ML20091C206

Rev 2 to Acceptability of Class I 24-Inch Feedwater Check Valves

ML20091C206
Person / Time
Site:	Limerick
Issue date:	05/31/1984
From:	Chung Y, Manley R, White R BECHTEL GROUP, INC.
To:
Shared Package
ML20091C186	List: ML20091C183 ML20091C206
References
1183-05EV, 1183-05EV-R02, 1183-5EV, 1183-5EV-R2, NUDOCS 8405300503
Download: ML20091C206 (32)

Text

{{#Wiki_filter:. ACCEPTABILII? OF CLASS 1 24-INCH FEEDWAThd CLIECK VALVES Report for Limerick Generating Station (Units 1 & 2) Prepared for W. C. McDaniel By M.. <-p p hung [ 4 .x Approved R. A. White Coatings and Corrosion Group Manager Approved M l R. A. Manley C [/ l Manager V i Materials and Quality Services Department Research and Engineering BECHTEL GROUP, INC. SAN FRANCISCO f l* l Job No. 8031-000 Tech Report No. 1183-05EV Revision 2 Log No. 114157 8405300503 840525 May 1984 PDRADOCK05000g E

ABSTRACT Toughness properties of A352 Grade LCB castings were evaluated to determine the acceptability of 24 inch Class I feedwater check valves for a 40*F lowest service temperature. A review of the foundry material test reports, deoxidation practice, heat treatment, and microstructures indicated sufficient grounds for acceptance of the valves to not only the ASME Section III Code requirements but also NRC's Standard Review Plan concerning fracture prevention. Various metallurgical ef fects on material toughness properties in terms of Charpy V-notch (CVN) impact values aed nil ductility transition temperatures (NDTT) were discussed. The NDTT data on steel castings in NUREG-0577 were found to be inappropriate for a generic evaluation of steel castings. Using other published NDTT data on steel castings and estimated NDTTs based on actual CVN data for the check valves, this report shows that the toughness of the 24-inch Class 1 feedwater check valve body castings satisfies the requirements of General Design Criteria concerning brittle fracture. A fracture mechanics analysis provided further confirmation that a brittle fracture l is unlikely in all of the eight 24-inch Class 1 feedwater check valves in the Limerick Generating Station Units 1 and 2. l r l l l l I I L 1 -i-l

i ACCEPTABILITY OF CLASS 1 24-INCH FEEDWATER CHECK VALVES CONTENTS Pag ABSTRACT................................................................. i CONTENTS................................................................. ii

1. 0 INTRODUCTION....................................................

1

2.0 CONCLUSION

S AND RECOMMENDATIONS................................. 1 3.0 M ATE R I A LS....................................................... 1 3.1 Description of Feedwater Check Valves...................... 1 3.2 Properties of Valve B ody Cas t ings.......................... 2 1 4.0 EVALU AT ION P ROCEDU RE S........................................... 2 4.1 To u ghne s s R eq u i r emen t s..................................... 2 4.1.1 ASME Section III Requirements....................... 3 4.1.2 TNDT Requirements................................... 3 4.2 TNDT Dat a fo r S te e l Cas t in g s............................... 4 4.2.1 N U RE G- 0 5 7 7 D a t a..................................... 4 4.2.2 SFSA Reports, MPC-13, and Steel Castings Handbook... 4 4.2.3 Quaker Alloy Data................................... 5 4.3 In-Situ Meta 11ography of Valve Body Castings 5 5.0 RE SU LTS OF EVALU ATION........................................... 5 5.1 Conformance to Toughness Requirements 5 5.2 A352 Grade LCB vs Other Steel Casting Grades............... 5 5.3 Applicability of TNDT Dat a in NU REG-05 7 7................... 5 5.4 Estimation of TNDT for A352 Grade LCB Check Valve Body Castings.............................................. 7 5.4.1 Effect of Quaker Alloy's Heat Treatment on TNDT****** 7 5.4.2 E f fec ts of Micros truc tures.......................... 7 5.4.3 TNDT Estimation..................................... 7 5.5 Permis sib le Lowes t Service Metal Temperat ure............... 8 5.6 Fracture Mechanics Evaluation.............................. 8 9 5.7 Acceptability of Check Valves RE FE RE N C E S............................................................... 11 APPENDIX A - Micros tructure of Valve Body Cas t ings....................... 19 -ii-

1.0 INTRODUCTION

General Design Criterion (CDC) 51 requires that "The reactor containment boundary shall be designed with sufficient margin" against brittle fracture. During a preliminary review of reactor pressure boundary materials for compliance to the CDC 51 requirements in June 1983, an NRC staff member calculated permissible lowest service metal temperatures (PLSMI) for various components including 24-inch Class 1 feedwater check valves (IF074 A and B, IF010 A and B, 2F074 A and B, 2F010 A and B). Using the nil ductility tzansition temperature (NDTI) data in NUREG-0577 (Issued for Comment in October 1979)li, his calculations showed +80*F as the PLSMT for the 24-inch feedwater check valve body castings in the Limerick Generating Station Units 1 and 2. This report presents the results of our assessment of the toughness of the particular valve castings using the material test reports from the foundry, published data in addition to those in NUREG-0577, and in-situ metallographic examination of four of the eight 24-inch Class 1 feedwater check valve body castings in Units 1 and 2 (Appendix A). The "new" data we have collected and the results of our analysis of these data showed that the PLSMT for these valves is lower than +40*F. This is the lowest service temperature (LST) anticipated for the check valves during the life of the plant.

2.0 CONCLUSION

S AND RECOMHENDATIONS (1) The basic toughness requirements that the 24-inch feedwater check valves must comply with are those in the 1977 Summer Addenda of the ASME Section III Code according to NRC's Standard Review Plan 6.2.7. The toughness of the check valve castings satisfies these requirements as well as those of the current Code. (2) In addition, an evaluation of nil ductility transition temperatures for the check valve castings indicates that these valves would meet a 40*F permissible lowest service metal temperature. (3) Brittle fracture is highly unlikely in these valves under an upset or normal conditions. (4) The toughness properties of these valves comply fully with the requirements of GDC 51. (5) It is recommended, therefore, that no system modification be made to raise the lowest service temperature from 40*F. 3.0 MATERIALS 3.1 Description of Feedwater Check Valves The following data were extracted from the design report of the 24-inch feedwater check valves prepared by Atwood & Morrill. (a) Valve Type: Swing check valve (Class 1). See Fig. 1. (b) Valve Nominal Size and Pressure Rating: 24 inch - 775# (c) Valve Inlet Diameter: 21 inches (d) Body Material: SA352 Grade LCB (e) Body Minimum Thickness: 2.083 inches including 0.120 inch corrosion allowance (f) Actual Minimum Body Vall Thickness: 2.313 inches. -., _.

(g) Bonnet Thickness tR: 3.5 inches (h) Disc Thickness t: 3.75 inches (i) Design Stress Intensity Valves S : 23.3 kai at room temperature m 18.9 kai at 500*F (j) Standard Calculation Pressure P,: 2100 psi (k) Lowest Initial Temperature: 40*F (1180 - 875 psi) 3.2 Properties of Valve Body Castings The valve bodies were cast in 1975 by Quaker Alloy to the requirements of SA352 Grade LCB. Table 1 is a compilation of data from Quaker Alloy's " Material Test Report" for six heats of the eight valve body castings. In addition to the chemical compositions and tensile properties, it shows Charpy V-notch (CVN) impact test results at 30*F. The contents of residual elements not in the original material test report were obtained only for Heat No. F6137 from the microfilm records in file at Quaker Alloy. The Quaker Alloy documentation revealed that the eight valve body castings received the following heat treatment: (1) Heated to 1690/1720*F, hold 4 hours 25 minutes, air cooled (4/18/75)* (2) Heated to 1630/1650*F, hold 6 hours, furnace cooled to 1440/1460*F, hold 2 hours, furnace cooled to 1420/1430*F, hold 40 minutes, and water quenched (4/22/75) (3) Tempered at 1200*F for 4 hours 15 minutes (4/25/75) (4) Stress relieved (af ter weld repairs) at 1200*F for 4 hours 15 minutes (7/15/75) The castings were examined using an X-ray source (for the Class II shrinkage acceptance level) and a magnetic particle method. 4.0 EVALUATION PROCEDURE 4.1 Toughness Requirements NRC's Standard Review Plan 2) 6.2.7 " Fracture Prevention of Containment Pressure Boundary" states the following acceptance criteria. "To meet the requirements of GDC 1,16 and 51, ferricic containment pressure boundary materials should meet the fracture toughness criteria for Class 2 components identified in the Summer 1977 Addenda of Section III of the ASME Code." This is the basic requirement that these Class 1 feedwater check valves must comply with. Where a pressure boundary component was not " fracture toughness tested", "the staff's assessment of the fracture toughness of materials" is " based -on the metallurgical characterization of these materials and fracture toughness data presented in NUREG-0577."

• The dates apply to Heat No. F5137.

4.1.1 ASME Section III Requirements The toughness requirements for valves in ACME Section III are grouped according to the nominal wall thickness of the connecting pipe. This is 1.812 inches for the 24-inch check valves. For the nominal pipe wall thickness between l-1/2 to 2-1/2 inches, the following requirements apply. 1977 Summer 1983 NC Subsection CVN impact testing Energy ft-lbs 35 average /30 min at the lowest service temperature (LST) or Mils Lateral 40 min 35 average /30 min lower E xpans ion 0,,,R, NDTT (TNDT) LST - TNDT>30*F (Appendix R) Thus, the only requirement for the check valves in the 1977 Summer Addenda of the Code is the 40 mils minimum lateral expansion in CVN tests at the lowest service temperature (LST)* or lower. The material test reports (Table 1) show that this requirement was satisfied. TNDT was not a requirement then. It is an option even in the 1983 Code. (For pressure retaining materials with a nominal wall thickness of connecting piping exceeding 2-1/2 inches, the (LST - TNDT >A) criterion applies.) 4.1.2 TNDT Requirements The valve castings were CVN tested with results that satis fy the toughness requirements of the 1977 Sununer Addenda of the Code. Still, the NRC staff used the TNDT data in NUREG-0577 in calculating the PLSMT. The acceptance criterion is basically the same as that in Appendix R of ASME Section III: the material temperature is at or above the TNDT P us the thickness correction (A) given in l Figure 2. The TNDT of a material can be determined as follows according to NRC's Standard Review Plan PSRP-5.3.4 ), 3 ( (1) Use the TNDT values in Table NF-2311(b)-1 in the Winter 1982 Addenda to the ASME Code. A352 Grade LCB is not included in this table or Table II of the review plan. I (2) The TNDT may be determined by testing. l The check valve castings were not tested for TNDT* (3) The TNDT may be estimated using the correlation given below. l For materials with a minimum specified yield strength between 35 ksi and 45 ksi, each of three Charpy specimens broken at TNDT should exceed 20 ft-lbs. We have used this method in estimating a TNDT l along with other data not presented in NUREG-0577. \\

• 40*F for Limerick -

4'.' 2 TNDT Data for Steel Castings 4.2.1 NUREG-0577 Data i Table 2 of this report represents a portion of the table in NUREG-0577 (or NUREG CR-3009) applicable to the TNDT of steel castings..According to this table, an average TNDT (or NDT) for A-27 and A-216 castings with thickness greater than 1-inch is 35'F and NDT+1. 36 is 57*F. We determined that this is based on the results of two heats (R3 and R4) of A-216 Grade WCB casting from one foundry in an investigation conducted by Banks et a1 ). Table 3 of this report 4 reproduced pertinent data from the paper by Banks et al. It will be shown later 4 that these data should not be used for a generic evaluation of A352 Grade LCB or other castings which were fracture toughness (CVN) tested. 4.2.2 SFSA Reports, MPC-13, and Steel Castings Handbook The Steel Founders ' Society of America (SFSA) issued two reports 5,6) on TNDT of steel castings. Some of the TNDT data in NUREG-0577 are referenced to SFSA (as personal communication rather than to any specific reports). The above two reports present the results of systematic laboratory tests as well as industry data. Although a large amount of data is presented in these reports, only a few are directly comparable to the check valve castings. Some of the highlights of these reports are as follows. (1) A typical TNDT is -10*F for carbon steel castings in a normalized and tempered condition. The range was -75'F to +90*F. (2) The best toughness was obtained in the quenched and tempered condition. (3) " Double normalized and tempered steels averaged 57 degrees lower in NDTT than comparable normalized and tempered counterparts." (4) Deoxidation with aluminum or aluminum-bearing alloy is beneficial to toughness, yielding fine grains. In the double normalized and tempered condition, aluminum lowered NDTT by about 12*F for every O.01% up to 0.035% aluminum. (5) "Small amounts of phosphorus and sulfur can significantly increase the NDTT." (6) "No loss in NDTT due to section size was observed for LCB steel." (7) Producer "A" reported -60*F to -75*F as TNDT for 8 heats of A352 Grade LCB in a double normalized and tempered condition. s'1 ) investigated the effects of section size on the 7 Wallace et toughness of various cast steels. Their results are presented in MPC-13. They found that "the toughness of normalized and tempered steels was only slightly affected by increasing mass." (The size effect was apparent in quenched and tempered cast steels with a high hardenability). The TNDT data for carbon steel castings in the report by Wallace ranged from -26*F for 5-inch thick WCB by one foundry to +32*F for both 1-inch thick WCA by another foundry and for 5-inch thick WCB by a third foundry (all in a normalized and tempered condition). This indicates the difficulty in generalizing the size effects on the toughness of steel cas tings..

Several TNDT data in the Steel Castings Handbook and its Supplement 5 were quoted from the above sources. 4.2.3 Quaker Alloy Data As mentioned before, Quaker Alloy did not conduct TNDT tests for the eight check valve castings as it was not a requirement in 1975. They conducted a few TNDT tests, however, for comparable grades (e.g. WCB) for Navy and Military requirements. Their records for nine heats of castings in the 1970s showed that the tests were conducted to satisfy a +10*F maximum TNDT requirement without actually determining a TNDT. For two heats, however, additional tests determined TNDT at 0 *F and -10 *F, respect ively. More recent data by Quaker Alloy showed -50*F and -60*F TNDT for arc melted heats as in the 1970s and even -70*F and -90*F for arc melted and A0D refined heats, all for normalized and tempered WCB. 4.3 In-Site Metallography of Valve Body Castings The valve body castings were metallographically examined in-situ by the Philadelphia Electric Company (Peco). The purpose was to verify the microstructures in relation to the chemical compositions and heat treatment. The results are presented in the Appendix A. 5.0 RESULTS OF EVALUATION 5.1 Conformance to Toughness Requirements It is clear that the 24-inch check valve castings produced by Quaker Alloy in 1975 complied with the toughness requirement in effect in 1975, 1977 Summer, and even 1983. The only toughness requirement that these castings needed to comply with is the 40 mil minimum lateral expansion in CVN tests. The material test reports for the castings list CVN properties at 30*F, which is 10*F lower than the lowest service temperature (40*F). The CVN values in the material test reports exceed the minimum requirements. This should satisfy the acceptance criteria of NRC's Standard Review Plan 6.2.7 for fracture toughness as well as ASME Section III. As discussed below, the toughness of these castings satisfy even the TNDT criteria. I 5.2 A352 Grade LCB vs Other Casting Grades Table 4 of this report presents a comparison of several carbon steel casting specifications. The grades of castings compared in this table have a specified minimum yield strength from 35 ksi to 45 ksi. A352 is one of only two l steel casting specifications that are specifically designated for low temperature l applications. They have a distinction of requiring CVN impact tests. Drop weight tests for TNDT are included in supplementary requirements which may be specified in a purchase order in addition to the " regular" requirements which i include CVN tests. Neither A-27 nor A-216 has a requirement for toughness. As will be elaborated later, therefore, A352 castings do not belong to the same category as regular A-27 and A-216 castings. 5.3 Applicability of TNDT Data in NUREG-0577 In NUREG-0577, A-216 Grade WCC is grouped together with A-27 Grade 70-40 for TNDT data presentation, simply because their chemical composition. -..

requirements are " virtually identical." In the case of these two grades, this kind of generalization may be reasonable merely based on chemical compositions. It would be, however, not reasonable to apply TNDT data from A-27 and A-216 castings to A352 Grade LCB. The two major variables on toughness of castings are heat treatment (or microstructures) and deoxidation in addition to chemical composition. Foundrymen would pay extra attention to chemical composition control and deoxidation practices well beyond what is apparent from ASTM chemical composition requirements when the castings are made for low temperature applications. None af the ASTM specifications on carbon steel castings have any stipulations on deoxidation practices. A wide variation in deoxidation practices and other controls among foundries is reflected on a wide variation in toughness data. The TNDT data in NUREG-0577 have only a narrow base and biased to the high side of TNDT for castings, particularly for those with thickness greater than 1 inch as explained below. As pointed out in Section 4.2.1 of this report, the TNDT data in NUREG-0577 for castings with thickness greater than 1-inch are based on the results of 12 tests from two heats of castings produced by one foundry. These are heat codes R3 and R4 in a normalized and tempered condition in Table 3 (a) of the report by Banks et al 4) The TNDT ranged from 0*F to +60*F with an average of 35*F. A review of other data [ Table 3(b)] indicates that R3 and R4 happen to be the worst two heats of the 22 heats investigated by them. Other comparable heat codes (D, K, and L) showed TNDT much lower than those for R3 and R4 for a comparable heat treatment condition (NWQT)*. The chemical compositions showed higher phosphorus and sulfur contents for R3 and R4 than other heats. Thus, the TNDT data for castings with thickness greater than 1-inch in NUREG-0577 are biased toward the high side. One of the conclusions by Banks et al waa that an " increase in section size has a generally degrading effect on NDTT when the current results are compared with other industry data for 1-in thick plate." This view is reflected in the data presented in NUREG-0577: -6*F as an average TNDT for 1-inch thick casting and 35*F as an average TNDT for thicker casting. A comparison of their data for various thicknesses (from 2-1/2 to 5-inches in 1/2-inch increments) show no size effects. In fact, in their later paper 8) Banks et al stated that "no discernable ef fect of plate ** thickness (1 to 3-inch) on toughness was noted" for steels including A352 Grade LCC. The SFSA reports and the MPC data support this latter view that the TNDT of steel castings is affected little by section size. There fore, thicker castings should not be " penalized" as much as the data in NUREG-0577 dictate. The paucity of tou2 uss data for steel castings, particularly in terms of h TNDT, and variations in Jeoxidation and other foundry practices make a statistical analysis dif ficult. It is inappropriate to apply the small TNDT data base in NUREG-0577 for generic applications. The above discussion presents a cufficient evidence why the TNDT data in NUREG-0577 should not apply to the A352 Grade LCB check valves. The TNDT of this casting may be estimated from the CVN data in the material test reports and comparison with TNDT data for comparable grades in sources other than NUREG-0577. Some of these data are considered "new ', as they have been published af ter NUREG-0577 was issued and have not been incorporated into NUREG/CR-3009. CNWQT = normalized, water quenched, and tempered

• • cast plate samples,.

l 5.4 Estimation of INDT for A352 Grade LCB Check Valve Body Castings Since the castings were not tested for TNDT and no samples are available for direct measurements, TNDT was estinted " based on the metallurgical characterization of this grade"2), 5.4.1 Effect of Quaker Alloy's Heat Treatment on TNDT The heat treatment condition of a material has a major effect on its toughness. The heat treatment for the check valves is given in Section 3.2 of this report. It is essentially equivalent to an NWQTT condition (normalizing, water quenching, and double tempering). The temperature step-down ( from 1630/1650*F to 1420/1430*F) before water quenching is somewhat unorthodox for low.:arbon steel castings. This was done perhaps to minimize quench cracking in a relatively thick and complex casting according to Quaker Alloy. The NRC staff considers that this is equivalent to normalzing rather than water quenching. Then, the castings should be considered as being orpivalent to a double normalized and double tempered condition, which according to the SFSA reports, relates to low TNDT values. Dilatometer studies in the SFSA reports indicated that the AR3 temperature is 1440*F for a low carbon steel casting with 0.16%C. The AR3 temperature for the check valve containing 0.20%C should be slightly lower. The final temperature (1420/1430*F) before water quenching in the Quaker Alloy's heat treatment would be just about the upper critical temperature. The effect of austenite decomposition should have been minimal. Therefore, the effect of the temperature step-down on the microstructure from water quenching should have been minimal. Thus, the comparison of the water quenching to a normalizing is conservative from the point of its ef fect on toughnest.. 5.4.2 Ef fects of Microstructures The micrographs in the Peco report (Appendix A) show fine grains. The beneficial ef fects of fine grains on the toughness of steels have been well recognized. The finer the grains, the higher the toughness or the lower the nil ductility transition temperature (NDTT). The fine grain structure of the valve castings is directly tied to the aluminum deoxidation practice as pointed out in 4.2.2 (4) and the heat treatment used. The grain aize is 8 or finer for these four valve castings according to the micrographs in the Appendix A. Reference 6 of the report cited the following NDTT for 0.2%C-1.0%Mn steel castings: -10*F for grain size 8 and -40*F for grain sizes 9.5 to 10. l 5.4.3 TNDT Estimation t f a1 ) determined CVN transition curves for heat code R4. 4 Banks et l This is reproduced as Figure 3 in this report. At +30*F, the highest curve indicates 35 ft-lbs for 3 inch thick R4 in an NWQT condition and the lowest at 15 ft-lbs for an NT condition. The data in Table 3(b) indicate TNDT of 0*F and 30*F, respectively, for these conditions. Therefore, for the check valve castings with an average CVN value of 34* ft-lbs at +30*F, the TNDT should be also 0*F. This is the lowest CVN values of all eight castings (Table 1). Therefore, this would correspond to the highest TNDT. This is supported by Quaker Alloy's data. Their other castings which are comparable to the check valve castings in the same period satisfied +10*F TNDT when the castings were produced with a low

• Average of 32-33-38 ft-lbs for Heat No. F6161. _

_, - - _ _ _ -. _ _ _ _ _ _ _. _ _ - - _____~___ ____

temperature toughness requirement, presumably using the scme deoxidation and other foundry control practices. Some of their heats in the 1970s showed 0*F and -10*F TNDT and some recent heats -50*F and -60* TNDT. These data demanstrate low TNDT for steel castings produced by Quaker Alloy. One of the methods provided by the US NRC Standard Review Plan (PSRP-5.4.3) for TNDT determination is given in Section 4.1.2(3) of this report. Assuming that the CVN transition curve of the check valve castings follows the trend given in Figure 3, the temperature at which CVN values of 20 ft-lbs are expected is about -15'F and -40*F for Heat No. F6161 and F6137, respectively. These are the TNDT for these castings according to this method. The SFSA report 6) derived regression equations for calculating TNDT based on chemical compositions. It provided the following (tentative) equations. (1) NNT Condition: TNDT = -9.5 - 26Ni - 46Mn + 53Mo + 56Si (2) NQT Condition: TNDT = -48. 2 - 32Ni - 305C - 1303Al + 78Si - 17Cr "These equations have been established with 99% confidence. Their standard error is 22*F and 29'F respectively." For the composition given in the material test report for Haat No. F6137, Equation (1) gives -20*F + 22*F and Equation (2) -80*F + 29'F. Since the heat treatment condition of the check valves is considered to be between the two, TNDT would be b( tween -20*F and -80*F according to these regression equations. All of the above TNDT estimates, although not perfect, provide a measure of conservatism, if one picks 0*F or -10*F as the TNDT for all of the eight check valve castings. 5.5 Permissible Lowest Service Metal Temperature i Figure 2 provides the following temperature correction values. 30*F up to 2-1/2 inch in thickness l 40*F for 3-inches in thickness 46*F for 3-1/2 inches in thickness 43*F for 3-3/4 inches in thickness These are the temperatures that must be added to TNDT to determine l the permissible lowest service metal temperature (PLSMT). According to ASME Section l III, the nominal wall thickness of the connecting pipe should be used for l determining the minimum CVN requirements or the PLSMT (Appendix R). Since the nominal wall thickness of the connecting pipe to the check valves is 1.812 inches, the PLSMT is 30*F for the check valves if their T NDT is 0*F. Even for the thickest portion of the casting which is 3-3/4 inches for the cover, a T NDT Of l l -10*F would give a 38'F PLSMT. All of these temperatures would satisfy the lowest service temperature of 40*F. i 5.6 Fracture Mechanics Evaluation i l The TNDT approach to designing against brittle fractures was originally developed for ship hull plate in the 1960s. When this method is applied l to castings of various heat treatment conditions without considering stress levels and flaw sizes, the results would become too restrictive. All three factors, l [ _

toughness, stress level, and flaw size should be evaluated together when considering a brittle fracture potential. The data presented in this report and discussion given would satisfy the minimum toughness requirement stipulated by ASME Section III and NRC's review plans. Let us now consider the other two factors, namely stress and flaw size. Since the check valves were tempered and stress relieved at 1200*F for 4 hours 15 minutes each, residual stresses should be ' low. Therefore, only the applied stresses are considered as an approximation. The check valves were designed to the Cla s s 1 requirements using an S value of 18.9 ksi. This is less than 50% of the m actual yield strength of the casting at 40'F. Also, during an upset condition, the maximum internal pressure is only 1180 psi as compared with the 2130 psi standard calculation pressure used. Therefore, the applied stresses from the internal pressure should be much lower than the design stresses. The following stress assumptions for calculating a stress intensity factor (K ) should be, then, much I higher than actual stresses: 6m (membrane stress) = 20 ksi and 6b (bending stress) = 20 ksi. (K ) equation in ASME IX, Appendix A, Using the stress intensity factor I Kt was calculated for an assumed surface flaw which is 3.5-inch long and 1-inch deep. The Class II shrinkage criteria for X-ray examination was conservatively equated to a 3.5-inch long surf ace crack for purpose of calculating Kr. This is an unlikely flaw size, since the castings were X-rayed and magnetic particle examined. AKI value of 90 ksiG was obtained for the above flaw under the assumed stress levels much higher than the maximum stresses expected during an upset condition. The fracture toughness (KIc) of the check valves may be estimated using its CVN data and an approximate correlation between KIc and CVN. For the transition zonc, this correlation is given by the following equation. Kye2 = 2 x E x (CVN)3/2, where E is Young's modulus A CVN energy abLorption of 34* and 74* f t-lbs is equivalent to 110 and 190 ksi G, respectively, according to the above equation. This means that a flaw l large as postulated above under high membrane and bending stresses would not lead as I to a brittle fracture. 5.7 Acceptability of Check Valves l l From the point of preventing brittle fractures, the 24-inch Class 1 l feedwater check valves fully meet the acceptance criteria of not only the ASME Section III Code but also the criteria in NRC's review plans. The foregoing discussions provide sufficient evidence that (1) The check valve castings meet the material toughness requirements of the 1977 Sununer Addenda of ASME Section III, (2) The TNDT evalaation satisfied the permissible lowest service metal temperature, and

• The lowest and the highest average CVN energy absorption values in Table 1. )

(3) Brittle fractures are highly unlikely as indicated by a fracture mechanics analysis. Therefore, the 24-inch Class 1 feedwater check valves satisfy the acceptoce criteria of GDC 51 concerning fracture prevention of the containment pressure boundary.

References:

1. NUREG-0577 (For Comment): Potential for Low Fracture Toughness and Lamellar Tearing on PWR Steam Generator and Reactor Coolant Pump Supports, October 1979 2. KUREG-0800 Standard Review Plan 6.2.7 Fracture Prevention of Containment Pressure Boundary, Rev. 0 - July 1981 3. NUREG-0800 Standard Review Plan PSRP-5.3.4 (Proposed New SRP Section) Fracture Toughness of Steam Generator and Reactor Coolant Pump Supports, Rev. 0 4. W. C. Banks, K. F, Skidmore, H. Schwartzbart: Mechanical Properties of Cast Carbon Steel in Heavy Sections, Journal of Pressure Vessel Technology, v. 96, n. 2, (May 1974), pp. 73-80 5. Steel Founders' Society of America: Research Report No. 75, The Nil Ductility Transition Temperature of Cast Steels, August 1971 6. Steel Founders' Society of America: Research Report No. 80, The Toughness of Cast Steels - NDTT, Charpy V-Notch and Dynamic Tear Tests, May 1974 7. R. Maino, J. Gomez-Gallardo, J. F. Wallace: Section Size Effects on Toughness of Various Cast Steel, in the Metal Properties Council MPC-13 Fracture Toughness of Wrought and Cast Steels, pp. 1-67, 1980 8. R. A. Douty, W. C. Banks, H. Schwartzbart: Mechanical Properties of Welded Cast Steels for Arctic Service, Welding Journal, v.55, n.8, (August 1976), pp. 661-670 TABIE 1 Chemical Compositions, Tensile Properties, and CVN Impact Values for the Check Valve Body Castings Heat No. Chemical Compositions Tensile Properties CVN lupact Properties at +30*F wt I T.S. Y.S. Elong RA Energy Lateral Expansion Shear Area C No Si P S kai kai I I f t-lbs mils F5178-7 0.18 0.70 0.40 0.016 0.019 79.0 53.5 29.0 66.2 55-58-56 42-44-42 40-40-40 F5178-8 0.18 0.70 0.40 0.016 0.019 79.0 53.5 29.0 66.2 64-67-62 53-55-48 90-90-80 F6137-1 0.20 0.76 0.46 0.017 0.013 71.5 42.5 30.0 53.3 43-45-42 41-44-43 40-40-40 F6140-2 0.21 0.79 0.48 0.018 0.012 71.5 42.5 31.5 59.4 68-57-56 59-50-45 30-30-30 F6152-1 0.19 0.82 0.50 0.021 0.015 93.0 61.0 28.0 63.8 47-58-57 41-47-45 40-40-40 F6161-1 0.22 0.72 0.48 0.017 0.015 71.0 42.0 31.5 50.5 32-33-38 40-45-43 40-40-40 F6589-5 0.22 0.95 0.49 0.020 0.011 78.0

48. 0 31.0 66.7 69-65-69 65-54-58 99-80-90 F6589-8 0.22 0.95 0.49 0.020 0.011 78.0

48. 0 31.0 66.7 69-65-69 65-54-58 99-80-90 SA 352 0.30 1.00 0.60 0.04 0.045 65.0 40.0 24 35 Cr. LCB max max max max max 90.0** min ein ein ASME III 3

1977 40 Summer min

• Residual elements:

0. 24% Cr. 0.19% Ni, <0.0012 Mo, 0.103 Cu, 0.04% A1

• • 90.0 kai maximum after A352-74.

CM The last digit (e.g. -7) denotes the casting number. _

-=. - = _. Table 2 1 2 T Data for Cast Steels in NUREG-0577* NDT Material NDT o NDT + 1.3c NDT + 2a Cast Steels. A-27, A-216 1" - 6'F 12'F 10*F 18'F = 35 17 57 69 (heat treated >1"/b condition) 1 A-352 max. -20 (LC 3). 4

• Corrected as revised in NUREG/CR-3009 J

4 i

Table 3 T Data for Various Heats of Carbon Steel Casting

• NDT (a)

(b) charpy V-seetch Impact energy and all-ductility tremeltion Charpy V.snetsh isspect esserfy med mil disetfHty tremettien teaspereteeres of the cast stoelo in the morneelised and teslepered tonoperatores og the seat stoese la une geseashed and toespered see. eendition ditlen Case

see, same CvN lages, tswy NDff.

Case

leesses, w

CYN lease tasy 'Off. No. Th.chasse, a feeermee w -20F, 6-Ibo I No. Theabases, pas, fassenese af -3tF. A-lhe f tonge avween twee a-yese $ O! I NY 9 se IS 13 $ 3 01 I wQ1sa 36 ao 39 31 0 2 10 me 30 30.0 -40

1. #3 212 N' 2 40

+40 3 15 s 30 22.7 -40 3 40

• SO 5

19 ee 27 24.5 30 3-t '2 a0 50 @ D2 1 NNQT 40 me e2 6L0 a 30 60 5 30 +40 3 50 se 72 5* 2 -60 s-l 't e0 + 50 2 ed se 7e 56 0 -ee 5 ed me 70 57.5 -50 $ 88 2-l/2 a0 0 3 50 + 3D

1. D2 1

N**Qf 58 45 ee 71 55 0 3 I 'l 50 + 3D 2 40 se 57 48 7 -60 e

s 0*

30 3 36 es 59 44 7 Sn 4-11 50

• 20 3

33 se 55 a3 0 50 5 40 20 0 D3 I wQfag 85 5 NaF58 I? ee 22 19 3 -40 2 ed on 49 47 0 3 30 so ** h.D +40 to 3 en el 5 t1.5 -10 5 23 es 56 37.3 -40 7 se 3' 30.7 -40 5 e 20 13.7 QI to awQtu 30 0 30 IF 5 S ee s 50 0 e El 1 NwQf5s 33 3 12 ee 23 to 0 -10 3 a5 m 50 80 0 22 se 36 29 5 -E 5 e ee 5 e0 - 30 0 L1 3-14

• 8eQf 53 s to 1

5 'e a 55 0 8 23 ee 27 25.3 20 12 ee 22 17.2 e0 81' S 3 *e 3 5 36 30 L2 bla 17 oo 3 21.3 30 8 15 to 22 19.0 -E ee2 3 32., es 30 0 -50 24 ee 33 F3 20

• *e t e t 0. 2 20 L3 3-l/2 19 me 35 29 2

-10 4 st3 3 '2.e 31 ft I -30 5 eD eeco 5 en la -40 El ht/2 NwQTSB di g 78 58 5 -20 s es 3 20 37 38 5 30 9 82 bl/2 59 % 73 e5.5 30 5 11 ee 36 l' 7 20 4 le se 34 36 5 10 5 55 ee 56 SS 5 -20

-v2

,;5

-o'

• '3

. m.-9.-_e. 5 10 .0 .n ...~...se.ne~,,.. n.. 4-1/2 7 ee 5. 5

7. 5

• 10 5 ee 3 4.1
• do g to 2-l/2 12 en 16.5 14.3 0

3 12 se 14 13 5 0 31/2 M se 18 17.5 0 a 13 se 19 M.0 -10 a-1/2 9 ee 23 14.5 .30 5 9 se 12 10.8

• 10 85 5

NwQTSR el se 47 44.3 -50 17 ee 22 20.1 .E to 3 5 to = 35 25.0 -50 87 5 5 en 5 5.0 50 es 3 21 se h 30.3 -40 5 23 ee 3B N.0 -E 15 se 23 17.0 09 5 810 5 12 ee 15 13 5 -e0 til 3 12 ee e9 31.0 .10 312 3 33 eo 48 200 -80 fl es N 33 0 -80 5 32 m 42 37.0 40 813 3 5 10 so 6e e2.3 -50 ele 3 37 se de 3B 7 50 15 se 29 20.2 30 5

• Omareel oney, se ws '

- lesesel esteen ears lee. e

• From Reference (4), which is the basis for Table 2.

Code numbers with e indicate chemical compositions equivalent to WCB... _

Table 4 A Comparison of ASTM Specifications for Steel Casting

• Required Charpy V-Notch Energy End Use ASTM Specification Grade Heat Treatment' ft-lb. (4 General Apphcation A 27 U70 40 A.N.NT. or QT None High Temp. Service A-216 WCC A,N, or NT None Pressure Vessels at A att u8 M "T*r4T is c.sw9 Low Temperature A-352 LCC

1. , NT or QT 15 (20) at -50 F (-46*C)

Pressure Service A-487 C NT None CN NT None CQ' QT None Heavy Section Pressure vessels A-643 Al NT or QT None. Pressure Vessels and Low Temperature A-757 A2Q QT 15 (20) at -50*F (-46*C) 'A = Anacated, N = Normahzed NT - Normahzed and Tmpered. QT = Quenched and Tempered. ' Grade CQ requires a tensile strength of 80 ksi (550 MPa). All other grades require.70 ksi (485 MPa). Yield and tensile ductahty requirements are identical for all.

• From Steel Castings Handbook, Table 15-31 A27: Specification for Mild-to Medium-Strength Carbon-Steel Castings for General Application A216: Specification for Carbon-Steel Castings Suitable for Fusion Welding for High-Temperature Service A352: Specification for Ferritic Steel Castings for Pressure-Containing Parts Suitable for Low-Temperature Service A487: Specification for Steel Castings Suitable for Pressure Service A643: Specification for Steel Castings, Heavy-Walled, Carbon and Alloy, For Pressure Vessels A757: Specification for Ferritic and Martensitic Steel Castings for Pressure-Containing and Other Applications for Low-Temperature Service -_-

~

-4 s 4-e r-m I ur a s _hvi l., N t 1 n J i% e / +4( ,t- .s g- {^ i A ~u.. A 1 / \\ ~ i a k. 4 U -s u r.: i= ~. 3,.. g _. e m ~ ~ ' v i . :=. .t 4 r* e h r.,*t i T v v i - s=M' -u X 3 ( Sest aan Tae Plos 8 I g dh.a 6-es O..e 4 I g setee ne .~L j:/ l I n.s. / ~ us t e / ,,i. I l mrtavecF ::= l . 2"=" l H d atL E..Eii:.=r:r::rern' . e. :.~.. =_ = g..,.a

1. N.ee N

ELil C l Q =e_%.4 .t :::~~ c 2142? l Fig. 1 Atwood & Morrill's 24 inch feedwater check valve l 1 l (

1 80 50 40 E J W t a 30 2= t 3 1: g20 10 I I 0 0 5/8 1 2 3 4 5 6 Thickness, inches Fig. 2 A temperature correction curve as a function of metal thickness (Reproduced from NUREG-0577 Rev.0, Fig. 1) ;

Si m E NPACT ENERGY

---

LATERAL EXPANSION gfj. ___.

l'l-d /. /, l* g:x?f/,2 / 77 2o a 3' STEEL R4 I 20 r, ,f ,/ HEAT TREATMENT:

• 2\\'NWQT 4 3' NWQT o 3' NT-2 0

-40 0 +40 +S0 +120 +MO +200 +240 +260 TEMPRRATURE, *F Fig. 3 Charpy V-Notch Transition Tamperature Curves. (Reproduced from Reference 4) x indicates the CVN Energy for the check valve. - - A transition curve postulated for the valve body casting Heat No. F6137..

APPENDIX A MICR0 STRUCTURES OF CLASS 1 FEEDWATER CHECK VALVE BODY CASTINGS In January 1984, Phildelphia Electric Company (PECo) performed an in-situ metallographic examination of the four Class 1 feedwater check valve body castings in Unit 1. This Appendix contains the results of that examination, Metallurgical Laboratory Note No. 84-758, February 6,1984. The microstructures of the four valve body castings support the conclusions of the body of this report. 1 l l l l l l I i

Energy Conversion Research Section S10-1, 2301 Market Street February 6,1984 FROM: J. F. DeLong T0: R. A. Mulford SUBJECT : Metallurgical Examination for Microstructure and Hardness of Four 24-inch Feedwater Check Valves, Limerick Generating Station Units 1 and 2 METALLURGICAL LABORATORY NOTE NO. 84-758 Per your request, the four valves were examined for microstruc-ture and hardness. The attached metallurgical laboratory note details the results of the examination. ] } ' {bA 'tDw 'y j (N(Ad <hn F. DeLong JFD/jm Attachment cc: J. P. Gibbons E. C. Kistner J. Moskowitz G. M. Leitch J. J. Clarey J. M. Corcoran J. W. Spencer P. R. Boyles R. H. Logue.

( Energy Conversion Research Section S10-1, 2301 Market Street February 6',1984 METALLURGICAL LARORATORY NOTE N0. 84-758 Metallurgical Examination for Micrnstructure and Hardness of Four 24-Inch Feedwater Check Valves, Limerick Generating Station Units 1 and 2

Purpose:

To determine the microstructure and hardness of each valve. Background : The following request was received from the Lime ick Project Manager: "In June 1983 the NRC reviewed Limerick carbon steel containment and pressure boundary components for conformance with GDC-51. That review identified a potential problem with the feedwater system isolation valves (IF074A and B and 1F010A and B). The NRC contended that the lowest acceptable operating temperature for these valves is 80*F. This would cause a problem when drawing HPCI and RCIC water from the condensate storage tank. The NRC contentions are based on generic data from an unpublished version of NUREG 0577 which has been greatly extrapolated to cover our valves. Please note that the specification and code acceptability of these valves is not in question. PECO requested Rechtel to investigate the specific metallurgical aspects of our valves and they have provided us with a report which establishes the accept-ability of our material at much lower temperatures. This report is l an attachment to BLP-31318 dated 12-3-83 We intend to use this report as t. basis for resolving the NRC concern. Bechtel and PEC0 metallurgists strongly feel that our position before the NRC can be greatly enhanced by knowledge of the actual grain structure of our valves. For this purpose, we request that arrangements be made to allow the PEC0 Metallurgical Lab to examine i l the four valves, which we understand have been turned over. This ~ examination would involve the polishing of a small area of the outside surface, then applying an etching solution and then a fac-i l simile impression tape. Upon completion of the examination, the results will be incorporated into our report and submitted to the Commission for their evaluation, it is necessary that this examina-tion be completed in early January so that the licensability of these valves can be confirmed with the Commission." l '- w, - ,+-.w-r e.,.,,e, m. r y_w.., ,~-,.,.t.__ ,_,y, ,, _ _,,.- -.-_ -,.__._ _,_, m-m_--.e --.-.--,,.-,-..,.r, .. _ ~,, _ _. -

• t -.

J I The check valve material is ASTM A352 grade LC8 with the following chemi-cal percent composition: i Carbon Manganese Silienn Phosphorus Sulfur 0.20% 0.76 0.46 0.017 0.013 I Chromium Nickel Molybdenum Copper Aluminum 0.24 0.19 less than 0.001 0.10 0.04-0.05 The valves, as reported by Quaker Alloy Heat Treatment, were heat-treated as follows: 1690-1720*F 4 hrs. 25 mins., air-cool 1630-1650'F 6 hrs 20 mins. Furnace Cool to: 1440-1460*F 2 hrs. Furnace Cool t'o: 1420-1430*F 40 mins., then water quench Temper: 1200 F 4 hrs. 15 mins. The NRC stated that the heat treatment is not sufficient to produce a quench and tempered microstructure but is equivalent to a double normalized and tempered heat treatment. The following microstructures were recorded on the feedwater check valves using the replica-tape technique and portable hardness equipment. l Conclusions : 1 I Valve No. Microstructure Hardness IF010A Fine pearlite in a R8 79 ferrite matrix l IF0108 Ferrite, pearlite and RR 81 some upper bainite IF074A Ferrite, upper bainite, RB 85 some pearlite 1F0748 Ferrite, upper bainite, RB 87 some pearlite l l !

i l

[ The microstructures consist of ferrite and pearlite, or a combination of ferrite, pearlite and upper bainite. These microstructures would be expected for a 0.20% carbon steel ASTM SA352 grade LCR water quenched from 1420-1430*F temperature. ' The following critical point temperatures are recorded for 1020 material in two source books: Grade 1020 AR3 1515'F(l) AR1 1270'F Grade 1020 fine grain AR3 1470'F(2) AR1 1340'F (1) " Practical Data for Phtallurgists", The Timken Roller Rearing Company Canton, OH, Mgy 1964, p. 50, (2) " Modern Steels and Their Properties", Bethlehem Steel, Sixth Edition, 1966, p. 52. i i

o

f

. The field _ metallography and hardness measurements were prepared by G. P. Monahan of the Materials Test Laboratory. s 04\\. bl m., (onF,DeLong JFD/jm Attachment 9 b 9 _

Microstructure: FEEDWATER CHECK VALVE 1F010A - PHOT 0MICR0 GRAPHS OF MICROSTRUCTURE -t ? " '. A l " & t if:, p,, p ,r; ',a ~ my ,-:v- -~w y;r po.,

y,

W~ ' y%pp

.>c-- t.R.yp p9). As. H / g f'W/Qdam ? sLn [^ ?h e%j,fAkn f $ k N?n g$YW e s 1 &$**1 Y W'N' iY 3,p;.vgg;fy^cg& ( L,'4 ?9 -l M.n.)JN, .$-.'f f**.j!' Q$ ~ }fh\\ '$ h>p p' $l;. N ^ dh 1 (M7% s r

c

. ~ - u Cg$$th .i h$ [Id d Ef N h')k},IM?k,. d, /Yl , sj -h k v Ih j $.gj$$ $if$1'N!?.:23 I E N :d [q)[k.L^d Magnification : 100X Magni fication : 400X r Wh J

3. )";-** e{,

~~~ w.v . \\yI m% eQm;y' ~ np ' $ }:'. }f.' p

• )

./ c r r Jb,,, l,.;ff,;f; Q?I fR,T'%h,nte rM '@ M '" t ' /- -/ b

g.,

,.,g ff M.h>%lfa\\* $ _j.j;p %;?v.3% y9.8',q%%,;g-n,2 ll?}',

'c c

s T ' 4L. M

• v Q?.T

. sky Rx '%Q. wha %.- $ . pp:r i n.;,e:g, ~ p p,N.- <g,,, v.,ja

n

.f..w *.. ?.. %,. ;h,y t.'$ $ { f.w ::. w :,a.s.y/;., [y,. Jf'j%w $, fd:.n y 4 .c x a .w,,- e i s, .,,. I. L m,- m w 9 Mr (4 *-~ s q qi -; y y :,, ', L;pf ;'1 q, vs ~a . lX6),.Q* g. ~~ kc,,g g;4tv.:n, 1;.. ,7, m "rL v' , x. ~, 2 M, :, >* c,.., ,, r.s 9 > Pr. e ,. \\.,.,.,,. .s 7- . y e, -/ e ,= e,f f 1.s 9,, w. m1 w{j... );A 7 Q, ~

k. - f' yfg'T.Y.V j Q ' c,, a.gp'.r#

s. <.

4 (t;4,,,9/C ( ' '!,

'. E,jP6 j. s ( j /'

• '.}o

e L

,,p, wk w.y? ,p ,1 4m,., I' ' T

1. 9, N:a v. ;

f.

• \\'-

b;) g,pl5 }k ~ 4 ... s % %;$ W:y}A..;!;ll Gd %l -Qf ~ . s./ \\ ) .A Magi fication : 800X 4agni fication : 1000X Etch: 2% Nital Hardness : RB 79 MICR0 STRUCTURE : Fine Pearlite and Ferrite /

1 FEEDWATER CHECK VALVE 1F010R - PHOT 0MICR0 GRAPHS OF MICROSTRUCTURE 94 x7;f,.'g[@g&p,s w L1 ~>s' p, .,,_yp;., ' *1 - } y-q ,,.ly;?., w).. g u,:pr. L, ~ v,. u ,.2 4> .x4 + A(.. A m, ~bg~h5], ~' .h k ks 'y>g~',. L.g , {h N "' }{ . '.. ' . 4 ' _ 3. ;:,' G., } } r. 4 Wy -[4 x n ^ ' 3R f 4 f

v. k. j

.j,M ,7 . ?L y u. v-t. hJ Q;p,.n. z$ - g* m, g ;,%,: ' /

2.,r -

. T s,; e~? Q.. m: ., yjQ [,th

Q } }.

+ ><g,v:l? ' $,;.g.g.iQ) . jJQ h

Q r

uT ';. ':: Magnification : 100X Magnification: 400X

s. *1f,P':. hm 4[:l*4o&'h$,},0fSl%

'O *-} 0$ eA*.$ fi  %. '"&Yg} Ii fy],.ls ,1 J 's, . f- " % %. K, f.U, %;b,..Q.~.[ j '-, ff f.k.@...,:::,.W~...blbg5p, jf. Y y q ] },? %

',&-.W.,n '+p.Q{@f, n.

:.

~..% {>>; n i 4:

fa -

s' ~y s i. $i N r9 %, c+ v.,,, .3--.v u, "..g .g:.e,, .r. 4 w ::'c + g.n,4.... v 4

n. - :.

..r.- c-r m ' 'Y 5 IO . ',Y u'Y 5 t e U E M # M E M'; W s # M.h/S.M NQ9 Magni fication : 800X Magni fication : 1000X Etch: 2% Nital Hardness : R B 81 MICROSTRUCTURE: Ferrite, pearlite, some bainite.

E a r FEEDWATER CHECK VALVE 1F074A - PHOT 0MICR0 GRAPHS OF MICROSTRUCTURE E r-s 9 j y a y,,,,_. - m:;.; m,c. ny , s,m.,.. a. n. ~., ".?, v,.,c. f. f a ;, . s.,.. m..

e. 7 7,

v m,

.e c,.r

,~ y . >,-{, ,.p 4 .s. j. ,t, my", ~ { , d,,f f Qf 3. t,ri . g.,, ,. m,m. ~-. c. 7. +J c,+. .- ~

• e..

4- ,o a ,, ; A >_ r.., .c.,.,s +,. , g>w.,c.e,g,g}.At y,.:ii ; ' > yl,.~jp *m. 3.... _ ... n.., .s .c. ., ~ ;, W, 9.a,w,U.. \\ u; r . n ;;.. e,, y a,.. J ' u-. ~..<- u e < a%. ; p A s,. u* ,, >. G,* g p:,

• r

,,,e s

,.,, a f.) s,c w

t s;\\c , 3,. c. m.. g, t.. y e v. 7, 3.

p... -,,;.; c;c.

s

• y.,,

? c a,.., w;.L, n.4,

u. y,

-a p.. A w> ,.t.., s; v 5,,. +c s e g..:, 4,- m'.'.r .. ~ . s.,r... y. ;..,. y~.v,,S, , y 5.a 7.g, _ - ,r _,. r.. /e e n n -w. u k., _.6, %,.. ~. e. v).;,. ,f,,~..,g- ,, y %. ~.... v u s j s ,,d *.g.. ; L g 3# . 7,. ,n n s. a e a = \\

• (=s..;

gg- .F. .a.,j, w. 4 -. m ,,3 n .w r. -

n v,

.u ._L.. z,w. r;. s. . '..,~, y y n. .,, 3. x.,e,, s .,J...c.u,

• q.-~,n^sa a.

w. .$.r y..,.. . ;c m u

;> g

. w n., r. ...n. \\ y; ,c ,q ,;a . g,e..[.

d :.7 9.,.; v.. *.....

mas .r-

.4

...r..- \\.. ,is .a .. ~ r.,, ,g a, t + , 2. i m.., w.v. 4 f- .. w ,i c ; +, ,:. -m.. % v, w, a,- 70 s w ,v ^ ' *." ., "2 4 s 's D ~ .- u, m, e,'u...,?I .*.j',- -\\. / s.. a..v .,.., + / w.,. -y,.z.,., Q vw. q '. ..+ r s ;.,, x. = ,n ~ m - g v e s..3 e t .t. o.,-;m., y,..,. o ~ y-,

9 o

r y,r,,ym ,,. y. s,x c. y _1 ., - a , ;p;.n x9 %; ; '.. ( '.'u!,,. , ; W.m,.cr.l

$. mly.c., J..,.j m.. m,.<.

.W :.; 4- %. g. ",, s. ' t

c.. -

.,c; 4 3, ll i., s.- <..-9 e .m .~ .~ ~ p .*,.? q., s., n Magni fication : 100X Magni fication : 400X =-- ' mm u m^ c EF ~ 7.:g w,' t ,y

• /

  ~

9 ' q, c. 1"4-x- ,,~ .A h,' ,g W;- .) g> ') .e ,f%;$ [ g,%v ,#W 's q, M,;.' 4\\ g.g }.-, .,,s.. g.. ~ em vs

• 9*y C i

%.,(% ^ * > m ? '~,, p._.%<V z 4 >W y W N A 6 %s g

• U '

t+L'C'h E b r. ^' u e p,f g.; - (t i \\ e FO' y., ,p;A 'u au s %- u 2 y4 + i ,, - 1r, + '~, Q . t;) t ( **_\\, y &.,. y,. r.,.;,: < %. c. ' - a ',. s s e.,p,f v i,' /,'.,. ! w e. y,, o g,. s ,,1, 4 ,2i 3, pg. j. "g r x ,*eL.g, E f, A' ( s r r s., .,c y,e .a. c s....Y.' i t e.' 4.r e 4 P -r s .e e~ja n 9 ..s y r *, to -%.g'r,Q (1;.}'g} Q, .f,s. 7 y '} *f a = [,-- . 4., Qs, y',, ) {.gQ'gu. s y /y,4 *g,W/,)%,.?.4y7,. :p., < \\'.F,, . ~,,. 9 ,y. p ::;ad, s .; v. ~,+ u.w, 7 ~., _ .;w ~, w n., ,,,; 7.y,.,(i. - y~.q *,. .~ r e ,, i ,y b t~

,,y ".)q g

i... J A. 3 I o yr.J, . g:, y @... p.,,, 7c (e.~<2.> = ,r 3 ..,.. e. p, ..., 1p.. ,p, n,,/' %q,P, ~ ( 4 = 'Q. A,pe'sj,M, +. g g',,) }..ty 4' 7 % g+:,% n, p .g , a 'J; : Jr /. y E- ,s C" 1 < :' .F -.,3, ,: ~ .,1A l.,' -., ;3 i r J

s. fy:&. 2.:

y ~ L.~.. ;(., y j

+

{ w. ,.. v. ~. i,. (4 8 !,[

7. [-

p

h..

w. A t. T c,....., <.< s a.. h... s, 1. >.o - Magni fication : 800X Magnification : 1000X Etch : 2% Nital m Hardness: RB 85 MICROSTRUCTURE : Ferrite, upper bainite, some pearlite I

O FEEDWATER CHECK VALVE 1F0748 - PHOT 0MICR0 GRAPHS OF MICROSTRUCTURE

qf*fi7Yfj*???.;35'iS%WIWh WiMfi h*MXI % a32 % (Y:( 1. 'W2%Q: ?-

k 'Y qf. h h ~ Y Yh2 CWd%; %,~ f%ppf #p d Eci4S[ amp d k} hflibUt N [f:%l'j M['b{3 })b I W$@,T'WN J^,,'.N'N 's: 5 M W' b t Y': p.*h, T % ' U1iM.t\\.. (W Q.fj p J ij Et.4 &a=, m-m'.v Q Thg 3 [M ? M S %chs,d,. %jcc$cyr'Q &jm.lg;h @p & it 4.M WTh, r E NLa3 S f,1 r.c p p:. w g, #

a ir f

4 tg 4 pj;v .m-9 h;d ?.f, M di-6. O %;,y.; y/.W S "V V W u. , NgmW rWM,f.3 Qi {gfnQ4rif@ ?$hf8$A }yRh.@MrWy$@h YASON-YU hn. ,7i.9 x DU N0G %d.h .9 .&j.W -(.:. I f,hh.?bi 'S ?+Y'k?h h:h.k t.h!i

hk&i&

^ hlif:$h

?h O,.' W w. n : k l'a +:-

k r<y #4 %v; w s N.s A,$ %ad thh$kghlm(@Uf&@M}UC$bs%W R . w :4  %, a, >' '.?l' % a* b=kf$p.Q'hk{h\\i((?lWl&ll$..&'i$;$1. ~ hy s N. l f l: h hn ~ m t J L} A;, Nt Q m4q (t.%p g.; p; &x M j. !. w *,':!r~fg% ' N aC :y,, t a o. :p :.u v .g 4 M Q > .,.D,1 W r.O a yf?% l i% dd(~ c fU,N.;f,4H g;;k.kgi%%G-'8I1 3 % sN M M '6Af/n,Nh'hS s i 4..){ a ?.C h.*)Y, fl*. Q, s...,;.d b' O % %.s:O 6[2; u w w w L uta w E w x !c'P@ Q.Y]R. f e $+tB 5 jM %y .s irs 9/.,... <. + :.2 A it._, hr Magni fication : 100X Magni fication : 500X y ~o, cr p) ' y) j. :-, ~..v;w - -.g l,l %&, yf 'i \\: & y W y};' Q,[')$ Y;~l 2, Q.mQ g-m :--.g p.p r u n c _'fi: I ,h(p.)a :[H pt; r V,.sy,4*c? :.4[r f g. dg/.$. M/ y ;,.f Miij .c h ;,,.N *, M y' k 5 w,.'l. d 'f?pl M w. D sD.4ww &; hi.5 )E }, h j,TD.k(,,%[,/1 .D 1 ',' f %,,,m a x.hYNb %n. fc?h.ff 'hhfk if ll h b'h $.,t)m y f p 4 sw % 'e a% 4 W.~iLf M S QSCW) V y::c;h;4 Nth,:#:>?'f % L nw$wm.,. { 4799)a . 4 M3hMi&Y ; } f.??. .V 4pf.n by ; . JhMqbhh,W.h,. ;. i $ $ $ Y (D Y $ 0.4'$ $ $;Tll b W b ' .% Q % 5 h k " h dL b h,5$ rik $$h: h ni Magni fication : 800X Magni fication : 1000X Etch: 2% Nital Hardness : RB 87 MICR0 STRUCTURE: Ferrite and bainite, some pearlite .}}