ML19064A590

Proposed Alternative Request IR-3-39, Alternative to ASME Code,Section XI, IWA-4221(C), to Permit Two Fillet Welds Not in Compliance with the Construction Code to Remain in Service

ML19064A590
Person / Time
Site:	Millstone
Issue date:	02/28/2019
From:	Mark D. Sartain Dominion Energy Nuclear Connecticut
To:	Document Control Desk, Office of Nuclear Reactor Regulation
References
Download: ML19064A590 (26)

Text

'*

Dominion Energy Nuclear Connecticut, Inc.

~ Dominion 5000 Dominion Boulevard, Glen Allen, VA 23060 DominionEnergy.com February 28, 2019 pr Energy U. S. Nuclear Regulatory Commission Serial No.19-055 Attention: Document Control Desk NSSL/MLC RO Washington, DC 20555 Docket No. 50-423 License No. NPF-49 DOMINION ENERGY NUCLEAR-CONNECTICUT, INC.

MILLSTONE POWER STATION UNIT 3 PROPOSED ALTERNATIVE REQUEST IR-3-39, ALTERNATIVE TO ASME CODE, SECTION XI, IWA-4221 (C), . TO PERMIT TWO FILLET WELDS NOT IN COMPLIANCE WITH THE CONSTRUCTION CODE TO REMAIN IN SERVICE Pursuant to 10 CFR 50.55a(z)(2), Dominion Energy Nuclear Connecticut, Inc. (DENG) requests Nuclear Regulatory Commission (NRG) approval of Alternative Request IR 39 for Millstone Power Station Unit 3 (MPS3). This alternative request would allow a deviation from the requirements of the American Society of Mechanical Engineers (ASME) Boiler and Pressure Vessel Code (Code),Section XI, IWA-4221 (c) to permit two fillet welds which are not in* compliance with the construction code to remain in-service.

In June 2017, it was identified that two inside diameter (ID) fillet welds, FW-12 and FW-30, fabricated during the 'B' Reactor Plant Component Cooling Water (CCP) heat exchanger (HX) replacement project in spring 2016, were n,ade without a 200°F preheat as required by the code of construction (ASME Ill, Table .. ND~4622.7 (b)-1, Exemptions to Mandatory PWHT [Post Weld Heat Treatment]) for P-:-1 materials greater than 1% inches thick. These welds are part of the double fillet-welded slip-on flange connections joining the 'B' CCP HX 24-inch diameter inlet and outlet nozzles to the Reactor Building Component Cooling Water piping headers. Following fabrication of these welds, final visual and magnetic particle examinations were performed. No recordable indications were found and in service leak testing confirmed no leakage.

Studies and tests conducted in support of this alternative request demonstrate that multi-pass, gas tungsten arc welding fillet welds FW-12 and FW-30, made without the 200°F preheat on 1.81-inch thick SA105 low carbon material, produce comparable/more refined microstructures and lower hardness of the weld and heat affected zone (HAZ) as compared to equivalent, Code-compliant, single-pass fillet welds made on 1.9-inch thick material with the 200°F preheat.

The overall global stresses in these flanges are low (limited to less than or equal to X of the allowable stress) and stresses in the inside corner (in the proximity of the ID fillet welds), are even lower and are predominantly compressive. This provides little, if any, driving force for either crack initiation or growth in these ID fillet welds or their HAZs. In the unlikely event of a failure at the ID fillet weld, the most likely path would either be through the weld throat or along the HAZ parallel to the weld fusion line. Neither of these paths would result in a breach of the flange wall or fluid boundary, and. the outside diameter fillet welds, which are fully Code-COmpliant, are adequate to carry u/?Jplf 7

//./(_

Serial No.19-055 Docket No. 50-423 Page 2 of3 to four times the anticipated loads and maintain the pressure boundary and structural integrity.

Testing has shown that fillet welds FW-12 and FW-30, made without the 200°F preheat, have sufficient strength, ductility, and toughness to perform satisfactorily, and are less prone to cracking as evidenced by the mock-up welds when compared to Code-compliant single-pass fillet welds. Because repair or replacement of these welds would present a hardship or unusual difficulty without a compensating increase in the level of quality and safety, DENG proposes continued use of the welds without repair or replacement. The supporting basis for this request is provided in the attachment to this letter.

DENG requests approval of this proposed alternative by March 1, 2020.

This proposed alternative request has been approved by the Millstone Facility Safety Review Committee.

If you have any questions in regard to this submittal, please contact Mr. Shayan Sinha at (804) 273-4687.

Sincerely, Mark D. Sartain Vice President - Nuclear Engineering & Fleet Support Commitments made in this letter: None

Attachment:

Proposed Alternative Request IR-3-39, Alternative to ASME Code,Section XI, IWA-4221 (c) to Permit Two Fillet Welds not in Compliance with the Construction Code to Remain in Service

Serial No.19-055

i. . . .

'.Ila.. Docket No. 50-423 Page 3 of 3 cc: U.S. Nuclear Regulatory Commission Region I 2100 Renaissance Blvd, Suite 100 King of Prussia, PA 19406-2713 R. V. Guzman NRC Senior Project Manager U.S. Nuclear Regulatory Commission, Mail Stop 08 C2 One White Flint North 11555 Rockville Pike Rockville, MD 20852-2738 NRC Senior Resident Inspector Millstone Power Station

Serial No.19-055 Docket No. 50-423 ATTACHMENT PROPOSED ALTERNATIVE REQUEST IR-3-39 ALTERNATIVE TO ASME CODE, SECTION XI, IWA-4221(Cl, TO PERMIT TWO FILLET WELDS NOT IN COMPLIANCE WITH THE CONSTRUCTION CODE TO REMAIN IN SERVICE MILLSTONE POWER STATION UNIT 3 (MPS3)

DOMINION ENERGY NUCLEAR CONNECTICUT, INC. (DENC)

Serial No.19-055 Docket No. 50-423 Attachment, Page 1 of 22 Proposed Alternative In Accordance with 10 CFR 50.55a(z)(2)

-- Hardship Without a Compensating Increase in Quality and Safety --

1.0 American Society of Mechanical Engineers Code Components Affected ASME Code Class: Code Class 3

Reference:

American Society of Mechanical Engineers (ASME)

Boiler and Pressure Vessel Code (Code),Section XI, IWA-4221(c)

Examination Category: N/A Item Number: See Table 1 below Table 1

• ' !< {r.,*i', i D~'sc;~bition

'CC~/:1~*,C FW-12 'B' CCP HX N-4 outlet slip-on flange to pipe ID fillet weld FW-30 'B' CCP HX N-3 inlet slip-on flange to pipe ID fillet weld Materials Involved: 24 inch diameter, 3/8" wall, SA-106 Grade B pipe, 24 inch diameter, 150 lb., SA-105 slip-on flange, ER70S-2, SFA 5.18, Gas Tungsten Arc Welding (GTAW) weld filler metal

Description:

Inside diameter (ID) fillet welds, FW-12 and FW-30,

• welded without 200°F preheat or post weld heat treatment (PWHT).

Components: 24 inch inlet and outlet slip-on {langes from the 'B' Reactor Plant Component Cooling Water (CCP) system heat exchanger (CCP*E1 B HX) to the Reactor Plant Component Cooling Water (RPCCW) system piping header.

System Design

Temperature/Pressure 150°F (inlet), 160°F (outlet)/ 185 psig

Serial No.19-055 Docket No. 50-423 Attachment, Page 2 of 22

2.0 Applicable Code Edition and Addenda

ASME Code,Section XI, 2004 Edition (No Addenda)

ASME Code, Section Ill, 1971 Edition with Addenda through Summer 1973 (original Code of Construction)

ASME Code, Section Ill, 2007 Edition through 2008 Addenda (Code of Construction used for Repair/Replacement Activity) 3.0 Applicable Code Requirements ASME Code,Section XI, 2004 Edition (No Addenda)

• IWA-4221 (c) states in part, "As an alternative to (b), the item may meet all or portions of the requirements of different Editions and Addenda of the Construction Code, or Section Ill when the Construction Code was not Section Ill ... Construction Code Cases may also be used."

ASME Code, Section Ill, 2007 Edition. through 2008 Addenda ASME Section . Ill, ND-4600 contains the following requirements regarding postweld heat treatment of ASME Ill Class 3 welds:

• ND-4622.1 states, "Except as otherwise permitted in ND-4622.7, all welds, including repair welds, shall be postweld heat treated."

• ND-4622.7 states, postweld heat treatment is not required for welds exempted in Table ND-4622.7(b)-1.

• Table ND-4622.7(b)-1, "Exemptions to Mandatory PWHT," requires that all welds in P No.1 material over 1% inches thick with a nominal thickness of%

inches or less are exempt from postweld heat treatment provided a minimum preheat of 200°F is applied.

4.0 Reason for Request

In June 2017, during work order preparation for the MPS3 'A' CCP HX replacement project, it was identified that two ID fillet welds, FW-12 and FW-30, fabricated during the 'B' CCP HX replacement project in spring 2016, were made without a 200°F preheat as required by the code of construction (ASME Ill, Table ND-4622.7(b)-1, Exemptions to Mandatory PWHT). As illustrated in Figure 1, these welds are part of the double fillet-welded slip-on flange connections joining the 'B' CCP HX 24-inch diameter inlet and outlet nozzles to the RPCCW piping headers. '

Serial No.19-055 Docket No. 50-423 Attachment, Page 3 of 22 Figure 1 Schematic of Slip-on Flange Welds These slip-on flanges are welded to 3/8-inch wall SA106 Grade B piping using an outside diameter (OD) 9/16-inch fillet weld and a 1/4-inch ID fillet weld. In this case, the larger OD fillet welds did not require a 200°F preheat since the thickness of the flange hub was less than 1% inches. However, the smaller ID fillet welds, FW-12 and FW-30, welded to the 1.81-inch thick flange required a 200°F preheat for PWHT exemption. Multi-pass GTAW was employed to complete these welds.

The completed welds satisfied Code-required visual and magnetic particle examinations and in service leak tests. These welds have been in place since the spring of 2016 when the 'B' CCP HX was replaced.

5.0 Proposed Alternative and Basis for Use 5.1 Proposed Alternative Pursuant to 10 CFR 50.55a(z)(2), DENC requests Nuclear Regulatory Commission (NRC) approval of Alternative Request IR-3-39 for MPS3. This alternative request would allow a deviation from the requirements of the ASME Code,Section XI, IWA-4221(c) to permit fillet welds, FW-12 and FW-30, which are not in compliance with the construction code to remain in-service.

DENC considered the option of replacing fillet welds FW-12 and FW-30; however, restoring these welds to full Code compliance would require the 'B' CCP HX to be out of service. As required by the MPS3 technical specifications, two CCP pumps and HXs must be operable during Modes 1, 2, 3, 4, 5, and 6 (until the water level above the reactor vessel flange is greater than or equal to 23 feet). Taking one CCP HX out of service for weld repair eliminates the system's defense-in-depth should either one of the other two GCP HXs become inoperable during this time.

Additionally, since the three MPS3 CCP pumps and HXs are located in the same area of the auxiliary building, repair of the 'B CCP HX fillet welds would require establishment of two protected equipment zones for the 'A' and 'C' CCP pumps and HXs. Because the weld repair evolution would require rigging and movement of large piping elements in an area where other safety related equipment is

Serial No.19-055 Docket No. 50-423 Attachment, Page 4 of 22 located, work in this area is difficult and poses an industrial safety risk to plant personnel and risk of damage to plant equipment.

The technical basis of the request demonstrates that fillet welds FW-12 and FW-30, made without the 200°F preheat, have sufficient strength, ductility, and toughness to perform satisfactorily, and are less prone to cracking as evidenced by the mock-up welds when compared to Code-compliant single-pass fillet welds.

Therefore, any decrease in the level of quality of the welds due to not performing the 200°F preheat is not commensurate with the hardship associated with replacing the welds. Thus, DENC proposes continued use of the welds without repair or replacement on the basis that repair or replacement would present a hardship or unusual difficulty without a compensating increase in the level of quality and safety.

5.2 Basis for Use The basis for this alternative request is supported by the following:

1) The subject welds have passed the Code-required final visual and magnetic

• particle examinations and post installation in-service leak testing. No cracking or other recordable indications were observed or are anticipated due to the omitted 200°F weld preheat.

2) Representative weld mock-ups and welding procedure qualification testing demonstrate that the subject welds and HAZs have acceptable properties with 75°F minimum preheat.

3) The stresses around the ID fillet welds are very low (<25% of the stress allowable) and are primarily compressive. Therefore, the chances of crack initiation or propagation are remote.

4) Code stress analysis demonstrates that if the ID fillet welds are not present (or are postulated to have failed completely) the OD fillet welds alone are fully capable of maintaining structural and pressure boundary integrity.

5) In the highly unlikely event of a failure of both the ID and OD fillet welds, the failure would be evident as a leak that would be detected during routine walkdowns before the failure would compromise the system or component structural integrity.

5.3 Code Requirements for Weld Qualification, Preheat/PWHT ASME Ill requires that welding procedure specifications (WPS) be qualified in accordance with ASME IX. The WPS used for making FW-12 and FW-30 was qualified for welding up to an 8-inch thick material without elevated preheat or PWHT. Specifically, Dominion Corporate Welding Manual procedure qualification

Serial No.19-055 Docket No. 50-423 Attachment, Page 5 of 22 record (PQR) 135 joined a 1%-inch SA-516 Grade 60/70 plate using the GTAW process with: ER70S-2 filler metal, a 70°F minimum preheat, and no PWHT. All mechanical test results, including impact tests of the welds and HAZs, met ASME IX and ASME Ill (Class 1, 2, and 3 requirements) for piping welds. Impact testing at -20°F averaged 90 ft-lbs. for the weld and 119 ft-lbs. for the HAZ compared to the ASME Ill requirement for SA-516 of 20 ft-lbs. and SA-105 of 15 ft-lbs. The system does not require impacts, but if it had been required, the ASME Ill (71/73) requirement would have been 15 ft-lbs. The WPS used for making FW-12 and FW-30 is fully qualified per ASME IX and Ill for welding with the minimum preheats used.

ASME Ill, Table ND-4622.7(b)-1 also specifies additional conditions on base metal thickness, weld thickness, preheat, and maximum carbon content for which mandatory PWHT may be exempt. DENC did not satisfy the minimum 200°F preheat required by this table for P-No. 1 (carbon steel) material greater than 1%

inches thick to be exempt from the PWHT requirement.

The preheat and PWHT rules of the Code are intended to be conservative because they must address multiple material specifications, products forms, chemistry variations, and weld joint configurations. ASME Ill, Sections NB, NC, and ND (Classes 1, 2, and 3) have similar conservative preheat and PWHT rules. A likely reason for establishing limits on thickness and carbon content was to address base material hardenability and the potential for hydrogen-induced cracking from monatomic hydrogen that is introduced during some welding processes.

Increasing carbon content increases hardenability. Thicker base metals are more effective heat sinks that cool the weld and HAZ faster. Rapid cooling rates in combination with higher hardenability may lead to undesirable hard welds and HAZ microstructures such as untempered martensite. These microstructures, when present, along with sufficient monatomic hydrogen and high tensile stresses, can lead to hydrogen-induced cracking in the weld and/or base metal HAZ. Application of preheat is one method to reduce the potential for hydrogen cracking because it slows the cooling rate offsetting a portion of the rapid cooling caused by thick materials. It may also drive off surface moisture (a potential source of hydrogen) and increases the temperature which may increase diffusion of hydrogen out of the weld and HAZ over time. Use of the GTAW process provides ultra-low hydrogen (typically <2 ml/100 grams) as another mitigating factor against hydrogen cracking.

Although the ASME Ill Code sets a thickness limit of 1% inches for P-No 1 materials, there is no step change in cooling rates and/or the resultant weld and HAZ microstructures that occur as thickness exceeds 1% inches. In

• fact, the mock-up test results described below demonstrate that the subject slip-on flange ID, multi-pass, GTAW fillet welds on the 1.9-inch nominal material made without the Code-required 200°F preheat will produce more optimal microstructures and lower HAZ hardness than Code-compliant single-pass fillet welds made with 200°F preheat under the same conditions.

Serial No.19-055 Docket No. 50-423 Attachment, Page 6 of 22 5.4 Mock-Up Testing Results Mock-ups were prepared for testing and evaluation ,of fillet weld specimens using the same welding technique and parameters used in the fabrication of fillet welds FW-12 and FW-30. A section of 24-inch diameter, 3/8-inch wall, SA106 Grade B pipe and a spare 24-inch, 150 lb., SA-105 slip-on flange* was selected for mock-up testing. The flange material matched the carbon content and slightly overmatched the carbon equivalent (calculated using ASME IX QW-403.26 formula) and strength of the production flanges. Sample specimens were fabricated which consisted of single pass and multi-pass fillet welds made with the same 75°F ambient preheat as the production welds, as well as single and multi-pass fillet welds with a 200°F preheat. Samples at 1%-inch thickness (max allowed without 200°F preheat) and 1 inch (representative of the flange hub weld joint) were also made for comparison. The carbon content and carbon equivalent for the installed flanges, the mock-up flange, the original PQR 135, and the SA 105 base material and filler metal are provided in Table 2.

Table 2 Material, Description, Heat No., Carbon Content, and Carbon Equivalent

.*J~t~r*:.*.1teni* * :car.hon Carbon Eqllf~it,~1,1t

• Jc~t, * *

"B" CCP HX 24" 150 LB Slip-On Flange SA105 0.19 0.42 Nozzles N3 & N4 Heat No. 217W891 24" 150 LB Slip-On Flange SA105 Mock-up 0.19 0.43 Heat No. 227J800 1 W' SA-516 Grade 60/70 PQR 135 0.19 0.42 Bethlehem/Lukins HT# R 4981 Base Material Specification SA105 0.35 0.67 Maximum Permissible ER70S-2 3/32 Production & Mock-up Weld Metal 0.04 0.26 ER70S-2 1/8 Production Weld Metal 0.05 0.25 ER70S-2 1/8 Production Weld Metal 0.05 0.26 ER70S-2 1/8 Mock-up Weld Metal 0.06 0.26 SFA 5.18 ER70S-2 Weld Filler Metal Spec Maximum 0.07 0.41

• One flange was used to mock up the ID fillet weld to 1.9" material, the OD fillet weld to 1" hub material, and a section was machined down to mock up an ID fillet weld to 1 Yi material.

Serial No.19-055 Docket No. 50-423 Attachment, Page 7 of 22 The hardness evaluations of the test specimens showed the highest levels of hardness to be in the coarse-grained HAZs (CGHAZ) of the flange (nearest the weld fusion line) of the single pass welds shown as 1A, 2A, 3A, and 4A in Table 3 and Figure 2 below. The hardness of the fine-grained HAZ (FGHAZ), the weld metal, and the pipe is substantially less than that in the CGHAZ of the heavy flange. Therefore, DENC's evaluation focused on the CGHAZ of the flange as it is the area where the highest hardness will occur and is thus the limiting area from the standpoint of excessive hardness causing degradation in the weldment.

Table 3 Mock-Up Vickers Micro-Hardness Results Weld Near FL Mockup Description Low Avg High Low Avg High Low Avg High Low Avg High lA Single pass 1.9" 75° 289 334 395 450 464 488 281 322 348 245 303 331 18 Multi-pass 1.9" 75° (Represents FW-12 & FW- 212 266 338 221 361 444 243 264 295 238 252 261 30) 2A Single pass 1.9" 200° 228 316 352 399 431 450 261 273 289 279 295 316 28 Multi-pass 1.9" 200° 193 234 289 209 264 355 217 247 309 235 245 252 3A Single pass 1 Yi" 75° 263 318 393 333 444 523 339 349 364 353 363 374 38 Multi-pass 1 Yi" 75° 209 245 285 223 305 409 255 272 293 250 263 271 4A Single pass 1" 75° 263 292 315 432 443 451 282 316 352 316 333 354 48 Multi-pass 1" 75° (Represents Installed 194 218 257 205 274 378 208 237 260 202 219 238 Compliant Hub OD Fillet Welds)

Figure 2 - Average Vickers Micro-Hardness Results 500 450

• Fine Grained HAZ 400 350

• Coarse Grained HAZ 300 250

• Weld Deposit Near Fusion line 200 150

• Weld Throat 100 so 0

lA 1B 2A 28 3A 38 4A 48

Serial No.19-055 Docket No. 50-423 Attachment, Page 8 of 22 5.5 Effect of Carbon Content on Peak HAZ Hardness ASME Ill does not specify hardness limitations or microstructural requirements for exemption from PWHT. The applicable PWHT exemptions provided by the Code in Table ND-4622.7(b)-1 are for P-No. 1 with carbon contents up to 0.35% as allowed in SA-105. Research by the Electric Power Research Institute has shown that this Table can be conservative when applied to materials with lower. carbon contents (<0.20%) as these have inherently lower hardenability (Ref. 1 and 2).

Using the Creusot-Loire equation (per ASME Pressure Vessel and Piping paper -

Ref. 3) to calculate predicted peak HAZ hardness of the production and mock-up flange materials and for SA 105 material with the maximum carbon contenUcarbon equivalent for comparison, provides the following predicted peak hardnesses:

TABLE4A Flange Base Material: Creusot-Loire Predicted Peak Hardness ti, 0 tJ .;,;;

lfsec!:t5t

}>,' ;

217W891 0.19 0.24 1.14 0.06 0.08 270 227J800 0.19 0.23 1.19 0.08 0.07 270 0.35 0.35 1.05 0.4 0.3 270 SA-105 Material Spec Max.

As shown in Table 4A above, the predicted peak hardness of the 0.35% maximum carbon content allowed by the SA-105 base material specification is 614 HVm.

This is 160 HVm higher than that of the 0.19% carbon content production and mock-up flanges. This 35% (i.e., 160/454 x 100%) increase in peak hardness could indicate a significant increase in the potential for brittle and crack susceptible microstructures in the CGHAZ. Thus, in comparing their peak hardness, the worst case allowed by the Code is predicted to be up to 35% higher than that of the mock-up and the actual production flanges.

It should be noted that the predicted peak hardness value of 454 HVm is very close to the actual measured average CGHAZ hardness values from Table 3 for the mock-up flange single pass fillet welds (464 - 431 HVm) with differences in cooling rates from the 200°F preheat in specimen 2A accounting for the largest deviation. This along with similar results comparing Table 48 predicted peak hardness to Table 3 measurements for the lower carbon equivalent weld metal single pass welds (363 - 295 HVm), confirms a strong positive correlation between the predicted and measured values with a Correlation Coefficient (R) of 0.96.

Serial No.19-055 Docket No. 50-423 Attachment, Page 9 of 22 Table4B Weld Filler Metal: Creusot-Loire Calculated Peak Hardness 1195V 0.05 0.52 1.09 0.03 0.04 Production Flanges 1/8 1263V 0.05 0.53 1.1 0.03 0.05 Production Flanges 1/8 1274V 0.06 0.52 1.08 0.03 0.03 270 }33f Mock-up 1/8 dia filler It is already well established that increasing carbon content increases hardenability and Martensite hardness. Therefore, no attempt was made to procure special material at or near the maximum carbon content to be used for a direct comparison as the above predicted peak hardness values and general metallurgical knowledge were considered sufficient to make it apparent that the effect of lowering the carbon content from the maximum allowed 0.35% to 0.19% drastically reduces the hardenability and peak hardness of the HAZ. In this case, the use of the lower carbon content (0.19%) is estimated to reduce peak hardness by 26% (i.e.,

160/614 X 100%).

Although this 26% reduction in peak hardness is reasonable and supported by literature, this value is not used as a quantitative measurement in support of this alternative request as it was not derived directly by experimental measurements.

It does, however, provide additional qualitative margin for- consideration when comparing the maximum reasonable expected hardness of a worst-case ASME Ill Code-acceptable weld to that of fillet welds FW-12 and FW-30.

5.6 Effect of Multi-pass vs. Single Pass on CGHAZ Hardness In addition to the benefits of the relatively low carbon .content and carbon equivalent, fillet welds FW-12 and FW-30 also benefited from the fact that they were deposited using multiple weld passes. Single-pass welds are not restricted

• by the Code. Single-pass welds are used successfully throughout the industry and were used to establish Code-acceptable bounding conditions for comparison to the multi-pass fillet welds FW-12 and FW-30.

The metallurgical microstructures and hardnesses that result from following ASME Ill Code rules were used to make a comparison of the metallurgical microstructures and hardness of ASME Ill Code-compliant welds with the results of the mockup representing non-compliant welds FW-12 and FW-30.

Table 5 below shows a comparison of the Vickers micro-hardness measurements of the CGHAZ of Code-compliant single-pass fillet welds with those of the mock-up representing FW-12 and FW-30 multi-pass fillet welds. Other variables not presented in Table 5 were maintained constant.

Serial No.19-055 Docket No. 50-423 Attachment, Page 10 of 22 Table 5 Comparison of Micro-Hardness for Code-Compliant Single-Pass Welds vs. Multi-Pass Production Weld Single/~ '.Thickness( 7; Preheat' . *~: AsME IU :,}sf IS Vickers Mi~ro!Hardness CGHAt' Multii:' (in.) *.*: ( °F) . Code~Complian(11 :, ,*, Min *.* ;: 'Avg.* .. Max *.*

2A Single 1.9 200 Yes 399 431 450 3A Single 1.5 75 Yes 333 444 523 4A Single 1.0 75 Yes 432 443 451 1B Multi 1.9 75 No 221 361 444 It is noted that in each column the hardness of the multi-pass fillet weld representing FW-12 and FW-30 (18), made without preheat, is lower than the corresponding hardness of the Code-compliant single-pass fillet welds, including the single-pass weld made with the 200°F preheat (2A).

The effect of multi-pass versus single-pass with all other variables being constant is calculated by comparing the CGHAZ hardness of the single-pass weld (A) to the multi-pass weld (8) for each set of conditions (1 through 4). The results are shown in Table 6 below.

Table 6 Comparison of Micro-Hardness for Single-Pass vs. Multi-Pass Welds Singl~{* : Thickness : *:\~reheat ~:iASME Ill ... ,\'/ickers Micro:IHardness C(i'HAZ.

M~ltij} .* (in:) . :1 *::: (l)F) 'd1~~CompliaM:l* :: -Avg,, ' {l,1. ' %chang~ft 1A Single 1.9 75 No 464 1B Multi 1.9 75 No 361

-103/464 -22%

~'.~ *,,

+':' "*

~* *.*:

2A Single 1.9 200 Yes 431 2B Multi 1.9 200 Yes 264

-167/431 -39%

t' ... ' *,

~J

• .*.1r* . :, I>< ..

. .r,.'

3A Single 1.5 75 Yes 444 3B Multi 1.5 75 Yes 305

-139/444 -31%

',;{ ' .t; *'  ;; .3

... *,;\.' .. / .*, '":') . .**,\

4A Single 1.0 75 Yes 443 4B Multi 1.0 75 Yes 274

-169/443 -38%

,,~"4. : .

• \ ~ l t;, ,;

JJ.* ,* ... 1*:,,; ,r",

• i;;s '

-578/1782 -32%

Serial No.19-055 Docket No. 50-423 Attachment, Page 11 of 22 The average effect of multi-pass welding versus single-pass welding is a 32%

reduction in hardness of the CGHAZ.

5.7 Effect of 200°F Preheat vs. Ambient (75°F)

The effect of preheat was calculated by comparing the average CGHAZ hardness for single-pass without preheat (1A) to single pass with 200°F preheat (2A) and multi-pass without preheat (1 B) to multi-pass with 200°F preheat (28) and averaging the two. The results are shown in Table 7 below.

Table 7 Comparison of Micro-Hardness for Single-Pass and Multi-Pass Welds With and Without Preheat

~:No~ Stngl~Zi y*+:ij;z i Thickn~ss* ***Preheat Multht (in.r  : *. (oF) lA Single 1.9 75 No 464

-7%

2A Single 1.9 200 Yes 431

-33/464

,:,.... . "';*f,j ..

• { .

18 Multi 1.9 75 No 361 28 Multi 1.9 200 Yes 264

-97/361 -27%

~:;, -'. )'.f I'"; ..

I*<

7

-130/825 -16%

The average effect of welding with the 200°F preheat as compared to welding without the 200°F preheat is a 16% reduction in hardness of the CGHAZ.

5.8 Comparison of the Effects of 200°F Preheat vs. Multi-Pass Welding From the tables above, the reduction in average CGHAZ hardness on P-No.1 material matching the carbon content and thickness of the actual production welds (FW-12 and FW-30) as a result of using 200°F preheat is 16%, over a range from 7% to 27%. This is specific to conditions representative of the actual production welds.

The effect of multi-pass welding was similarly shown to reduce the average CGHAZ hardness by 32%, over a range from 22% to 39%. While the ranges have some overlap, the effect of multi-pass welding was significantly greater than that of the 200°F preheat for conditions bounding the actual production welds (FW-12 and FW-30).

This is evident in the direct comparison of the three ASME Ill Code-comp_liant single-pass fillet welds 2A, 3A, and 4A with the multi-pass weld 1B representative of the actual production welds (FW-12 and FW-30):

Serial No.19-055 Docket No. 50-423 Attachment, Page 12 of 22 Table 8 Comparison of Micro-Hardness for Code-Compliant Single-Pass Welds vs. Multi-Pass Production Weld (same data as Table 5) s_,

Single/::: . Thk P,reheat iASME 111 , Vickers Micro Har~ness CGHAZ *

l\lo. **(in;) Avg~,**

-;-:;, Multi}{

,~f(9F) CodE!,~ompliant ;'.i[,)llin MaxJ/

2A Single 1.9 200 Yes 399 431 450 3A Single 1.5 75 Yes 333 444 523 4A Single 1.0 75 Yes 432 443 451

,:;;,.,.: ~~*tl

• j  :,

,.*'l.p' /.J  ;,'.l:,:1,; ),it ,. .,h;i lB Multi 1.9 75 No 221 361 444 In each case, the hardness of the multi-pass fillet weld (18), representing FW-12 and FW-30 made without preheat, is lower than the corresponding hardness of the Code-compliant single-pass fillet welds, including the single-pass weld made with the 200°F preheat (2A).

While it is recognized that the highest reduction in CGHAZ hardness is achieved by combining both 200°F preheat and multi-pass welding, as in specimen 28, requiring both 200°F preheat and multi-pass welding exceeds the requirements of the applicable code. In a direct comparison of the effectiveness of 200°F preheat versus that of multi-pass GTAW welding for the specific conditions representing FW-12 and FW-30, multi-pass welding was determined to be approximately twice as effective at reducing the average CGHAZ hardness.

Thus, it has been demonstrated that the CGHAZ hardness of the 18 multi-pass fillet weld, representative of the production welds (FW-12 and FW-30), is well within the range of hardness found in the other three ASME Ill Code-compliant single-pass fillet welds (2A, 3A, and 38). As such, this evaluation of representative welds concludes that the omission of ASME Ill Code-required 200°F preheat while depositing FW-12 and FW-30 will not result in excessive hardness or degradation beyond conditions bounded by the three Code-compliant single pass fillet welds.

Serial No.19-055 Docket No. 50-423 Attachment, Page 13 of 22 6.0 Photomicrographs Single Pass Multi-Pass Toe - last bead Multi-Pass Root - first bead 19,. . CGHAZ 1.9" 75° F Multi-Pass Toe 18- CGHAZ 1.9" 75aF Multi-Pass Root 2A- CGHAZ 1.9" 200a F Single Pass All Micrographs are 40 0x prior to reproduction 3% Nital Etch

Serial No.19-055 Docket No. 50-423 Attachment, Page 14 of 22 Single-Pass Multic.Pass Toe - fast bead Multi-Pass Root - first bead All Microg phs are 400x prior fo reproduction 3% Nita.1Etc*

Serial No.19-055 Docket No. 50-423 Attachment, Page 15 of 22 6.1 Evaluation of Photomicrographs and Microstructure The microstructures of all single pass welds (left column - 1A, 2A, 3A, and 4A) consist of a phase mixture of low carbon Martensite, 8ainite, and Ferrite. The corresponding multi-pass welds (middle and right columns - 18, 28, 38, and 48) all show the same phases to varying degrees with evidence of grain refinement commensurate with the number of subsequent weld passes (higher at the root and less at the toe).

Comparing the individual photomicrographs for each row moving from left to right, the initial pass entitled, "Single Pass" has coarser and predominantly columnar grains with a mix of different structures and phases. As the images are viewed from left to right, the grains become more refined, with smaller and much more equiaxed grains in the last column. These smaller, more equiaxed grains are evidence of grain refinement from the multiple reheat cycles experienced during multi-pass GTAW welding as numerous small beads are deposited one over another. The resulting fine equiaxed grains contribute to the high values for impact properties produced in the HAZ of the multi-pass GTAW weld as reported in PQR 135.

For comparison, the microstructures in photomicrographs 18, which are representative of the actual production welds (FW-12 and FW-30), have a more uniform and more refined appearance than the Code-compliant single-pass fillet welds shown in 2A, 3A, and 4A. There is none of the lighter colored phase present in the 18 photomicrographs. The presence of carbon and/or carbides darkens the image, thus the lack of the lighter phase areas would be indicative of increased tempering-releasing carbon from the interstitial sites to either dislocations or other higher energy locations in the matrix. This is significant in that it provides evidence that there is no untempered Martensite.

6.2 Combination of Hardness and Microstructural Evaluations Hardness alone is not a good predictor of strength, ductility, or toughness unless the microstructure from which it was measured is known. In carbon steels, the best combination of strength, ductility, and toughness comes from fine grained tempered Martensite. Evaluation of the micrographs and micro-hardness measurements together indicate that the CGHAZ microstructures contained some Martensite since neither 8ainite nor Ferrite can reach the peak hardnesses that were measured.

Combining the contributions of both the mock-up hardness and microstructural evaluations, it can be seen that the 18 specimens (representing the actual production multi-pass fillet welds FW-12 and FW-30) have both a finer, more equiaxed grain structure with no indications of any untempered Martensite and lower hardness compared to the Code-compliant single-pass fillet welds in 2A, 3A, and 4A. The combination of improved microstructure and reduced hardness

Serial No.19-055 Docket No. 50-423 Attachment, Page 16 of 22 provides confidence that fillet welds FW-12 and FW-30 have a better combination of mechanical properties and are less prone to contain small areas of brittle or crack-susceptible microstructural elements than the Code-compliant

. single-pass fillet welds represented by specimens 2A, 3A and 4A. As such, the mock-up has demonstrated that production welds FW-12 and FW-30 have a combination of microstructures and micro-hardness properties that provide assurance of mechanical properties and resistance to cracking at least equal to that of ASME Ill Code-compliant single-pass fillet welds represented by specimens 2A, 3A, and 4A.

7.0 Stress Evaluation To determine the global stress on the 24-inch diameter inlet and outlet nozzle slip-on flanges, the ASME Code pipe stress equations for dead load, thermal, occasional upset, and occasional faulted were used. Stresses for the specified inlet and outlet flanges on the MPS3 'B' CCP heat exchanger are not explicitly stated in the pipe stress calculation NP(B)-X7202, Rev. 4 (Ref. 4); however, they may be derived from the moments calculated within the pipe stress programmatic output. These Code equations calculate pipe stress based upon longitudinal stress (pressure) and bending stress (loading). As the longitudinal stress is constant at a specific location and pressure, variations in the bending moment have non-linear effects on the total stress.

The piping flanges at the specified location in NP(B)-X7202, Rev. 4 are represented by node 110/115. To determine the maximum bending moment (MA, Ms, or Mc), the directional moments are combined using the Square Root of the Sum Squares (SRSS) method. The moments for each are specified below.

Calculation NP(B)-X7202, Rev. 4 shows the absolute maximum moments for the flanges at nodes 110/115 for each scenario.

Table 9 Absolute Maximum Moments for the Flanges at Nodes 110/115 Deadload (ft-lb) lhermal (ft-lb) OBEI (ft-lb) OBEA (ft-lb) SSEI (ft-lb) SSEA(ft-lb)

M, I M, I M: M. I M.* I M, M, I M, I M, M.IM,IM: M. I My I M: M, IMvlM.

6,138 1 :z.19s 1 6,777 5,643 116,1351 6,328 5,s10 I s,4n I 4,22s 195 J 231 1 209 8,285 112.1061 6,278 121s I n9 I 639 MA = 9,404 Mc = 18,227

=

OBE OBEI + OBEA SSE= SSEI +SSEA M. I My I M, M. I M, I M, 5,765 I 8,714 I 4,437 9,498 I 12,885 I 6,917 Msup.set = 11,351 MaFa'l.ltted = 17,438 Where:

I= Inertia A = Anchor Motion QBE= Operating Basis Earthquake SSE= Safe Shutdown Earthquake

Serial No.19-055 Docket No. 50-423 Attachment, Page 17 of 22 The stress equations are as follows from Section Ill of the 1971 Edition of the Code, up to and including, summer of 1973 addendum. For conservatism, the section modulus of the pipe is used below when in actuality the section modulus of the flange is much larger, which would reduce the calculated stresses. One half of the OBEA/SSEA moment is used in Equation 9. This is consistent and acceptable for piping design at MPS3 (Ref. 6).

Equation 8 - Sustained Loads (Deadload) 185 ..!!!._

• 24.0 in PDq 0.75iMA in2 112,847 in-lb lb lb

- + =:; 1.0Sh, + , 3,657- :5: 15,000-4tn z 4. 0.375 in 161.9 in3 in2 in2 Where:

P = Internal design pressure, 185 psig (Ref. 4)

Do= Outside diameter of pipe, 24.0 in tn = Nominal wall thickness, 0.375 in i = Stress intensification factor Fig. NC-3672.9{a)-1, 0.75i ~ 1.0, Conservative Slip-on Flange SIF assuming single welded slip-on flange i = 1.3 :. 0.75*1.3 ::::1.0, Flange assembly is a double welded slip-on flange, which per Table NC-3672.9(a)-1 would utilize an SIF of 1.2.

MA= Resultant moment loading due to weight and other sustained primary loads, 112,847 in-lb, 3

Z = Section modulus of pipe, 161.9 in (Ref. 5),

Sh= Basic material allowable stress at maximum (hot) temperatures for SA-181 Gr. 1, 15,000 psi (Ref. 7).

Note: Piping material is ASME SA-106 Gr. B, Sc/Sh= 15,000 psi, (Ref. 7)

Flange material is ASME SA-105, Sc/Sh= 15,000 psi Equation 9 (Upset)

The effects of pressure, weight, other sustained mechanical loads, and occasional loads, including Operating Basis Earthquake (QBE) must meet the requirements of Equation 9 from NC-3652.2. As noted above, one half of the OBEA moment is used in Equation 9.

185 .!!!._ *24.0 in PmdJo *(MA+MB-U). in2 249,065in-lb lb lb

+ 0. 75z :5: 1.2Sh* . + , 4,498---:- :5: 18,000 -

4tn Z 4. 0.37 5 m 161.9 in3 in2 in2 Where:

Pmax = Maximum operating pressure for Normal or Upset operating conditions, Design P conservatively used.

Me-u = Resultant moment loading due to occasional loads, 136,218 in-lb.

Serial No.19-055 Docket No. 50-423 Attachment, Page 18 of 22 Equation 9 (Faulted)

The effects of pressure, weight, other sustained mechanical loads, and occasional loads, including Design Basis Earthquake (DBE) must meet the requirements of Equation 9 from NC-3652.2. As noted above, one half of the SSEA moment is used in Equation 9.

lb 185-

• 24.0 in p mo./) o *(MA + M8-F) in2 322,102 in - lb lb lb 4tn + 0.75i Z =:;; 2.4Shi 4. 0.375 in + 161.9 in3 '4,95~::;; 36,000 in2 Where:

Pmax = Maximum operating pressure for Normal or Upset operating conditions, Design P conservatively used.

Ma-F:; Resultant moment loading due to occasional loads, 209,255 in-lb.

Equation 10 The requirements of either Equation 10 or 11 from NC-3652.3 must be met. The effects of thermal expansion may be met by satisfying the requirements of Equation 10, as follows:

iMc 1.3

• 218,725 in-lb lb lb

-=:;;sA, , 1,756--::;; 22,500-z 161.9 in3 in 2 in2 Where:

SA= Allowable stress range for expansion stress= 1.25 Sc+ 0.25 Sh.

Sc= Basic material allowable stress at minimum (cold) temperatures, 15,000 psig.

Mc= Range of resultant moment loading due to thermal expansion, 218,725 in-lb.

A summary of the stresses calculated above and the ratio of applied stress versus the allowable is provided in Table 1O below.

Table 10 Calculated Stresses for 24-lnch Diameter Slip-On Flange and Ratio of Applied Stress vs. Allowable Stress Condition Allowable Stress (psi) Maximum Stress (psi) Ratio Equation 8 1.0 sh= 1s,ooo 3,657 0.24 Equation 9U 1.2 sh= 1s,ooo 4,498 0.25 Equation 9F 2.4 sh= 36,ooo 4,950 0.14 Equation 10 SA= 22,500 1,756 0.08

Serial No.19-055 Docket No. 50-423 Attachment, Page 19 of 22 The stresses associated with the ID fillet welds (shown schematically below) were considered . The overall global stresses in the flange are low (limited to less than or equal to% of the allowable stress) and most of the bolting , tensile , and bending loads are transferred through the outer portion of the flange , into the hub, across the OD fillet weld , and into the pipe. The stresses in the inside corner of the ID fillet weld are even lower and are predominantly compressive. These stresses are a fraction of the allowable stresses for the materials used . In the unlikely event of a failure at the ID fillet weld, the most likely path would either be through the weld throat or along the HAZ parallel to the weld fusion line. Neither of these paths would result in a breach of the flange wall or fluid boundary. Instead , a failure would simply result in a leak path past the ID fillet weld into the annulus between the pipe and the flange . Even in the unlikely event that the ID weld should fail , the OD fillet weld , which is fully Code-compliant, is more than adequate to carry all of the anticipated loads and maintain the pressure boundary and structural integrity. A pressure boundary leak would require breaching both the ID and the OD fillet welds and is extremely unlikely.

Figure 3 Kinematic Sketch of Stresses v-,-~~~ stud loads Upwards red arrow shows kinematic material displacements outwards, resulting in tensile hoop stress region Pipe Pressure Loading Pipe Wall Downwards blue arrow shows kinematic material displacements inwards, resulting in compressive hoop stress region Pipe Centerline 8.0 Evaluation of Potential Corrosion in Annulus Between Pipe and Flange The MPS3 RPCCW system is treated with Hydrazine at 10 to 30 ppm to protect the carbon steel material surfaces by creating a protective film that prevents oxidation and corrosion . The water is also maintained at a pH of around 9.7 with low oxygen . This environment limits the amount of oxidation and corrosion to protect the exposed and wetted carbon steel surfaces.

In the event that the ID fillet weld should fail and allow a path for RPCCW treated water to enter the annulus between the pipe and the flange , the void would fill

Serial No.19-055 Docket No. 50-423 Attachment, Page 20 of 22 with treated water. Any oxygen in the annulus would be used to depletion resulting in a very limited amount of general corrosion that would slow to zero as the oxygen is depleted. Since there is no exchange of water out of the annulus, what enters will remain trapped in that space. Eventually the Hydrazine will decompose into ammonia, but the environment will remain quasi-static with just a very slow corrosion rate as only a limited amount of oxygen enters the space through the path in the ID fillet weld. This process is expected to be very slow and would produce minimal wall loss over the remaining life of the plant, including license extensions.

In the extremely unlikely event that a corrosive environment develops and ion or other exchange feeds reactants directly into a corrosion cell, the majority of that corrosion would still be localized in the immediate area nearest the source of oxygen. This kind of localized corrosion has been observed in other systems with much more corrosive environments such as in aerated seawater service.

The resulting localized corrosion cell might eventually produce a small slowly growing leak; however, this component is located in a routinely accessed area of the auxiliary building. Therefore it would be expected that any significant leak would be detected during routine Operations activities before it can cause any kind of loss of system function or component integrity.

9.0 Conclusions Studies and tests conducted in support of this alternative request demonstrate that multi-pass, GTAW fillet welds, FW-12 and FW-30, fabricated without the 200°F preheat on 1.81-inch thick SA105 low carbon material, produce comparable or more refined microstructures and lower hardness (of the weld and HAZ) as compared to equivalent, Code-compliant, single-pass fillet welds made on 1.9-inch thick material with 200°F preheat and on material :s; 1%-inch thick without preheat.

During installation, fillet welds FW-12 and FW-30 were final visual and magnetic particle examined and no recordable indications were found. Post installation in-service leak testing confirmed no leakage.

The overall global stresses in these flanges are low (limited to less than or equal to % of the allowable stress) and are transferred through the outer portion of the flange, into the hub, across the OD fillet weld, and into the pipe. The stresses in the inside corner of the ID fillet welds are even lower and are predominately compressive. This provides little, if any, driving force for either crack initiation or growth for these ID fillet welds or their HAZs. In the unlikely event of a failure at the ID fillet weld, the most likely path would either be through the weld throat or along the HAZ parallel to the weld fusion line. Neither of these paths results in a breach of the flange wall or fluid boundary. In the unlikely event that the ID weld should fail, the OD fillet weld, which is fully Code-compliant, is adequate to carry up to four times the Code design loads and will maintain the pressure boundary and structural integrity.

Serial No.19-055 Docket No. 50-423 Attachment, Page 21 of 22 In the extremely unlikely event that the ID fillet weld fails and the OD fillet weld also fails, the only conceivable failure would produce a small slowly growing leak.

This component is located in a routinely accessed area of the auxiliary building and any significant leak would be detected during routine Operations activities before causing loss of system function or component integrity.

Testing has shown that the fillet welds made on 1.9-inch thick material without 200°F preheat (or PWHT) will continue to provide sufficient strength, ductility, and toughness to perform satisfactorily. These fillet welds are less prone to cracking as evidenced by the welding process (ultra-low hydrogen multi-pass GTAW), microstructures (primarily tempered Martensite improved by grain refinement), and hardness (reduced peak and average) when compared to Code-compliant single-pass fillet welds.

Repair or replacement of these welds online or during an outage eliminates the CCP system's defense-in-depth should either one pf the other two CCP pumps or HXs become inoperable during this time. Additionally, due to the physical location of the CCP HXs, the tight work area poses an industrial safety risk to plant personnel and risk of damage to plant equipment. Based on mockup testing and evaluation of the production welds, DENG concludes that repair or replacement of the FW-12 and FW-30 welds presents a hardship without a compensating increase in the level of quality of safety. Thus, DENG proposes continued use of the welds without repair or replacement.

10.0 Duration of the Proposed Request This proposed alternative is requested for the remaining useful life of fillet welds FW-12 and FW-30.

11.0 Precedent This proposed alternative request is similar to an alternative request submitted by Entergy on August 19, 2013 for River Bend Station Unit 1 (ADAMS Accession No. ML13239A074). This alternative request was approved by the NRC on January 28, 2014 (ADAMS Accession No. ML13353A608).

12.0 References

1. EPRI Technical Report 1008277, "PWHT Exemptions for Low Hardenability Materials"

2. EPRI Technical Report 1022883, "2009-2010 Post Weld Heat Treatment Exemption Thickness Test Results"

3. American Society of Mechanical Engineers PVP2012-78571, "Alternative

Serial No.19-055

" '* , t Docket No. 50-423 Attachment, Page 22 of 22 Approach for Qualification of Temperbead Welding in the Nuclear Industry."

4. Dominion Calculation NP(B)-X7202, Rev. 004-00, "Reactor Plant Component Cooling Piping: Auxiliary Building."

5. Crane Technical Paper 410, "Flow of Fluids through Valves, Fittings, and Pipe," 1981;

6. Millstone Specification, SP-ME-572, Rev. 003-02, "Specification for Piping Classes for Millstone Unit 3."

7. ASME Boiler & Pressure Vessel Co9e, Section Ill, Division 1, "Rules for Construction of Nuclear Power Plant Components," 1971 Edition, addenda through Summer 1973.