Engineering Process-Zone Model for Evaluation of Structural Strength of Fuel Channel Annulus Spacers in CANDU Nuclear Reactors

Author(s):  
Douglas Scarth ◽  
Steven Xu ◽  
Cheng Liu

The core of a CANDU(1) (CANada Deuterium Uranium) pressurized heavy water reactor consists of a lattice of either 390 or 480 horizontal Zr-Nb pressure tubes, depending on the reactor design. These pressure tubes contain the fuel bundles. Each pressure tube is surrounded by a Zircaloy calandria tube that operates at a significantly lower temperature. Fuel channel annulus spacers maintain the annular gap between the pressure tube and calandria tube throughout the operating life. To meet this design requirement, annulus spacers must have adequate structural strength to carry the interaction loads imposed between the pressure tube and calandria tube. Crush tests that have been performed on specimens from as-received and ex-service Inconel X-750 alloy spacers have demonstrated that the structural strength of Inconel X-750 spacers has degraded with operating time due to irradiation damage. There was a need for an engineering model to predict the future maximum load carrying capacity of the spacer coils for use in Fitness-for-Service evaluations of spacer structural integrity. An engineering process-zone model has been developed and used to analyze the spacer crush test results, and provide predictions of the Inconel X-750 spacer coil future maximum load carrying capacities. The engineering process-zone model is described in this paper. The process-zone model is based on the strip-yield approach of a process zone with a uniform restraining stress representing the fracture region that is surrounded by elastic material.

Author(s):  
Cheng Liu ◽  
Leonid Gutkin ◽  
Douglas Scarth

The core of a CANDU®(1) pressurized heavy water reactor consists of a lattice of either 390 or 480 horizontal Zr-Nb pressure tubes, depending on the reactor design, which contain the nuclear fuel. Each pressure tube is surrounded by a Zircaloy calandria tube that operates at a significantly lower temperature. Fuel channel annulus spacers maintain the annular gap between the pressure tube and the calandria tube throughout the reactor operating life. To meet this design requirement, the annulus spacers must have adequate structural strength to carry the interaction loads between the pressure tube and the calandria tube. Crush tests performed on specimens from Inconel X-750 spacers, both non-irradiated and ex-service, have demonstrated that their structural strength had degraded with operating time due to irradiation damage. An engineering process-zone model was developed and used to analyze the spacer crush test results, and to predict the maximum load carrying capacities of the Inconel X-750 spacer coils, as described in the companion paper “Engineering Process-Zone Model for Evaluation of Structural Strength of Fuel Channel Annulus Spacers in CANDU Nuclear Reactors” presented at the PVP2017 Conference. The developed model is based on the strip-yield approach of a process zone with a uniform restraining stress that represents the fracture region surrounded by elastic material. This baseline process-zone model has been improved by allowing the restraining stress to evolve with the variation in the opening displacement in accordance with a traction-separation constitutive relation. The development of this improved engineering process-zone model incorporating a non-trivial traction-separation constitutive relation is described in this paper.


Author(s):  
Steven X. Xu ◽  
Dennis Kawa ◽  
Jun Cui ◽  
Heather Chaput

In-service flaws in cold-worked Zr-2.5 Nb pressure tubes in CANDU(1) reactors are susceptible to a phenomenon known as delayed hydride cracking (DHC). The material is susceptible to DHC when there is diffusion of hydrogen atoms to a service-induced flaw, precipitation of hydrides on appropriately oriented crystallographic planes in the zirconium alloy matrix material, and development of a hydrided region at the flaw tip. The hydrided region could then fracture to the extent that a crack forms and DHC is said to have initiated. Examples of in-service flaws are fuel bundle scratches, crevice corrosion marks, fuel bundle bearing pad fretting flaws, and debris fretting flaws. These flaws are volumetric in nature. Evaluation of DHC initiation from the flaw is a requirement of Canadian Standards Association (CSA) Standard N285.8. This paper describes the validation of the weight function based process-zone model for evaluation of pressure tube flaws for DHC initiation. Validation was performed by comparing the predicted threshold load levels for DHC initiation with the results from DHC initiation experiments on small notched specimens. The notches in the specimens simulate axial in-service flaws in the pressure tube. The validation was performed for both un-irradiated and pre-irradiated pressure tube material.


Author(s):  
Cheng Liu ◽  
Leonid Gutkin ◽  
Douglas Scarth

Zr-2.5Nb pressure tubes in CANDU 1 reactors are susceptible to hydride formation when the solubility of hydrogen in the pressure tube material is exceeded. As temperature decreases, the propensity to hydride formation increases due to the decreasing solubility of hydrogen in the Zr-2.5Nb matrix. Experiments have shown that the presence of hydrides is associated with reduction in the fracture toughness of Zr-2.5Nb pressure tubes below normal operating temperatures. Cohesive-zone approach has recently been used to address this effect. Using this approach, the reduction in fracture toughness due to hydrides was modeled by a decrease in the cohesive-zone restraining stress caused by the hydride fracture and subsequent failure of matrix ligaments between the fractured hydrides. As part of the cohesive-zone model development, the ligament thickness, as represented by the radial spacing between adjacent fractured circumferential hydrides, was characterized quantitatively. Optical micrographs were prepared from post-tested fracture toughness specimens, and quantitative metallography was performed to characterize the hydride morphology in the radial-circumferential plane of the pressure tube. In the material with a relatively low fraction of radial hydrides, further analysis was performed to characterize the radial spacing between adjacent fractured circumferential hydrides. The discrete empirical distributions were established and parameterized using continuous probability density functions. The resultant parametric distributions of radial hydride spacing were then used to infer the proportion of matrix ligaments, whose thickness would not exceed the threshold value for low-energy failure. This paper describes the methodology used in this assessment and discusses its results.


Author(s):  
Steven X. Xu ◽  
Jun Cui ◽  
Douglas A. Scarth ◽  
David Cho

Flaws found during in-service inspection of Zr-2.5Nb pressure tubes in CANDU(1) reactors include fuel bundle scratches, debris fretting flaws, fuel bundle bearing pad fretting flaws and crevice corrosion flaws. These flaws are volumetric and blunt in nature. A key structural integrity concern with in-service blunt flaws is their susceptibility to delayed hydride cracking (DHC) initiation, particularly for debris fretting flaws under flaw-tip hydride ratcheting conditions. Hydride ratcheting conditions refer to situations when flaw-tip hydrides do not completely dissolve at normal operating temperature, and accumulation of flaw-tip hydrides occurs with each reactor heat-up/cool-down cycle. A significant number of in-service flaws are expected to be under hydride ratcheting conditions at late life of pressure tubes. DHC initiation evaluation procedures based on process-zone methodology for flaws under hydride ratcheting conditions are provided in CSA (Canadian Standards Association) N285.8-15. The process-zone model in CSA N285.8-15 predicts whether DHC initiation occurs or not for given flaw geometry and operating conditions, regardless of the number of reactor heat-up and cool-down cycles. There has been recent new development. Specifically, a cycle-wise process-zone model has been developed as an extension to the process-zone model in CSA N285.8-15. The cycle-wise process-zone model is able to predict whether DHC initiation occurs or not during a specific reactor heat-up and cool-down cycle under applied load. The development of the cycle-wise process-zone model was driven by the need to include flaw-tip stress relaxation due to creep in evaluation of DHC initiation. The technical basis for the development of the cycle-wise process-zone model for prediction of DHC initiation under flaw-tip hydride ratcheting conditions is described in this paper.


Author(s):  
Jun Cui ◽  
Gordon K. Shek ◽  
Douglas A. Scarth ◽  
William K. Lee

Flaws in Zr-2.5 Nb alloy pressure tubes of CANDU nuclear reactors are susceptible to a crack initiation and growth mechanism called Delayed Hydride Cracking (DHC), which is a repetitive process that involves hydrogen diffusion, hydride precipitation, growth of the hydrided region and fracture of the hydrided region at the flaw-tip. The presence of small surface irregularities, or secondary flaws, at the bottom of service-induced fretting flaws in pressure tubes requires an integrity assessment in terms of DHC initiation. Experimental data and analytical modeling are required to predict whether DHC initiation can occur from the secondary flaws. In the present work, an experimental program was carried out to examine the impact of small secondary flaws with sharp radii on DHC initiation from simulated fretting flaws. Groups of cantilever beam specimens containing blunt notches with and without secondary flaws were prepared from unirradiated pressure tube materials hydrided to a nominal concentration of 50 wt ppm hydrogen. The specimens were subjected to multiple thermal cycles to form hydrides at the flaw-tip at different applied stress levels, which straddled the threshold value for DHC initiation. The threshold conditions for DHC initiation were established for different simulated fretting and secondary flaws. The experimental results are compared with predictions from the engineering process-zone DHC initiation model.


Author(s):  
C. K. Chow ◽  
S. J. Bushby ◽  
H. F. Khartabil

The CANDU®-Supercritical Water Reactor (CANDU-SCWR) is one of the six reactor concepts being considered by the Generation-IV International Forum (GIF) for international collaborative R&D. With SCW coolant, the thermodynamic efficiency is increased to over 40%. The CANDU-SCWR is moderated using heavy water, and it has fuel bundles residing inside horizontal pressure tubes, similar to the current CANDU design. The coolant, however, is light water at 25 MPa, with an inlet temperature of 350°C and an outlet temperature of 625°C. Because of the high temperature and high pressure of the coolant, the standard CANDU pressure tube design cannot be used. This paper presents one of the insulated pressure tube designs being considered for the CANDU-SCWR fuel channels. Unlike current CANDU reactors, the proposed CANDU-SCWR fuel channel does not use calandria tubes to separate the pressure tubes from the moderator. Each pressure tube is in direct contact with the moderator, which operates at an average temperature of about 80°C. The pressure tube is thermally insulated from the hot coolant by a porous ceramic insulator. A perforated metal liner protects the insulator from being damaged by the fuel bundles and erosion by the coolant. The coolant pressure is transmitted through the perforated metal liner and insulator and applied directly to the relatively cold pressure tube. The material selection for each fuel channel component depends on its function. The fuel sheaths and the perforated liner must have high corrosion resistance in SCW, although their resident times are significantly different. The insulator must have high thermal resistance and corrosion resistance in SCW, plus sufficient strength to bear the weight of the fuel bundles without significant thickness reduction during its design life. The pressure tube is the pressure boundary material, so it must have high strength to contain the coolant. One common requirement for all in-core fuel channel components is that they should be as neutron transparent as possible. The irradiation deformation of all these components must also be considered in their design. This paper presents the design of this fuel channel, reviews existing data for materials, indicates where more data are required, and summarizes our plans to obtain these data.


Author(s):  
Brian W. Leitch

The CANDU power generation system is based on a natural uranium fuelled reactor with a heavy water moderator. A unique feature of the CANDU reactor is the horizontal fuel channel that allows on-line re-fuelling and fuel management. Pressure tubes containing the fuel bundles and pressurized heavy water coolant are the in-core component of the fuel channel assemblies. Calandria tubes span the length of the reactor core and provide passageways for the pressure tubes through the reactor core. The calandria and pressure tubes are each approximately 6 meters long. The calandria tube separates the heavy water moderator (∼80°C) from the pressure tube (∼300°C). Both tubes are subjected to gravity loads but the pressure tube carries the additional load of the fuel bundles as well as experiencing high temperature and irradiation induced material effects. The pressure tube deflects under the combined loading and areas of the pressure tube could come into contact with the calandria tube. This contact would limit the operating efficiency and lifetime of the fuel channel. To maintain a gap between the pressure and calandria tubes, helical springs manufactured from rectangular cross-section wire are placed over the pressure tube. These helical springs are known as garter springs and four such springs are spaced along the pressure tube. Initially, there is no contact between the springs and the calandria tube, but as gravity forces and creep effects begin to act, the pressure tube sags and garter spring/calandria tube contact occurs. As the pressure tube continues to deform, a portion of the pressure tube weight, fuel and coolant is transmitted through the garter spring onto the calandria tube. The calandria tube, in turn, begins to deflect under the applied stresses. This creep deformation of the fuel channel takes place over many thousands of operating hours. Eventually, creep induces a permanent vertical deformation (sag) in the fuel channel. The sag of a fuel channel is an important factor in the operation of the structure and many methods are used to determine the general response of the pressure tube/calandria tube/garter spring system. These methods assume the garter spring is a rigid component. This paper specifically examines the garter spring behaviour with respect to the non-linear material and contact response between the pressure tube/garter spring/calandria tube components. A three dimensional (3-D) finite element solid model of the garter spring is used to determine the non-linear response of the helical garter spring to the transverse forces applied from 3-D shell finite element models of the pressure and calandria tubes. Comparison with experimental, crushing tests on garter springs illustrate the analytical model is well behaved. Applying the operating load to the 3-D model shows that the garter spring’s transverse deformation is small and that assuming the garter spring is a rigid component is valid.


Author(s):  
Douglas A. Scarth ◽  
Joanna Wu ◽  
Ted Smith ◽  
Dennis M. Kawa

Delayed Hydride Cracking (DHC) in Zr-2.5 Nb alloy material is of interest to the CANDU (Canada Deuterium Uranium) industry in the context of the potential to initiate DHC at a blunt flaw in a CANDU reactor pressure tube. The material is susceptible to DHC when there is diffusion of hydrogen atoms to the flaw, precipitation of hydride platelets, and development of a hydrided region at the flaw tip. The hydrided region can then fracture to the extent that a crack forms, and is able to grow by the DHC crack growth mechanism. An engineering process-zone model for evaluation of DHC initiation at a blunt flaw that takes into account flaw geometry has been developed. The model is based on representing the stress relaxation due to hydride formation, and crack initiation, by an infinitesimally thin process zone. Application of the engineering process-zone model requires calculation of the stress intensity factor, and the crack-mouth opening displacement, for a fictitious crack at the tip of a blunt flaw. In the current model, calculation of these quantities is based on a cubic polynomial fit to represent the stress distribution ahead of the blunt flaw tip, where the stress distribution is generally calculated by finite element analysis. However, the cubic polynomial is not always an optimum fit to the stress distribution for very small root radius flaws, due to the large stress gradients near the flaw tip. Application of the weight function method will enable a more accurate representation of the flaw-tip stress distribution for the calculation of the stress intensity factor and the crack-mouth opening displacement. Weight functions for a crack at the tip of a blunt flaw in a thin wall cylinder have been developed for implementation into the engineering process-zone model. These weight functions are applicable to a wide range of blunt flaw depths and root radii, as well as a wide range of flaw-tip crack depths. The development and verification of the weight functions is described in this paper. The verification calculations are in reasonable agreement with alternate solutions, and have confirmed that the weight functions have reasonable accuracy for engineering applications of the process-zone methodology.


Author(s):  
Gordon K. Shek ◽  
Jun Cui ◽  
Douglas A. Scarth ◽  
Steven Xu

The Zr-2.5Nb pressure tubes of CANDU reactor are susceptible to a cracking mechanism known as Delayed Hydride Cracking (DHC), which is a repetitive process that involves hydrogen diffusion, hydride precipitation and fracture at a stress concentrator such as a flaw or a crack. Service-induced flaws are present in some pressure tubes and they need to be assessed for susceptibility to DHC initiation. An engineering procedure based on the process-zone methodology has been developed and incorporated into the Canadian standard to determine the susceptibility of flaws in pressure tubes to DHC initiation. The engineering procedure was validated against experiments on flaws which were oriented in the axial direction of the pressure tube. However, many of the service-induced flaws are oriented at some oblique angle with respect to the axial direction of the tube and they may have higher threshold stresses for DHC initiation than those of the axial flaws. It would be advantageous to apply the process-zone evaluation procedure to such angled blunt flaws. For this purpose, an experimental study was carried out to measure the threshold stresses for DHC initiation from machined V-notches with different orientations (0°, 15°, 30° and 45°) with respect to the axial direction of an unirradiated pressure tube. The experimental results were used to support the development of the evaluation procedure for angled blunt flaws. The experimental program and the validation of the engineering procedure for angled blunt flaws are described in this paper.


Author(s):  
Preeti Doddihal ◽  
Dennis Kawa ◽  
Douglas Scarth ◽  
Yu Chen

Abstract The core of a CANDU (CANada Deuterium Uranium) pressurized heavy water reactor includes several hundred horizontal fuel channels that pass through a calandria vessel containing the heavy water moderator. In each fuel channel, annulus spacers are used to maintain the gap between the cold calandria tube and the hot pressure tube, a pressurized vessel containing the nuclear fuel in contact with heavy water coolant. In order to carry the loads between the pressure tube and calandria tube, the annulus spacers are required to possess adequate structural strength throughout the operating life of the reactor. The Inconel X-750 spacers used in some reactor units are susceptible to irradiation induced degradation. As irradiation fluence increases with operating time, material embrittlement has been observed due to helium bubble formation in the X-750 spacer material. An engineering approach for assessing the structural strength of CANDU annulus spacers has been recently developed. When the ductility of the material is relatively low, the region susceptible to fracture under applied tensile stress may be adequately idealized as a strip-yield process zone surrounded by elastic material and associated with restraining stress. The engineering approach is based on applying the strip-yield process zone methodology to fracture at a nominally smooth surface. Finite element modeling was undertaken to simulate the strip-yield based fracture process zone. The finite element analyses and results are presented in this paper. The finite element results verify the engineering equations developed to assess the structural strength of annulus spacers.


Sign in / Sign up

Export Citation Format

Share Document