@ recsegsnreeretctmansemeans bw et Residual stresses in indian pressurised heavy water reacior rolled joints, ‘Singh RK, Khan M.A. Charan J Kushwaha HS, Saha RK. haba Aton Remar Crt Tate ABSTRACT. Evaluation of residual stresses in rolled joints of PHWR primary coolant components such as ‘pressure tube, liner tube and sparger tube is caried out by combined experimental and analytical the linear elastic theory is shown to be adequate to asessthe residual stresses in pressure tube and liner tube, deformation plasticity theory with monotone loading assumption is shown to be effective for asessng residual stresses in nonlinear range for sparger tbe. The sssumptions of the analysis are validated by strain data and clastplasic stesses. It is demonstrated that low tensile esdualstesss are developed in these rolled joints which are not significant from the point of view of flaw propagation and delayed hydride cracking LINTRODUCTION Studies on residual stresses in Indian Pressuited Heavy Water Reactor (PHWR) components have been continuing research and development activity, Rolled joins sre commonly used in PHWRs to join tubular components to tube sheets or end fitngs: Presse cube, liner tube and sparger tube are three such components employing rolled joins for their attachments, The information on residual streses in these rolled joins have been genersted for assuring their satisfactory Iong_ term performance from the consideration of flaw propagation due to cyclic Jas, delayed hydride eracking (DHC) and sess relaxation in the mld joints leading to leakage. The pressure tube rolled joins have been modified from low clearance joints for Zitealoy-2 tubes to nearly zero clearance rolled joint for 22.5% Nb tubes Stainless stel (AIST 410 grade) liner tubes are used as a part of the end fitting assembly ofthe coolant channels, ‘These tes are joined tothe martensitic stainless stecl (AIST 403 rade) end-fiting assembly through rolled joins. Recently a development work has been carried ott t0 install Zitaloy 2 sparger tubes (0 serve as moderator inlet tothe core with rolled joins in he SS 304 L calandsia tube sheet. This arrangement gives uniform moderator flow and eliminates the ned for routing ‘moderator inlet piping. through the inaccesibe rion of the ealandria val.‘The present paper brings out combined experimental and analytical cise studies of residual stress nalyss for different rlled joint designs of pressure tube, liner tube and sparger tube. For the ‘measurement of residual stresses in Zircaloy 2/ Zr- 2366 Nb pressre tube, a combination of ‘machining operations such as sleve removal, sit cuting and hole diling technique are sed on the test assemblies which are instrumented with strain gauges. Thee categories of safety ‘concems are addressed inthis study. In the first eategory the concer isto control the level of tensile residual stresses in Zircaloy 2/ 72.58% Nb pressure tubes to avoid on set of DC ‘mechanism near the rolled joint region of the pressure tube, In this case the end fiting of relatively high strength maierial undergoes only & minor plastic deformation. Linear clastic theory is used to predict the residual stresses in this cas, It is demonstrated that the compressive stresses on account of contact pressure developed between pressure tube and end-fiting assembly are dominant over stesses developed due to local plastic deformation by rolling. This feature results in a very low level of tesie residual stresses. The second category of concem is relevant for line tubes In this case is shown through measurements that the tensile residual stresses ate gain local in nature and are of low magnitude compared to the yied strength of SS grade 410 liner tube material. The third cass of problem is that ofthe Ziresoy 2 sparge tube rolled in SS 304. calandria side tube sheet. Here the concern isto check the level of tensile residal stresses in tube sheet during and after competion of rolling. Measurements cn test assemblies have been carried out to check the level of residual stests,Inthis cas the tube sheet undergoes significant plasti deformation and due to contact typeof rolling lod the tube shoot remains in compress State of stress. Evaluation of residual stresses inthis case was carried out by development of an algortim based on monotonic deformation plasticity theory wit. isotopic strain hardening assumpsions to conver the measured strains to stresses. The present study demonstrates that die ‘o rolling the tube sheet mains in compressive state of residual stress which will not allow any rack propagation in tube sheet 2 RESIDUAL STRESSES IN PRESSURE TUBES ‘The pressure tubes of Ziraloy 2 and Zi-2.5% Nb ae installed in endings by colling the inside sutface of pressure tubes to cause plastic deformation against a rigid end-fiting of stainless stel AISI 403 grad (fig. 1). An optimum level of plastic deformation is imparted tothe tube material uring the tube rolling operation so as to firmly seoare the tubes te ge leakproof joint along. ‘with a high pull out strength However, excessive tensile residual sresses could be detrimental due to increased susceptibilities to DHC in the loa) areas close 1) the rolled join, Cases of failure of Zr- 2.5% Nb pressure tubes due to tensile residual stresses have been reported earlier [1]. In Indian PHWRs development of pressure tbe rolled joints have been successively carted ‘ut to achieve adequate contact pressure without any appreciable plasie deformation of the end Fiings. [Evaluation of residual stresses in test assemblies of Zrcaloy-2 and Zs-2.5% Nb pressure tubes rolled toa short piece of end-fiting was catied out through a sequence of sleeve removal, slit cutting. and hole drilling methods (fig2). Measurements were performed on the inside and ‘outside surfaces of presure tube with the help of normal stain rosettes and hole diling type steain rosettes [2]. A typical layout of sain rsetes of hole diling type is shown in fg.3. The strain data thus obtained is indicative of overall residual stresses due to contac pressure 402developed between the sleve andthe tube alongwith the Toca streses developed on account of, rolling opecation. The strain data were collected after sleeve removal, slit cuting and finally uring the hole éiling operation by inserting an air driven turbine cutter tool through the slit {dri a small hole (1.59 mm nominal dia) atthe ceatre ofthe stan rset. In the following paragraph residual sess data derived from measurement on a nearly zero clearance rolled joint of Zr-2.5% Nb pressure tube of SOO MWe PHWR design is described ‘The distribution of suesses along the pressure tube axis due to contact pressure obtained during sleeve removal is shown in fig. Overall bending suesses obtained by slit cutting method are shown in fig. The combined residual stresses duc to sleeve removal and slit cutting ate shown in fig.6 may be noticed that the dominant stesses in rolled portion are compressive and overall bending stresses in untoled portion of tube are of very small magnitude This indicates tha the rolling tol is well within the ené-fiting and insignificant tensile stesses due to hydrostatic ‘component of load are developed inthe tube Local surface stresses in the tbe are obtained by hole drilling and the distribution of these slresses are shown in fg. This measurement gives the local tensile stresses in unrolled portion fof the tbe, It indicates thatthe tensile stresses are developed during rolling process due to multiple lobes emerging under the local roller forces. Table 1 shows the total residual stress pictuce and its orientation with respect tothe longitudinal axis by combining stesses due to all the three operations namely sleeve removal, sit cuting and hole deiling. In the table circumferential and axial steses aze also shown, Fig shows the distibation of the total fesidual stesses and the maximum tensile stess of 208 MPa is noticed. The dominant stresses fare compressive and are governed by contact pressure. ‘Similar studies were also cared out on rolled joints of Zcaloy-2 presse tube with end-fiting test assemblies with high clearance jint desig. In this ease measuremets on both outside and inside surfaces of pressure be were taken up and it was shown tha: the maximum tensile stresses on outside surface in rolled portion is $17 MPa and 621 MPa on the two specimens respectively while on inside surface itis ofthe order of 358 Mpa (table2). Measurements made for Zireloy-2 and 21-25% Nb tubes ata few other locations along the eircumference indicated that stresses vary within 12-18% and hence it may be concluded thatthe tbe isnot subjected to ‘any torque during the rolling operation. This is clear from the orientation of principal stresses Which are nearly axial and circumferential (fig 9) as shown in table | for Z-2.55 Nb tube ‘experiment. From the above two studies itis demonstrated that 2- 2.55 Nb pressure tube with low clearance joint results into low level of tensile residual stresses. All the new 229 MWe and 500 MWe PHWR plants in India have been designed with Z-2.5% Nb pressure cabs with interference fit rolled joints, This will result in still lower level of tensile residual stresses, Thus the total tensile ‘svesses will be dominated by operating sreses and there will be no appreciable change in applied stress intensity due to residual stresses ina good rolled joint design 403‘3 RESIDUAL STRESSES IN LINER TUBE Lines tube of stainless steel grade AISI 410 is rolled to the end-fiting and acts es transition ‘between pressure tube and primary heat transport feeders (ig!) Residual stress evaluation liner tube is of interest o ensure its performance for cyclic load and proper leak tightnes, Inthe resent study residual stresses were determined during and after the rolling fora wall thickness reduction of 12 %. Table 3 shows the strains and calculated wresses. during and after the rolling operation on the tube OD. The maximum residual tensile svess of 221 MPa inthe liner tube is noticed in eircumfereail direction after the completion of ring. The maximum tensile stress of 485 MPa is developed during the rolling process when te tool is located inside the tube. The longitudinal sresses are compressive and both the peak stesses decay rapidly within a distance of 35 to-43 mm from the sleeve edge. 4 RESIDUAL STRESSES IN SPARGER TUBE. ‘The arler two examples of residual stress analysis were based on linear elastic theory. Zircaloy 2, 2-2.5% Nb and SS 410 have yield stress comparable tothe end fiting sleeve materiel and Dt ratio is small. This results in very low level of tensile stresses compared to the yield strength of ‘ube material and the sleeve remains in nearly elastic state, For rolled joints with large Dit rato such as calanria tube and sparger tube, the tbe is frst rolled in the tbe shect and subsequently an inser is rolled to acieve adequate strength as shown in fig. 10. this case the tube sheet of Stainless stel grade AISTSS 304 L is comparatively ductile nd has lower yield strength. Here the concern sto check he level of residual stresses in tbe sheet during and after the completion of rolling [Experiments on two test assemblies of Ziresloy 2 sparger tube roled into stainless stel 3081. sleeve were cared out from zero nip o a maximam of 12% wall thickness reduction in steps of| 26. Subsequently inserts were put and further rolling was carted ot from zero nip to 10% wall thickness reduction in the first experiment and 8% wall thickness reduetion in the second experiment. The strain date were collected at 69 mm, 77 mm and 85 mn from centre line of sleeve on diametrically opposite locations. The two sirain data wee within 10% of difference which demonstrates the accuracy of data collected. The repestabilty of strain data was also confirmed on two experimental sets set I & set I) well within lint of 10%. Peak strain values in the range of 1.4% to 1.59% were noted. In order to obtain stresses from the ‘measured strains a computer program ROSEPL was developed forthe determination of elaso plastic stresses from recorded strain data based on Von Mises flow rule and isotropic strain hardening criteria (3,4). This is based on deformation plasticity theory since the strains during rolling and after completion of rolling fora patculr level of wall thickness reduction did not Aiffer signiicanly A simplified assumption of monotonic loading is therefore adequate, Table 5 shows the stress distribution in tube sheet fo the two experiments(st 1 & set I) before and after the rolling operation. The maximum circumferential tensile sess of 119 MPa at a distance of 24 rm from sleeve ID is observed. Near the sleeve ID surface these siesses are compressive, Aer completion of rolling the maximum residual tensile stress is 65 MPaat a distance of 24 min from the sleeve surface. 404tis concluded from the above study thatthe contact lad developed duct rolling of ductile tube sheet of $S 304L results in high compressive suesses near the sleeve surface which are balanced by a tensile ste field of lower magnitude over a wide distance (S). These experiments were done in sleeve of 190 mm ID and with large solid material surrounding the bore in actual tube sheets the overall state of compression willbe significant, Tis stady demonstrates the adequsey ‘of rolled jointsand no flaw wil be initiated during rolling operation, S CONCLUSIONS In this paper evaluation of residual stresses for rolled joints on Zitcloy 2 and 21-25% Nb pressure tubes, $$410 Tinear tube have been made and itis shown tht the tensile stresses in tubes near the sleeve edge is within elastic limit, For large Dit ratio tues such as spargr tube {he dominant elastoplastic stresses in tbe sheet are compresive. Evaluation ofthese stresses have been carried out using deformation based plasticity theory. The monotonic loading assumption has been validated with the stresses evaluated during and after the rolling. This study {is helpfl to assess the feasibility of rolling tubes for repair or re-engineering work of old plant Flaw propagation or onset of DHC mechanism is unlikely on account of low level of resi tensile stresses for PHWR tabla rolled joints. 6 REFERENCE, 1. Pecryman ECW (1978). "Pickering Pressure Tube Cracking Experience”, Nuclear Energy, 17 No2 pp 95-105 2. Khan MA, Charan 1, Singh RK, Kushvaha HS, (1991) "Residual Stress Measurement in 500 MWe PHWR Z:-2.5% Nb Pressure Tube Rolled Joint, TasklO081" 3. Mendelson A (1983) Plasticity: Theory and Application, Krieger Publishing, 4, Owen DRI, Hinton F, (1980) Finite Blement in Plasticity: Theory and Practice, Pineridge Press, Swansea, UK 5. Khan MA, Charan J, Sigh RK, Kushwsha HS,1991) “Report on Strain Measurement on Rolled Joint on MAPS Sparger Tube, Annexure-IC™ ‘Table 1 Residual Stresses inside surface) Distribution in 500 MWe PHWR_ Zr-2.5% Nb Pressure Tube Rolled Joint with Low Clearance Joint. ‘Strain Gauge Principal Sires Principal Stress Reference Angle Circumferentil- Longitudinal: No. -1(MPa) -2(MPs) (Degrees) Stress (MPa) Sires (MPs) 1 6b 56.7 113 118 46.4 2 80 “37 251 “437 “80 3 ns 310 102 560 4 2080 3 15 82 3 1610-5040 98 879 6 “197 3960 04 96.8 1 76 4020 24 1560 8 16 3480 39 11650-1260 405‘Table 2 Maximum Tensile Residual Stresses in Zirealoy 2 Pressure Tube Rolled Joint Assemblies with High clearance Joint. Rolled portion out side surface 517-621 MPa Unvolled Portion inside surface 558 MPa ‘Table 3 Micro Strains and Stresses (MPa) in Liner tube of $00 MW PHWR during and after Rolling (12% Wall Reduction) 1 -BS denote strain gauge nos. a= axial direction, c= circumferential direction Strsin —Mierosinin Stresses Micro-sain Stresses Gauge No, Daring Rolling ‘Alter Rolling | Bla “19-4810 +1953 3168 © 320 485.5 1612 m2 Baa “1689-3070 1681 3085 © Te 617 700 443 B3a 969 203.4 1197 248.6 ¢ 85 444 146 46.0 Baa 378 830 392 $95 © “23 BA 7 220 Bsa 143 352 5 194 © 6 29 4 223 B64 n 187 25 35 e 3 158 a 20 Bra 0 60 4 49 ° 36 39 29 12 Ba 08 020 “16 25 © 25 49 15 22 ‘Table 4 Residual Stresses (MPa) in Sparger Tube Sleeve Based on Elasto-Plastic Theory Data B1.962E05 MPa, v=03, 6j=23387MPa, H=2528 MPa Experiment Set I. Experiment Set 11 Distance (mim) During Rolling After Rolling During Rolling After Rolling from Sleeve Radial Hoop Radial Hoop Radial Hoop Radial ‘Hoop 8 2816 363 2855 664 2846 577 2883-705 2825 371 2857 S01 2765 18 23 498 16 27S 203 504 265 2337 82-2557 AL ‘241 302 2388 28 251 264 "2452 127 4 “1321 643 “M421 1801337774 125182 477 191 AS13653 1263 455-1255 16 [Notes.() Two experiment seis T& TI were dane on test pieces from same batch of material for sleeve, tbe & inset (Gi) Two values of residual stresses on sleeve at 8 mm ,16 mm & 24 mm are at diametrically ‘opposite locations as shown in fig 10FIG, 3 STRAN GAUGE LocaTion IN PRESSURE TJBE FG, 19 ROLLING OF SPARGER TUBESe2 = Siess 2 ee Principat Stesses in. pa, aa RESIOUAL "STRESSES (SLEEVE, REMOYAL) 18500 ‘ie Paws 20250 no Tue SE Fig. © misoual Srrescee GLEBE SU cur) ? Principal Stresses in pa eae Fig. 8 500. sine Pre Gt Be hw Tune Distance in'mm 408 Stresses in Moo. i Principal Stresses in po $ 4 é Fa ® FesoUaL STRESSES ( sur curTig IN 300 Be Pewte 2, 2 se ME i Ly cman eee) Once in Sm ResouAL StRESSES (HOLE DRLLNG) 300 ie Brn 20 2m NTU Ditonee init geswUal SiRESS” W500 We Pati 2 2 en TURE