Polishing method, pipework and circuits resulting from its implementation

Abstract

The invention relates to a method (56 to 59) for polishing a previously levelled weld bead, made of a material sensitive to corrosion when under stress and / or subjected to fatigue, comprising at least one climb milling pass (58), on the previously levelled bead, using an abrasive strip having abrasive tungsten carbide grains on the surface having a size of between 40 microns and 200 microns, at an abrasive grain speed of between 380 and 680 millimetres per second, at an abrasive belt feed rate of between 20 and 150 millimetres per minute, and at a pass depth of between 0.5 and 2 millimetres.

Description

[0001] DESCRIPTION

[0002] TITLE OF THE INVENTION: POLISHING METHOD, PIPING AND CIRCUITS RESULTING FROM ITS IMPLEMENTATION

[0003] Technical field of the invention

[0004] The present invention relates to a polishing process, piping, and circuits resulting from its implementation. The invention is particularly applicable to the polishing of internal welds in high-pressure fluid transport piping made of materials susceptible to stress corrosion cracking. The invention is also applicable to welds subjected to fatigue phenomena, and more specifically to thermal fatigue in mixing zones. The invention is particularly applicable to the fabrication and maintenance of stainless steel piping for a safety injection system (RIS) and a reactor shutdown cooling system (RRA) of a nuclear reactor.

[0005] State of the art

[0006] Stress corrosion cracking (SCC) of a metal or alloy generally results from the combined action of three factors:

[0007] - a tensile mechanical stress (residual or applied stress), - an aggressive surrounding environment, and

[0008] - a material sensitive to the CSC phenomenon, at least under certain conditions.

[0009] In the nuclear industry, the acronym PWSCC is sometimes used, for Primary water stress-corrosion cracking, which can be translated as "primary water stress-corrosion".

[0010] SCC is a particularly insidious phenomenon: initially very subtle, its speed and intensity are subsequently difficult to predict. It is feared because it can lead to a sudden and "catastrophic" failure of the material (even at a value well below its normal mechanical strength limit).

[0011] It is generally estimated that three conditions must be met for the occurrence (and subsequent spread) of a stress corrosion cracking phenomenon:

[0012] - mechanical stress (applied or residual, i.e., resulting, for example, from welding, bending, work hardening, or cold rolling of the material), - aggressive physicochemical conditions of the environment: temperature (which can also be a source of thermal shock, expansion or contraction phenomena, and pressure variations in the environment); hydrogen potential; composition of the environment and possible irradiation (neutron, high-energy), and

[0013] - sensitivity of the material to stress corrosion cracking (under certain stress and environmental conditions).

[0014] For example, chloride stress corrosion cracking can occur in austenitic stainless steels under tensile stress in the presence of oxygen, chloride ions, and high temperatures. This is one of the most significant forms of stress corrosion cracking affecting the nuclear industry.

[0015] Other conditions that increase the probability of CSC cracking or increase their growth rate include, for example, the type and shape of the material (installation design), fluence (under neutron irradiation), the carbon content of the steel (if applicable), the type of welding process...

[0016] It is demonstrated that high-energy neutron irradiation (such as that which takes place in the core of a light water or heavy water reactor) increases the susceptibility to stress corrosion cracking (SCC) of stainless steels, including austenitic steels.

[0017] The structural corrosion cracking (SCC) of austenitic stainless steel causes significant cracks, which can reach five to six millimeters in depth around the entire circumference of the pipe. A temporary solution proposed to address emergencies is the "sleeving" of the damaged sections of pipe.

[0018] As an example, the reactor feed pipe, called the "ARE" pipe, carries pressurized water at 83 bar and superheated to 225°C. The steam pipe carrying steam to the turbine, known as the "WP pipe", carries pressurized steam between 55 and 70 bar, with the steam having a temperature between 270 and 297°C.

[0019] These extreme stresses on the pipes can lead to erosion / degradation of the piping, particularly welds such as those joining different pipe sections. Weld degradation can manifest as cracks or variations in material density and / or thickness.

[0020] Regarding weld fatigue, cracks are initiated at the toe of the weld bead and are just as difficult to detect as SCC (Stress Cracking). Fatigue cracking of a weld occurs when the geometry of the weld bead (particularly a bead of excessive penetration) exhibits a concentration of tensile stresses, and when repeated thermal and / or mechanical stresses (including vibration) are applied to it.

[0021] As far as remediation is concerned, to date, no other means of remediation is known than the removal of the cracked parts and their replacement with new parts.

[0022] Description of the invention

[0023] The present invention aims to remedy all or part of the drawbacks of the prior art.

[0024] To this end, according to a first aspect, the present invention relates to a method of polishing a weld bead previously ground down in a material sensitive to the phenomenon of stress corrosion cracking and / or subjected to a phenomenon of fatigue, and more particularly to thermal fatigue, a method which comprises at least one pass, on the ground bead, in the downswing, of an abrasive belt having, on the surface, abrasive grains of tungsten carbide, of size between 40 microns and 200 microns, with a speed of the abrasive grains between 380 and 680 millimeters per second, with a feed speed of the abrasive belt between 20 and 150 millimeters per minute, and with a depth of pass between 0.5 and two millimeters.

[0025] Thanks to these arrangements, residual compressive stresses are generated on the internal surface of the weld, which eliminates one of the factors in the appearance of stress corrosion cracks and / or weld fatigue.

[0026] In optional embodiments, the process includes at least one cycle of two successive passes, first in opposition and then in swallow, of the abrasive belt on the flush bead with, for each pass, a speed of the abrasive grains between 380 and 680 millimeters per second, with a feed speed of the abrasive belt between 20 and 150 millimeters per minute, and with a depth of pass between 0.5 and two millimeters.

[0027] In optional embodiments, the rotation speed of the abrasive belt is between 150 and 260 rpm.

[0028] In optional embodiments, the tungsten carbide abrasive grains (40) have a size between 100 microns and 150 microns. In optional embodiments, the feed speed of the abrasive belt is between 65 and 75 millimeters per minute.

[0029] In optional embodiments, the depth of cut is between one and 1.5 millimeters.

[0030] The inventors discovered that each of these optional features increases the residual compressive stresses.

[0031] In optional embodiments, the weld bead is pre-formed between two stainless steel elements.

[0032] According to a second aspect, the present invention relates to a method of assembling two metal elements, which includes a step of welding these elements, a step of mechanically grinding the weld bead, and then a polishing which is the subject of the invention, as succinctly presented above.

[0033] In optional embodiments, during the welding step, an electric arc welding process under inert gas protection with a refractory electrode is implemented.

[0034] The advantages, purposes and particular characteristics of this assembly process being similar to those of the polishing process which is the subject of the invention, they are not recalled here.

[0035] According to a third aspect, the present invention relates to piping obtained by welding piping elements made of material sensitive to stress corrosion cracking and / or subject to fatigue, this welding generating an internal weld bead in the piping, grinding of this bead and polishing of the ground weld bead by implementing the polishing process which is the subject of the invention as briefly described above.

[0036] According to a fourth aspect, the present invention relates to a safety injection circuit (RIS) of a nuclear reactor, which includes at least one piping of the object of the invention, as briefly described above.

[0037] According to a fifth aspect, the present invention relates to a reactor shutdown (RRA) cooling circuit of a nuclear reactor, which includes at least one piping of the invention, as briefly described above.

[0038] The advantages, purposes, and specific characteristics of this piping and these circuits being similar to those of the process that is the subject of the invention, they are not repeated here. Brief description of the figures

[0039] Other advantages, purposes and specific features of the invention will become apparent from the following non-limiting description of at least one particular embodiment of the device, method and piping that are the subject of the invention, with reference to the accompanying drawings, in which:

[0040] Figure 1 shows, in axial view, a particular embodiment of a device of the invention in operation within a pipe. Figure 2 shows, in lateral view, a tool of a particular embodiment of a tool of the device of the invention.

[0041] Figure 3 is a partial enlargement of [Fig. 2],

[0042] Figure 4 shows, in perspective, an abrasive strip of the tool illustrated in [Fig. 2], and

[0043] Figure 5 represents, in the form of a flowchart, a succession of implementation steps in a particular embodiment of the process that is the subject of the invention.

[0044] Description of the implementation methods

[0045] The present description is given by way of non-limiting attribution, each feature of an embodiment being able to be advantageously combined with any other feature of any other embodiment.

[0046] It should be noted from the outset that the figures are not to scale. The tool and abrasive belt of the device that is the subject of the invention are sized and adapted for a predetermined pipe diameter.

[0047] Figure 1 shows a pipe 10 inside which a weld bead 11 has been formed and is to be ground down and polished. It is worth recalling here that ground down a weld bead is an operation that consists of mechanically removing the excess weld (bead) and thus bringing the weld bead flush with the internal surface of the pipe. Mechanical polishing is a finishing process that removes, by abrasion, defects, scratches, or roughness from the part. The aim is to obtain a high-quality surface finish, characterized in particular by the roughness index (Ra) and homogeneity.

[0048] A grinding cutter (not shown), rotated by a tool 12, travels along the weld 11 to grind its surface. Then, a polishing abrasive belt 13, also rotated by the tool 12, travels along the weld 11 to polish its surface. A robot (not shown) sets the tool 12 in motion. In the first grinding phase, this movement positions the grinding cutter opposite the weld bead 11, then places the grinding cutter against the weld bead 11, and simultaneously rotates the tool 12 and the grinding cutter while maintaining a predetermined, adjustable grinding depth.In a first phase, of polishing, the movement of the tool 12 puts the abrasive belt 13 in relation to the flush weld bead 11, then puts the abrasive belt 13 in contact with the flush weld bead 11, and simultaneously rotates the tool 12 and the abrasive belt 13 while maintaining a parameterizable polishing depth, of a value determined beforehand, as described in detail below.

[0049] For example, piping 10 is a safety injection circuit (RIS) or reactor shutdown cooling (RRA) circuit piping of a pressurized light water reactor (PWR). For example, piping 10 has a diameter of ten inches, with a nominal diameter of 250 mm, an outside diameter of 273.1 mm, and a wall thickness of 25.4 mm. For example, the piping is made of stainless steel.For example, weld 11 was performed using a TIG (Tungsten Inert Gas) welding process, also known as GTAW (Gas Tungsten Arc Welding), a gas-shielded electric arc welding process with a refractory electrode (process "141," which uses a non-consumable tungsten electrode in a shield of inert gases such as argon, helium, or mixtures thereof, but not nitrogen) for the root and support passes. These passes can be performed manually, automatically, or mechanically. The fill passes can be performed using the same welding process or another (e.g., stick welding). The weld connects two pipe elements of the same diameter and of the same material, i.e. 316L stainless steel.

[0050] 316L grade stainless steel, or other stainless steels designated in the nuclear field as: "Z2 CND 17.12", "Z2 CND 18.12 with controlled nitrogen", "X2CrNiMo17-12-2", "X2CrNiMo17-12-3", "X2 CrNiMo17-12-2 with controlled nitrogen", are austenitic stainless steels. Their main alloying constituents, after iron, are chromium (between 16 and 18%), nickel (10 to 12%), and molybdenum (2 to 3%), up to 2% manganese, with small amounts (<1%) of silicon, phosphorus, and sulfur also present. The addition of molybdenum provides high corrosion resistance, compared to localized corrosive attack by chlorides and general corrosion, while sulfur is added to improve its machinability and welding capability.

[0051] The present invention relates in particular to the parameterization of the polishing phase in order to generate, on the polished weld surface 11, a significant residual compressive surface stress in the axial direction, i.e., parallel to the axis of the pipe 10. This compressive surface stress is, by convention, negative, preferably with an absolute value greater than 100 MPa and, more preferably, greater than 200 MPa. This surface stress reduces the risk of stress corrosion cracking and / or fatigue cracking of the weld. This surface stress is determined by X-ray diffraction (XRD), an analytical technique based on the diffraction of X-rays by matter. The general method consists of bombarding the sample with X-rays and analyzing the intensity of the scattered X-rays according to their orientation in space.X-ray diffraction is a well-known and accurate method for measuring residual stresses and residual austenite in crystalline materials. This method allows for the evaluation of surface stresses (ranging from a few micrometers to approximately one millimeter) in crystalline materials.

[0052] To position the abrasive belt 13 against the weld bead 11, the tool 12 includes a pusher 17 with a rigid roller 18, for example, a metallic one. The pusher 17 is deployed by a radial linear motor or a radial cylinder 19. Rollers 14 and 15 are positioned on either side of the abrasive belt 13 (see Figures 2 and 3). The axes of rotation of the roller 18, the tool 12, the abrasive belt 13 and the rollers 14 and 15 are parallel to the axis of the pipe 10 at the location of the weld 11. To adjust the depth of the grinding and polishing performed by the grinding cutter, then the abrasive belt 13, a linear motor or a cylinder 16 radially moves the shaft of the grinding cutter and then the shaft of the hub 42 (see figure 4) of the abrasive belt 13.

[0053] A particular embodiment of the tool 12 is shown in Figure 2, with an enlargement of its central part, the tool 12, in Figure 3. There, surrounding a central body 24, are a front module 21 and a flange 27 for connection to a robot (not shown) allowing the positioning of the module 24 at the weld of the pipe 10. The front module 21 is rigid and has, at its end furthest from the flange 27, a generally conical shape whose apex is located outside (in front) of the tool 12. The front module 21 remains fixed in the pipe 10 during the grinding and then polishing of the weld bead 11, that is to say during the rotation of the grinding cutter and then the abrasive belt 13, respectively, and of the central body 24.To lock the front module 21 in position, it includes an inflatable balloon 22, for example made of silicone, which is inflated until it reaches the inner surface of the pipe 10 and exerts sufficient pressure to ensure adequate static friction. Furthermore, the balloon 22 prevents the dispersion of particles from the grinding and polishing of the weld 11 into the pipe 10. When this balloon 22 is deflated, as illustrated in Figure 2, rollers 28 prevent the front module 21 from rubbing against the inner wall of the pipe 10.

[0054] A flexible coupling 23, for example made of polyurethane, connects the front module 21 and the central body 24. Similarly, a flexible coupling 25, for example made of polyurethane, connects the central body 24 to the flange 27. The flexible couplings also serve as conduits for fluids (pressurized air) and electrical connections for powering the components of the tool 12 and for transmitting control signals or signals from sensors (not shown) integrated into the tool 12, for example, cameras. The flexibility of the flexible couplings 23 and 25 facilitates the movement of the tool 12 through a bend in the pipe. Optionally, at least one freewheel ring 26 31 is positioned on at least one of the flexible couplings 23 and 25 to prevent friction between each flexible coupling and the inner wall of the pipe 10. The flange 27 is equipped with rollers 29 for the same reason.The axis of rotation of the rollers 28, 29 and 31 is perpendicular to the longitudinal axis 34 of the tool 12.

[0055] The central body 24 includes the pusher 17, the roller 18, the radial linear motor or radial cylinder 19, the rollers 14 and 15, the linear motor or cylinder 16 and the abrasive belt 13. It can be seen in Figure 3 that the rollers 14 and 15 have a central groove 30 of sufficient size for the weld bead 11 to pass through this groove 30 before it is ground down.

[0056] While the central body 24 is rotated (through at least 360 degrees to traverse the entire weld 11) relative to the pipe 10, the abrasive belt 13 is rotated relative to the central body 24, first in opposition on the forward movement, then in downward on the return movement. Figure 4 shows an abrasive belt 13 having abrasive grains 40 (symbolized in Figure 4 by a connected dotted line) on the surface of a resin layer 41, on a means 42. In the axial direction, the dimension of the abrasive belt 13 is, for example, 35 mm. The diameter of the abrasive belt 13 is, for example, 50 mm. The abrasive grains 40 are made of tungsten carbide, of size (according to a standardized measurement, for example the intercept measurement method or the planimetric method) between 40 microns and 200 microns and, preferably, between 100 and 150 microns, for example 120 microns.

[0057] The hub has a diameter of, for example, 35.5 millimeters, and the resin layer is seven millimeters thick. For example, the resin is an RTV FA type, a two-component silicone elastomer that cures at room temperature through a polyaddition reaction. These resins are initially fluids that, after curing, transform into elastic and resistant materials.

[0058] Preferably, the two components of this resin are:

[0059] - for part A, the product marketed by the company Bouygues Construction Services Nucléaires® (BCSN®) under the reference "RTV FA 876 part A", a mixture of polyorganosiloxanes, fillers, and additives, including cristobalite (CAS no. 14464-46-1, EC no. 238-455-4) in a weight concentration between 10% and 20%, and

[0060] - for part B, the product marketed by the company Bouygues Construction Services Nucléaires® under the reference "RTV FA 876 part B", a mixture of Polyorganosiloxanes, fillers, and additives, including cristobalite in a weight concentration between 10% and 20% and dodecamethylcyclohexasiloxane (no. CAS 540-97-6, no. EC 208-762-8), in a weight concentration between 0.1% and 1%.

[0061] The belt, comprising the hub 42 and the resin 41, is supplied by Bouygues Construction Services Nucléaires®, under reference 2022-655-DT-80192A. It is first cast in a mold by injection. The abrasive grains 40, made of tungsten carbide, are then bonded to this abrasive belt.

[0062] The resin is made of PMUC material (an acronym used to define Products and Materials Used in Nuclear Power Plants). PMUC also refers to the "PMUC" standard implemented by the EDF Group as part of its nuclear power plant operation and maintenance activities. This technical standard specifies a number of chemical requirements mandated by the site operator to prevent contamination of surfaces that must resist corrosion, including the austenitic stainless steels of reactor circuits carrying primary fluid or fluids injected into the primary. The chemical analyses required for EDF's PMUC approval involve determining the halogen (fluorides, chlorides, bromides), sulfur, and other harmful species content of Products and Materials Usable in Nuclear Power Plants in order to verify compliance with the PMUC chemical specifications defined by EDF.Indeed, chemical elements in the halogen family and sulfur compounds can, under certain conditions, cause various forms of corrosion of metallic alloys and could endanger nuclear installations at the heart of power plants.

[0063] Furthermore, these laboratory analyses are used to determine the presence of prohibited metals such as lead, mercury, iron, etc. This analysis is divided into two successive phases: mineralization followed by quantification, which leads to the determination of the total halogen and sulfur content of the products. The purpose of mineralization is to dissolve all the chemical species to be quantified in the product into an aqueous solution.

[0064] The inventors have determined the values ​​of the operating parameters of the central body 24 which enable the generation, on the polished weld surface 11, of a significant residual compressive surface stress in the axial direction, i.e. parallel to the axis of the pipe 10. This compressive surface stress is negative, preferably with an absolute value greater than 100 MPa and, more preferably, greater than 200 MPa.

[0065] The weld is ground down by milling (using an HSS milling cutter, for High Speed ​​Steel). This involves removing the excess weld bead (burr) in successive 0.5 mm deep passes until the weld toe is reached. The tool grinds in a conventional manner; only the final pass is made in a climb cut to ensure the cleanest possible finish. This last pass also removes any remaining chips or filings.

[0066] Polishing ensures that the desired residual stresses are achieved.

[0067] The succession of grinding and polishing phases, carried out with the parameters listed in Table 1 below, makes it possible to achieve the target residual stress criteria. By comparing the two MTO-MIA-028 half-coupons, the inventors observed, firstly, that the absence of a finishing grinding pass has no effect on the stress state and, secondly, that the use of a milling cutter, preferably a chip breaker to reduce the size of the generated chips, makes it possible to achieve the target residual stress criteria.

[0068] The inventors observed that only the last tool used on the material (polishing strip) produces the "signature" of the surface stress state. Regarding the number of polishing passes, this parameter does not appear to have a significant effect on residual stresses. Compared to the MTO-MIA-026 coupon, a lower feed rate seems to influence residual stresses more favorably than a higher feed rate. Increasing the polishing time is beneficial to the stress state. The grinding and polishing parameters used on the MTO-MIA-028 coupon have no impact on the hardness values, either at the surface or in depth.

[0069] [Table 1]

[0070]

[0071] In a series of tests similar to the one described above, the inventors discovered that the target criterion in terms of residual stresses was met, with the parameter values ​​mentioned below, by obtaining axial compressive residual surface stress values ​​between -233 MPa and -470 MPa. The operating parameter values ​​of the device during the leveling phase are preferably the following:

[0072] - Depth of cut (equivalent to a support pressure) in roughing (opposition), between 0.5 and two millimeters, and preferably between 1 and 1.5 millimeters.

[0073] - Depth of pass (equivalent to a support pressure) in finishing (swallowing), between 0.02 and 0.2 millimeters, and preferably between 0.03 and 0.07 millimeters.

[0074] The operating parameter values ​​of the device that is the subject of the invention are, during the polishing phase, at least one of the following, and preferably all of the following:

[0075] - The speed of the abrasive surface is between 380 and 680 millimeters per second, preferably with a rotation speed of the abrasive belt 13 between 150 and 260 rpm,

[0076] - Feed speed of the abrasive belt 13 (rotation of the central body 24), between 20 and 150 millimeters per minute, and preferably between 65 and 75 mm / min,

[0077] - Depth of cut (equivalent to a support pressure), between 0.5 and two millimeters, and preferably between 1 and 1.5 millimeters,

[0078] - Number of cycles (going in opposition and returning in downstream) greater than or equal to one, preferably between one and three round trips.

[0079] These values ​​relating to the polishing phase ensure that the stress release phenomenon which promotes stress corrosion cracking of welds cannot occur, and facilitate fluid circulation, and consequently prevent the cracking phenomenon, thus maintaining the safety of the piping, particularly of a nuclear reactor.

[0080] The device and method of the invention are particularly applicable to assembly welds for pipe replacement operations and to work on existing pipes at a nuclear site. They can also be used for prefabricated welds in the workshop, thus avoiding certain constraints encountered during on-site work.

[0081] Of course, the present invention is not limited to the pipe diameters and materials described above, but also applies to the polishing of any flush weld on metallic components susceptible to stress corrosion cracking, that is, welds that present a risk of stress corrosion cracking, even if the surface compressive stresses may have a lower absolute value than those described above. These materials include stainless steels and other austenitic materials. Besides nickel alloys, examples of materials susceptible to stress corrosion cracking include carbon steels and aluminum alloys of the 2XXX, 5XXX, and 7XXX series.

[0082] As explained above, the invention also applies to the polishing of weld beads subjected to a fatigue phenomenon and more particularly to thermal fatigue or, at the same time, to a stress corrosion cracking phenomenon and a fatigue phenomenon.

[0083] As mentioned above, the device and method of the invention reduce both the risks of cracking due to structural cracking (SCC) and those induced by weld fatigue. In the latter case, the object of the invention is to eliminate the tensile stress concentration that allows crack initiation.

[0084] Figure 5 illustrates successive steps of a 50 process for grinding and polishing an internal weld of a pipe.

[0085] In step 51, stainless steel parts are welded using a method known per se, for example, the TIG welding process. In step 52, the grinding cutter is positioned opposite the weld bead. In step 53, values ​​are assigned to the grinding parameters, including the grinding cutter's rotation speed, feed rate, and depth of cut, as described above. In step 54, the weld bead is mechanically ground down until it is flush with the inside diameter of the pipe. In step 55, the grinding cutter is removed. In step 56, the abrasive belt is positioned opposite the ground weld bead. In step 57, values ​​are assigned to the polishing parameters, including the abrasive belt rotation speed, its feed rate and the depth of cut, as described above.During step 58, a polishing pass is performed on the flush weld bead using the assigned parameter values. As symbolized by the dashed arrow looping back to step 58, several successive passes can be performed. If only one pass 58 is performed, it is done in the downstroke direction. Otherwise, preferably, at least one cycle of two successive passes 58 is performed, first in the upstroke direction and then in the downstroke direction, with the abrasive belt 13 on the flush weld bead. During step 59, the abrasive belt is removed.

Claims

DEMANDS 1. A method (56 to 59) for polishing a weld bead (11) previously ground down in a material susceptible to stress corrosion cracking and / or subject to fatigue, characterized in that it comprises at least one pass (58), on the previously ground down weld bead, in the downslope direction, of an abrasive belt (13) having, on the surface, tungsten carbide abrasive grains (40), of size between 40 microns and 200 microns, with an abrasive grain speed between 380 and 680 millimeters per second, with an abrasive belt feed speed between 20 and 150 millimeters per minute, and with a depth of pass between 0.5 and two millimeters.

2. Method (56 to 59) according to claim 1, which comprises at least one cycle of two successive passes (58), first in opposition and then in swallow, of the abrasive belt (13) on the flush bead with, for each pass, a speed of the abrasive grains of between 380 and 680 millimeters per second, with a feed speed of the abrasive belt of between 20 and 150 millimeters per minute, and with a depth of pass of between 0.5 and two millimeters.

3. Method (56 to 59) according to any one of claims 1 or 2, wherein the rotation speed of the abrasive belt (13) is between 150 and 260 rpm.

4. Method (56 to 59) according to any one of claims 1 to 3, wherein the abrasive tungsten carbide grains (40) have a size between 100 microns and 150 microns.

5. Method (56 to 59) according to any one of claims 1 to 4, wherein the feed speed of the abrasive belt is between 65 and 75 millimeters per minute.

6. Method (56 to 59) according to any one of claims 1 to 5, wherein the depth of cut is between one and 1.5 millimeters.

7. A method (56 to 59) according to any one of claims 1 to 6, wherein the weld bead is pre-formed between two stainless steel elements.

8. A method (50) for assembling two metal elements, comprising a step (51) of welding these elements, a step (52 to 55) of mechanically grinding the weld bead, and then polishing (56 to 59) according to a method according to any one of claims 1 to 7.

9. Method (50) according to claim 8, wherein during the welding step, an electric arc welding process (51) under inert gas protection with a refractory electrode is implemented.

10. Piping obtained by welding (51) piping elements made of material susceptible to stress corrosion cracking and / or subject to fatigue, this welding generating an internal weld bead (11) in the piping, grinding (52 to 55) of this bead and polishing of the ground weld bead by implementing the polishing process according to one of claims 1 to 9.

11. Safety injection circuit (RIS) of a nuclear reactor, comprising at least one piping according to claim 10.

12. Reactor shutdown cooling circuit (RRA) of a nuclear reactor, comprising at least one pipe according to claim 10.

Images