in the non-instrumented test (KV) for a given machine and energy level; (3) the relative differences between KV and Wt increased with increasing absorbed energy levels; (4) variations in maximum force are lower than variations in absorbed energy values; (5) instrumented data indicate that the variation in the curves is very small up to maximum force, and that differences in absorbed energy mainly occur during fracture propagation (post-maximum force data); (6) data from these two independent measures of absorbed energy indicate that scatter is due primarily to material variability; and (7) the bias between the two machines is significantly reduced when the same striker is used for testing.

Analysis of the Belgian Surveillance Fracture Toughness Database Using Conventional and Advanced Master Curve Approaches

The “classical” regulatory approach to the analysis of surveillance capsules in nuclear power plants entails an indirect estimate of the fracture toughness of the beltline materials, by inferring rather than measuring their toughness properties. Indeed, the irradiation-induced shift of the fracture toughness curve is assumed to be equal to the shift of the Charpy absorbed energy transition curve at a predefined level (41 J). An alternative surveillance approach, primarily based on direct fracture toughness measurements in the ductile-to-brittle transition region using the Master Curve procedure, has been applied to surveillance materials from several Belgian nuclear power plants in the past 15 years. This has led to the establishment of a significant database, consisting of 292 fracture toughness data points for 23 material conditions (unirradiated materials and surveillance capsules). In this study, different temperature normalization approaches are applied to the available data. The analyses show that data clearly follow the Master Curve formalism. Moreover, it is confirmed that both the static (KIc) and the dynamic (KIR) curves of the ASME Code Section XI provide an effective lower bound to the measured results, although more conservatism is evident when using RTNDT as the normalization parameter. Both the conventional (ASTM E1921-08, ‘‘Standard Test Method for Determination of Reference Temperature, T0, for Ferritic Steels in the Transition Range’’) and advanced (Multi-Modal) Master Curve analyses of the database clearly demonstrate that normalizing data by (T−RTT0) provides the best rationalization of the available information and the most effective representation of the experimental scatter.

Miniature Compact Tension Specimens for Upper Shelf Fracture Toughness Measurements on RPV Steels

Miniature compact tension, MC(T), specimens (thickness B = 4.15 mm) have been recently investigated at SCK⋅CEN for measuring the fracture toughness of steels under fully ductile conditions (upper shelf regime). The first results of this assessment, made by comparison with results obtained from 1TC(T) samples of various RPV steels with quite different toughness properties, have been described in a previous paper by the same authors. It was observed that, although a systematic and significant underestimation of both critical toughness values and crack propagation resistance was observed for the MC(T) samples with respect to the bigger specimens, it is possible to establish an empirical correlation between the two geometries which allows deriving acceptable estimates of the material’s critical toughness. In this work, further investigations are presented, related to: (1) the behavior of MC(T) and 1TC(T) specimens in the crack tip blunting phase; (2) the role of work hardening in decreasing the tearing resistance of MC(T) specimens; and (3) the use of alternative parameters (crack-tip opening displacement, crack-tip opening angle, Ernst’s modified J-integral JM). Evidence is also presented which could justify a revision of the specimen measuring capacity (Jmax) in the current ASTM Standard E1820-05, which appears too severe and could be relaxed. Further work will include validating on irradiated materials the empirical correlations established between 1TC(T) and MC(T) test results.

Towards Crack Arrest Testing Using Miniature Specimens

Crack arrest is an important concept that can be useful to guarantee the safety of reactor pressure vessels. In case of a pressurized thermal shock, a postulated crack can initiate and arrest due to a decrease in driving force combined with an increase in material toughness due to the thermal and neutron embrittlement gradient along the thickness. Due to the size of the specimen and the difficulty in obtaining valid results according to the ASTM E1221-06 standard, “Standard Test Method for Determining Plane-Strain Crack-Arrest Fracture Toughness, KIa, of Ferritic Steels,” current efforts to develop this technique for irradiated materials are very limited. However, advances in dynamic fracture modeling and elastic-plastic fracture mechanics open the possibility of using miniature crack arrest specimens. We have developed a stiff setup to perform crack arrest tests on precracked Charpy (PCCv) specimens. Different strategies were evaluated to provide enough reduction in driving force to produce arrest. Special attention is given to the starter notch and different starter notch preparations are investigated: precracking, chevron, and brittle weld. Testing configurations were found that guarantee crack arrest with sufficient remaining ligament. The dynamic loading condition is investigated experimentally using strain gauges. Finite element calculations of an arresting crack are also performed in order to provide more insight of the loading after the crack arrest event. Although it is not yet possible to derive the arrest fracture toughness from the PCCv testing, results are encouraging and open new perspectives in the field of crack arrest determination.

Fracture Toughness Characterization of High-pressure Pipe Girth Welds Using Single-edge Notched Tension [SE(T)] Specimens

The safety and reliability of large-diameter pipelines for the transport of fluid hydrocarbons is being improved by the development of high-strength steels, advanced weld technologies, and strain-based design (SBD) methodologies. In SBD, a limit is imposed on the applied strains rather than the applied stresses. For high-pressure pipelines, SBD requires an assured strength overmatch for the weld metal as compared to the base material, in order to avoid strain localization in the weldment during service. Achieving the required level of strength overmatch, as well as acceptable ductility and low-temperature fracture toughness, is a challenge as the pipe strength increases. Published studies show that low constraint geometries such as single-edge tension [SE(T)] or shallow-notched single-edge bend [SE(B)] specimens represent a better match to the constraint conditions of surface-breaking circumferential cracks in large-diameter pipelines during service (Shen, G., Bouchard, R., Gianetto, J. A., and Tyson, W. R., “Fracture Toughness Evaluation of High Strength Steel Pipe,” Proceedings of PVP2008, ASME Pressure Vessel and Piping Division Conference, Chicago, IL, July 27–31, ASME, New York, 2008). However, the SE(T) geometry is not included in any of the most widely used elastic-plastic fracture mechanics (EPFM) test standards. A procedure has been developed for performing and analyzing SE(T) toughness tests using a single-specimen technique that includes formulas for calculating the J-integral and crack-tip opening displacement, as well as for estimating crack size using rotation-corrected elastic unloading compliance. Here, crack-resistance curves and critical toughness values obtained from shallow-crack SE(T) specimens (a0/W ≈ 0.25) are compared to shallow-crack (a0/W ≈ 0.25) SE(B) specimens. We believe that the SE(T) methodology is mature enough to be considered for inclusion in future revisions of EPFM standards such as ASTM E1820 and ISO 12135, although additional work is needed to establish validity limits for SE(T) specimens.

Certification of NIST Room Temperature Low-Energy and High-Energy Charpy Verification Specimens.

The possibility for NIST to certify Charpy reference specimens for testing at room temperature (21 °C ± 1 °C) instead of −40 °C was investigated by performing 130 room-temperature tests from five low-energy and four high-energy lots of steel on the three master Charpy machines located in Boulder, CO. The statistical analyses performed show that in most cases the variability of results (i.e., the experimental scatter) is reduced when testing at room temperature. For eight out of the nine lots considered, the observed variability was lower at 21 °C than at −40 °C. The results of this study will allow NIST to satisfy requests for room-temperature Charpy verification specimens that have been received from customers for several years: testing at 21 °C removes from the verification process the operator’s skill in transferring the specimen in a timely fashion from the cooling bath to the impact position, and puts the focus back on the machine performance. For NIST, it also reduces the time and cost for certifying new verification lots. For one of the low-energy lots tested with a C-shaped hammer, we experienced two specimens jamming, which yielded unusually high values of absorbed energy. For both specimens, the signs of jamming were clearly visible. For all the low-energy lots investigated, jamming is slightly more likely to occur at 21 °C than at −40 °C, since at room temperature low-energy samples tend to remain in the test area after impact rather than exiting in the opposite direction of the pendulum swing. In the evaluation of a verification set, any jammed specimen should be removed from the analyses.

Investigation of Transition Fracture Toughness Variation within the Thickness of Reactor Pressure Vessel Forgings

An investigation has been conducted on the influence of the through-thickness sampling position on the tensile and fracture toughness properties of reactor pressure vessel forgings, using material from two actual pressurized water reactor vessels. The aim was to quantify the safety margins entailed by extracting surveillance samples from the 1/4T and 3/4T positions, as recommended by the current legislation. For each forging, seven layers have been considered: Inner surface, 1/8T, 1/4T, 1/2T, 3/4T, 7/8T, and outer surface; for each position, two tensile tests at room temperature and 16 fracture toughness tests in the ductile-to-brittle transition region have been performed. In terms of tensile properties, for both forgings the strength is higher at the surfaces than in the centre, while ductility (elongations and reduction of area) is substantially unaffected. For both materials, fracture toughness is better at the surfaces than in the central portion, although differences in terms of Master Curve reference temperatures are statistically not relevant at the 95 % confidence level. This effect is more pronounced for one of the materials, due to the larger amount of material removed with respect to the original heat-treated forging, and is qualitatively confirmed by metallographic observations. The results obtained from this study are in substantial agreement with similar studies found in the literature, although most authors have reported larger differences between surface and central layers.

IAEA Coordinated Research Project on Master Curve Approach to Monitor Fracture Toughness of RPV Steels: Final Results of the Experimental Exercise to Support Constraint Effects

The precracked Charpy single-edge notched bend, SE(B), specimen (PCC) is the most likely specimen type to be used for determination of the reference temperature, T0 , with reactor pressure vessel (RPV) surveillance specimens. Unfortunately, for many RPV steels, significant differences have been observed between the T0 temperature for the PCC specimen and that obtained from the 25-mm thick compact specimen [1TC(T)], generally considered the standard reference specimen for T0 . This difference in T0 has often been designated a specimen bias effect, and the primary focus for explaining this effect is loss of constraint in the PCC specimen. The International Atomic Energy Agency (IAEA) has developed a coordinated research project (CRP) to evaluate various issues associated with the fracture toughness Master Curve for application to light-water RPVs. Topic Area 1 of the CRP is focused on the issue of test specimen geometry effects, with emphasis on determination of T0 with the PCC specimen and the bias effect. Topic Area 1 has an experimental part and an analytical part. Participating organizations for the experimental part of the CRP performed fracture toughness testing of various steels, including the reference steel JRQ (A533-B-1) often used for IAEA studies, with various types of specimens under various conditions. Additionally, many of the participants took part in a round robin exercise on finite element modeling of the PCVN specimen, discussed in a separate paper. Results from fracture toughness tests are compared with regard to effects of specimen size and type on the reference temperature T0 . It is apparent from the results presented that the bias observed between the PCC specimen and larger specimens for Plate JRQ is not nearly as large as that obtained for Plate 13B (−11°C vs −37°C) and for some of the results in the literature (bias values as much as −45°C). This observation is consistent with observations in the literature that show significant variations in the bias that are dependent on the specific materials being tested. There are various methods for constraint adjustments and two methods were used that reduced the bias for Plate 13B from −37°C to −13°C in one case and to − 11°C in the second case. Unfortunately, there is not a consensus methodology available that accounts for the differences observed with different materials. Increasing the Mlim value in the ASTM E-1921 to ensure no loss of constraint for the PCC specimen is not a practicable solution because the PCC specimen is derived from CVN specimens in RPV surveillance capsules and larger specimens are normally not available. Resolution of these differences are needed for application of the master curve procedure to operating RPVs, but the research needed for such resolution is beyond the scope of this CRP.Copyright