ML12088A424: Difference between revisions

From kanterella
Jump to navigation Jump to search
(Created page by program invented by StriderTol)
(Created page by program invented by StriderTol)
Line 18: Line 18:


=Text=
=Text=
{{#Wiki_filter:c Indian Point Unit 2 CHECWORKS Power Uprate Analysis Calculation No. 04071 1-02 Revision 0 Issued For-Use March 23,2005 prepared for: Entergy Nuclear Northeast 295 Broadway Suite 3 PO Box 308 Buchanan, NY 1051 1-0308 ENTERGY NUCLEAR NORTHEAST prepared by: TECHNOLOGIES, INC. E. Main St., Suite 215 Dundee, IL 60118 ACCEPTED ACCEPTED AS NOTED RESUBMITTAL NOT REQUIRED ACCEPTED AS NOTED RESUBMITTAL REQUIRED 3 4 NOT ACCEPTED CSI TECHNOLOGIES.
{{#Wiki_filter:ENT000072 Submitted: March 28, 2012 Indian Point Unit 2 CHECWORKS Power Uprate Analysis Calculation No. 040711-02 Revision 0 Issued For-Use March 23,2005 preparedfor:
INC . IP2 CHECWORKS Power Uprate Analysis 1 . 2 . 3 . 4 . 5 . Table of Contents INTRODUCTION  
Entergy Nuclear Northeast 295 Broadway Suite 3 PO Box 308 Buchanan, NY 10511-0308 ENTERGY NUCLEAR NORTHEAST prepared by:
..............................................................................................................................................
ACCEPTED ACCEPTED AS NOTED RESUBMITTALNOT REQUIRED TECHNOLOGIES, INC.
2 PURPOSE ...........................................................................................................................................................
3 ACCEPTED AS NOTED RESUBMITTAL REQUIRED E. Main St., Suite 215 4 NOT ACCEPTED                                    Dundee, IL 60118 c
3 SCOPE ................................................................................................................................................................
 
4 ASSUMPTIONS AND MODELING DECISIONS  
CSI TECHNOLOGIES. INC               .                                                                     IP2 CHECWORKS Power Uprate Analysis Table of Contents
.........................................................................................
: 1. INTRODUCTION           ..............................................................................................................................................         2
5 METHODOLOGY  
: 2. PURPOSE...........................................................................................................................................................               3
............................................................................................................................................
: 3. SCOPE ................................................................................................................................................................           4
8 5.1. 5.2. 5.3. 5.4. 5.5. 5.6. 5.7. INPUT SPU GLOBAL DATA ...........................................................................................................................
: 4. ASSUMPTIONS AND MODELING DECISIONS.........................................................................................                                                       5
8 UPDATE MODEL BASED ON SPU OPERATIONAL OR CONFIGURATION CHANGES ........................................
: 5. METHODOLOGY ............................................................................................................................................                         8 5.1. INPUTSPU GLOBALDATA...........................................................................................................................                             8 5.2. REDEFINKION        OF CHECWORKS LINES....................................................................................................                                 12 5.3. UPDATE MODELBASEDON SPU OPERATIONAL OR CONFIGURATION CHANGES ........................................                                                                     13 5.4. UPDATE  NETWORK            FLOWANALYSIS                D E F I N ~ O......................................................................................
13 UPDATE NETWORK FLOW ANALYSIS DEFIN~ONS
NS                                                                                          13 5.5. IMPLEMENT      THE ADVANCED              RUNDEFINITION                ...........................................................................................       13 5.6. PERFORM WEAR RATE ANALYSIS...............................................................................................................                                 15 5.7. QUANTIFY    EFFECTOF STRETCH                  POWERUPRATE ........................................................................................                       16
......................................................................................
: 6. RESULTS  .........................................................................................................................................................           17 6.1. COMPONENT        LEVELWEARRATECHANGES                                DUE TO SPU...........................................................................                 17 6.2. STEAMCYCLELEVELWEARRATECHANGES                                          DUE TO SPU ........................................................................               17
13 PERFORM WEAR RATE ANALYSIS ...............................................................................................................
: 7. REFERENCES        .................................................................................................................................................           18 APPENDIX  A:       CHECWORKS MODELED                               LINES       ..............................................................................................     21 APPENDIX B:       COMPONENT LEVELWEAR RATE CHANGES DUE TO SPU                                                           ........................................................ 33 APPENDIX c :        STEAM CYCLE LEVELWEAR RATE CHANGES DUE TO SPU                                                             ..................................................... 35 ATTACHMENT   A: REFERENCED CORRESPONDENCE AND COMMUNICATIONS                                                             ....................................................... 39
15 QUANTIFY EFFECT OF STRETCH POWER UP RATE ........................................................................................
              .
16 6 . RESULTS .........................................................................................................................................................
Calculation No 040711.02. Revision 0                                                                                                                                      Page 1 of 44
17 COMPONENT LEVEL WEAR RATE CHANGES DUE TO SPU ...........................................................................
 
17 STEAM CYCLE LEVEL WEAR RATE CHANGES DUE TO SPU ........................................................................
CS/TECHNOLOGIES, /NC.                                       IP2 CHECWORKS Power Uprate Analysis
17 7 . REFERENCES
( 1. Introduction This calculation documents the revision of the Indian Point Unit 2 CHECWORKS model to predict Flow-Accelerated Corrosion (FAC) wear rate changes due to Stretch Power Uprate (SPU). The Indian Point Unit 2 SPU will change feedwater and steam flow rates, temperatures, and enthalpies, which in turn change local chemistry values. All of these factors affect wear rates due to FAC. As a result of the uprate, some lines will experience accelerated rates of FAC, while others will have reduced rates. The impact on each line depends on the complex interaction of changes in flow rate, pressure, temperature, enthalpy, steam quality, and chemistry on the FAC degradation mechanism.
.................................................................................................................................................
18 REDEFINKION OF CHECWORKS LINES ....................................................................................................
12 IMPLEMENT THE ADVANCED RUN DEFINITION
...........................................................................................
13 6.1. 6.2. APPENDIX A: CHECWORKS MODELED LINES ..............................................................................................
21 APPENDIX B: APPENDIX c: COMPONENT LEVEL WEAR RATE CHANGES DUE TO SPU ........................................................
33 STEAM CYCLE LEVEL WEAR RATE CHANGES DUE TO SPU .....................................................
35 ATTACHMENT A: REFERENCED CORRESPONDENCE AND COMMUNICATIONS  
.......................................................
39 Calculation No . 040711.02.
Revision 0 Page 1 of 44 CS/ TECHNOLOGIES, /NC. IP2 CHECWORKS Power Uprate Analysis ( 1. Introduction This calculation documents the revision of the Indian Point Unit 2 CHECWORKS model to predict Flow-Accelerated Corrosion (FAC) wear rate changes due to Stretch Power Uprate (SPU).
The Indian Point Unit 2 SPU will change feedwater and steam flow rates, temperatures, and enthalpies, which in turn change local chemistry values.
All of these factors affect wear rates due to FAC. As a result of the uprate, some lines will experience accelerated rates of FAC, while others will have reduced rates. The impact on each line depends on the complex interaction of changes in flow rate, pressure, temperature, enthalpy, steam quality, and chemistry on the FAC degradation mechanism.
Indian Point 2 had previously developed a CHECWORKS model of FAC-susceptible piping. However, the previous model did not address the changes that will result from Appendix K Uprate and Stretch Power Uprate. This calculation details the process required to revise the CHECWORKS model so that it correctly reflects all plant power levels (the original power level, Appendix K Uprate, and Stretch Power Uprate). Also documented are the changes in FAC wear rates due to the SPU.
Indian Point 2 had previously developed a CHECWORKS model of FAC-susceptible piping. However, the previous model did not address the changes that will result from Appendix K Uprate and Stretch Power Uprate. This calculation details the process required to revise the CHECWORKS model so that it correctly reflects all plant power levels (the original power level, Appendix K Uprate, and Stretch Power Uprate). Also documented are the changes in FAC wear rates due to the SPU.
Note that historical @re-uprate and Appendix K Uprate) operating conditions remain within the model, associated to the applicable operating cycles.
Note that historical @re-uprate and Appendix K Uprate) operating conditions remain within the model, associated to the applicable operating cycles. The SPU operating conditions are associated to the calendar time that those conditions are scheduled to occur, starting in Cycle 17 [7.3.1]. In this way, the models predictions of total current and future wear will be as accurate as possible because the predictions will be based on both historical and expected future operating conditions.
The SPU operating conditions are associated to the calendar time that those conditions are scheduled to occur, starting in Cycle 17 [7.3.1]. In this way, the model's predictions of total current and future wear will be as accurate as possible because the predictions will be based on both historical and expected future operating conditions.
Calculation No. 040711-02, Revision 0                                                     Page 2 of 44 I                                   -~
Calculation No. 040711-02, Revision 0 Page 2 of 44 I -~
 
CSI TECHNOLOGIES, INC. IP2 CHECWORKS Power Uprate Analysis ( 2. Purpose The purposes of the power uprate analysis in CHECWORKS are as follows:
CSI TECHNOLOGIES, INC.                                       IP2 CHECWORKS Power Uprate Analysis
0 To quantify changes in FAC wear rates due to the Stretch Power Uprate (both increases and decreases). To gain the ability to describe the effects of power uprate on FAC-susceptible piping in the Licensing Submittal. To ensure that the CHECWORKS model reflects current plant conditions going forward. 0 0 Calculation No. 040711-02, Revision 0 Page 3 of 44 cs/ TECHNOLOGIES, /NC. IP2 CHECWORKS Power Uprate Analysis I 3. Scope The scope of this power uprate analysis was determined from the input Indian Point 2 CHECWORKS model
( 2. Purpose The purposes of the power uprate analysis in CHECWORKS are as follows:
[7.2]. However, not all lines and components in the input model were included in this analysis. Only those lines and components that were assigned to one of the Wear Rate Analysis run definitions in Table 1 .O of document number 001 30- TR-001 were analyzed as part of this project [7.10]. The CHECWORKS model also contains "non-modeled" lines and components (typically assigned to CHECWORKS lines with the prefix  
0   To quantify changes in FAC wear rates due to the Stretch Power Uprate (both increases and decreases).
'NCW' for "Non-CHECWORKS"). No analysis was performed on these "NCW' lines or any other plant lines and components that were not part of the "official" CHECWORKS model. Note that Table 1.0 of document 00130-TR-001 has two hand-written entries "x-under w/ exp joints" and "CND FWH 22 to FWH 23" [7.10]. Lines and components that belong in these two locations were also included in this analysis.
To gain the ability to describe the effects of power uprate on FAC-susceptible piping in the Licensing Submittal.
This analysis was performed using CHECWORKS FAC version 1 .OG. Assumptions and modeling decisions made during this analysis are documented in Section 4. The methodology employed during this analysis is detailed in Section 5. Results obtained are listed in Section 6 and in the Appendices. Finally, Section 7 includes a list of all references used in this analysis.
To ensure that the CHECWORKS model reflects current plant conditions going forward.
Calculation No. 040711-02, Revision 0 Page 4 of 44 I--
Calculation No. 040711-02, Revision 0                                                   Page 3 of 44
cs/ TECHNOLOGIES, /NC. IP2 CHECWORKS Power Uprate Analysis i 4. Assumptions and Modeling Decisions The following assumptions and modeling decisions apply to the updated CHECWORKS model. 4.1. All data in the CHECWORKS model received as input (the as-received model)
 
cs/TECHNOLOGIES, /NC.                                         IP2 CHECWORKS Power Uprate Analysis I 3. Scope The scope of this power uprate analysis was determined from the input Indian Point 2 CHECWORKS model [7.2]. However, not all lines and components in the input model were included in this analysis. Only those lines and components that were assigned to one of the Wear Rate Analysis run definitions in Table 1.O of document number 00130-TR-001 were analyzed as part of this project [7.10]. The CHECWORKS model also contains non-modeled lines and components (typically assigned to CHECWORKS lines with the prefix NCW for Non-CHECWORKS). No analysis was performed on these NCW lines or any other plant lines and components that were not part of the official CHECWORKS model. Note that Table 1.0 of document 00130-TR-001 has two hand-written entries x-under w/ exp joints and CND FWH 22 to FWH 23 [7.10].
Lines and components that belong in these two locations were also included in this analysis.
This analysis was performed using CHECWORKS FAC version 1.OG.
Assumptions and modeling decisions made during this analysis are documented in Section 4. The methodology employed during this analysis is detailed in Section 5.
Results obtained are listed in Section 6 and in the Appendices. Finally, Section 7 includes a list of all references used in this analysis.
Calculation No. 040711-02, Revision 0                                                 Page 4 of 44
--I
 
cs/ TECHNOLOGIES, /NC.                                       IP2 CHECWORKS Power Uprate Analysis
: 4.     Assumptions and Modeling Decisions The following assumptions and modeling decisions apply to the updated CHECWORKS model.
4.1. All data in the CHECWORKS model received as input (the as-received model)
[7.2] was assumed to be complete and accurate.
[7.2] was assumed to be complete and accurate.
4.2. Exceptions were made to Assumption 4.1 in the following cases. These exceptions appear in this section, as they are not a standard part of a CHECWORKS power uprate analysis.
4.2. Exceptions were made to Assumption 4.1 in the following cases. These exceptions appear in this section, as they are not a standard part of a CHECWORKS power uprate analysis.
4.2.1. 4.2.2. 4.2.3. 4.2.4. 4.2.5. 4.2.6. The line on the CHECWORKS Heat Balance Diagram (HBD) representing the Moisture Separator/Moisture Preseparator drain lines was remodeled so the downstream connection of the line was the Heater Drain Tank instead of Feedwater Heater
4.2.1. The line on the CHECWORKS Heat Balance Diagram (HBD) representing the Moisture Separator/MoisturePreseparator drain lines was remodeled so the downstream connection of the line was the Heater Drain Tank instead of Feedwater Heater 25. The correct configuration was determined from flow diagrams [7.3.3] and the SPU heat balance diagram C7.1.31.
: 25. The correct configuration was determined from flow diagrams [7.3.3] and the SPU heat balance diagram C7.1.31. The CHECWORKS HBD was corrected so that the steam driven Feed Pump supply originated in the Main Steam system upstream of LP Turbine as opposed to Main Steam upstream of the HP Turbine. Plant layout shows steam supply as coming from both locations; however, the SPU HBD shows that supply is from Main Steam upstream of LP Turbine under normal operating conditions [7.1.3]. Reheat Steam to the Moisture Separator Reheater on the CHECWORKS HBD was remodeled so the source was from the Main Steam system upstream of the HP Turbine as opposed to downstream of the HP Turbine. The correct configuration was determined from flow diagrams [7.3.3]
4.2.2. The CHECWORKS HBD was corrected so that the steam driven Feed Pump supply originated in the Main Steam system upstream of LP Turbine as opposed to Main Steam upstream of the HP Turbine. Plant layout shows steam supply as coming from both locations; however, the SPU HBD shows that supply is from Main Steam upstream of LP Turbine under normal operating conditions [7.1.3].
and the SPU heat balance diagram [7.1.3]. The condensate pump was added to the CHECWORKS HBD; the condensate pump had been omitted previously. The input CHECWORKS model did not have components in the Condensate system between Feedwater Heaters 22 and Feedwater Heaters 23. All components in these lines were modeled as part of this analysis. Component information and configuration was determined by using the existing FAC isometric for Unit 3 as a guide [7.8]. Lines were named using Sketch 80A as the sketch number. The FAC Program Owner supplied component names
4.2.3. Reheat Steam to the Moisture Separator Reheater on the CHECWORKS HBD was remodeled so the source was from the Main Steam system upstream of the HP Turbine as opposed to downstream of the HP Turbine.
[7.9]. An error stating, "The input hydrazine conc, at SG Steam outlet caused a mass in-balance in the Water Chemistry Analysis" was generated when performing water chemistry analysis for Cycle
The correct configuration was determined from flow diagrams [7.3.3] and the SPU heat balance diagram [7.1.3].
: 14. The hydrazine Calculation No. 040711-02, Revision 0 Page 5 of 44 CSI TECHNOLOGIES, /NC.
4.2.4. The condensate pump was added to the CHECWORKS HBD; the condensate pump had been omitted previously.
IP2 CHECWORKS Power Uprate Analysis 4.3. 4.4. 4.5. 4.2.7. 4.2.8. concentration at the Steam Generator Outlet was lowered from 18 ppb to 15 ppb for this water treatment.
4.2.5. The input CHECWORKS model did not have components in the Condensate system between Feedwater Heaters 22 and Feedwater Heaters
This action enabled water chemistry analysis to run without error. The hydrazine concentration at the SG outlet and MSR drain locations were not specified for Cycle 16 water treatment.
: 23. All components in these lines were modeled as part of this analysis.
If this data is not entered, CHECWORKS calculates a Concentration of zero at these locations. Therefore, estimated hydrazine concentrations were entered at these locations by using the "rules of thumb" for a Recirculating Steam Generator
Component information and configuration was determined by using the existing FAC isometric for Unit 3 as a guide [7.8]. Lines were named using Sketch 80A as the sketch number. The FAC Program Owner supplied component names [7.9].
[7.6]. Based on the "rules of thumb", the concentration of hydrazine at the Steam Generator Outlet was assumed to be 60% of the final feedwater concentration, while the concentration of hydrazine at the MSR Drain was assumed to be 120% of the final feedwater concentration.
4.2.6. An error stating, "The input hydrazine conc, at SG Steam outlet caused a mass in-balance in the Water Chemistry Analysis" was generated when i
During WRA, an error message appeared for component MS- 1 B 14P- 1 US stating that, "Orifice size 5.761" cannot be greater than pipe inside diameter 5.61 1"". The initial thickness of this component (0.507") was deleted; nominal thickness was then used in this calculation. WRA was able to proceed without further error. For a number of lines on the SPU Heat Balance Diagram [7.1.3] and Appendix K Heat Balance Diagram [7.1.2], thermodynamic and flow values (pressure, enthalpy, and flow rate) were listed separately for the steam phase and the water phase or for each train in a parallel train configuration.
performing water chemistry analysis for Cycle 14. The hydrazine Calculation No. 040711-02, Revision 0                                                   Page 5 of 44
The overall flow rate, pressure, and enthalpy of these lines were calculated and entered in the CHECWORKS Steam Cycle (see Section 5.1.3). The combined flow rate was calculated as the sum of the liquid and steam flow rates (or the sum of multiple trains), the combined pressure was calculated as the average of all pressures, and the enthalpy was calculated as the weighted average of liquid and steam enthalpy (or the weighted average of multiple trains). These calculations were performed based on EPRI's Guidelines for Plant Modeling and Evaluation of Component Inspection Data [7.6].
 
Plant period data was estimated for the cycle where the SPU will occur. Start and end dates were estimated based on anticipated dates [7.3.1].
CSI TECHNOLOGIES, /NC.                                         IP2 CHECWORKS Power Uprate Analysis concentration at the Steam Generator Outlet was lowered from 18 ppb to 15 ppb for this water treatment. This action enabled water chemistry analysis to run without error.
An estimation of operating hours was calculated from these dates based on calendar days.
4.2.7. The hydrazine concentration at the SG outlet and MSR drain locations were not specified for Cycle 16 water treatment. If this data is not entered, CHECWORKS calculates a Concentration of zero at these locations.
The Water Treatment for this cycle was modeled as equivalent to the most recent complete operating cycle. The model can be updated with actual values for these inputs when this data becomes available. Due to the use of the Advanced Run Definition feature in this model, which is required for accurate modeling of the power uprate condition, CHECWORKS FAC Version 1 .OG cannot accurately represent the operating conditions in tees that combine or split flow. In order to maintain simplicity in the CHECWORKS model, tees were modeled with a flow rate equal to the highest flow rate present in Calculation No. 040711-02, Revision 0 Page 6 of 44 ._ -- ,-
Therefore, estimated hydrazine concentrations were entered at these locations by using the rules of thumb for a Recirculating Steam Generator [7.6]. Based on the rules of thumb, the concentration of hydrazine at the Steam Generator Outlet was assumed to be 60% of the final feedwater concentration, while the concentration of hydrazine at the MSR Drain was assumed to be 120% of the final feedwater concentration.
CSI TECHNOLOGIES, /NC. IP2 CHECWORKS Power Uprate Analysis the tee. Therefore, the predicted wear rates for tees should be used with caution.
4.2.8. During WRA, an error message appeared for component MS- 1B 14P-1 US stating that, Orifice size 5.761 cannot be greater than pipe inside diameter 5.61 1. The initial thickness of this component (0.507) was deleted; nominal thickness was then used in this calculation. WRA was able to proceed without further error.
Calculation No. 040711-02, Revision 0 Page 7 of 44 CSI TECHNOLOGIES, INC. IP2 CHECWORKS Power Uprate Analysis 100.00 101.19 104.48 i 5. Methodology Additional input data was entered into the CHECWORKS model to model the power uprate. A number of tasks were required to convert the previous Indian Point Unit 2 model into a format that was compatible with a power uprate. The following section describes the tasks performed to enter power uprate data and to convert the model into a form compatible with multiple power levels. 3090.2 Cycles 1-16A Original Power Level 3127.0 Cycle 16B Appendix K Uprate 3228.5 Cycle 17 to End of Life Stretch Power Uprate 5.1. Input SPU Global Data Additional CHECWORKS Power Levels representing the SPU and Appendix K operating conditions were defined in the model and associated to the applicable Plant Period.
4.3. For a number of lines on the SPU Heat Balance Diagram [7.1.3] and Appendix K Heat Balance Diagram [7.1.2], thermodynamic and flow values (pressure, enthalpy, and flow rate) were listed separately for the steam phase and the water phase or for each train in a parallel train configuration. The overall flow rate, pressure, and enthalpy of these lines were calculated and entered in the CHECWORKS Steam Cycle (see Section 5.1.3). The combined flow rate was calculated as the sum of the liquid and steam flow rates (or the sum of multiple trains), the combined pressure was calculated as the average of all pressures, and the enthalpy was calculated as the weighted average of liquid and steam enthalpy (or the weighted average of multiple trains). These calculations were performed based on EPRIs Guidelines for Plant Modeling and Evaluation of Component Inspection Data [7.6].
The following sections detail the steps involved in this task. 5.1.1. CHECWORKS Heat Balance Diagram The CHECWORKS Heat Balance Diagram (HBD) was reviewed to ensure that it correctly portrayed plant configuration. The CHECWORKS HBD was compared to the plant Heat Balance Diagrams
4.4. Plant period data was estimated for the cycle where the SPU will occur. Start and end dates were estimated based on anticipated dates [7.3.1]. An estimation of operating hours was calculated from these dates based on calendar days. The Water Treatment for this cycle was modeled as equivalent to the most recent complete operating cycle. The model can be updated with actual values for these inputs when this data becomes available.
[7.1] and flow diagrams [7.3.3].
4.5. Due to the use of the Advanced Run Definition feature in this model, which is required for accurate modeling of the power uprate condition, CHECWORKS FAC Version 1.OG cannot accurately represent the operating conditions in tees that combine or split flow. In order to maintain simplicity in the CHECWORKS model, tees were modeled with a flow rate equal to the highest flow rate present in Calculation No. 040711-02, Revision 0                                                       Page 6 of 44
Discrepancies were noted and corrected as discussed in Section 4.2. i 5.1.2. Power Level Data A Plant Power Level is defined for each power level at which the plant is operated for a significant period of time. Two new power levels were added to the model representing the SPU and Appendix K conditions.
,-                                         --                                 ._
Table 5.1 lists all power levels in the model and the operating cycles to which they apply. Table 5.1 Power Level History Data was entered for the new power levels on the Power Level Form in accordance with the CHECWORKS User's Guide [7.5].
 
Table 5.2 summarizes the Appendix K power level input data and the source of this data. Calculation No. 040711-02, Revision 0 Page 8 of 44 -----
CSI TECHNOLOGIES, /NC.                                       IP2 CHECWORKS Power Uprate Analysis the tee. Therefore, the predicted wear rates for tees should be used with caution.
CSI TECHNOLOGIES, INC. IP2 CHECWORKS Power Uprate Analysis Temp (F) Blowdown Rate (Mlb/hr>
Calculation No. 040711-02, Revision 0                                                     Page 7 of 44
Table 5.2 Appendix K Power Level Input Data 513.1 7.1.3 0.0542 7.1.3 INotes: Stretch Power Uprate. 3 127.OMWt x - Field should be left blank for a PWR. Carryover
 
(%) Feedwater Vent Rate f%> Table 5.1 summarizes the Appendix K power level input data and the source of this data. 0.02 7.1.3 X 7.5 Table 5.3 SPU Power Level Input Data I Steam Rate (Mlb/hr> I 13.903750 I 7.1.3 I I Pressure (mia) I 765.0 I 7.1.3 I I Reheater Vent Rate (%) I X I 7.5 I IMoisture Smarator Carrvunder
CSI TECHNOLOGIES,       INC.                                     IP2 CHECWORKS Power Uprate Analysis i 5.     Methodology Additional input data was entered into the CHECWORKS model to model the power uprate. A number of tasks were required to convert the previous Indian Point Unit 2 model into a format that was compatible with a power uprate. The following section describes the tasks performed to enter power uprate data and to convert the model into a form compatible with multiple power levels.
(%> I X I 7.5 I /Notes: Stretch Power Umate.
5.1. Input SPU Global Data Additional CHECWORKS Power Levels representing the SPU and Appendix K operating conditions were defined in the model and associated to the applicable Plant Period. The following sections detail the steps involved in this task.
3228.5 MWt I x - Field should be left blank for a PWR. 5.1.3. Steam Cycle Data Steam Cycle Data is used by CHECWORKS to calculate chemistry conditions during wear rate analysis.
5.1.1. CHECWORKS Heat Balance Diagram The CHECWORKS Heat Balance Diagram (HBD) was reviewed to ensure that it correctly portrayed plant configuration. The CHECWORKS HBD was compared to the plant Heat Balance Diagrams [7.1] and flow diagrams [7.3.3]. Discrepancies were noted and corrected as discussed in Section 4.2.
It is also used to calculate operating conditions when the Advanced Run Definition feature is implemented (see Section 5.5). Steam cycle data was entered for each heat balance item for the SPU power level. All data was entered in accordance with EPRI's "Guidelines for Plant Modeling and Evaluation of Component Inspection Page 9 of 44 Calculation No. 040711-02, Revision 0 cs/ TECHNOLOGIES, /NC. IP2 CHECWORKS Power Uprate Analysis Data" [7.6] and the CHECWORKS User's Guide [7.5]. Table 5.4 summarizes SPU steam cycle input data and the source of this data. Table 5.4 Appendix K Steam Cycle Input Data (t x = No value entered (not required by CHECWORKS).
5.1.2. Power Level Data i                          A Plant Power Level is defined for each power level at which the plant is operated for a significant period of time. Two new power levels were added to the model representing the SPU and Appendix K conditions.
Table 5.1 lists all power levels in the model and the operating cycles to which they apply.
Table 5.1 Power Level History 100.00    3090.2      Cycles 1-16A          Original Power Level 101.19    3127.0        Cycle 16B            Appendix K Uprate 104.48    3228.5 Cycle 17 to End of Life    Stretch Power Uprate Data was entered for the new power levels on the Power Level Form in accordance with the CHECWORKS Users Guide [7.5].
Table 5.2 summarizes the Appendix K power level input data and the source of this data.
Calculation No. 040711-02, Revision 0                                                       Page 8 of 44
-
 
CSI TECHNOLOGIES, INC.                                               IP2 CHECWORKS Power Uprate Analysis Table 5.2 Appendix K Power Level Input Data INotes: Stretch Power Uprate. 3127.OMWt x - Field should be left blank for a PWR.
Table 5.1 summarizes the Appendix K power level input data and the source of this data.
Table 5.3 SPU Power Level Input Data I Steam Rate (Mlb/hr>                     I 13.903750 I   7.1.3     I IPressure (mia)                           I     765.0 I   7.1.3     I Temp (F)                                       513.1      7.1.3 Blowdown Rate (Mlb/hr>                         0.0542      7.1.3 Carryover (%)                                  0.02      7.1.3 Feedwater Vent Rate f%>                            X        7.5 I Reheater Vent Rate (%)                  I        X  I    7.5      I IMoisture Smarator Carrvunder (%> I                X  I    7.5      I
                        /Notes: Stretch Power Umate. 3228.5 MWt                              I x - Field should be left blank for a PWR.
5.1.3. Steam Cycle Data Steam Cycle Data is used by CHECWORKS to calculate chemistry conditions during wear rate analysis. It is also used to calculate operating conditions when the Advanced Run Definition feature is implemented (see Section 5.5). Steam cycle data was entered for each heat balance item for the SPU power level. All data was entered in accordance with EPRIs Guidelines for Plant Modeling and Evaluation of Component Inspection Calculation No. 040711-02, Revision 0                                                       Page 9 of 44
 
cs/TECHNOLOGIES,         /NC.                                         IP2 CHECWORKS Power Uprate Analysis Data [7.6] and the CHECWORKS Users Guide [7.5]. Table 5.4 summarizes SPU steam cycle input data and the source of this data.
Table 5.4 Appendix K Steam Cycle Input Data (t
x = No value entered (not required by CHECWORKS).
(1) The HBD Item name is automatically generated by CHECWORKS. Feedwater heaters are numbered sequentially in reverse flow order. Feedwater Heater 1 is the feedwater heater closest to the steam generator (equivalent to heater 26 at Indian Point 2). Extraction lines are numbered sequentially in order of decreasing pressure.
(2) MSEP 1 represents the conditions in both the moisture separator and moisture pre-separator drain lines as recommended by EPFU Guidelines [7.6].
(3)HPEXTLINE 2 is a fictitious high-pressure extraction line representing the steam lines between the pre-separator and main separator as recommended by EPFU Guidelines [7.6].
(4) Flow rate is for exiting steam flow was entered as zero as recommended by EPRI Guidelines [7.6]. Pressure and enthalpy were obtained from the Appendix K PEPSE model [7.1.2].
(5) Enthalpy calculated as the weighted average of the steam and liquid phases. Steam phase enthalpy was obtained directly from the PEPSE diagram as the enthalpy after moisture removal in the LP Turbine. Liquid phase enthalpy was calculated as the enthalpy of saturated liquid at the pressure given on the PEPSE diagram.
(6) Reheater drain flow was entered as the sum of the flow through the reheater drain tanks and the vent chamber drains [7.1.2]. Pressure and enthalpy were entered as the conditions through the reheater drain tank [7.1.2].
Table 5.5 summarizes SPU steam cycle input data and the source of this data.
Calculation No. 040711-02, Revision 0                                                                  Page 10 of 44
 
cs/TECHNOLOGIES,/NC.                                                    IP2 CHECWORKS Power Uprate Analysis If Table 5.5 SPU Steam Cycle Input Data i,
x = N o value entered (not required by CHECWORKS).
(1) The HBD Item name is automatically generated by CHECWORKS. Feedwater heaters are numbered sequentially in reverse flow order. Feedwater Heater 1 is the feedwater heater closest to the steam generator (equivalent to heater 26 at Indian Point 2). Extraction lines are numbered sequentially in order of decreasing pressure.
(1) The HBD Item name is automatically generated by CHECWORKS. Feedwater heaters are numbered sequentially in reverse flow order. Feedwater Heater 1 is the feedwater heater closest to the steam generator (equivalent to heater 26 at Indian Point 2). Extraction lines are numbered sequentially in order of decreasing pressure.
(2) MSEP 1 represents the conditions in both the moisture separator and moisture pre-separator drain lines as recommended by EPFU Guidelines
[7.6]. (3) HPEXTLINE 2 is a fictitious high-pressure extraction line representing the steam lines between the pre-separator and main separator as recommended by EPFU Guidelines [7.6].
(4) Flow rate is for exiting steam flow was entered as zero as recommended by EPRI Guidelines [7.6]. Pressure and enthalpy were obtained from the Appendix K PEPSE model [7.1.2]. (5) Enthalpy calculated as the weighted average of the steam and liquid phases. Steam phase enthalpy was obtained directly from the PEPSE diagram as the enthalpy after moisture removal in the LP Turbine. Liquid phase enthalpy was calculated as the enthalpy of saturated liquid at the pressure given on the PEPSE diagram.
(6) Reheater drain flow was entered as the sum of the flow through the reheater drain tanks and the vent chamber drains [7.1.2]. Pressure and enthalpy were entered as the conditions through the reheater drain tank [7.1.2]. Table 5.5 summarizes SPU steam cycle input data and the source of this data. Calculation No. 040711-02, Revision 0 Page 10 of 44 cs/ TECHNOLOGIES, /NC. IP2 CHECWORKS Power Uprate Analysis Table 5.5 SPU Steam Cycle Input Data If x = No value entered (not required by CHECWORKS). (1) The HBD Item name is automatically generated by CHECWORKS. Feedwater heaters are numbered sequentially in reverse flow order. Feedwater Heater 1 is the feedwater heater closest to the steam generator (equivalent to heater 26 at Indian Point 2). Extraction lines are numbered sequentially in order of decreasing pressure.
(2) MSEP 1 represents the conditions in both the moisture separator and moisture pre-separator drain lines as recommended by EPRI Guidelines [7.6].
(2) MSEP 1 represents the conditions in both the moisture separator and moisture pre-separator drain lines as recommended by EPRI Guidelines [7.6].
(3) HPEXTLINE 2 is a fictitious high-pressure extraction line representing the steam lines between the pre-separator and main separator as recommended by EPRI Guidelines
(3) HPEXTLINE 2 is a fictitious high-pressure extraction line representing the steam lines between the pre-separator and main separator as recommended by EPRI Guidelines [7.6].
[7.6]. (4) Flow rate is for exiting steam flow was entered as zero as recommended by EPRI Guidelines [7.6]. Pressure and enthalpy were obtained from the SPU PEPSE model [7.1.3].
(4) Flow rate is for exiting steam flow was entered as zero as recommended by EPRI Guidelines [7.6]. Pressure and enthalpy were obtained from the SPU PEPSE model [7.1.3].
(5) Enthalpy calculated as the weighted average of the steam and liquid phases. Steam phase enthalpy was obtained directly from the PEPSE diagram as the enthalpy after moisture removal in the LP Turbine. Liquid phase enthalpy was calculated as the enthalpy of saturated liquid at the pressure given on the PEPSE diagram.
(5) Enthalpy calculated as the weighted average of the steam and liquid phases. Steam phase enthalpy was obtained directly from the PEPSE diagram as the enthalpy after moisture removal in the LP Turbine. Liquid phase enthalpy was calculated as the enthalpy of saturated liquid at the pressure given on the PEPSE diagram.
i,, 5.1.4. Plant Period Data Each power level identified in Table 5.1 was associated to the correct operating cycle or cycles. The Appendix K Uprate start date was obtained through correspondence with plant personnel
5.1.4. Plant Period Data Each power level identified in Table 5.1 was associated to the correct operating cycle or cycles. The Appendix K Uprate start date was obtained through correspondence with plant personnel [7.3.3].The SPU is scheduled for the start of Cycle 17. To include this power level in the model, plant periods were created that have not yet occurred. Start dates, Calculation No. 040711-02, Revision 0                                                                   Page 11 of 44 LI  _ -
[7.3.3]. The SPU is scheduled for the start of Cycle 17. To include this power level in the model, plant periods were created that have not yet occurred. Start dates, Calculation No. 040711-02, Revision 0 Page 11 of 44 --- - LI_-
                                      ---         -
CSI TECHNOLOGIES, INC. IP2 CHECWORKS Power Uprate Analysis end dates, operating hours, and chemistry data was estimated for these periods (see Section 4.4). 5.2. Redefinition of CHECWORKS Lines The CHECWORKS model was prepared for use of the Advanced Run Definition (see Section 5.5). This task consisted of redefining the CHECWORKS lines within the model so that all components on a given line are subject to the same thermodynamic conditions. Appendix A contains a listing of all lines that appear in the CHECWORKS model. For conservatism, the tee where flow rate changed was associated to the line having the greatest flow rate.
 
Note that the model will over-predict the wear for some sections of tees (see Section 4.5). Figure 5- 1 illustrates the procedure used to group components into lines. In the diagram, seven lines exist, each with a different percentage of the total flow. Note that the tee is grouped with the line number seeing the greatest percentage of flow. Figure 5-1 Diagram of Line Redefinition Procedure 1 EQUIPMENT OUTPUT V Calculation No. 040711-02, Revision 0 Page 12 of 44 (_I_____ --
CSI TECHNOLOGIES,       INC.                                   IP2 CHECWORKS Power Uprate Analysis end dates, operating hours, and chemistry data was estimated for these periods (see Section 4.4).
CSI TECHNOLOGIES, INC. IP2 CHECWORKS Power Uprate Analysis 5.3. 5.4. 5.5. Components were grouped into lines by comparing the input CHECWORKS model [7.2] with FAC isometrics [7.7]
5.2. Redefinition of CHECWORKS Lines The CHECWORKS model was prepared for use of the Advanced Run Definition (see Section 5.5). This task consisted of redefining the CHECWORKS lines within the model so that all components on a given line are subject to the same thermodynamic conditions. Appendix A contains a listing of all lines that appear in the CHECWORKS model.
and flow diagrams [7.3.3].
For conservatism, the tee where flow rate changed was associated to the line having the greatest flow rate. Note that the model will over-predict the wear for some sections of tees (see Section 4.5).
New lines were given names according to the naming convention, below. AABBC-D-E AA BB C D E = Brief description of the line = = = = Abbreviation of the system (ex: CD = Condensate, EX = Extraction Steam, etc.) Sketch number the line begins on (ex: 01, 80, etc.) Sketch letter if sketch number includes a letter (ex: A for sketch 80A) Sequential number for each line on one sketch, numbered in flow order New line names were created as required by CHECWORKS, not where plant line names changed. Therefore, lines may contain components located on different sketches but in all cases the sketch number corresponds to the first component in the line. For example, line name "CD80-1-FWH 23A to FWH 24A" is located in the Condensate system, on FAC sketch 80, and runs from Feedwater Heater 23A to Feedwater Heater 24A.
Figure 5- 1 illustrates the procedure used to group components into lines. Inthe diagram, seven lines exist, each with a different percentage of the total flow. Note that the tee is grouped with the line number seeing the greatest percentage of flow.
Update Model Based on SPU Operational or Configuration Changes With an uprate in power, lines with partial operation may increase or decrease their operation frequency (such as additional trains in operation or bypass lines now operated during full power). Additionally, uprates may call for design changes that may result in piping configuration changes.
Figure 5-1 Diagram of Line Redefinition Procedure 1     EQUIPMENT OUTPUT V
Calculation No. 040711-02,Revision 0                                                       Page 12 of 44
(_I_____ --
 
CSI TECHNOLOGIES,       INC.                                   IP2 CHECWORKS Power Uprate Analysis Components were grouped into lines by comparing the input CHECWORKS model [7.2] with FAC isometrics [7.7] and flow diagrams [7.3.3]. New lines were given names according to the naming convention, below.
AABBC-D-E AA         = Abbreviation of the system (ex: CD = Condensate, EX = Extraction Steam, etc.)
BB        = Sketch number the line begins on (ex: 01, 80, etc.)
C        = Sketch letter if sketch number includes a letter (ex: A for sketch 80A)
D        = Sequential number for each line on one sketch, numbered in flow order E        = Brief description of the line New line names were created as required by CHECWORKS, not where plant line names changed. Therefore, lines may contain components located on different sketches but in all cases the sketch number corresponds to the first component in the line.
For example, line name CD80-1-FWH 23A to FWH 24A is located in the Condensate system, on FAC sketch 80, and runs from Feedwater Heater 23A to Feedwater Heater 24A.
5.3. Update Model Based on SPU Operational or Configuration Changes With an uprate in power, lines with partial operation may increase or decrease their operation frequency (such as additional trains in operation or bypass lines now operated during full power). Additionally, uprates may call for design changes that may result in piping configuration changes.
No operational or configuration changes occurred due to the SPU [7.3.2].
No operational or configuration changes occurred due to the SPU [7.3.2].
Update Network Flow Analysis Definitions Network Flow Analysis (NFA) is a module within CHECWORKS that calculates flow rate, pressure drops, temperature changes, and steam quality changes based on piping configuration and source/sink conditions.
5.4. Update Network Flow Analysis Definitions Network Flow Analysis (NFA) is a module within CHECWORKS that calculates flow rate, pressure drops, temperature changes, and steam quality changes based on piping configuration and source/sink conditions. NFA is used where operating conditions are not completely known. A different set of source/sirik conditions is entered for each NFA per power level. Additional input data must be added to each NFA to reflect the SPU power level identified in Section 5.1.2.
NFA is used where operating conditions are not completely known. A different set of source/sirik conditions is entered for each NFA per
The Indian Point 2 CHECWORKS model did not contain any NFA definitions; therefore, no updates to the model were performed during this task [7.2].
5.5. Implement the Advanced Run Definition This task involved a redefinition of the source in which CHECWORKS obtains thermodynamic conditions (pressure, enthalpy, temperature, and quality) and flow rate conditions for a component. Previously all thermodynamic and flow rate conditions had been entered individually for each component on the component data forms. However, the component form allows only one set of thermodynamic Calculation No. 040711-02, Revision 0                                                        Page 13


==Subject:==
==Subject:==
RE: RAI FAC-1 Harry, There are no operational changes for Steam Plant Systems, such as use of additional trains or use of bypass lines not currently in operation, associated with uprate of IP2 / IP3. Changes due to the uprate which affect FAC are primarily changes in system flowrates and temperatures, which are documented in the Heat Balance calculations. Impact of these changes on piping velocities and temperatures is addressed in the applicable sections of the BOP Uprate Engineering Reports, forwarded to Entergy for review. Modifications associated with the uprate of IP2 / IP3 include:
RE: RAI FAC-1 Harry, There are no operational changes for Steam Plant Systems, such as use of additional trains or use of bypass lines not currently in operation, associated with uprate of IP2 / IP3. Changes due to the uprate which affect FAC are primarily changes in system flowrates and temperatures, which are documented in the Heat Balance calculations. Impact of these changes on piping velocities and temperatures is addressed in the applicable sections of the BOP Uprate Engineering Reports, forwarded to Entergy for review.
(1) MSR internal moisture separation system replacement, (2) HP Turbine rotor replacement, (3) HP turbine interstage drain piping modification, and (4) Relocation of HP turbine 1'' stage pressure taps.
Modifications associated with the uprate of IP2 / IP3 include: (1) MSR internal moisture separation system replacement, (2) HP Turbine rotor replacement, (3) HP turbine interstage drain piping modification, and (4)
Regards, Glenn C. %em*m* , 1 From: Hartjen, Harry
Relocation of HP turbine 1'' stage pressure taps.
[mailto:HHartje@entergy.com] Sent: Wednesday, October 06,2004 1 :30 PM Page 41 of 44 Calculation No. 04071 1-02, Attachment A, Revision 0 CSI TECHNOLOGIES, INC. IP2 CHECWORKS Power Uprate Analysis To: Cunningham, Glenn  
Regards, Glenn C.
  %em*m*
,
1 From: Hartjen, Harry [mailto:HHartje@entergy.com]
Sent: Wednesday, October 06,2004 1:30 PM Calculation No. 040711-02, Attachment A, Revision 0                                           Page 41 of 44
 
CSI TECHNOLOGIES, INC.                                             IP2 CHECWORKS Power Uprate Analysis To: Cunningham, Glenn


==Subject:==
==Subject:==
RE: RAI FAC-1 Glen: This is to confirm that the wear rate comparison will be completed by 12/31/2004.
RE:RAI FAC-1 Glen:
As I mentioned to you we are having a contractor perform the update to our Checworks models.
This is to confirm that the wear rate comparison will be completed by 12/31/2004.
One of the inputs I have to provide to them is: The listing of all Unit 2 and Unit 3 operational and/or configuration changes due to the power uprate (i.e., additional trains in operation, bypass lines operated at full power, etc.) Do you have this information or can you direct my question to someone who has this information.
A s I mentioned to you we are having a contractor perform the update to our Checworks models. One of the inputs I have to provide to them is:
Thanks, Harry Hartjen (914) 736-8356 From: Cunningham, Glenn [mailto:glenn.cunningham@shawgrp.com]
The listing of all Unit 2 and Unit 3 operational and/or configuration changes due to the power uprate (i.e.,
Sent: Wednesday, October 06, 2004 1250 PM To: Hartjen, Harry  
additional trains in operation, bypass lines operated at full power, etc.)
Do you have this information or can you direct my question to someone who has this information.
: Thanks, Harry Hartjen (914) 736-8356 From: Cunningham, Glenn [mailto:glenn.cunningham@shawgrp.com]
Sent: Wednesday, October 06, 2004 1250 P M To: Hartjen, Harry


==Subject:==
==Subject:==
FW: RAI FAC-1 y_^*- From: Cunningham, Glenn Sent: Wednesday, October 06,2004 1241 PM To: 'hartje@entergy.com'  
FW: RAI FAC-1 y_^*-
From: Cunningham, Glenn Sent: Wednesday, October 06,2004 1241 P M To: 'hartje@entergy.com'


==Subject:==
==Subject:==
RAI FAC-1 Harry, As I indicated, I made a few edits in your input.
RAI FAC-1
Please confirm that the wear rate comparison discussed in the last sentence will be completed by the end of 2004. Thanks, Internet Email Confidentiality Footer******************  
: Harry, As I indicated, I made a few edits in your input. Please confirm that the wear rate comparison discussed in the last sentence will be completed by the end of 2004.
***************** Privileged/Confidential Information may be contained in this message. If you are not the addressee indicated in this message (or responsible for delivery of the message to such person), you may not copy or deliver this message to anyone. In such case, you should destroy this message and notify the sender by reply email. Please advise immediately if you or your employer do not consent to Internet email for messages of this kind. Opinions, conclusions and other information in this message that do not relate to the official business of The Shaw Group Inc. or its subsidiaries shall be understood as neither given nor endorsed by it. The Shaw Group Inc. http:/hnrww.shawgrp.com Calculation No. 040711-02, Attachment A, Revision 0 Page 42 of 44 _-.
: Thanks,
CSI TECHNOLOGIES, INC. IP2 CHECWORKS Power Uprate Analysis --__l-l-_l_
*****************Internet Email Confidentiality Footer******************
_lll~____- - - I Internet Email Confidentiality Foofer--*****-***  
Privileged/Confidential Information may be contained in this message.
***************** ( PrivilegedIConfidential Information may be contained in this message. If you are not the addressee indicated in this message (or responsible for delivery of the message to such person), you may not copy or deliver this message to anyone. In such case, you should destroy this message and notify the sender by reply email. Please advise immediately if you or your employer do not consent to Internet email for messages of this kind. Opinions, conclusions and other information in this message that do not relate to the official business of The Shaw Group Inc.
If you are not the addressee indicated in this message (or responsible for delivery of the message to such person), you may not copy or deliver this message to anyone. In such case, you should destroy this message and notify the sender by reply email. Please advise immediately if you or your employer do not consent to Internet email for messages of this kind. Opinions, conclusions and other information in this message that do not relate to the official business of The Shaw Group Inc. or its subsidiaries shall be understood as neither given nor endorsed by it.
or its subsidiaries shall be understood as neither given nor endorsed by it. The Shaw Group Inc.
The Shaw Group Inc.
http://w.shawgrp.com c Page 43 of 44 Calculation No. 040711-02, Attachment A, Revision 0 I I- -_
http:/hnrww.shawgrp.com Calculation No. 040711-02, Attachment A, Revision 0                                           Page 42 of 44
cs/ TECHNOLOGIES, /NC. IP2 CHECWORKS Power Uprate Analysis Reference 7.3.3 Email from Ron Macina (IP3) to Brian Trudeau (CSI Technologies), dated 1/10/2005, regarding addition a1 Heat Balance Diagrams and uprate start dates, CSI Doc. No 04071 140.
_-.
CSI Doc. No. 04071 140 Brian, Please find below the requested Heat BalancesAnformation.
 
: 1) IP3 Pre and Post Appendix K Heat Balances <<IP3TuningR4 Pre Appendix K (3037 NSS Pwr).pdf>>
CSI TECHNOLOGIES, INC.                                             IP2 CHECWORKS Power Uprate Analysis
4P3TuningR4 Post Apendix K.pdf>> 2) IP2 Pre Appendix K Heat Balance <<IP2-TuningR4 Pre Appendix K (3090 NSS Pwr).pdP>
                  -  I
: 3) Start Date for IP3 Appendix K uprate was 12/22/02 4) Start Date for IP2 Appendix K uprate was 5/23/03 Thanks, Ron Macina 91 4-736-8363  
                                                              -       - - _  _    l
-----0rig i nal Message----- From: Brian Trudeau
-l ll l-~  __ _
[mailto:btrudeau@csitechnologies.com]
l__ _
Sent: To: Macina, Ron  
( *****************Internet Email Confidentiality Foofer--*****-***
PrivilegedIConfidentialInformation may be contained in this message.
If you are not the addressee indicated in this message (or responsible for delivery of the message to such person), you may not copy or deliver this message to anyone. In such case, you should destroy this message and notify the sender by reply email. Please advise immediately if you or your employer do not consent to Internet email for messages of this kind. Opinions, conclusions and other information in this message that do not relate to the official business of The Shaw Group Inc. or its subsidiaries shall be understood as neither given nor endorsed by it.
The Shaw Group Inc.
http://w.shawgrp.com c
Calculation No. 040711-02, Attachment A, Revision 0                                     Page 43 of 44
                                        -_                            II-
 
cs/ TECHNOLOGIES, /NC.                                         IP2 CHECWORKS Power Uprate Analysis Reference 7.3.3 Email from Ron Macina (IP3)to Brian Trudeau (CSITechnologies), dated 1/10/2005, regarding addition a1 Heat Balance Diagrams and uprate start dates, CSI Doc. No 04071 140.
CSI Doc.No. 04071 140 Brian, Please find below the requested Heat BalancesAnformation.
: 1) IP3 Pre and Post Appendix K Heat Balances
  <<IP3TuningR4 Pre Appendix K (3037 NSS Pwr).pdf>> 4P3TuningR4 Post Apendix K.pdf>>
2)IP2 Pre Appendix K Heat Balance
  <<IP2-TuningR4 Pre Appendix K (3090 NSS Pwr).pdP>
: 3) Start Date for IP3 Appendix K uprate was 12/22/02
: 4) Start Date for IP2 Appendix K uprate was 5/23/03 Thanks, Ron Macina 914-736-8363
  -----0rig i nal Message-----
From: Brian Trudeau [mailto:btrudeau@csitechnologies.com]
Sent: Tuesday, December 28,2004 3:52 P M To:       Macina, Ron


==Subject:==
==Subject:==
Request for Additional HBDs Tuesday, December 28,2004 3:52 PM CSI Doc. No. 04071 137 Ron, Based on our discussion earlier today, it appears that we have some additional work to do on this end. We plan on modeling all three power levels (100% - the Original power level, -101.4% - the Appendix K power level, and 105% - the SPU power level) for each unit. In doing this, we will capture plant conditions at each point in history. To complete this task, we will need the following input: (1) Unit 3 PEPSE or HBD at -101.4% power (the Appendix K power uprate)
Request for Additional HBDs CSI Doc. No.04071137 Ron, Based on our discussion earlier today, it appears that we have some additional work to do on this end. We plan on modeling all three power levels (100% - the Original power level, -101.4% - the Appendix K power
(2) Unit 2 Original PEPSE or HBD at 100% power (pre-Appendix K) (3) Unit 3 Start date for the Appendix K Uprate (4) Unit 2 Start date for the Appendix K Uprate Please let me know if you have any questions.
                      -
We will be contacting Harry to inform him of this plan. Thank you, Brian Trudeau CSI Technologies, Inc.
level,and 105% the SPU power level) for each unit. In doing this, we will capture plant conditions at each point in history.
(847) 836-3000 ext. 71 7 www.csitechnologies.com i Calculation No. 040711-02, Attachment A, Revision 0 Page 44 of 44 "."
To complete this task, we will need the following input:
CALCULATION APPROVAL COVER SHEET Document Title: Indian Point Unit 2 CHECWORKS Power Uprate Analysis Document No.: 04071 1-02 Client: Entergy Nuclear Northeast Project No.: 04071 1 This calculation has been prepared in accordance with Section 4.3 of the CSI Quality System Manual, Revision 2. Revision:
(1) Unit 3 PEPSE or HBD at -101.4% power (the Appendix K power uprate)
0 Prepared:
(2)Unit 2 Original PEPSE or HBD at 100% power (pre-Appendix K)
Date: Verified:
(3) Unit 3 Start date for the Appendix K Uprate (4) Unit 2 Start date for the Appendix K Uprate Please let me know if you have any questions. We will be contacting Harry to inform him of this plan.
Date: Approved: Date:
Thank you, Brian Trudeau CSI Technologies, Inc.
3/2Yk5i--
(847) 836-3000 ext. 717 www.csitechnologies.com i
Daniel R. Poe (CSI) Approved:
Calculation No. 040711-02, Attachment A, Revision 0                                   Page 44 of 44
Date:
                                                                            "."
Revision Control Sheet Rev. 0 Project Number: 04071 1 Project Name: Client: Entergy Nuclear Northeast Document Title: Document Number: 04071 1-02 Indian Point Units 2 & 3 CHECWORKS Power Uprate Analysis Indian Point Unit 2 CHECWORKS Power Uprate Analysis Date Description of Changes CSI Client Approval Approval 3/23/05 Initial Issue - For Use}}
 
CALCULATION APPROVAL COVER SHEET Document Title:         Indian Point Unit 2 CHECWORKS Power Uprate Analysis Document No.:           04071 1-02 Client:                 Entergy Nuclear Northeast Project No.:             040711 This calculation has been prepared in accordance with Section 4.3 of the CSI Quality System Manual, Revision 2.
Revision:   0 Prepared:                                             Date:
Verified:                                           Date:
Approved:                                             Date: 3/2Yk5i--
Daniel R. Poe (CSI)
Approved:                                             Date:
 
Revision Control Sheet Project Number:       040711 Project Name:         Indian Point Units 2 & 3 CHECWORKS Power Uprate Analysis Client:                Entergy Nuclear Northeast Document Title:        Indian Point Unit 2 CHECWORKS Power Uprate Analysis Document Number:      040711-02 Rev.      Date                     Description of Changes           CSI   Client Approval Approval 0      3/23/05 Initial Issue - For Use}}

Revision as of 07:09, 12 November 2019

Entergy Pre-Filed Evidentiary Hearing Exhibit ENT000072 - IP2 Checworks Power Uprate Analysis, Calculation No. 040711-02, Rev. 0
ML12088A424
Person / Time
Site: Indian Point  Entergy icon.png
Issue date: 03/28/2012
From:
CSI Technologies
To:
Entergy Nuclear Northeast, Entergy Nuclear Operations, Atomic Safety and Licensing Board Panel
SECY RAS
Shared Package
ML12088A422 List:
References
RAS 22102, 50-247-LR, 50-286-LR, ASLBP 07-858-03-LR-BD01 040711-02, Rev 0
Download: ML12088A424 (47)


Text

ENT000072 Submitted: March 28, 2012 Indian Point Unit 2 CHECWORKS Power Uprate Analysis Calculation No. 040711-02 Revision 0 Issued For-Use March 23,2005 preparedfor:

Entergy Nuclear Northeast 295 Broadway Suite 3 PO Box 308 Buchanan, NY 10511-0308 ENTERGY NUCLEAR NORTHEAST prepared by:

ACCEPTED ACCEPTED AS NOTED RESUBMITTALNOT REQUIRED TECHNOLOGIES, INC.

3 ACCEPTED AS NOTED RESUBMITTAL REQUIRED E. Main St., Suite 215 4 NOT ACCEPTED Dundee, IL 60118 c

CSI TECHNOLOGIES. INC . IP2 CHECWORKS Power Uprate Analysis Table of Contents

1. INTRODUCTION .............................................................................................................................................. 2
2. PURPOSE........................................................................................................................................................... 3
3. SCOPE ................................................................................................................................................................ 4
4. ASSUMPTIONS AND MODELING DECISIONS......................................................................................... 5
5. METHODOLOGY ............................................................................................................................................ 8 5.1. INPUTSPU GLOBALDATA........................................................................................................................... 8 5.2. REDEFINKION OF CHECWORKS LINES.................................................................................................... 12 5.3. UPDATE MODELBASEDON SPU OPERATIONAL OR CONFIGURATION CHANGES ........................................ 13 5.4. UPDATE NETWORK FLOWANALYSIS D E F I N ~ O......................................................................................

NS 13 5.5. IMPLEMENT THE ADVANCED RUNDEFINITION ........................................................................................... 13 5.6. PERFORM WEAR RATE ANALYSIS............................................................................................................... 15 5.7. QUANTIFY EFFECTOF STRETCH POWERUPRATE ........................................................................................ 16

6. RESULTS ......................................................................................................................................................... 17 6.1. COMPONENT LEVELWEARRATECHANGES DUE TO SPU........................................................................... 17 6.2. STEAMCYCLELEVELWEARRATECHANGES DUE TO SPU ........................................................................ 17
7. REFERENCES ................................................................................................................................................. 18 APPENDIX A: CHECWORKS MODELED LINES .............................................................................................. 21 APPENDIX B: COMPONENT LEVELWEAR RATE CHANGES DUE TO SPU ........................................................ 33 APPENDIX c : STEAM CYCLE LEVELWEAR RATE CHANGES DUE TO SPU ..................................................... 35 ATTACHMENT A: REFERENCED CORRESPONDENCE AND COMMUNICATIONS ....................................................... 39

.

Calculation No 040711.02. Revision 0 Page 1 of 44

CS/TECHNOLOGIES, /NC. IP2 CHECWORKS Power Uprate Analysis

( 1. Introduction This calculation documents the revision of the Indian Point Unit 2 CHECWORKS model to predict Flow-Accelerated Corrosion (FAC) wear rate changes due to Stretch Power Uprate (SPU). The Indian Point Unit 2 SPU will change feedwater and steam flow rates, temperatures, and enthalpies, which in turn change local chemistry values. All of these factors affect wear rates due to FAC. As a result of the uprate, some lines will experience accelerated rates of FAC, while others will have reduced rates. The impact on each line depends on the complex interaction of changes in flow rate, pressure, temperature, enthalpy, steam quality, and chemistry on the FAC degradation mechanism.

Indian Point 2 had previously developed a CHECWORKS model of FAC-susceptible piping. However, the previous model did not address the changes that will result from Appendix K Uprate and Stretch Power Uprate. This calculation details the process required to revise the CHECWORKS model so that it correctly reflects all plant power levels (the original power level, Appendix K Uprate, and Stretch Power Uprate). Also documented are the changes in FAC wear rates due to the SPU.

Note that historical @re-uprate and Appendix K Uprate) operating conditions remain within the model, associated to the applicable operating cycles. The SPU operating conditions are associated to the calendar time that those conditions are scheduled to occur, starting in Cycle 17 [7.3.1]. In this way, the models predictions of total current and future wear will be as accurate as possible because the predictions will be based on both historical and expected future operating conditions.

Calculation No. 040711-02, Revision 0 Page 2 of 44 I -~

CSI TECHNOLOGIES, INC. IP2 CHECWORKS Power Uprate Analysis

( 2. Purpose The purposes of the power uprate analysis in CHECWORKS are as follows:

0 To quantify changes in FAC wear rates due to the Stretch Power Uprate (both increases and decreases).

0 To gain the ability to describe the effects of power uprate on FAC-susceptible piping in the Licensing Submittal.

0 To ensure that the CHECWORKS model reflects current plant conditions going forward.

Calculation No. 040711-02, Revision 0 Page 3 of 44

cs/TECHNOLOGIES, /NC. IP2 CHECWORKS Power Uprate Analysis I 3. Scope The scope of this power uprate analysis was determined from the input Indian Point 2 CHECWORKS model [7.2]. However, not all lines and components in the input model were included in this analysis. Only those lines and components that were assigned to one of the Wear Rate Analysis run definitions in Table 1.O of document number 00130-TR-001 were analyzed as part of this project [7.10]. The CHECWORKS model also contains non-modeled lines and components (typically assigned to CHECWORKS lines with the prefix NCW for Non-CHECWORKS). No analysis was performed on these NCW lines or any other plant lines and components that were not part of the official CHECWORKS model. Note that Table 1.0 of document 00130-TR-001 has two hand-written entries x-under w/ exp joints and CND FWH 22 to FWH 23 [7.10].

Lines and components that belong in these two locations were also included in this analysis.

This analysis was performed using CHECWORKS FAC version 1.OG.

Assumptions and modeling decisions made during this analysis are documented in Section 4. The methodology employed during this analysis is detailed in Section 5.

Results obtained are listed in Section 6 and in the Appendices. Finally, Section 7 includes a list of all references used in this analysis.

Calculation No. 040711-02, Revision 0 Page 4 of 44

--I

cs/ TECHNOLOGIES, /NC. IP2 CHECWORKS Power Uprate Analysis

4. Assumptions and Modeling Decisions The following assumptions and modeling decisions apply to the updated CHECWORKS model.

4.1. All data in the CHECWORKS model received as input (the as-received model)

[7.2] was assumed to be complete and accurate.

4.2. Exceptions were made to Assumption 4.1 in the following cases. These exceptions appear in this section, as they are not a standard part of a CHECWORKS power uprate analysis.

4.2.1. The line on the CHECWORKS Heat Balance Diagram (HBD) representing the Moisture Separator/MoisturePreseparator drain lines was remodeled so the downstream connection of the line was the Heater Drain Tank instead of Feedwater Heater 25. The correct configuration was determined from flow diagrams [7.3.3] and the SPU heat balance diagram C7.1.31.

4.2.2. The CHECWORKS HBD was corrected so that the steam driven Feed Pump supply originated in the Main Steam system upstream of LP Turbine as opposed to Main Steam upstream of the HP Turbine. Plant layout shows steam supply as coming from both locations; however, the SPU HBD shows that supply is from Main Steam upstream of LP Turbine under normal operating conditions [7.1.3].

4.2.3. Reheat Steam to the Moisture Separator Reheater on the CHECWORKS HBD was remodeled so the source was from the Main Steam system upstream of the HP Turbine as opposed to downstream of the HP Turbine.

The correct configuration was determined from flow diagrams [7.3.3] and the SPU heat balance diagram [7.1.3].

4.2.4. The condensate pump was added to the CHECWORKS HBD; the condensate pump had been omitted previously.

4.2.5. The input CHECWORKS model did not have components in the Condensate system between Feedwater Heaters 22 and Feedwater Heaters

23. All components in these lines were modeled as part of this analysis.

Component information and configuration was determined by using the existing FAC isometric for Unit 3 as a guide [7.8]. Lines were named using Sketch 80A as the sketch number. The FAC Program Owner supplied component names [7.9].

4.2.6. An error stating, "The input hydrazine conc, at SG Steam outlet caused a mass in-balance in the Water Chemistry Analysis" was generated when i

performing water chemistry analysis for Cycle 14. The hydrazine Calculation No. 040711-02, Revision 0 Page 5 of 44

CSI TECHNOLOGIES, /NC. IP2 CHECWORKS Power Uprate Analysis concentration at the Steam Generator Outlet was lowered from 18 ppb to 15 ppb for this water treatment. This action enabled water chemistry analysis to run without error.

4.2.7. The hydrazine concentration at the SG outlet and MSR drain locations were not specified for Cycle 16 water treatment. If this data is not entered, CHECWORKS calculates a Concentration of zero at these locations.

Therefore, estimated hydrazine concentrations were entered at these locations by using the rules of thumb for a Recirculating Steam Generator [7.6]. Based on the rules of thumb, the concentration of hydrazine at the Steam Generator Outlet was assumed to be 60% of the final feedwater concentration, while the concentration of hydrazine at the MSR Drain was assumed to be 120% of the final feedwater concentration.

4.2.8. During WRA, an error message appeared for component MS- 1B 14P-1 US stating that, Orifice size 5.761 cannot be greater than pipe inside diameter 5.61 1. The initial thickness of this component (0.507) was deleted; nominal thickness was then used in this calculation. WRA was able to proceed without further error.

4.3. For a number of lines on the SPU Heat Balance Diagram [7.1.3] and Appendix K Heat Balance Diagram [7.1.2], thermodynamic and flow values (pressure, enthalpy, and flow rate) were listed separately for the steam phase and the water phase or for each train in a parallel train configuration. The overall flow rate, pressure, and enthalpy of these lines were calculated and entered in the CHECWORKS Steam Cycle (see Section 5.1.3). The combined flow rate was calculated as the sum of the liquid and steam flow rates (or the sum of multiple trains), the combined pressure was calculated as the average of all pressures, and the enthalpy was calculated as the weighted average of liquid and steam enthalpy (or the weighted average of multiple trains). These calculations were performed based on EPRIs Guidelines for Plant Modeling and Evaluation of Component Inspection Data [7.6].

4.4. Plant period data was estimated for the cycle where the SPU will occur. Start and end dates were estimated based on anticipated dates [7.3.1]. An estimation of operating hours was calculated from these dates based on calendar days. The Water Treatment for this cycle was modeled as equivalent to the most recent complete operating cycle. The model can be updated with actual values for these inputs when this data becomes available.

4.5. Due to the use of the Advanced Run Definition feature in this model, which is required for accurate modeling of the power uprate condition, CHECWORKS FAC Version 1.OG cannot accurately represent the operating conditions in tees that combine or split flow. In order to maintain simplicity in the CHECWORKS model, tees were modeled with a flow rate equal to the highest flow rate present in Calculation No. 040711-02, Revision 0 Page 6 of 44

,- -- ._

CSI TECHNOLOGIES, /NC. IP2 CHECWORKS Power Uprate Analysis the tee. Therefore, the predicted wear rates for tees should be used with caution.

Calculation No. 040711-02, Revision 0 Page 7 of 44

CSI TECHNOLOGIES, INC. IP2 CHECWORKS Power Uprate Analysis i 5. Methodology Additional input data was entered into the CHECWORKS model to model the power uprate. A number of tasks were required to convert the previous Indian Point Unit 2 model into a format that was compatible with a power uprate. The following section describes the tasks performed to enter power uprate data and to convert the model into a form compatible with multiple power levels.

5.1. Input SPU Global Data Additional CHECWORKS Power Levels representing the SPU and Appendix K operating conditions were defined in the model and associated to the applicable Plant Period. The following sections detail the steps involved in this task.

5.1.1. CHECWORKS Heat Balance Diagram The CHECWORKS Heat Balance Diagram (HBD) was reviewed to ensure that it correctly portrayed plant configuration. The CHECWORKS HBD was compared to the plant Heat Balance Diagrams [7.1] and flow diagrams [7.3.3]. Discrepancies were noted and corrected as discussed in Section 4.2.

5.1.2. Power Level Data i A Plant Power Level is defined for each power level at which the plant is operated for a significant period of time. Two new power levels were added to the model representing the SPU and Appendix K conditions.

Table 5.1 lists all power levels in the model and the operating cycles to which they apply.

Table 5.1 Power Level History 100.00 3090.2 Cycles 1-16A Original Power Level 101.19 3127.0 Cycle 16B Appendix K Uprate 104.48 3228.5 Cycle 17 to End of Life Stretch Power Uprate Data was entered for the new power levels on the Power Level Form in accordance with the CHECWORKS Users Guide [7.5].

Table 5.2 summarizes the Appendix K power level input data and the source of this data.

Calculation No. 040711-02, Revision 0 Page 8 of 44

-

CSI TECHNOLOGIES, INC. IP2 CHECWORKS Power Uprate Analysis Table 5.2 Appendix K Power Level Input Data INotes: Stretch Power Uprate. 3127.OMWt x - Field should be left blank for a PWR.

Table 5.1 summarizes the Appendix K power level input data and the source of this data.

Table 5.3 SPU Power Level Input Data I Steam Rate (Mlb/hr> I 13.903750 I 7.1.3 I IPressure (mia) I 765.0 I 7.1.3 I Temp (F) 513.1 7.1.3 Blowdown Rate (Mlb/hr> 0.0542 7.1.3 Carryover (%) 0.02 7.1.3 Feedwater Vent Rate f%> X 7.5 I Reheater Vent Rate (%) I X I 7.5 I IMoisture Smarator Carrvunder (%> I X I 7.5 I

/Notes: Stretch Power Umate. 3228.5 MWt I x - Field should be left blank for a PWR.

5.1.3. Steam Cycle Data Steam Cycle Data is used by CHECWORKS to calculate chemistry conditions during wear rate analysis. It is also used to calculate operating conditions when the Advanced Run Definition feature is implemented (see Section 5.5). Steam cycle data was entered for each heat balance item for the SPU power level. All data was entered in accordance with EPRIs Guidelines for Plant Modeling and Evaluation of Component Inspection Calculation No. 040711-02, Revision 0 Page 9 of 44

cs/TECHNOLOGIES, /NC. IP2 CHECWORKS Power Uprate Analysis Data [7.6] and the CHECWORKS Users Guide [7.5]. Table 5.4 summarizes SPU steam cycle input data and the source of this data.

Table 5.4 Appendix K Steam Cycle Input Data (t

x = No value entered (not required by CHECWORKS).

(1) The HBD Item name is automatically generated by CHECWORKS. Feedwater heaters are numbered sequentially in reverse flow order. Feedwater Heater 1 is the feedwater heater closest to the steam generator (equivalent to heater 26 at Indian Point 2). Extraction lines are numbered sequentially in order of decreasing pressure.

(2) MSEP 1 represents the conditions in both the moisture separator and moisture pre-separator drain lines as recommended by EPFU Guidelines [7.6].

(3)HPEXTLINE 2 is a fictitious high-pressure extraction line representing the steam lines between the pre-separator and main separator as recommended by EPFU Guidelines [7.6].

(4) Flow rate is for exiting steam flow was entered as zero as recommended by EPRI Guidelines [7.6]. Pressure and enthalpy were obtained from the Appendix K PEPSE model [7.1.2].

(5) Enthalpy calculated as the weighted average of the steam and liquid phases. Steam phase enthalpy was obtained directly from the PEPSE diagram as the enthalpy after moisture removal in the LP Turbine. Liquid phase enthalpy was calculated as the enthalpy of saturated liquid at the pressure given on the PEPSE diagram.

(6) Reheater drain flow was entered as the sum of the flow through the reheater drain tanks and the vent chamber drains [7.1.2]. Pressure and enthalpy were entered as the conditions through the reheater drain tank [7.1.2].

Table 5.5 summarizes SPU steam cycle input data and the source of this data.

Calculation No. 040711-02, Revision 0 Page 10 of 44

cs/TECHNOLOGIES,/NC. IP2 CHECWORKS Power Uprate Analysis If Table 5.5 SPU Steam Cycle Input Data i,

x = N o value entered (not required by CHECWORKS).

(1) The HBD Item name is automatically generated by CHECWORKS. Feedwater heaters are numbered sequentially in reverse flow order. Feedwater Heater 1 is the feedwater heater closest to the steam generator (equivalent to heater 26 at Indian Point 2). Extraction lines are numbered sequentially in order of decreasing pressure.

(2) MSEP 1 represents the conditions in both the moisture separator and moisture pre-separator drain lines as recommended by EPRI Guidelines [7.6].

(3) HPEXTLINE 2 is a fictitious high-pressure extraction line representing the steam lines between the pre-separator and main separator as recommended by EPRI Guidelines [7.6].

(4) Flow rate is for exiting steam flow was entered as zero as recommended by EPRI Guidelines [7.6]. Pressure and enthalpy were obtained from the SPU PEPSE model [7.1.3].

(5) Enthalpy calculated as the weighted average of the steam and liquid phases. Steam phase enthalpy was obtained directly from the PEPSE diagram as the enthalpy after moisture removal in the LP Turbine. Liquid phase enthalpy was calculated as the enthalpy of saturated liquid at the pressure given on the PEPSE diagram.

5.1.4. Plant Period Data Each power level identified in Table 5.1 was associated to the correct operating cycle or cycles. The Appendix K Uprate start date was obtained through correspondence with plant personnel [7.3.3].The SPU is scheduled for the start of Cycle 17. To include this power level in the model, plant periods were created that have not yet occurred. Start dates, Calculation No. 040711-02, Revision 0 Page 11 of 44 LI _ -

--- -

CSI TECHNOLOGIES, INC. IP2 CHECWORKS Power Uprate Analysis end dates, operating hours, and chemistry data was estimated for these periods (see Section 4.4).

5.2. Redefinition of CHECWORKS Lines The CHECWORKS model was prepared for use of the Advanced Run Definition (see Section 5.5). This task consisted of redefining the CHECWORKS lines within the model so that all components on a given line are subject to the same thermodynamic conditions. Appendix A contains a listing of all lines that appear in the CHECWORKS model.

For conservatism, the tee where flow rate changed was associated to the line having the greatest flow rate. Note that the model will over-predict the wear for some sections of tees (see Section 4.5).

Figure 5- 1 illustrates the procedure used to group components into lines. Inthe diagram, seven lines exist, each with a different percentage of the total flow. Note that the tee is grouped with the line number seeing the greatest percentage of flow.

Figure 5-1 Diagram of Line Redefinition Procedure 1 EQUIPMENT OUTPUT V

Calculation No. 040711-02,Revision 0 Page 12 of 44

(_I_____ --

CSI TECHNOLOGIES, INC. IP2 CHECWORKS Power Uprate Analysis Components were grouped into lines by comparing the input CHECWORKS model [7.2] with FAC isometrics [7.7] and flow diagrams [7.3.3]. New lines were given names according to the naming convention, below.

AABBC-D-E AA = Abbreviation of the system (ex: CD = Condensate, EX = Extraction Steam, etc.)

BB = Sketch number the line begins on (ex: 01, 80, etc.)

C = Sketch letter if sketch number includes a letter (ex: A for sketch 80A)

D = Sequential number for each line on one sketch, numbered in flow order E = Brief description of the line New line names were created as required by CHECWORKS, not where plant line names changed. Therefore, lines may contain components located on different sketches but in all cases the sketch number corresponds to the first component in the line.

For example, line name CD80-1-FWH 23A to FWH 24A is located in the Condensate system, on FAC sketch 80, and runs from Feedwater Heater 23A to Feedwater Heater 24A.

5.3. Update Model Based on SPU Operational or Configuration Changes With an uprate in power, lines with partial operation may increase or decrease their operation frequency (such as additional trains in operation or bypass lines now operated during full power). Additionally, uprates may call for design changes that may result in piping configuration changes.

No operational or configuration changes occurred due to the SPU [7.3.2].

5.4. Update Network Flow Analysis Definitions Network Flow Analysis (NFA) is a module within CHECWORKS that calculates flow rate, pressure drops, temperature changes, and steam quality changes based on piping configuration and source/sink conditions. NFA is used where operating conditions are not completely known. A different set of source/sirik conditions is entered for each NFA per power level. Additional input data must be added to each NFA to reflect the SPU power level identified in Section 5.1.2.

The Indian Point 2 CHECWORKS model did not contain any NFA definitions; therefore, no updates to the model were performed during this task [7.2].

5.5. Implement the Advanced Run Definition This task involved a redefinition of the source in which CHECWORKS obtains thermodynamic conditions (pressure, enthalpy, temperature, and quality) and flow rate conditions for a component. Previously all thermodynamic and flow rate conditions had been entered individually for each component on the component data forms. However, the component form allows only one set of thermodynamic Calculation No. 040711-02, Revision 0 Page 13 of 44

, _1_- --

cs/TECHNOLOGIES, /NC. IP2 CHECWORKS Power Uprate Analysis and flow rate conditions to be entered (Le. from one power level). Therefore, use of the component form as the input for thermodynamic and flow rate conditions is not valid, as it does not reflect both pre-uprate and post-uprate conditions.

Instead, thermodynamic and flow conditions were entered globally (see Section 5.1) and linked to components through the association of a line to the CHECWORKS HBD (except 2-type lines). The following sections detail the steps performed to implement the Advanced Run Definition.

5.5.1. Enter Flow Factors On the CHECWORKS HBD level, flow rates are expressed in totals rather than for each train. For example, feedwater flow rate might be entered as 10 million pounds per hour, where each train of a three-train system sees 3.33 million pounds per hour. As a result, flow multipliers had to be entered for the lines so that the actual flow rate is used to calculate wear rate at the component level. Thus for each line a flow multiplier, or flow factor, was calculated. The flow factor is used to adjust the CHECWORKS HBD calculated flow rate. The calculated flow factor for each line was entered on the ARD form.

There are some exceptions to the use of flow factors. The first is for lines and flow segments where NFA would be used to calculate operating conditions and flow rate. For these the train flow is directly entered into the NFA definitions. Therefore, these lines the assigned flow factor is 1.O.

Other exceptions are made for some lines and flow segments where the ARD form is used as the source of operating conditions. In some cases, if the input source (PEPSE or HBD) already listed flow rate per train, then the flow factor is set to 1.O and the train flow rate is entered.

Flow factors were calculated by consulting the CHECWORKS HBD, the plant heat balance diagrams [7.1], and the flow diagrams [7.3.3]. Flow factors for each line appear in Appendix A.

5.5.2. Enter Duty Factors The duty factor is used to specify the fraction of the total plant operating hours that a given line was in operation. For full-time lines, the duty factor is 1.O. For part-time lines, the duty factor is set to a value less than one based on operation. For example, if a line has full flow half of the time and zero flow half of the time, then the lines would be modeled with full flow and the duty factor would be set to 0.5. Use of the duty factor is in accordance with the recommendations of the EPRI Guidelines for Plant Modeling and Evaluation of Component Inspection Data [7.6].

Duty factors were taken from the input CHECWORKS model [7.2]. Duty factors for each line appear in Appendix A.

Calculation No. 040711-02, Revision 0 Page 14 of 44

CSI TECHNOLOGIES, INC. IP2 CHECWORKS Power Uprate Analysis 5.5.3. Complete Advanced Run Definition Form for Z-Type Lines Lines not associated to the CHECWORKS HBD are called Z-type lines.

Because they are not associated to the HBD, CHECWORKS cannot automatically calculate chemistry and operating conditions for these lines.

Therefore, when using the ARD function, the user must input not only flow factors and duty factors but also thermodynamic conditions, flow rate, and chemistry conditions for each operating cycle.

Z-type lines were created due to limitations in the CHECWORKS HBD.

In these cases, the computer model does not obtain the data fiom the correct location on the HBD, or the CHECWORKS program did not allow the correct data to be entered. For example, there is no global input into CHECWORKS to specify the pressure, temperature, enthalpy, or quality in feedwater heater drain lines. Instead the model calculates the conditions in the shell side drain as being equivalent to tube side heater outlet. This is incorrect, so the CHECWORKS HBD was not used as the source of operating conditions for heater drain lines. Instead, operating conditions for heater drain lines were entered on the ARD form.

Appendix A lists all the lines in the model and includes whether or not the line is a Z-type line.

For Z-type lines, thermodynamic data and flow rate was obtained from the Heat Balance Diagrams [7.1].

5.5.4. Set Wear Rate Analysis Source of Data Option The CHECWORKS model allows the user to specifL the source of component operating conditions. Component operating conditions can come from one of four locations: the CHECWORKS HBD, the Component form, an NFA, or the ARD. During wear rate analysis, CHECWORKS can use the operating conditions stored at the component level (COMP), determine the operating conditions based upon steam cycle data and Advanced Run Definition Flow Factors (HBD), use the operating conditions entered on the Advanced Run Definition form only (ARD), or to use the operating conditions calculated using an NFA (NFA). For all cases, the option NFA->HBD->ARD->COMP was selected. This directs CHECWORKS to preferentially use Network Flow Analysis first (if it exists for the line), followed by the ARD (for Z-type lines), the HBD (for all remaining lines), and finally the component.

The option NFA->HBD->ARD->COMP was selected for all lines since the model includes multiple power levels.

5.6. Perform Wear Rate Analysis Wear Rate Analysis (WRA) is a module within CHECWORKS that provides I predicted wear rates and remaining service lives for each modeled component.

WRA was run on every component so that the predicted wear rates include the Calculation No. 040711-02, Revision 0 Page 15 of 44

, - I

CSI TECHNOLOGIES, INC. IP2 CHECWORKS Power Uprate Analysis SPU conditions. WRA includes an error-trapping routine, so that discrepancies will be identified and corrected. WRA was performed successfully without error.

5.7. Quantify Effect of Stretch Power Uprate An analysis was performed to calculate the change in CHECWORKS predicted wear rates due to the SPU conditions. Wear Rate Analysis was performed for two periods representing the original power level and the SPU power level. The water treatment for these two periods was modeled as identical, so wear rate changes are due to SPU conditions only. In both cases, Wear Rate Analysis was performed using Pass 2 methods, where predictions are calibrated to inspection history.

The analysis obtained both actual results and percentage differences for representative components and lines so that detailed comparisons could be made.

The analysis was limited to non-Chromium containing components only, so average values would not be skewed by these components. In addition the changes due to SPU for some of the dominant parameters affecting FAC wear rates (temperature, steam quality, and flow rate) were determined.

i Calculation No. 040711-02, Revision 0 Page 16 of 44

CSI TECHNOLOGIES, INC. IP2 CHECWORKS Power Uprate Analysis

6. Results 6.1. Component Level Wear Rate Changes due to SPU An analysis was performed on a sample of some of the components in the model most susceptible to FAC. Pass 2 Wear Rate Analysis was performed at the pre-uprate, original power level and the SPU power level. The samples were determined as follows:

0 The five components in the model with the highest wear rates for the original, pre-uprate power level were selected.

0 The five components that experienced the greatest percent increase in wear rate due to the SPU (excluding chromium containing components).

The results of this analysis appear in Appendix B 6.2. Steam Cycle Level Wear Rate Changes due to SPU An analysis was performed comparing Steam Cycle Level changes in wear rate predictions due to the SPU. Lines with similar thermodynamic conditions were grouped together and the average component wear rate was calculated for each grouping, called a Steam Cycle Location. In addition some of the dominant parameters affecting FAC wear rates (temperature, steam quality, and flow rate) were determined. The results of this analysis appear in tabular form in Appendix C. In addition, results are summarized on a steam cycle drawing in Appendix C.

Calculation No. 040711-02, Revision 0 Page 17 of 44

--.-* _ - - -- .- -

CSI TECHNOLOGIES,INC. IP2 CHECWORKS Power Uprate Analysis i 7. References 7.1. Indian Point 2 Heat Balance Diagrams 7.1.1. Original HBD, 3090.2 MWt: Indian Point 2 Nuclear Power Plant Benchmark PEPSE Model Tuned to 1-22-03 Data, Sheets 1-6, Run Date 1/ 10/05.

7.1.2. Appendix K HBD, 3 127.0 MWt: Indian Point 2 Nuclear Power Plant Benchmark PEPSE Model Tuned to 1-22-03 Data, Sheets 1-6, Run Date 11/11/04.

7.1.3. Stretch Power Uprate HBD, 3228.5 MWt: Indian Point 2 Nuclear Power Plant Uprate PEPSE Model with New HP Turbine, Sheets 1-6, S&W Calc 58030-HU(S)-001 Rev. 0, Attachment 8.3 pages 1-6.

7.2. Indian Point 2 CHECWORKS FAC model, input model (as-received), electronic files provided to CSI on 10/5/2004.

7.3. Referenced Correspondence and Communications (see Attachment A) 7.3.1. Email from Harry Hartjen (IP) to Daniel R. Poe (CSI Technologies), dated i 10/12/2004, regarding SPU implementation dates, CSI Doc. No.

0407 1111.

7.3.2. Email from Harry Hartjen (IP) to Daniel R.Poe (CSI Technologies), dated 10/18/2004,regarding operational and configuration changes due to SPU, CSI Doc. No. 04071 113.

7.3.3. Email from Ron Macina (IP3) to Brian Trudeau (CSI Technologies), dated 1/10/2005,regarding addition a1 Heat Balance Diagrams and uprate start dates, CSI Doc. No 04071 140.

7.4. Indian Point 2 Flow Diagrams Main Steam, Dwg No.9321-F-2017, Rev. 83 Condensate & Boiler Feed Pump Suction, Dwg No.9321-2018, Rev. 137 Boiler Feedwater, Dwg No.9321-F-2019, Rev. 110 Extraction Steam, Dwg No.9321-F-2020, Rev.41 Heater Drain & Vents, Dwg No.9321-F-2022-52, Rev. 52 Moisture Separator and Reheater Drains & Vents, Dwg No.9321-F-2023-21, Rev. 31 Boiler Feed Pump Turbine Steam Lines Drains & Vents, Dwg No.9321-H-2024-23, Rev. 23 Steam Supply & Condensate Return, Dwg No.9321-F-2027, Rev. 61 Main Steam, Dwg No.227780, Rev. 50 Moisture PreSeparator,Dwg No.A-228272, Rev. 15 Condensate & Boiler Feed Pump Suction, Dwg No.A-235307, Rev. 29 Calculation No. 040711-02, Revision 0 Page 18 of 44

-- I_

CSI TECHNOLOGIES, INC. IP2 CHECWORKS Power Uprate Analysis Heater Drain & Vents, Dwg No. A-235304,Rev. 23 Steam Generator Blowdown & Blowdown Sample System, Dwg No. 9321-F-2729,Rev. 66 7.5. CHECWORKS Flow-Accelerated Corrosion Application, Version 1.OG User Guide, Document TR- 103198-P 1-R1, October 2000.

7.6. CHECWORKS StedFeedwater Application, Guidelines for Plant Modeling and Evaluation of Component Inspection Data, Doc. No. 1009599, Final Report, September 2004.

7.7. Indian Point 2 FAC Program Isometrics/Sketches SK-1Rev D SK-33ARev C SK-50B Rev C SK-2Rev D SK-33BRev C SK-50C Rev C SK-3Rev D SK-34ARev C SK-51Rev D SK-4 Rev C SK-34BRev C SK-52Rev C SK-5Rev E SK-35ARev C SK-53Rev C SK-6Rev E SK-35BRev C SK-54 Rev C SK-7Rev C SK-36ARev C SK-56Rev D SK-7ARev B SK-36BRev C SK-57Rev C SK-8Rev D SK-37ARev C SK-66ARev C SK-9Rev D SK-37BRev C SK-68ARev D SK-12Rev D SK-38ARev C SK-69A Rev D SK-19Rev C SK-38BRev D SK-71Rev C SK-20Rev E SK-39Rev B SK-72Rev B SK-21A Rev C SK-40Rev B SK-73Rev D SK-21B Rev C SK-41 R e v B SK-74Rev C SK-22ARev C SK-42 Rev B SK-75Rev C SK-22BRev C SK-43 Rev B SK-76Rev C SK-23A Rev C SK-44 Rev B SK-77Rev C SK-23BRev D S K 4 A Rev C SK-78Rev C SK-24ARev B SK-45B Rev B SK-80Rev B SK-24BRev C SK-45CRev B SK-81R e v B SK-25ARev B SK-45D Rev B SK-82Rev C SK-25B Rev C SK-46A Rev C SK-83Rev C SK-26ARev B SK-46B Rev C SK-84Rev C SK-26B Rev C SK-47 Rev B SK-92Rev C SK-27 Rev C SK-48A Rev C SK-93Rev B SK-28Rev C SK-48B Rev D SK-94Rev B SK-29Rev C SK-49A Rev D SK-95Rev B SK-30Rev C SK49B Rev B SK-242Rev A SK-31Rev C SK-49C Rev D SK-243Rev A SK-32Rev C SK-5OARev D SK-244Rev A 7.8. Indian Point 3 FAC Program Isometric, Erosion Corrosion Inspection Turbine Building & Heater Bay Condensate System Piping Isometric from F.W. Heaters 32A, B & C to F.W. Heaters 33A, B, C, Drawing No. EC-H-50061, Revision 1.

Calculation No. 040711-02, Revision 0 Page 19 of 44

-, ..___

CsI TECHNOLOGIES, INC. IP2 CHECWORKS Power Uprate Analysis 7.9. Component Names for Newly Modeled Lines Between Feedwater Heaters 22 and Feedwater Heaters 23, electronic file UNIT2 COMP NOS.xls provided to CSI on 11/15/2004.

7.10. Flow Accelerated Corrosion Program CHECWORKS Analysis Enhancement, Technical Report No. 00 130-TR-001, Volume 1 of 1, Revision 0, December 2000.

Calculation No. 040711-02, Revision 0 Page 20 of 44 I. . -. -.I

.-

CSI TECHNOLOGIES, INC. IP2 CHECWORKS Power Uprate Analysis Appendix A CHECWORKS Modeled Lines Calculation No. 040711-02, Appendix A, Revision 0 Page 21 of 44

--

,--- -- .-

CSI I ECHNOLOGIES, /NC. IP3 CHECWORKS Power Uprate Analysis CD8OA-2-FWH 22B to HEADER Feedwater Heater 228 to Header A235307-29 HBD 7 0.333 1.000 CD80A-3-FWH 22C to HEADER Feedwater Heater 22C to Header A235307-29 HBD 7 0.333 1.000 Feedwater Heater 22 Outlet Header Between 22BT CD80A-4-FWH 22 OUTLET HEADER and 22CT A235307-29 HBD 7 0.667 1.000 Feedwater Heater 22 Outlet Header to Feedwater CD80A-5-FWH 22 to FWH 23 HEAD Heater 23 Inlet Header A235307-29 HBD 7 1.000 1.000 Feedwater Heater 22 Inlet Header Between 23CT CD80A-6-FWH 23 INLET HEADER and 23BT A235307-29 HBD 7 0.667 1.000 Feedwater Heater 23 Inlet Header to Feedwater CD80A-7-HEADER to FWH 23A Heater 23A A235307-29 HBD 7 0.333 1.000 Feedwater Heater 23 Inlet Header to Feedwater CD80A-8-HEADER to FWH 23B Heater 23B A235307-29 HBD 7 0.333 1.000 Feedwater Heater 23 Inlet Header to Feedwater CD8OA-9-HEADER to FWH 23C Heater 23C A235307-29 HBD 7 0.333 1.000 CD81-1-FWH 24A to FWH 25A Feedwater Heater 24A to Feedwater Heater 25A A235307-29 HBD 5 0.333 1.000 CD81-2-FWH 24B to FWH 25B Feedwater Heater 24B to Feedwater Heater 25B A235307-29 HBD 5 0.333 1.000 CD81-3-FWH 24C to FWH 25C Feedwater Heater 24C to Feedwater Heater 25C , A235307-29 HBD 5 0.333 1.000 CD82-1-FWH 25A to HDR Feedwater Heater 25A to Header A235307-29 HBD 4 0.333 1.000 CD82-2-FWH 258 to HDR Feedwater Heater 258 to Header A235307-29 HBD 4 0.333 1.000 CD82-3-FWH 25C to HDR Feedwater Heater 25C to Header A235307-29 HBD 4 0.333 1.000 Feedwater Heater 25 Outlet Header Between 258 CD82-4-HDR 25BT to 25CT Tee and 25C Tee A235307-29 HBD 4 0.667 1.000 Feedwater Heater 25 Outlet Header Between 2%

CD82-5-HDR 25CT to HDP OUT ,Teeand Heater Drain Pump Outlet Connection A235307-29 HBD 4 1.000 1.000 Feedwater Heater 25 Outlet Header Between Connection from HDP Discharge and Boiler Feed CD83-1-HDR HDP to BFP21T Pump 21 Tee A235307-29 Z-type 3 1.000 1.000 CD83-2-HDR to BFP21 Feedwater Heater Header to Boiler FeedPump21 A235307-29 Z-type 3 0.500 1.000

______

Calculation No040711-02,Appendix A, Revision 0

~~

Page 22 of 44

, -

.-- -7 Ch tECHNOLOGIES, INC. IP3 CHECWORKS Power Uprate AIMYS~S 5th Point Extraction Steam Header Between Tee to to BT Feedwater Heater 25C and Feedwater Heater 256 9321-F-202041 HBD 18 0.667 1.000 Calculation No. 040711-02, Appendix A, Revision 0 Page 23 of 44

,~ .--  %. - - -

QI I ECHNOLOGIES. INC. IP3 CHECWORKS Power Uprate Analysis 5th Point Extraction Steam Header to Feedwater ES7-4-5THPT ESHDR to FWH 25B Heater 25B 9321-F-202041 HBD 18 0.333 1.000 5th Point Extraction Steam Header to Feedwater ES7-5-5THPT ESHDR to FWH 25A Heater 25A 9321-F-202041 HBD 18 0.333 1.000 Moisture PreseparatorSeparating Tank A Vent to ES7A-1 -SEP TKA VNT to FWH25 Header Upstream of Feedwater Heaters 25A,B,C A228272-15 HBD 18 0.500 1.000 Moisture Preseparator Separating Tank B Vent to ES7A-2-SEP TKB VNT to FWH25 Header Upstream of Feedwater Heaters 25A,B,C A228272-15 HBD 18 0.500 1.000 ES8-I-6THPT ES to HDR HP Turbine 6th Point Extraction Steam to Header 9321-F-2020-41 HBD 17 0.500 1.000 ES8-2-6THPT ES to HDR HP Turbine 6th Point Extraction Steam to Header 9321-F-202041 HBD 17 0.500 1.000 HP Turbine 6th Point Extraction Steam Header to ES8-3-6THPT ESHDR to FWH 26 Feedwater Heater 26A,B,C 9321-F-20204 1 HBD 17 1.000 1.000 HP Turbine 6th Point Extraction Steam Header to ES8-4-6THPT ESHDR to FWH 26C Feedwater Heater 26C 9321-F-202041 HBD 17 0.333 1.000 HP Turbine 6th Point Extraction Steam to Header Between Tee to Feedwater Heater 26C and ES8-5-6THPT ESHDR 26CT to BT Feedwater Heater 268 9321-F-2020-41 HBD 17 0.667 1.000 HP Turbine 6th Point Extraction Steam Header to ES8-6-6THPT ESHDR to FWH 268 Feedwater Heater 268 9321-F-202041 HBD 17 0.333 1.000 HP Turbine 6th Point Extraction Steam Header to ES8-7-6THPT ESHDR to FWH 26A Feedwater Heater 26A 9321-F-2020-41 HBD 17 0.333 1.000 FW71-1-BFP21 DISCH to HDR Boiler Feedpump 21 Discharge to Header 9321-F-2019-110 Z-type 2 0.500 1.000 FW72-1-BFP22 DISCH to HDR Boiler Feedpump 22 Discharge to Header 9321-F-2019-110 Z-type 2 0.500 1.000 Boiler Feedpump Header to Feedwater Heater 26A,B,C Between BFP Discharge Tee and High FW73-1-BFPHDR to FWH26ABC Pressure Feedwater Heater Bypass Tee 9321-F-2019-110 Z-type 2 1.000 1.000 Boiler Feedpump Header to Feedwater Heater 26A,B,CBetween High Pressure Feedwater Heater FW73-2-BFPHDR to FWH26ABC Bypass Tee and Feedwater Heater 26C Tee 9321-F-2019-110 Z-type 2 1.000 1.000 IFW73-3-BFPHDR to FWH26C Boiler Feedpump Header to Feedwater Heater 26C 9321-F-2019-110 2-type 2 0.333 , 1.000 J Calculation No. 040711-02, Appendix A, Revision 0 Page 24 of 44

- 1

.?z cb I ECHNOLOGIES, /NC. IP3 CHECWORKS Power Uprate Analysis Boiler Feedpump Header to Feedwater Heater 26A,B,CBetween Feedwater Heater 26C Tee and

'W73-4-BFPHDR to FWH26ABC Feedwater Heater 26B Tee 9321-F-2014110 Z-type 2 0.667 1.000

'W73-5-BFPHDR to FWH26B Boiler Feedpump Header to Feedwater Heater 26B 9321-F-2019-110 2-type 2 0.333 1.000

'W73-6-BFPHDR to FWH26A Boiler Feedpump Header to Feedwater Heater 26A 9321-F-2019-110 Z-type 2 0.333 1.000

'W74-1-FWH26A to DISHDR Feedwater Heater 26A to Heater Discharge Header 9321-F-2019-110 HBD 1 0.333 1.000

'W74-2-FWH26B to DISHDR Feedwater Heater 268 to Heater Discharge Header 9321-F-2019-110 HBD 1 0.333 1.000 Feedwater Heater 26 Discharge Header Between

'W74-3-FWH26 to DISHDR 268 Tee and 26C Tee 9321-F-2019-110 HBD 1 0.667 1.000 Feedwater Heater 26C to Heater Discharge

'W74-4-FWH26C to DISHDR Header 9321-F-2019-110 HBD 1 0.333 1.000 Feedwater Heater 26 Discharge Header Between FW74-5-FWH26 to DISHDR 26C Tee and Steam Generator Takeoff 9321-F-2019-110 HBD 1 1.000 1.000 Feedwater Heater 26 Discharge Header to Steam FW75-I-DISHDR to SG21 Generator 21 9321-F-2019-110 HBD 1 0.250 1.000 Feedwater Heater 26 Discharge Header to Steam FW76-1-DISHDR to SG22 Generator 22 9321-F-2019-110 HBD 1 0.250 1.000 Feedwater Heater 26 Discharge Header to Steam FW77-1-DISHDR to SG24 Generator 24 9321-F-2019-110 HBD 1 0.250 1.000 Feedwater Heater 26 Discharge Header to Steam FW78-1-DISHDR to SG23 Generator 23 9321-F-2019-110 HBD 1 0.250 1.000 Feedwater Heater 26A to CV Upstream of Heater HD12-I -FWH26A to CV Drain Tank 9321-F-2022-52 Z-type 11 0.333 1.000 Feedwater Heater 268 to CV Upstream of Heater HD12-2-FWH26B to CV Drain Tank 9321-F-2022-52 2-type 11 0.333 1.000 Feedwater Heater 26C to CV Upstream of Heater HD12-3-FWH26C to CV Drain Tank 9321-F-2022-52 Z-type 11 0.333 1.000 HD12-4-FWH26A CV to HTR DR TK Feedwater Heater 26A CV to Heater Drain Tank 9321-F-2022-52 Z-type 11 0.333 1.000 Calculation No. 040711-02, Appendix A, Revision 0 Page 25 of 44

- ,--, --*

ch I ECHNOLOGIES, INC. IP3 CHECWORKS Power Uprate Analysis iD12-5-FWH26B CV to HTR DR TK Feedwater Heater 26B CV to Heater Drain Tank 9321-F-2022-52 Z-type 11 0.333 1.000 iD12-6-FWH26C CV to HTR DR TK Feedwater Heater 26C CV to Heater Drain Tank 9321-F-2022-52 2-type 11 0.333 1.000 iD19-1-HDT to HDP 21 SUCT Heater Drain Tank to Heater Drain Pump 21 9321-F-2022-52 HBD 10 0.500 1.000 iD19-2-HDT to HDP 22 SUCT Heater Drain Tank to Heater Drain Pump 22 9321-F-2022-52 HBD 10 0.500 1.000 Heater Drain Pump 21 to Boiler Feed Pump iD20-1-HDP21to BFP SUCTION Suction 9321-F-2022-52 HBD 9 0.500 1.000 Heater Drain Pump 22 to Boiler Feed Pump iD20-2-HDP22to BFP SUCTION Suction 9321-F-2022-52 HBD 9 0.500 1.000 Heater Drain Pump Discharge Tee to Boiler Feed iD20-3-HDP DIS T to BFP SUC Pump Suction 9321-F-2022-52 HBD 9 1.000 1.000 Feedwater Heater 24A to CV Upstream of iD21A-I-FWH24A to CV Feedwater Heater 23A A235304-23 Z-type 13 0.333 1.000 Feedwater Heater 24A CV to Feedwater Heater iD21A-2-FWH24ACV to FWH23A 23A A235304-23 Z-type 13 0.333 1.000 Feedwater Heater 248 to CV Upstream of iD22A-l-FWH24B to CV Feedwater Heater 23B A235304-23 Z-type 13 0.333 1.000 Feedwater Heater 24B CV to Feedwater Heater HD22A-2-FWH24B CV to FWH23B 238 A235304-23 Z-type 13 0.333 1.000 Feedwater Heater 24C to CV Upstream of HD23A-1-FWH24C to CV Feedwater Heater 23C A235304-23 Z-type 13 0.333 1.000 Feedwater Heater 24C CV to Feedwater Heater HD23A-2-FWH24C CV to FWH23C 23C A235304-23 Z-type 13 0.333 1.000 Feedwater Heater 23A CV to Feedwater Heater HD242A-l-FWH23ACV to FWH22A 22A A235304-23 Z-type 14 0.333 1.000 Feedwater Heater 236 CV to Feedwater Heater HD243A-l-FWH23B CV to FWH22B 22B A235304-23 Z-type 14 0.333 1.000 Feedwater Heater 23C CV to Feedwater Heater HD244A-l-FWH23CCV to FWH22C 22C A235304-23 Z-type 14 0.333 1.000 Feedwater Heater 23A to CV Upstream of HD24A-q-FWH23A to CV Feedwater Heater 22A A235304-23 Z-type 14 0.333 1.000 Feedwater Heater 238 to CV Upstream of HD25A-l-FWH23Bto CV Feedwater Heater 22B A235304-23 Z-type 14 0.333 1.000 Calculation No. 040711-02, Appendix A, Revision 0 Page 26 of 44

\

-,, ---

cs, I ECHNOLOGIES, INC. IP3 CHECWORKS Power Uprate Analysis Feedwater Heater 23C to CV Upstream of HD26A-l-FWH23C to CV Feedwater Heater 22C A235304-23 Z-type 14 0.333 1.000 HD9-1-FWH25A to HTR DRN TK Feedwater Heater 25A Drain to Heater Drain Tank 9321-F-2022-52 Z-type 12 0.333 1.000 HD9-2-FWH25B to HTR DRN TK Feedwater Heater 256 Drain to Heater Drain Tank 9321-F-2022-52 Z-type 12 0.333 1.000 HD9-3-FWH25C to HTR DRN TK Feedwater Heater 25C Drain to Heater Drain Tank 9321-F-2022-52 Z-type 12 0.333 1.000 Pre-Separator to Moisture Separator Reheater - A MS56-1-PRESEP to MSR-A (Line 1 of 2) A235308 Z-type 27 0.250 1.000 Pre-Separator to Moisture Separator Reheater - A MS56-2-PRESEP to MSR-A (Line 2 of 2) A235308 Z-type 27 0.250 1.000 MS56-3-PRESEP to MSR-A Pre-Separator to Moisture Separator Reheater - A A235308 Z-type 27 0.500 1.000 Pre-Separator to Moisture Separator Reheater -

MS56-4-PRESEP to MSR23A 23A A235308 Z-type 27 0.167 1.000 MS56-5-PRESEP to MSR-A Pre-Separator to Moisture Separator Reheater - A A235308 Z-type 27 0.333 1.000 Pre-Separator to Moisture Separator Reheater -

MS56-6-PRESEP to MSR22A 22A A235308 Z-type 27 0.167 1.000 Pre-Separator to Moisture Separator Reheater -

MS56-7-PRESEP to MSR21A 21A A235308 Z-type 27 0.167 , 1.000 Pre-Separator to Moisture Separator Reheater - B MS57-1-PRESEP to MSR-B (Line 1 of 2) A235308 Z-type 27 0.250 1.000 Pre-Separator to Moisture Separator Reheater - B MS57-2-PRESEP to MSR-B (Line 2 of 2) A235308 Z-type 27 0.250 1.000 MS57-3-PRESEP to MSR-B Pre-Separator to Moisture Separator Reheater - B A235308 Z-type 27 0.500 1.000 Pre-Separator to Moisture Separator Reheater -

MS57-4-PRESEP to MSR23B 238 A235308 Z-type 27 0.167 1.000 MS57-5-PRESEP to MSR-B Pre-Separator to Moisture Separator Reheater - B A235308 Z-type 27 0.333 1.000 Pre-Separator to Moisture Separator Reheater -

MS57-6-PRESEP to MSR22B 226 A235308 Z-type 27 0.167 1.000 Pre-Separator to Moisture Separator Reheater -

MS57-7-PRESEP to MSR21B 21B A235308 Z-type 27 0.167 1.000 Moisture Separator 21A to Header Upstream of MSD27-1-MS21A to MSDT 21A Moisture Separator Drain Tank 21A (Line 1 of 3) 9321-F-2023-31 Z-type 19 0.056 1.000 Calculation No. 040711-02, Appendix A, Revision 0 Page 21 of 44

CL I ECHNOLOGIES, INC. IP3 CHECWORKS Power Uprate h a i y s i s Moisture Separator 21A to Header Upstream of MSD27-2-MS21A to MSDT 21A Moisture Separator Drain Tank 21A (Line 2 of 3) 9321-F-2023-31 Z-type 19 0.056 1,000 Moisture Separator 21A to Header Upstream of MSD27-3-MS21A to MSDT 21A Moisture Separator Drain Tank 21A (Line 3 of 3) 9321-F-2023-31 Z-type 19 0.056 1.000 Moisture Separator 21A Header Upstream of MSD27-4-MS21A to MSDT 21A Moisture Separator Drain Tank 21A 9321-F-2023-31 Z-type 19 0.112 1.000 Moisture Separator 21A Header to Moisture MSD27-5-MS21A to MSDT 21A Separator Drain Tank 21A 9321-F-2023-31 Z-type 19 , 0.167 1.000 Moisture Separator 22A to Header Upstream of MSD28-1-MS22A to MSDT 22A Moisture Separator Drain Tank 22A (Line 1 of 3) 9321-F-2023-31 Z-type 19 0.056 1.000 Moisture Separator 22A to Header Upstream of MSD28-2-MS22A to MSDT 22A Moisture Separator Drain Tank 22A (Line 2 of 3) 9321-F-2023-31 Z-type 19 0.056 1.000 Moisture Separator 22A to Header Upstream of MSD28-3-MS22A to MSDT 22A Moisture Separator Drain Tank 22A (Line 3 of 3) 9321-F-2023-31 Z-type 19 0.056 1.000 Moisture Separator 22A Header Upstream of MSD28-4-MS22A to MSDT 22A Moisture Separator Drain Tank 22A 9321-F-2023-31 Z-type 19 0.112 1.000 Moisture Separator 22A Header to Moisture MSD28-5-MS22A to MSDT 22A Separator Drain Tank 22A 9321-F-2023-31 Z-type 19 0.167 1.000 Moisture Separator 23A to Header Upstream of MSD29-1-MS23A to MSDT 23A Moisture Separator Drain Tank 23A (Line 1 of 3) 9321-F-2023-31 Z-type 19 0.056 1.000 Moisture Separator 23A to Header Upstream of MSD29-2-MS23A to MSDT 23A Moisture Separator Drain Tank 23A (Line 2 of 3) 9321-F-2023-31 Z-type 19 0.056 1.000 Moisture Separator 23A to Header Upstream of MSD29-3-MS23A to MSDT 23A Moisture Separator Drain Tank 23A (Line 3 of 3) 9321-F-2023-31 Z-type 19 0.056 1.000 Moisture Separator 23A Header Upstream of MSD29-4-MS23A to MSDT 23A Moisture Separator Drain Tank 23A 9321-F-2023-31 Z-type 19 0.112 1.000 Moisture Separator 23A Header to Moisture MSD29-5-MS23A to MSDT 23A Separator Drain Tank 23A 9321-F-2023-31 Z-type 19 0.167 1.000 Moisture Separator 21B to Header Upstream of MSD30-1 -MS21B to MSDT 21B Moisture Separator Drain Tank 21B (Line 1 of 3) 9321-F-2023-31 Z-type 19 0.056 1.000 Moisture Separator 21B to Header Upstream of MSD30-2-MS21 B to MSDT 21B Moisture Separator Drain Tank 218 (Line 2 of 3) 9321-F-2023-31 Z-type 19 0.056 1.000

~ ~~

Calculation No. 040711-02, Appendix A, Revision 0 Page 28 of 44

- _-I

%--

Ch I ECHNOLOGIES, INC. IP3 CHECWORKS Power Uprate Analysis Moisture Separator 21B to Header Upstream of MSD30-3-MS21 B to MSDT 21B Moisture Separator Drain Tank 218 (Line 3 of 3) 9321-F-2023-31 Z-type 19 0.056 1.000 Moisture Separator 21B Header Upstream of MSD30-4-MS21 B to MSDT 21B Moisture Separator Drain Tank 21B 9321-F-2023-31 Z-type 19 0.112 1.000 Moisture Separator 21B Header to Moisture MSD30-5-MS21 B to MSDT 21B Separator Drain Tank 21B 9321-F-2023-31 Z-type 19 0.167 1.000 Moisture Separator 22B to Header Upstream of MSD31-1-MS22B to MSDT 22B Moisture Separator Drain Tank 22B (Line 1 of 3) 9321-F-2023-31 Z-type 19 0.056 1.000 Moisture Separator 228 to Header Upstream of MSD31-2-MS22B to MSDT 228 Moisture Separator Drain Tank 228 (Line 2 of 3) 9321-F-2023-31 Z-type 19 0.056 1.000 Moisture Separator 22B to Header Upstream of MSD31-3-MS22B to MSDT 22B Moisture Separator Drain Tank 22B (Line 3 of 3) 9321-F-2023-31 Z-type 19 0.056 1.000 Moisture Separator 22B Header Upstream of MSD31-4-MS22B to MSDT 22B Moisture Separator Drain Tank 228 9321-F-2023-31 Z-type 19 0.112 1.000 Moisture Separator 22B Header to Moisture MSD31-5-MS22B to MSDT 228 Separator Drain Tank 22B 9321-F-2023-31 Z-type 19 0.167 1.000 Moisture Separator 23B to Header Upstream of MSD32-1-MS23B to MSDT 23B Moisture Separator Drain Tank 238 (Line 1 of 3) 9321-F-2023-31 Z-type 19 0.056 1.000 Moisture Separator 23B to Header Upstream of MSD32-2-MS23B to MSDT 23B Moisture Separator Drain Tank 238 (Line 2 of 3) 9321-F-2023-31 Z-type 19 0.056 1.000 Moisture Separator 238 to Header Upstream of MSD32-3-MS23B to MSDT 23B Moisture Separator Drain Tank 238 (Line 3 of 3) 9321-F-2023-31 Z-type 19 0.056 1.000 Moisture Separator 238 Header Upstream of MSD32-4-MS23B to MSDT 238 Moisture Separator Drain Tank 23B 9321-F-2023-31 Z-type 19 0.112 1.000 Moisture Separator 238 Header to Moisture MSD32-5-MS23B to MSDT 238 Separator Drain Tank 23B 9321-F-2023-31 Z-type 19 0.167 1.000 Moisture Separator Drain Tank 21A to Heater MSD33A-1-MSDT 21A to HDT Drain Tank 9321-F-2023-31 Z-type 19 0.167 1.000 Moisture Separator Drain Tank 22A to Heater MSD34A-1-MSDT 22A to HDT Drain Tank 9321-F-2023-31 Z-type 19 0.167 1.000 Moisture Separator Drain Tank 23A to Heater MSD35A-1-MSDT 23A to HDT Drain Tank 9321-F-2023-31 Z-type 19 0.167 1.000 Calculation No. 040711-02, Appendix A, Revision 0 Page 29 of 44

QI r-I ECHNOLOGIES, INC.

h

--~

IP3 CHECWORKS Power Uprate Analysis Moisture Separator Drain Tank 21B to Heater MSD36A-1-MSDT 21B to HDT Drain Tank 9321-F-2023-31 Z-type 19 0.167 1.000 Moisture Separator Drain Tank 22B to Heater MSD37A-1-MSDT 22B to HDT Drain Tank 9321-F-2023-31 Z-type 19 0.167 1.000 Moisture Separator Drain Tank 23B to Heater MSD38A-1-MSDT 23B to HDT Drain Tank 9321-F-2023-31 Z-type 19 0.167 1.000 Moisture Separator Reheater 21A to Moisture MSD39-I-RHTR21A to RHDT 21A Separator Reheater Drain Tank 21A 9321-F-2023-31 HBD 20 0.167 1.000 Moisture Separator Reheater 22A to Moisture MSD40-1-RHTR 22A to RHDT 22A Separator Reheater Drain Tank 22A 9321-F-2023-31 HBD 20 0.167 1.000 Moisture Separator Reheater 23A to Moisture MSD41-1-RHTR 23A to RHDT 23A Separator Reheater Drain Tank 23A 9321-F-2023-31 HBD 20 0.167 1.000 Moisture Separator Reheater 21B to Moisture MSD42-1-RHTR 21B to RHDT 21B Separator Reheater Drain Tank 218 9321-F-2023-31 HBD 20 0.167 1.000 Moisture Separator Reheater 228 to Moisture MSD43-1-RHTR 228 to RHDT 228 Separator Reheater Drain Tank 22B 9321-F-2023-31 HBD 20 0.167 1.000 Moisture Separator Reheater 238 to Moisture MSD44-1-RHTR 23B to RHDT 238 Separator Reheater Drain Tank 238 9321-F-2023-31 HBD 20 0.167 1.000 Reheater Drain Tank 21A to CV Upstream of MSD45A-I-RHDT21Ato CV Feedwater Heater 26 Header 9321-F-2023-31 HBD 20 0.167 1.000 Reheater Drain Tank 21A CV to Feedwater Heater MSD45B-l-RHDT21ACV to FWH26 26 Header 9321-F-2023-31 HBD 20 0.167 1.000 Reheater Drain Tank A Header to Feedwater Heater 26 Between RHDT 22A and RHDT 23A MSD45C-1-RHDT A HDR to FWH26 Connections 9321-F-2023-31 HBD 20 0.167 1.000 Reheater Drain Tank A Header to Feedwater Heater 26 Between RHDT 23A and RHDT 21A MSD45C-2-RHDT A HDR to FWH26 Connections 9321-F-2023-31 HBD 20 0.333 1.000 Reheater Drain Tank A Header to Feedwater Heater 26 Between RHDT 21A Connection and MSD45C-3-RHDT A HDR to FWH26 Feedwater Heater 26C Tee 9321-F-2023-31 HBD 20 0.500 1.000 Reheater Drain Tank A Header to Feedwater

.. - . -

MSD45C-4-RHDT A HDR to FWH26C Heater 26C 9321-F-2023-31 HBD 20 0.167 1.000 Calculation No. 040711-02, Appendix A, Revision 0 Page 30 of 44

-

-. -\

c& I ECHNOLOGIES, /NC. IP3 CHECWORKS Power Uprate Analysis Reheater Drain Tank A Header Between Feedwater Heater 26C Tee and Feedwater Heater MSD45C-5-RHDT A HDR to FWH26 26B Tee 9321-F-2023-31 HBD 20 0.333 1.000 Reheater Drain Tank A Header to Feedwater MSD45D-1-RHDT A HDR to FWH26B Heater 26B 9321-F-2023-31 HBD 20 0.167 1.000 Reheater Drain Tank A Header to Feedwater MSD45D-2-RHDT A HDR to FWH26A Heater 26A 9321-F-2023-31 HBD 20 0.167 1.000 Reheater Drain Tank 22A to CV Upstream of MSD46A-l-RHDT22A to CV Feedwater Heater 26 Header 9321-F-2023-31 HBD 20 0.167 1.000 Reheater Drain Tank 22A CV to Feedwater Heater

...-

MSD46A-2-RHDT22A CV to FWH26 26 Header 9321-F-2023-31 HBD 20 0.167 1.000 Reheater Drain Tank 23A to CV UDstream of MSD47-1 -RHDT23A to CV Feedwater Heater 26 Header 9321-F-2023-31 HBD 20 0.167 1.000 Reheater Drain Tank 23A CV to Feedwater Heater MSD47-2-RHDT23A CV to FWH26 26 Header 9321-F-2023-31 HBD 20 0.167 1.000 Reheater Drain Tank 21B to CV Upstream of MSD48A-1 -RHDT21B to CV Feedwater Heater 26 Header 9321-F-2023-31 HBD 20 0.167 1.000 Reheater Drain Tank 21B CV to Feedwater Heater MSD48B-1-RHDT21 B CV to FWH26 26 Header 9321-F-2023-31 HBD 20 0.167 1.000 Reheater Drain Tank B Header to Feedwater Heater 26 Between RHDT 23B and RHDT 228 MSD48B-2-RHDT B HDR to FWH26 Connections 9321-F-2023-31 HBD 20 0.333 1.000 Reheater Drain Tank 228 to CV Upstream of MSD49A-l-RHDT22B to CV Feedwater Heater 26 Header 9321-F-2023-31 HBD 20 0.167 1.000 Reheater Drain Tank 22B CV to Feedwater Heater MSD49B-1 -RHDT22B CV to FWH26 26 Header 9321-F-2023-31 HBD 20 0.167 1.000 Reheater Drain Tank B Header to Feedwater Heater 26 Between RHDT Connections and MSD49C-I-RHDT B HDR to FWH26 Feedwater Heater 26 Connections 9321-F-2023-31 HBD 20 0.500 1.000 Reheater Drain Tank B Header to Feedwater MSD49C-2-RHDT B HDR to FWH26C Heater 26C 9321-F-2023-31 HBD 20 0.167 1.000 Calculation No. 040711-02, Appendix A, Revision 0 Page 31 of 44

CSI I ECHNOLOGIES, INC. IP3 CHECWORKS Power Uprate Analysis Reheater Drain Tank B Header Between Feedwater Heater 26C Tee and Feedwater Heater MSD49C-3-RHDT B HDR 26B Tee 9321-F-2023-31 HBD 20 0.333 1.000 Reheater Drain Tank B Header to Feedwater MSD49C-4-RHDT B HDR to FWH26B Heater 26B 9321-F-2023-31 HBD 20 0.167 1.000 Reheater Drain Tank B Header to Feedwater MSD49C-5-RHDT B HDR to FWH26A Heater 26A 9321-F-2023-31 HBD 20 0.167 1.000 Reheater Drain Tank 236 to CV Upstream of MSD50A-l-RHDT23B to CV Feedwater Heater 26 Header 9321-F-2023-31 HBD 20 0.167 1.000 Reheater Drain Tank 238 CV to Feedwater Heater MSDSOC-1-RHDT23B CV to FWH26 26 Header 9321-F-2023-31 HBD 20 0.167 1.000 Steam Generator Blowdown from SG 21 to SG SG51-1-CONT PEN to SGBFTK Blowdown Flash Tank 9321-F-272%66 Z-type 30 0.250 1.000 Steam Generator Blowdown from SG 22 to SG SG52-1 -CONT PEN to SGBFTK Blowdown Flash Tank 9321-F-2729-66 Z-type 30 0.250 1.000 Steam Generator Blowdown from S G 23 to SG SG53-1-CONT PEN to SGBFTK Blowdown Flash Tank 9321-F-2729-66 Z-type 30 0.250 1.000 Steam Generator Blowdown from SG 24 to SG SG54-1-CONT PEN to SGBFTK Blowdown Flash Tank 9321-F-2729-66 Z-type 30 0.250 1.000 Calculation No. 040711-02, Appendix A, Revision 0 Page 32 of 44

cs/ TECHNOLOGIES, /NC. IP2 CHECWORKS Power Uprate Analysis Appendix B Component Level Wear Rate Changes due to SPU Calculation No. 040711-02, Appendix B,Revision 0 Page 33 of 44 I

___ ---

Ip2 CHECWORKS Power Uplrte Analysis CSI TECHNOLOGIES,INC.

Page 34 of 44 Calculation No. 040711-02, Appendix B, Revision 0

CSI TECHNOLOGIES, INC. IP2 CHECWORKS Power Uprate Analysis Appendix C Steam Cycle Level Wear Rate Changes due to SPU Calculation No. 040711-02, Appendix C, Revision 0 Page 35 of 44

CSI TECHNOLOGIES. INC. Ipz CHECWORKS Power Upnte Analysis c

3 CD Connecbm from HDP to EFP 53 14% 14% 14% 01 47 3773 37 , 0, 0' 1396 53%

FWH 5 to Connecbon from 4 CD ,HDP 36 0 5% 0 5% 0 5% 0 02 53 3824, 2 9, 0; 01 977 35%

5 CD FWH4toFWH5 44 2 8% 2 9% 2 8% 02 62 3005 2 2: 0; 0 977 35%

6 CD ,FWH3toFWH4 36 4 1%' 4 1% 4 1% 02 61 2555 18' 0: oi 9 n 35%

7 ' CD FWH2bFWH3 94 3 6% 4 6% 3 5% 03 87 2043 17 0, 0, 977 35%

HDP Outlet to CD System 9 HD ,Connochon 44 7 1% 10 3 1 , 6 9% 06 89 3654 6 5, 0, 01 419 97%

10 HD HDTanktoHDP 15 10 3% 10 4% 10 3% 04 47 3848 6 5: 0: 0 419 97%

-0 5%

~

11 HD FWH6toHDTank 36 -4 1% -4 0% -4 1% -0 2 37 3893 41 Oi 0, 1 72 12 ' HD FWH5toHDTank 54 2 3% 2 3%, 2 3X: 01 37 3863 33 0; 0: 108 57%

13 , HD FWH4toFWH3 89 5 5%, 5 7% 4 4% 03 51 2633 2 3, O/ oi 046 41%

14 ' HD 'FWH3toFWH2 147 58% 5 8%' 4 7% 02 31 2120 21 0 0, 099 35%

-08; 080

~

17 EX ,HPExtrachontoFWH6 75 7 0% 10 3%- 0 0% 01 20 4448, 70 920 153%

18 EX ,PresepExtracbontoFWH5 17 17 1% 18 7% 15 4% 08 53 3879 34 922 -07' 108 57%

Unlike Unlt 3, does not represent condibon in both Molsture Separator and Monture Preseparator Drains as Moisture Preseparator Drains are not modeled in Moist Separator Drains to HD CHECWORKS Greatest percent change 19 MSD Tank 248 46 3% 66 7% 33 3% 05 15 3853 44 0 0 106 43 2% increase in wear rate (post-SPU)

Greatest percent change decrease in wear 20 RHD .Reheater Drain to FWH 6 530 -69% -3 6% 7 0% -0 3 44 4959 62 33, -05, 0 92 -1 1 1% rate (post-SPU)

Locabon contains superheated steam both 21 , EX LPExtmcbontoFWH4 97 0 0% 0 0% 0 0% 00 00 3338, 83 1000 0 046 41%preandpostSPU 22 , EX LPExhacbontoFWH3 81 -62% 11% -7 3% -0 7 103 261 1 21 923 08 053 30%

27 , MS ,HP Turbine to Moist Separator 77 15 4% 16 1% 1 3 2 3 6 ~ 0 01 01 3879 34 9051 -081 1217 62%

Wear rate change due to slight chemisw 30 SG SteamGenElowdown 117 0 4% 0 5% 0 3% 0 01 16 5114 0 0, 0 0 05 0 0% change, not power uprate NO components from this i d o n in me 8 CD FWHltoFWH2 0 - - - - - 1580 26 0 0 g 77 3 5% CHECWORKS model No Components from this locabon in the 15 HD ,FWHZtoFWHl 0 - - - - - 1693 26 0 0 150 2 7% CHECWORKS model No Components from this locabon in the 16 HD ,FWH 1 tocondenser 0 - - - - - 1000 7 6: 0 0 2 21 0 5% CHECWORKS model No Components from this location in the 23 EX LPExtractiWtoFWH2 0 - - - - - 2115 18, 780, 04, 0 51 12% CHECWORKS model No Components from this locatton in the 24 , EX LPExtracbontoFWH1 0 - - - - - 1684 19 753 -14 0 70 -3 8% CHECWORKS model No Components from this locabon in the 25 , MS SteamGentoHPTurbine 0 - - - - - 4978 -08, 997 0 12 97 6 7% CHECWORKS model No Components from this locabon in the c 26 , MS steam Gen to Reheater Shell 0 - - - - - 5089 13 999 0 0 92 -11 1% CHECWORKS model Cakulrtion No. 040711-02, Appendix C, Revision 0 Page 36 of 42

CSI TECHNOLOGIES,IWC. IF2 CHECWORKS Power Uprate Anslyru 1

- -

28 , MS MoistSeparatortoReheater 0 - - - - - - - 1- -I -

mcawn (roes nor contain

- components pipiiiy NOComponents from this locabon in the 29 MS ReheatertoLPExbactron 0 - - - - - 4851 101 1000 0, 9 87 3 4% CHECWORKS model No Components from this locabon in the 31 MS Boiler Feed Pump Turbine Drain 0 - - - - - 101 7 , 0 868 02 0 15 10 7% CHECWORKS model Cakulstiion No. 04071142, Appendix C ,Revision 0 Page 37 of 42

r

_______________

a I I n I w P

9 I

I I

I I

I I

I I

I I

I'

- I I

-w

" .. l - +

I I

I b I

'I w -I Im I

I I

I I

I

-I 0

.I IIJ Ii

!I 19 If A, I c i I n t I .

l I N I

CSI TECHNOLOGIES, INC. IP2 CHECWORKS Power Uprate Analysis Attachment A Referenced Correspondence and Communications Calculation No. 040711-02, Attachment A, Revision 0 Page 39 of 44

CSI TECHNOLOGIES, INC. IP2 CHECWORKS Power Uprate Analysis Reference 7.3.1 Email from Harry Hartjen (IP)to Daniel R. Poe (CSITechnologies), dated 10/12/2004, regarding SPU implementation dates, CSI Doc. No. 04071111.

Dan:

Following are input to the DDIR:

No.8A SPU operations are expected to be implemented on 11/22/2004at IP2.

No. 9A SPU operations are expected to be implemented on 04/08/2005at IP3.

Harry Hartjen (914) 736-8356 Calculation No. 040711-02, Attachment A, Revision 0 Page 40 of 44 I_

-

I

CsI TECHNOLOGIES, INC. IP2 CHECWORKS Power Uprate Analysis Reference 7.3.2 S

Email from Harry Hartjen (IP)to Daniel R. Poe (CSITechnologies), dated 10/18/2004, regarding operational and configuration changes due to SPU, CSI Doc. No. 04071113.

Dan:

Attached is the response from our engineer, the Shaw Group to your request for a listing of all operational and/or configuration changes due to the SPU.

Note that there are no operational and/or configuration changes to IPZIP3 due to the SPU.

This information was required for DDIR No. 6 and 7.

Per your request, a hard copy set of IP3 FAC isometric drawings will be mailed today to your attention.

Note that drawing EC-H-50077 is not yet revised in this mailing. For correct piping configuration, see the information I sent you for the 3R12 Checworks update.

Included in this mailing is:

1. Flow Accelerated Corrosion Program Checworks Analysis Enhancement, Technical Report No. 00130-TR-001 Revision 0 Volume 1 of 1. December, 2000. This report serves as the Unit 2 Checworks Model documentation. In addition to the Checworks Wear Rate Analysis Run Definitions listed in Table 1.O, there are the following two additions:
a. X-under w/exp joints
b. CND FWH 22 to FWH 23; this was added due to the SPU.

This information was required for DDIR No. 4 Harry Hartjen x ~-xI-L.-- --- I_-"-^

From: Cunningham, Glenn [1]

Sent: Thursday, October 07,2004 2:45 PM To: Hartjen, Harry Cc: Scanlon, Michael; Chakrabarti, Syamal

Subject:

RE: RAI FAC-1 Harry, There are no operational changes for Steam Plant Systems, such as use of additional trains or use of bypass lines not currently in operation, associated with uprate of IP2 / IP3. Changes due to the uprate which affect FAC are primarily changes in system flowrates and temperatures, which are documented in the Heat Balance calculations. Impact of these changes on piping velocities and temperatures is addressed in the applicable sections of the BOP Uprate Engineering Reports, forwarded to Entergy for review.

Modifications associated with the uprate of IP2 / IP3 include: (1) MSR internal moisture separation system replacement, (2) HP Turbine rotor replacement, (3) HP turbine interstage drain piping modification, and (4)

Relocation of HP turbine 1 stage pressure taps.

Regards, Glenn C.

%em*m*

,

1 From: Hartjen, Harry [2]

Sent: Wednesday, October 06,2004 1:30 PM Calculation No. 040711-02, Attachment A, Revision 0 Page 41 of 44

CSI TECHNOLOGIES, INC. IP2 CHECWORKS Power Uprate Analysis To: Cunningham, Glenn

Subject:

RE:RAI FAC-1 Glen:

This is to confirm that the wear rate comparison will be completed by 12/31/2004.

A s I mentioned to you we are having a contractor perform the update to our Checworks models. One of the inputs I have to provide to them is:

The listing of all Unit 2 and Unit 3 operational and/or configuration changes due to the power uprate (i.e.,

additional trains in operation, bypass lines operated at full power, etc.)

Do you have this information or can you direct my question to someone who has this information.

Thanks, Harry Hartjen (914) 736-8356 From: Cunningham, Glenn [3]

Sent: Wednesday, October 06, 2004 1250 P M To: Hartjen, Harry

Subject:

FW: RAI FAC-1 y_^*-

From: Cunningham, Glenn Sent: Wednesday, October 06,2004 1241 P M To: 'hartje@entergy.com'

Subject:

RAI FAC-1

Harry, As I indicated, I made a few edits in your input. Please confirm that the wear rate comparison discussed in the last sentence will be completed by the end of 2004.
Thanks,
                                  • Internet Email Confidentiality Footer******************

Privileged/Confidential Information may be contained in this message.

If you are not the addressee indicated in this message (or responsible for delivery of the message to such person), you may not copy or deliver this message to anyone. In such case, you should destroy this message and notify the sender by reply email. Please advise immediately if you or your employer do not consent to Internet email for messages of this kind. Opinions, conclusions and other information in this message that do not relate to the official business of The Shaw Group Inc. or its subsidiaries shall be understood as neither given nor endorsed by it.

The Shaw Group Inc.

http:/hnrww.shawgrp.com Calculation No. 040711-02, Attachment A, Revision 0 Page 42 of 44

_-.

CSI TECHNOLOGIES, INC. IP2 CHECWORKS Power Uprate Analysis

- I

- - - _ _ l

_ -l ll l-~ __ _

l__ _

( *****************Internet Email Confidentiality Foofer--*****-***

PrivilegedIConfidentialInformation may be contained in this message.

If you are not the addressee indicated in this message (or responsible for delivery of the message to such person), you may not copy or deliver this message to anyone. In such case, you should destroy this message and notify the sender by reply email. Please advise immediately if you or your employer do not consent to Internet email for messages of this kind. Opinions, conclusions and other information in this message that do not relate to the official business of The Shaw Group Inc. or its subsidiaries shall be understood as neither given nor endorsed by it.

The Shaw Group Inc.

http://w.shawgrp.com c

Calculation No. 040711-02, Attachment A, Revision 0 Page 43 of 44

-_ II-

cs/ TECHNOLOGIES, /NC. IP2 CHECWORKS Power Uprate Analysis Reference 7.3.3 Email from Ron Macina (IP3)to Brian Trudeau (CSITechnologies), dated 1/10/2005, regarding addition a1 Heat Balance Diagrams and uprate start dates, CSI Doc. No 04071 140.

CSI Doc.No. 04071 140 Brian, Please find below the requested Heat BalancesAnformation.

1) IP3 Pre and Post Appendix K Heat Balances

<<IP3TuningR4 Pre Appendix K (3037 NSS Pwr).pdf>> 4P3TuningR4 Post Apendix K.pdf>>

2)IP2 Pre Appendix K Heat Balance

<<IP2-TuningR4 Pre Appendix K (3090 NSS Pwr).pdP>

3) Start Date for IP3 Appendix K uprate was 12/22/02
4) Start Date for IP2 Appendix K uprate was 5/23/03 Thanks, Ron Macina 914-736-8363

0rig i nal Message-----

From: Brian Trudeau [4]

Sent: Tuesday, December 28,2004 3:52 P M To: Macina, Ron

Subject:

Request for Additional HBDs CSI Doc. No.04071137 Ron, Based on our discussion earlier today, it appears that we have some additional work to do on this end. We plan on modeling all three power levels (100% - the Original power level, -101.4% - the Appendix K power

-

level,and 105% the SPU power level) for each unit. In doing this, we will capture plant conditions at each point in history.

To complete this task, we will need the following input:

(1) Unit 3 PEPSE or HBD at -101.4% power (the Appendix K power uprate)

(2)Unit 2 Original PEPSE or HBD at 100% power (pre-Appendix K)

(3) Unit 3 Start date for the Appendix K Uprate (4) Unit 2 Start date for the Appendix K Uprate Please let me know if you have any questions. We will be contacting Harry to inform him of this plan.

Thank you, Brian Trudeau CSI Technologies, Inc.

(847) 836-3000 ext. 717 www.csitechnologies.com i

Calculation No. 040711-02, Attachment A, Revision 0 Page 44 of 44

"."

CALCULATION APPROVAL COVER SHEET Document Title: Indian Point Unit 2 CHECWORKS Power Uprate Analysis Document No.: 04071 1-02 Client: Entergy Nuclear Northeast Project No.: 040711 This calculation has been prepared in accordance with Section 4.3 of the CSI Quality System Manual, Revision 2.

Revision: 0 Prepared: Date:

Verified: Date:

Approved: Date: 3/2Yk5i--

Daniel R. Poe (CSI)

Approved: Date:

Revision Control Sheet Project Number: 040711 Project Name: Indian Point Units 2 & 3 CHECWORKS Power Uprate Analysis Client: Entergy Nuclear Northeast Document Title: Indian Point Unit 2 CHECWORKS Power Uprate Analysis Document Number: 040711-02 Rev. Date Description of Changes CSI Client Approval Approval 0 3/23/05 Initial Issue - For Use