ML100850406

From kanterella
Jump to navigation Jump to search
Calculation LE-0113, Rev. 0, Reactor Core Thermal Power Uncertainty Calculation Unit 1
ML100850406
Person / Time
Site: Limerick Constellation icon.png
Issue date: 12/02/2009
From:
Exelon Generation Co
To:
Office of Nuclear Reactor Regulation
Shared Package
ML100850379 List:
References
LE-0113, Rev 0
Download: ML100850406 (94)


Text

ATTACHMENT 11 Exelon Generation Company, LLC Calculation LE-0113, Rev. 0, Reactor Core Thermal Power Uncertainty Calculation Unit 1.

Exel n Design Analysis Major Revision Cover Sheet Design Analysis (Major Revision)

Last Page No. 93 Analysis No.:

l LE-0113 Revision:

0

Title:

Reactor Core Thermal Power Uncertainty Calculation Unit 1 ECIECR No.:

LG 09-00096 Revision:

001 1

Station(s):

Limerick Component(s):

Unit No.:

1 N/A

Discipline:

LEDE Code/Keyword:

Descrip.

N/A

Safety/QA Class:

N 006. 041. 042, 043. 044.

System Code:

047 zz Structure:

N/A

CONTROLLED DOCUMENT REFERENCES Document No.:

From/To Document No.:

From/To LEAE-MUR-0001, Bounding Uncertainty fm L-S-11, Rev. 15, DBD Feedwater System From Analysis for Thermal Power determination LM-0552, Rev 7, Reactor Heat Balance From L S-15, Rev. 10, DBD Control Rod Drive From Calculation Limerick Units 1 & 2 System LM-553. Rev. 0. Determination of the RPV Heat L-S-19. Rev. 10. DBD Recirculation System From From Loss LE-Ol 16, Reactor Dome Narrow Range From L-S-36, Rev. 10, DBD Reactor Water Makeup From Pressure Measurement Uncertainty System LM-562, Rev 2. CRD Flow Rates and System Limerick PEPSE M MUR PU and EPU Heat From From Pressures Balances

- Eval 2009-03880 RO LEAM-MUR-0038 (Reactor Heat Balance>

To LEAM-MUR0046 (TS Instrument Setpoints)

To LEAM-MUR-0041 (Neutron Monitoring System LEAM MUR 0039 (Power Flow Map>

To w RBM)

LEAM-MUR-0048 (Generic Disposition Applicability Confirmationi Is this Design Analysis Safeguards Information?

Yes [1 No If yes, see SY-AA-101 -1 06 Does this Design Analysis contain Unverified Yes [1 No If yes Assumptions?

1 ATl/AR This_Design Analysis SUPERCEDES:

NA In its entirety.

Description of Revision (Ii.t ffct I p.igt fr at s lt I su t.

(

ri.,,.,- i 1

r a

(

t t

sa r F JF/

ver i I

r

tz

.4 thod of Review Detailed Review [.l Alternate Calculations (attached) [

Testing [ i eever

LGS Caic No. LEO113 Exe fl Reactor Core Thermal Power Uncertainty Revision 0 Review Notes:

Independent review Peer review El Performed a line by line review of the calculation, All comments were incorporated to the satisfaction of the reviewer. This calculation supports the margin uncertainty recapture (MURI by reducing the measurement uncertainty uf Core Thermal Power from 2 % to 0.35.

For External Analyses Only External Approver:

Richard Brusato Print Name SigriName Date Exelon Reviewer:

MitulAjmera Print Name SignName Date Independent 3i Party Review Required?

Yes No fl Exelon Approver:

Raymond George I (

Print Name Sign Name Date

2.1 2.2 2.4 2.5 3.0 4.0 K

'TU S.

4.3 4.5 K

F 4,

4.9 7.1 i

7"93 7.4 7,6 7.7 Reactor Core Thermal Power Uncertainty Calculation Unit 1 Revision 0

.s

.............................i1tkllY "KK*YkpYtiY..........c.ft.............................s......s..

..............w.......................................

CONFIGURATION......h-....................w......>>............- -............-.1........ 5 IS...........................................F......-.Ff.........K" fY.b.c....k.....x..............c....t....:.x..............t ".....kK+Yw....

INPUTS..................................................................

...1....... Y.......R

.....,.-.........e........................x......xt-UNCERTAINTY..........a.................'........................8 REACTOR CLEANUP SYSTEM TEMPERATURE

...lwwv. as.. "KN.RkKK....sKK....Yk.hkhi".cNk"TY.4...sw>>.-kw.w..,....... "...s..r.. 12

  • E UNCERTAINTY Y.w.w......s....a........s.sw.. ".F.....rt1k.NKt....K.)...".f1. aklPC1[##f..# "31.1.a.... "x...Yrrxx...e........... 1 RECIRCULATION PUMP MOTOR UNCERTAINTY.....................

........w................... ".T4................fbf.........Y 1

.KFb..>>VRF##M..h.w. " h.A...rtxh-s..wwx..>>e.w...a-#... "."F. ".wx yyyy">>.YV4. ".Pt9llcK.3EYYAFK..Y..................c EFERENCES........................

.'..1P

........iK.K.....o....h.......f3f.A1fK...Y......tkf..."##s....w........s.ansa....xa...w.......axaf4...........w........

METHODOLOGY...................................

Y.

.... "....f...A.bfi...F....kfk.....>>.........................x................

.................24 PROCEDURES

.................................................................... drtv.....t...................................

.........s-... 4 NT.............as.................. "...-................N.......h"..isRilIHi..............................--.......24

., CODES,

....'s.......a..........v.......xs.........f............ "........

a............. 4

............K..Y.k..........cii........................................................

... "4-.bf........-....24 GENERAL ELECTRIC (GE) DOCUMENTS............f.......>....s....x.......w........a.'.........

...a.k...a............................ 25 CALCULATION and engine n

` ;

REFERENCES....x ".....e:......s.,.......................e.....r>>.....sw,a....

.......I........ 1Fr 771 S

.....................N ifik w.

a

.ation ANALYSIS....KKa. N iJ+ki..................

.-.. Nf

.IKe.lf

. e R"

.. e`..¢ w Methodology...t.... "..........rt................-..............

LE OF CONTENTS

~

...........n....

... a+....yF-.bx...

....e.r.........#w-................w..-.,.................

..25 6

......rtsa.ww.a..w:w.................wxk...... :........ -.k'-.............Y... -

CALCULATION:..

..................s.....................

..a1...2F

....................'(B..AF4il..YiY

.. a Pkt..

........- ".besw. "..s

............tx............................- eao-.-Y35 dwater Flaw Uncertainty.....e..............T.......k.-........iiifbif.kic..fTea1....R.as.rxs...rn.n..............---....................x3 Steam Dome Pressure Measurement Uncertain Reactor Water Clean-up (R CU Pew Loop Uncertainty...

U Temperature Loop Uncertainty....a.......YF1aK.wf.Klff..KKi.eEw.R..1.rt.......................4....

.............. ua........... 42 CRD Flow Rate Uncertainty................,.......................... F. il p HEAT UNCERTAIN Page 3 of 93 i.>n.....................................BNrt-...............4....................27 LE-0113

....................................................25

. F. b....... -..x#*....... oa..'LF

.................. -b.... -----.27

.,....................................................3

.........ins..a-www...... "............................... 35

...7....re"T....a. vra.n x

...................... 45

f..r

....................w..........

..... 51

.........FA..

Determination P Urortainty........................... n................-.-..h...4.................a............v.syr>>n.x. ".2st...t Total CT7 Uncertainty Ca ulation.........

.........Va..Nf..i...Kk.+l.a..9ktteF.i.6bffi$c :.,.a.erx....s..............................-.....6 N............a,ee.arezree..

.........ssss.a..a..a

..a..............ttY4L4aN1..tYy.F4ib.94.#.....-.-..........w.v.........

............. aw........ 64

..wxx..w.w........t "...............................

.a>>. a--.y-F....f.....fff..kK............ axe....-.............-n........I....................

Exel()n.

Reactor Core Thermar Power Uncertainty Calculation Unit 1 LE-0113 Revision 0 1.0 1.1 2.0 2.1 22 2.3 2.4 2.5 3.0 4.0

..t1 4.2 4.3 4.4 4.5 4.6 4]

4.8 4.9 5.0 a.o 6.1 8.2 7.0 7.1 7.2 7.3 7.4 7,5 7.6 7.7 7.8 B.O 9.0 TABLE OF CONTENTS FUNCTIONAL DESCRIPTION AND CONFlGUAATION ou uH ******************* HO H

H 5

INPUTS *.*......u H

5 REACTOR WATER ClEANUP (AWCU) FLOW LOOP UNCERTAINTY 8

REACTOR CLEANUP SYSTEM TEMPERATURE 12 CAD ROW RATE UNCERTAiNTY 15 RECIRCULATION PUMP MOTOR UNCERTAINTY mHO.m m *** ou................. *** m. 19 ASSUMPT10NS AND UMITATIONS H H

0

.,.22 DESIGN BASIS OOCUMENTS u

    • ~

H

~~.u **u **..,

24 SPECIFrCATIONSt CODES~ & STANDARDS u

m 24 LIMERICK STATION DRAWiNGS n

n.'

n H

24 GENERAL ELECTRIC (GE) DOCUMENTS H

hn n

25 CALCULATIONS and engineering analysis moo H ***h."

25 IDENTIFICATION OF COMPUTER PROORAMS 27 Feedwster Flow Uncertainty 35 Steam Dome Pressure Measurement UnoertaintyHHO H

35 Reactor Water Clean--up (RWCU) Flow Loop Uncertainty H ******m S5 RWCU Tel'11perature loop Uncertainty

..,*.** H H~****n 42 CRD Flow Rate Uncertainty H

+

~

45 RecircUlation Pump HEAT UNCERTAINTY 00+

51 Determination of CTP Uncertainty

"'MH' 55 Total CTP Ureenainly Ca.tcufation H

61 C*ONCLUSIONS H""

  • M' H

64 Page 3 of 93 Exel()n.

Reactor Core Thermar Power Uncertainty Calculation Unit 1 LE-0113 Revision 0 1.0 1.1 2.0 2.1 22 2.3 2.4 2.5 3.0 4.0

..t1 4.2 4.3 4.4 4.5 4.6 4]

4.8 4.9 5.0 a.o 6.1 8.2 7.0 7.1 7.2 7.3 7.4 7,5 7.6 7.7 7.8 B.O 9.0 TABLE OF CONTENTS FUNCTIONAL DESCRIPTION AND CONFlGUAATION ou uH ******************* HO H

H 5

INPUTS *.*......u H

5 REACTOR WATER ClEANUP (AWCU) FLOW LOOP UNCERTAINTY 8

REACTOR CLEANUP SYSTEM TEMPERATURE 12 CAD ROW RATE UNCERTAiNTY 15 RECIRCULATION PUMP MOTOR UNCERTAINTY mHO.m m *** ou................. *** m. 19 ASSUMPT10NS AND UMITATIONS H H

0

.,.22 DESIGN BASIS OOCUMENTS u

    • ~

H

~~.u **u **..,

24 SPECIFrCATIONSt CODES~ & STANDARDS u

m 24 LIMERICK STATION DRAWiNGS n

n.'

n H

24 GENERAL ELECTRIC (GE) DOCUMENTS H

hn n

25 CALCULATIONS and engineering analysis moo H ***h."

25 IDENTIFICATION OF COMPUTER PROORAMS 27 Feedwster Flow Uncertainty 35 Steam Dome Pressure Measurement UnoertaintyHHO H

35 Reactor Water Clean--up (RWCU) Flow Loop Uncertainty H ******m S5 RWCU Tel'11perature loop Uncertainty

..,*.** H H~****n 42 CRD Flow Rate Uncertainty H

+

~

45 RecircUlation Pump HEAT UNCERTAINTY 00+

51 Determination of CTP Uncertainty

"'MH' 55 Total CTP Ureenainly Ca.tcufation H

61 C*ONCLUSIONS H""

  • M' H

64 Page 3 of 93

Exelon, Reactor Core Thermal Power Uncertainty Calculation Unit 1 Revision Q TABLE F TABLES LE-0113 TaNe 2-1 " Design I rpxts...................s..........................

...................... -..........- k.........x......s.....F.................................... 6 Table 2-2. RWCU System Inlet Flow le nit I...............2..............a............s..................................P....+.....-..8 Tab.'.' 2

. RWCU System Wt Fl Differential Pressure Transmitter..............

...........2i......+..N.s..M'4.......-f".....---...9 Table

. FIWCUI System PPC Precision Signal Resistor Unit................NA.......A..s1A........k.............--...............s... 10 Table 2-6. R System PPC Analog Input Card Unit I...+......+.....i

..........x.....-.:...s......

..........................-ka". 11 Table -6. RWCU System LocV Service

~1FentR7N~...NA"ss...+....x....s.w..as...s..s.....x.a.. " a.....YYFdS.+FF..6ftk.S....NSi.........--A Tables. 2-7. RWCU System Inlet Thermocouple s......w.....x.....s......s....wsx.......+1..YF...fFaPk..k-.F.1l..ki5 "..kkY...............................-....12 Table 2-8. RWyste.O!u't Themkxmple......-..........Y...".FF.............illfN.a~1..k."i4fr........v.......s...cr.-.........-.............13 Table 2-9. RWCU System The Analog Input Ca Unit 1...ss......s.............s...ss.............. --.-......... 14 Table 2-mrocouple Local Service Environments...........s................."V>.ai...........................1 4 Tabfe 2-1 Table 2-1 Table 2-13. CRC Hydraulic System PPC Precision Sign Resistor Unit.............................,.........................."......17 I'" Ioraulic System Fl Element - Unk

.......aF....W1.iki*...r.........

CRD Hydraulic System Differential Pressure T Wor.........F...............--....................................... 1 Table 2-'14. CRD Hyd ulic y em PPC Analog Input Card Unit 1..............

Table,2-16.

ecirculation Pump Motor Waft Transducer.........................'......................k........+.............."...

Page 4 of 93 Table 2-1 S. CRD Hydraulic System Local Service Environments..............s....F......F............."..k.a..............+3a............

.........s 19 Table -17N Recirculation Pump Motor Watt Transducer PPC Precision Signal Resistor Unit................................ 20 Vale 2-18. Recirculat Pump Motor Watt Tt r ducer PPC Analog Input Card unlit 1............................. -......-.20 Table A7-1.

Sensltivity Analysis...............x..................................

......"..s.+f..F...-..-.........--

........... 76 Table

. Relationship, between a,, Y(aj ), ', and f /...............s.........Fl...k......+ "..UYfik4..bk........-i............................ 79 Exelon.

Reactor Core Thermal Power Uncertainty Calculation Unit 1 TABU! OF TABLES LE-0113 Revision 0 Table 2-2. RWCU System In~t Flow Eler'l\\ent - Unit 1 u

~

n 8

Table 2-3. RWCU System Inlet Flow Differential Pressure Transmitter mw u

..,w 9

TabIe2-4. RWCU System PPC Precision Signal Resistor Unit Hm m

      • u *****w on U

H 10 Table 2*5. AWCU System PPC Analog Input Card Unit 1

~

H 11 Table 2-6. RWCU S'YStem LooaI service Envirooments H

un

'H 12 Table 2*7. RWCU System Inlet Thermocouple n

n 12 Table 2-8. RWCU System Outtet TherTn()COlJpl$

13 Table 2..9. RWCU System Therrnocoople PPC Analog Input Card Unit 1 u

mmm. on H 14 Table.2..10. RWCU System Thermocouple Local Service Environments n

~

~u m*

H 14 Table 2..11. CAD HycIraulic System Flow Elel118nt - Unit 1..*~

n H,.

H 15 Table 2..12. CRD Hydraulic System Differential Pressure Transmilter.m m

u 16 Table 2..13. CRD Hydraulic System PPC Precision Signal Resistor Unit..Hn u

~*'

17 Table 2..14. CAD Hydraulic System PPC Analog Input Card Unit 1

++>

17 Table 2..15. CAD Hydraulic System locaJ Service Environmenfs m

      • on u

18 Table 2*16. Recirculation Pump Motor Watt Transctucer

'1'+

19 Table 2..11. Recirculation Pump Motor Watt Transducer PPC Precision Signal Resistor Unit 20 Table 2..18. Recirculation Pump Motor Watt Transducer PPC Analog Input Card Unit 1 m

20 Table A7-1. CTP CaJcufatlon sensitivity Allalysis u

~**+

H 1e TabJe AS*1. Relationship between (f,. Y(G1), O')(J and n HH H

79 Page 4 of 93 Exelon.

Reactor Core Thermal Power Uncertainty Calculation Unit 1 TABU! OF TABLES LE-0113 Revision 0 Table 2-2. RWCU System In~t Flow Eler'l\\ent - Unit 1 u

~

n 8

Table 2-3. RWCU System Inlet Flow Differential Pressure Transmitter mw u

..,w 9

TabIe2-4. RWCU System PPC Precision Signal Resistor Unit Hm m

      • u *****w on U

H 10 Table 2*5. AWCU System PPC Analog Input Card Unit 1

~

H 11 Table 2-6. RWCU S'YStem LooaI service Envirooments H

un

'H 12 Table 2*7. RWCU System Inlet Thermocouple n

n 12 Table 2-8. RWCU System Outtet TherTn()COlJpl$

13 Table 2..9. RWCU System Therrnocoople PPC Analog Input Card Unit 1 u

mmm. on H 14 Table.2..10. RWCU System Thermocouple Local Service Environments n

~

~u m*

H 14 Table 2..11. CAD HycIraulic System Flow Elel118nt - Unit 1..*~

n H,.

H 15 Table 2..12. CRD Hydraulic System Differential Pressure Transmilter.m m

u 16 Table 2..13. CRD Hydraulic System PPC Precision Signal Resistor Unit..Hn u

~*'

17 Table 2..14. CAD Hydraulic System PPC Analog Input Card Unit 1

++>

17 Table 2..15. CAD Hydraulic System locaJ Service Environmenfs m

      • on u

18 Table 2*16. Recirculation Pump Motor Watt Transctucer

'1'+

19 Table 2..11. Recirculation Pump Motor Watt Transducer PPC Precision Signal Resistor Unit 20 Table 2..18. Recirculation Pump Motor Watt Transducer PPC Analog Input Card Unit 1 m

20 Table A7-1. CTP CaJcufatlon sensitivity Allalysis u

~**+

H 1e TabJe AS*1. Relationship between (f,. Y(G1), O')(J and n HH H

79 Page 4 of 93

.1 INPUTS Exel6n.

1.0 PURPOSE 1.1 FUNCTIONAL DESCRIPTION AND CONFIGURATION 2.0 DESIGN BASIS Reactor Core Thermal Power LE-0113 Uncertainty Calculation Unit 1 Revision 0 The purpose of this calculation is to determine the uncertainty in the reactor core thermal power (heat balance) calculation performed by the Plant Process Computer (PPC). This calculation will evaluate the contribution of the different instrument channel loop uncertainties to the uncertainty of the Core Thermal Power (CTP) value using the reactor heat balance relationship when the plant is operating at 100% rated power under steady state conditions.

This calculation is being performed in support of the licensing amendment for Measurement Uncertainty Recapture (MUR) power uprate. This calculation applies to Limerick Generating Station Unit 1.

Limerick Generation Station (LGS) Unit 1 will be installing highly accurate ultrasonic feedwater flow meters per Engineering Change Request (ECR) LG 09-00096. This calculation will determine the uncertainty in Core Thermal Power calculation when the reactor heat balance is performed using the process computer with the feedwater flow and temperature measurement input supplied by the Caldon@ Leading Edge Flow Meters Check Plus (LEFMV+) System Ultrasonic Flow Meters (UFM).

Various plant parameters are monitored by the NSSS computer to develop the reactor core thermal power calculation. On June 1, 2000 Appendix K to Part 50 of Title 10 of the Code of Federal Regulations was changed to allow licensees to use a power uncertainty of less than 2 % in their LOCA analysis. The change allowed licenses to recapture power by using state-of-art devices to more precisely measure feedwater flow. Feedwater flow inaccuracy is a large contributor in the uncertainty determination of reactor power. This calculation is being performed in the support of a License Amendment Request (LAR) for a MUR power uprate.

Table 2-1 lists the parameters which specify input to the core thermal power calculation, their uncertainty values, and the source of these values.

The values for Feedwater Flow, Feedwater Temperature, and Reactor Narrow Range Dome Pressure are specified by separate calculations as follows (Ref. 4.8.6 thru 4.8.8) :

" LEAF-MUR-0001, Bounding Uncertainty Analysis for Thermal Power determination LE-0116, Reactor Dome Narrow Range Pressure Measurement Uncertainty The uncertainties for Reactor Water Clean-up (RWCU) Flow Rate, RWCU Inlet Temperature Thermocouple, Control Rod Drive (CRD) Flow Rate, and Recirculation Pump Power are calculated in individual sections of this calculation. Recirculation Pump Efficiency is given in calculation LM-0552 (Ref. 4.8.2). The thermal loss due to radiated heat loss to the drywall is specified by separate calculation LM-553 (Ref. 4.8.3) for calculating the reactor heat balance by hand.

Other inputs and the related source references are listed in the Table 2-1.

Page 5 of 93 Exelon.

1.0 PURPOSE Reactor Core Thermal Power Uncertainty Calculation Unit 1 LE-0113 Revision 0 The purpose of this calculation is to determine the uncertainty in the reactor core thermal power (heat balance) calculation performed by the Plant Process Computer (PPC). This calculation will evaluate the contribution of the different instrument channel loop uncertainties to the uncertainty of the Core Thermal Power (CTP) value using the reactor heat balance relationship when the plant is operating at 1000k rated power under steady state conditions.

This calculation is being performed in support of the licensing amendment for Measurement Uncertainty Recapture (MUR) power uprate. This calculation applies to Limerick Generating Station Unit 1.

1.1 FUNCTIONAL DESCRIPTION AND CONFIGURATION Limerick Generation Station (LGS) Unit 1 will be installing highly accurate ultrasonic feedwater flow meters per Engineering Change Request (ECA) LG 09-00096. This calculation will determine the uncertainty in Core Thermal Power calculation when the reactor heat balance is performed using the process computer with the feedwater flow and temperature measurement input supplied by the Caldon Leading Edge Flow Meters Check Plus (LEFM-/+) System Ultrasonic Flow Meters (UFM).

2.0 DESIGN BASIS Various plant parameters are monitored by the NSSS computer to develop the reactor core thermal power calculation. On June 1, 2000 Appendix K to Part 50 of Title 10 of the Code of Federal Regulations was changed to allow licensees to use a power uncertainty of less than 2 %

in their LOCA analysis. The change allowed licenses to recapture power by using state-ot-art devices to more precisely measure feedwater flow. Feedwater flow inaccuracy is a large contributor in the uncertainty determination of reactor power. This calculation is being performed in the support ot a License Amendment Request (LAR) for a MUR power uprate.

2.1 INPUTS Table 2-1 lists the parameters which specify input to the core thermal power calculation, their uncertainty values, and the source of these values.

The values for Feedwater Flow. Feedwater Temperature, and Reactor Narrow Range Dome Pressure are specified by separate calculations as follows (Ref. 4.8.6 thru 4.8.8):

LEAE-MUR-0001, Bounding Uncertainty Analys;s for Thermal Power determination LE-0116, Reactor Dome Narrow Range Pressure Measurement Uncertainty The uncertainties for Aeactor Water Clean-up (RWCU) Flow Rate, RWCU Inlet Temperature Thermocouple. Control Rod Drive (CRD) Flow Rate, and Recirculation Pump Power are calculated in individual sections of this calculation. Recirculation Pump Efficiency is given in calculation LM-0552 (Ref. 4.8.2). The thermal lass due to radiated heat loss to the drywell is specified by separate calculation LM-553 (Ref. 4.8.3) for calculating the reactor heat balance by hand.

Other inputs and the related source references are listed in the Table 2-1.

Page 5 of 93

Exelvn.

Reactor Core Thermal Power Uncertainty Calculation Unit 1 Revision 0 Table 2-1.

Design Inputs (Ref. 4.8.5, Sec. 2.0)

LE-0113 Page 6 of 93 Feedwater Enthalpy NIA N/A 409.71 BtuAbm (430.8 °F and 1155 psig)

0.

Bt005 005 (Ref. 4.9.3)

(Ref. 4.9.3, Attachment 6)

Feedwater Mass 15.09 Mlbmfir Flow Rate (LEFM 10-C986 NIA t 0.32 %

(Ref. 4.$.6)

~+ System)

(Ref. 4.8.9, Table 6-11 a)

Peeess~ure r N/A N/A 1155 PSIG t 10 psig (Section 3.11)

(Ref.4.4.1)

Feedwater TE-006-A1744 thru 430.8 °F Temperature 1 N041 A-F A1750 (Ref. 4.8.9, Table 6-11 a) t 0.57 °F

( Ref. 4.8.6 }

Radiated Reactor 0.89 MW (U 1)

Pressure Vessel (RPV) Heat Loss NIA NIA 1.Q4 MW U2 10%

(Section 3.12)

(Ref. 4.8.3, See. 2.0) 1043 PSIG Reactor Dome PT-042-El 234 (1057.7 psia, Ref. 4.8.9, t 20 prig (Ref. 4.8.8)

Pressure 1 N008 Table 6-11 a)

Saturated Steam 1191.1 Btu/Ibm Enthalpy N/A NIA (1043 psig, sat) t 0.05 Btullbm

{Ref. 4.9.3}

(Ref. 4.9.3, Attachment 5)

Recirculation Pump Motor 1 A(B)-P201 NIA 94.8 °la (Attachment 10 &

N/A N/A efficiency Ref. 4.8.2)

Recirculation Pump Motor 1A(B)-P201 NIA 7700 Hp (5.74 MW) t 1.4 °lo (Section 7.6.8)

Power (Ref. 4.3.3) 419.83 Btulibm RWCU Discharge Enthalpy NIA N/A

( 440 °F and 1168 psi g) t 0.005 BtuAbm (Ref. 4.9.3)

(Ref. 4.9.3, Attachment 6)

Description Inst. Tog No.

Computer Point Nominal Value Uncertainty Uncertainty Basis CRD Enthalpy NIA N/A 71.93 Btu/Ibm (100 °F and 1448 prig) t 0.005 BtuAbm (Ref. 4.9.3)

(Ref. 4.9.3, Attachment 6)

CRD Water Flow Discharge TE-046-103 A1201 100°F(Ref. 4.4.1) t 0.7 °F (Ref. Attachment 1)

Temperature Nominal Flow CRD Water Flow Rate FT-046-1 N004 A1711 105 GPM 5.4%

(Section 7.5.6)

Exelon.

Reactor Core Thermal Power Uncertainty Calculation Unit 1 Table 2-1.

Design Inputs LE-0113 Revision 0 Description Inst. Tag Computer Nominal Value Uncertainty Uncertainty No.

Point Basis 71.93 Btu/Ibm

+/- 0.005 CRD Enthalpy N/A NlA (100 OF and 1448 psig)

Btullbm (Ref. 4.9.3)

(Ref. 4.9.3, Attachment 6)

CRD Water Flow Discharge TE..046*103 A1201 100°F(Ref. 4.4.1)

+/- 0.7 OF (Ref. Attachment 1)

Temperature Nominal Flow CRD Water Flow FT-046-A1711 105 GPM 5.4%

(Section 7.5.6)

Rate 1N004 (Ref. 4.8.5, Sec. 2.0)

Feedwater 409.71 Btullbm

+/-0.005 Enthalpy NlA NlA (430.8 OF and 1155 psig)

Btu/Ibm (Aef. 4.9.3)

(Ref. 4.9.3, Attachment 6)

Feedwater Mass 15.09 Mlbmlhr Flow Rate (LEFM 10-C986 NlA (Ref. 4.8.9, Table 6-11a)

+/-0.32 %

(Ref. 4.8.6)

-/+ System)

Feedwater N/A NlA 1155 PSIG

+/- 10 psig (Section 3.11)

Pressure (Ref.4.4.1)

Feedwater TE-Ooe-A1744 thru 430.8 OF

+/- 0.57 OF (Ref. 4.8.6)

Temperature 1N041A*F A1750 (Ref. 4.8.9, Table 6-11a)

Radiated Reactor 0.89 MtN (U1)

Pressure Vessel N/A NlA 1.04 MW (U2)

+/- 10%

(Section 3.12)

(RPV) Heat Loss (Ref. 4.8.3, Sec. 2.0) 1043 PSIG Reactor Dome PT-042-E1234 (1057.7 psia, Ref. 4.8.9,

20 pslg (Ref. 4.8.8)

Pressure 1N008 Table 6-11a)

Saturated Steam 1191.1 BtuJIbm Enthalpy N/A N/A (1043 psig, sat)

+/- 0.05 Btu/Ibm (Ref. 4.9.3)

(Ref. 4.9.3, Attachment 5)

Recirculation 94.8 % (Attachment 10 &

Pump Motor 1A(B)-P201 N/A N/A N/A efficiency Ref. 4.8.2)

Recirculation 7700 Hp (5.74 MW)

Pump Motor 1A(B)~P201 N/A

t 1.4 %

(Section 7.6.8)

Power (Ref. 4.3.3) 419.83 BtuJIbm AWCU Discharge N/A NlA (440 OF and 1168 psig)

+/- 0.005 (Ref. 4.9.3)

Enthalpy Btu/lbm (Ref. 4.9.3, Attachment 6)

Page 6 of 93

Exelc'~

Reactor Core Thermal Power Uncertainty Calculation Unit 1 Revision 0 Table 20.

CaNgn Inputs LE-0113 Page 7 of 93 Description Inst. Tag Computer Nominal Value Uncertainty Uncertainty No.

Point Basis RWCU Discharge Temperature TE-044-I N015 A1742 440 OF (Section 2.3.2) 4.37 (Section 7.4.6)

RWCU Inlet Flow Rate FT-044-A171$

360 GPM (Max) 2.3 %

(Section 7.3.6)

I N036A RA 1111)

RWCU Regen Heat Exchanger TE-044-I N004 A1741 530 OF 4.37 OF (Section 7.4.6)

Inlet Temperature (Section 2.3.1 )

RWCU Suction Enthalpy WA N/A 524.39 l3tu/Ibm 0.005 (Ref. 4.9.3)

(530 OF and 1060 prig}

13tu/Ibm Exelon.

Reactor Core Thermal Power Uncertainty Calculation Unit 1 Table 2-1.

Design Inputs LE-Q113 Revision 0 Description Inst. Tag Computer Nominal Value Uncertainty Uncertainty No.

Point Baais RWCU Discharge TE-044-440°F (Section 7.4.6)

Temperature 1N015 A1742 (Section 2.3.2)

t:4.37 RWCU Inlet Flow FT*044*

360 GPM (Max)

Rate 1N036A A1718 Ref. 4.5.11)

1:2.3%

(Section 7.3.6)

AWCU Regen TE*044*

530 of Heat Exchanger 1N004 A1741 (Section 2.3.1) 4.37 of (Section 7.4.6)

Inlet Temperature RWCU Suction 524.39 Btulfbm

% 0.005 Enthalpy NlA N/A (530 of and 1060 psig)

Btu/Ibm (Ref. 4.9.3)

Page 7 of 93

Exelc°~n.

2.2 REACTOR WATER CLEANUP (RWCU) FLOW LOOP UNCERTAINTY 2.2.1 Reactor Water Cleanup System Equipment Design Data (Ref. 4.3.4)

Main Cleanup Recirculation Pumas "A" Pump "B" & "C" Pumps Number required 1

2 Capacity, % (each) 100 50 "A" pump capacity is greater that the combined capacity of the "B" and "C" pumps 2.2.2 RWCU Flow Measurement Loop Diagram RWCU flow is measured by an orifice plate (FE-044-1 N035) located on the suction side of the RWCU Recirculation Pumps, which provides a OP signal to a Rosemount transmitter (FT-044-1 N036A). The transmitter supplies a milliamp signal to the PPC for display in the Control Room.

The instrument loop consists of the following : flow element, flow transmitter, a signal resistance unit and a PPC input/output (i/O) module. The loop configuration is shown below (Ref. 4.5.10) :

The loop components evaluated in this document (the applicable performance specifications and process parameter data) :

2.2.3 RWCU System Inlet Flow (Ref. 4.4.1, 4.5.6, 4.5.10, 4.5.11, 4.5.16, 4.6.4, and 4.9.1)

Reactor Core Thermal Power LE-0113 Uncertainty Calculation Unit 1 Revision 0 Table 2-2.

RWCU System Inlet Flow Element - Unit 1 Page 8 of 93 Component I.D. :

FE-044-1 N035 "RWCU SUCTION FLANGE UPSTREAM OF VALVE MV-44-1 F001 Device Type :

Orifice Plate Manufacturer/Model No. :

Vickery Simms Inc

./145C3227P037 Reference Accuracy (A1) :

+/-1.50°la of actual flow rate

~ Installation Accuracy:

Environmental` Conditions (Temp.):

+/-0.5%

40"F (min.), to 156°F (max.)

System flow rate (Ibm/hr)

Normal operation "A" pump 154,000 Normal operation "B" plus "C" pump 133,000 Maximum operation 180,000 Exelen.

Reactor Core Thermal Power Uncertainty Calculation Unit 1 LE-0113 Revision 0 2.2 REACTOR WATER CLEANUP (RWCU) FLOW LOOP UNCERTAINTY 2.2.1 Reactor Water Cleanup System Equipment Design Data (Ref. 4.3.4)

System flow rate (Ibm/hr)

Normal operation"A lf pump Normal operation "B" prus "e" pump Maximum operation 154,000 133,000 180,000 Main Cteanup Recirculation Pumps Number required Capacity, Ok (each)

"Aft Pump 1

100 "B" &"C" Pumps 2

50 "A" pump capacity is greater that the combined capacity of the "B" and "C" pumps 2.2.2 RWCU Flow Measurement Loop Diagram RWCU flow is measured by an orifice plate (FE-044-1 N035) located on the suction side of the RWCU Recirculation Pumps, which provides a i.\\P signal to a Rosemount transmitter (FT-044-1N036A). The transmitter supplies a milliamp signal to the PPC for display in the Control Room.

The instrument loop consists of the following: flow element, flow transmitter, a signal resistance unit and a PPC input/output (I/O) module. The loop configuration is shown below (Ref. 4.5.10):

FE-044-1N035 FT-044-1N036A PPC A1718 The loop components evaluated in this document (the applicable performance specifications and process parameter data):

2.2.3 RWCU System Infet Flow (Ref. 4.4.1, 4.5.6, 4.5.10, 4.5.11, 4.5.16, 4.6.4, and 4.9.1)

Table 2-2.

RWCU System Inlet Flow Element - Unit 1 J Com;~~ent 1.0.:

'<<'>>+><<0>:

,~""","?"""-

'~"".'>>:>:<<

j::"."",,,

FE-044-1N035 uRWCU SUCTION FLANGE UPSTREAM OF VALVE HV-44-1F001"

>'=,~.,~"'<&.,..

Device Type:

Orifice Plate Manufacturer/Model No.:

Vickery Simms Inc.l145C3227P037

~eference Accuracy (A1):

+/-1.500/0 of actual ftow rate jlnstaliatiOn Accuracy.

+/- 0.50/0 I Environmental Cond~jons (Temp.):

40°F (min.) to 156°F (max.)

c<.'.-~.,___~ ~

...."""'.>Y:"".;"'...,""....

_",,"*'WNN'","-",",,*"

Page 8 of 93 Exelen.

Reactor Core Thermal Power Uncertainty Calculation Unit 1 LE-0113 Revision 0 2.2 REACTOR WATER CLEANUP (RWCU) FLOW LOOP UNCERTAINTY 2.2.1 Reactor Water Cleanup System Equipment Design Data (Ref. 4.3.4)

System flow rate (Ibm/hr)

Normal operation"A lf pump Normal operation "B" prus "e" pump Maximum operation 154,000 133,000 180,000 Main Cteanup Recirculation Pumps Number required Capacity, Ok (each)

"Aft Pump 1

100 "B" &"C" Pumps 2

50 "A" pump capacity is greater that the combined capacity of the "B" and "C" pumps 2.2.2 RWCU Flow Measurement Loop Diagram RWCU flow is measured by an orifice plate (FE-044-1 N035) located on the suction side of the RWCU Recirculation Pumps, which provides a i.\\P signal to a Rosemount transmitter (FT-044-1N036A). The transmitter supplies a milliamp signal to the PPC for display in the Control Room.

The instrument loop consists of the following: flow element, flow transmitter, a signal resistance unit and a PPC input/output (I/O) module. The loop configuration is shown below (Ref. 4.5.10):

FE-044-1N035 FT-044-1N036A PPC A1718 The loop components evaluated in this document (the applicable performance specifications and process parameter data):

2.2.3 RWCU System Infet Flow (Ref. 4.4.1, 4.5.6, 4.5.10, 4.5.11, 4.5.16, 4.6.4, and 4.9.1)

Table 2-2.

RWCU System Inlet Flow Element - Unit 1 J Com;~~ent 1.0.:

'<<'>>+><<0>:

,~""","?"""-

'~"".'>>:>:<<

j::"."",,,

FE-044-1N035 uRWCU SUCTION FLANGE UPSTREAM OF VALVE HV-44-1F001"

>,=,~.,~",<&.,..

Device Type:

Orifice Plate Manufacturer/Model No.:

Vickery Simms Inc.l145C3227P037

~eference Accuracy (A1):

+/-1.500/0 of actual ftow rate jlnstaliatiOn Accuracy.

+/- 0.50/0 I Environmental Cond~jons (Temp.):

40°F (min.) to 156°F (max.)

c<.'.-~.,___~~

_",,"*'WNN'","-",",,*"

Page 8 of 93

Reactor Core Thermal Power Uncertainty Calculation Unit 1 Revision 0 Table 20.

RWCU System Inlet Flow Element - Unit 1 2.2.4 RWCU System Inlet Flow (Ref. 4.5.6,4.5.10,4.6.4, 4.7.2,4.7.6,4.7.7,4.9.5, and 4.9.7)

Table 2-3.

RWCU System Inlet Flow Differential Pressure Transmitter Page 9 of 93 LE-01 13 Environmental Conditions (Press.) :

(-) 1.0 (min.) to (+) 7.0 inches H20 (max.)

Environmental Conditions (RH 20 (min.) to 90 (max.)

Pipe Size:

6 inch schedule 80 Flange Rating :

600#

Pipe Class Service No. :

DCA-101 "RWCU from Recirc. Pump Suction Valve F004 Normal Operating Temperature :

530 Design Temperature :

582 OF Maximum Operating Temperature :

582 OF Normal Operating Pressure :

1060 prig Design Pressure :

1250 psig Maximum Operating Pressure :

1360 psig Normal flow Rate:

360 gpm Maximum Flow Rate :

477 gpm AP 0 Max. Flow Rate :

200 inches H20, nameplate data : 1 178 psig & 545 OF Component I.D. :

FT-044-1 N036A"REACTOR WATER CLEANUP INLET" Location (AREA / EVEL / RM) :

016 1283' / 506 Device Type :

Differential Pressure Transmitter Manufacturer/Model No. :

Rosemount/1 153DB5RCN0039 Quality Classification :

Q (Not Required)

Accident Service:

N/A Seismic Category :

N/A

Tech Spec Requirement

N/A Upper Range Limit 68 OF:

750 inches H20 Lower Range Limit @ 68 'IF :

0-125 inches H20

~

Calibrated Range :

0 0083 inches H20 - static pressure corrected Operating Range :

0 to 220 inches H20 at 1060psig, Exelon.

Reactor Core Thermal Power Uncertainty Calculation Unit 1 Table 2-2.

RWCU System Inlet Flow Element - Unit 1 LE-0113 Revision 0 Environmental Conditions (Press.):

(-) 1.0 (min.) to (+) 7.0 inches H20 (max.)

Environmental Conditions (RH %):

20 (min.) to 90 (max.)

Pipe Size:

6 inch schedule 80 Flange Rating:

600#

Pipe Class Service No.:

DCA-101 "AWCU from Recirc. Pump Suction Valve FOO4 Normal Operating Temperature:

539°F Design Temperature:

582 OF Maximum Operating Temperature:

582 OF Normal Operating Pressure:

1060 psig Design Pressure:

1250 psig Maximum Operating Pressure:

1360 psig Normal flow Rate:

360gpm Maximum Flow Rate:

4ngpm L\\P @ Max. Flow Rate:

200 inches H20, nameplate data: 1178 psig & 545 OF 2.2.4 RWCU System Inlet Flow (Ref. 4.5.6, 4.5.10,4.6.4,4.7.2,4.7.6,4.7.7,4.9.5, and 4.9.7)

Table 2-3.

RWCU System InJet Flow Differential Pressure Transmitter Component 1.0.:

FT-044-1 N036A "REACTOR WATER CLEANUP INLET" Location (AREA I EVEL I AM):

016/283' / 506 Device Type:

Differential Pressure Transmitter Manufacturer/Model No.:

Rosemountl1153DB5RCNOO39 Quality Classification:

a (Not Required)

Accident Service:

N/A

-_.--~_.~--

Seismic Category:

N/A ITech Spec Requirement:

N/A t----

-,-~---,-

IUpper Range Limit @ 68 OF:

750 inches H2O ILower Range Limit @ 68 OF:

0-125 inches H20

[c;librated Range:

,,~----

oto 218.3 inches H20 - static pressure corrected

,-~-,

IOperating Range:

oto 220 inches H20 at 1060psig L""''''''''W"",,,_mH_'

~

~

~

~

H

~

1V,..,.,.",,"'*-""'.*,W.*N-""'.*.w.*NN,W.*..'.*.w.*,*,*,*,*.*.*g*,***.*

....,**,**,.........w........wn.*.*,.-,-_nf:

Page 9 of 93 Exelon.

Reactor Core Thermal Power Uncertainty Calculation Unit 1 Table 2-2.

RWCU System Inlet Flow Element - Unit 1 LE-0113 Revision 0 Environmental Conditions (Press.):

(-) 1.0 (min.) to (+) 7.0 inches H20 (max.)

Environmental Conditions (RH %):

20 (min.) to 90 (max.)

Pipe Size:

6 inch schedule 80 Flange Rating:

600#

Pipe Class Service No.:

DCA-101 "AWCU from Recirc. Pump Suction Valve FOO4 Normal Operating Temperature:

539°F Design Temperature:

582 OF Maximum Operating Temperature:

582 OF Normal Operating Pressure:

1060 psig Design Pressure:

1250 psig Maximum Operating Pressure:

1360 psig Normal flow Rate:

360gpm Maximum Flow Rate:

4ngpm L\\P @ Max. Flow Rate:

200 inches H20, nameplate data: 1178 psig & 545 OF 2.2.4 RWCU System Inlet Flow (Ref. 4.5.6, 4.5.10,4.6.4,4.7.2,4.7.6,4.7.7,4.9.5, and 4.9.7)

Table 2-3.

RWCU System InJet Flow Differential Pressure Transmitter Component 1.0.:

FT-044-1 N036A "REACTOR WATER CLEANUP INLET" Location (AREA I EVEL I AM):

016/283' / 506 Device Type:

Differential Pressure Transmitter Manufacturer/Model No.:

Rosemountl1153DB5RCNOO39 Quality Classification:

a (Not Required)

Accident Service:

N/A

-_.--~_.~--

Seismic Category:

N/A ITech Spec Requirement:

N/A t----

-,-~---,-

IUpper Range Limit @ 68 OF:

750 inches H2O ILower Range Limit @ 68 OF:

0-125 inches H20

[c;librated Range:

,,~----

oto 218.3 inches H20 - static pressure corrected

,-~-,

IOperating Range:

oto 220 inches H20 at 1060psig L""''''''''W"",,,_mH_'

~

~

~

~

H

~

1V,..,.,.",,"'*-""'.*,W.*N-""'.*.w.*NN,W.*..'.*.w.*,*,*,*,*.*.*g*,***.*

....,**,**,.........w........wn.*.*,.-,-_nf:

Page 9 of 93

Exelon, Reactor Core Thermal Power Uncertainty Calculation Unit 1 Revision 0 Table 2-3.

RWCU System Inlet Flow Differential Pressure Transmitter 2.2.5 RWCU System Signal Resistor Unit (Ref. 4.1.1, 4.5.10, 4.5.17, and 4.7.3) :

Table 2-4.

RWCU System PPC Precision Signal Resistor Unit Page 1 0 of 93 LE-0113 Calibration Span:

218.3 inches H2O Output Signal :

4 - 20 mA Setpoint :

N/A Calibration Period:

24 months Accuracy (A2) :

t 0.25 % calibrated span (see note)

Calibration Accuracy:

t 0.5 %

Stability (Drift, D2):

t 0.2 % URL for 30 months f2cs]

Temperature Effect (DTE1), per 100°F t (0.75 % of upper range limit + 0.5 % span)

Temperature Normal Operating Limits : 40 to 200 °F Overpressure Effect :

Maximum zero shift of t 1.0 % URL above 2000 psig Static Pressure Zero Effect:

+/-0.2 % of upper range limit Static Pressure Span Effect (SPNE2) : t 0.5 % input reading per 1, 000 psi.

Seismic (vibration) Effect (SEIS2) :

Accuracy within f0.5 % of upper range limit during and after a seismic disturbance defined by a required response spectrum with a ZPA of 4 g's.

Power Supply Effect (PSE2) :

< 0.005 % of calibrated span per volt Mounting Position Effect :

No span effect. Zero shift of up to 1.5 inH20 EMI/RFI Effect :

Not Specified Response time (damping) :

Code N - Adjustable damping ; max. 0.8 seconds Harsh temperature effect (HTE2) :

Accuracy within +/- 5.0 % of URL during and after exposure to 265 °F (129.5 °C), 24 psig, for 35 hours4.050926e-4 days <br />0.00972 hours <br />5.787037e-5 weeks <br />1.33175e-5 months <br />.

Humidity limits :

0 to 100 % Relative Humidity (RH)

Safety Classification:

Application - Non-safety-Related Radiation Effect (e2R) :

Accuracy within t 4.0 % of URL during and after exposure to 2.2 x107 rads, TID of gamma Note : Includes combined effects of linearity, hysteresis, and repeatability Dwg. Designation :

SRU-1 Device Type :

^

Precision signal resistor unit (Manufacturer/Model No. :

Bailey Type 766 Exelon.

Reactor Core Thermal Power Uncertainty Calculation Unit 1 LE-0113 Revision 0 SRU*1 Table 2-4.

RWCU System PPC Precision Signal Resistor Unit Table 2*3.

RWCU System Inlet Flow Differential Pressure Transmitter Calibration Span:

218.3 inches H2O Output Signal:

4-20 mA Setpoint:

N/A Calibration Period:

24 months Accuracy (A2):

+/- 0.25 % calibrated span (see note)

Calibration Accuracy:

+/- 0.50/0 Stability (Drift, 02):

t 0.2 % URL for 30 months [20-]

Temperature Effect (DTE1), per 100°F +/- (0.75 % of upper range limit + 0.5 % span)

Temperature Normal Operating Limits: 40 to 200 of Overpressure Effect:

Maximum zero shift of +/- 1.0 % URl above 2000 psig Static Pressure Zero Effect:

+/-O.2 % of upper range limit Static Pressure Span Effect (SPNE2):

+/- 0.5 % input reading per 1,000 psi.

Seismic (vibration) Effect (SEIS2):

Accuracy within +/-O.5 Ok of upper range limit during and after a seismic disturbance defined by a required response spectrum with a ZPA of 4 g's.

Power Supply Effect (PSE2):

< 0.005 % of calibrated span per volt Mounting Position Effect:

No span effect. Zero shift of up to 1.5 inH20 EMIIRFI Effect:

Not Specified Response time (damping):

Code N

  • Adjustable damping; max. 0.8 seconds Harsh temperature effect (HTE2):

Accuracy within +/- 5.0 % of URL during and after exposure to 265 of (129.5 °C), 24 psig. for 35 hours4.050926e-4 days <br />0.00972 hours <br />5.787037e-5 weeks <br />1.33175e-5 months <br />.

Humidity limits:

oto 100 % Relative Humidity (RH)

Safety Classification:

Application - Non-safety*Related Radiation Effect (e2R):

Accuracy within :t 4.0 Ok of URl during and after exposure to 2.2 x1 07 rads. TID of gamma Nota: Includes combined effects of linearity, hysteresis, and repeatabUity 2.2.5 RWCU System Signal Resistor Unit (Ref. 4.1.1,4.5.10,4.5.17, and 4.7.3):

IDwg. Designation:

ro~vice Type:

  • ------_'----+~p-re-c-is-iO-n-Sj-gn-a-l-re-s'--is-to-r-u-n-it----"--'--------i 1",~~_n_u_fa_c~~,:.r._1M_o_d_e!_~~::_,~_~._,~ _

__L__8a_i_le~y_T_yp_e",_.~7__6,.,,6

.."WN...W,.,_.

N'"""_._**__.,,'"


J Page 10 of 93

Exelun.

Table 2-4.

RWCU System PPC Precision Signal Resistor Unit 2.2.6 RWCU System Computer Point-Plant Process Computer (PPC) (Ref. 4.5.10 and Attachment 4) :

The PPC calculates Core Thermal Power based in part on the measurement of Reactor Water Cleanup flow. The PPC uses an analog input card, which read the voltage drop across a precision 250 ohm resistor.

Table 2-5.

RWCU System PPC Analog Input Card Unit 1 Page 1 1 of 93 LE-0113 Reactor Core Thermal Power Uncertainty Calculation Unit 1 Revision 0 Component I.D. :

A1718 Location :

10-0603 (H12-P603)

Device Type :

PPC - Potentiometer (Analog) Input Card Manufacturer/Model No. :

Analogic/ANDS5500 Quality Classification:

N/A Accident Service:

N/A Seismic Category:

N Tech Spec Requirement:

N/A Selected Full Scale Span :

t 5 VDC Calibration Span :

(,) 5 VDC to (+) 5 VDC Calibration Period :

24 months Accuracy (A3) :

f 0.5 % of full scale span Input Impedance (resistance) :

10 Meg Ohms Analog to digital Converter.

Not Specified Power Supply Effect (PSE3) :

N/A EMI/RFI Effect:

N/A Response time (damping) :

N/A Operating Temperature Limits :

32 to 122 °F (0 to 50 "C)

Humidity limits :

Not Specified Safety Classification :

Non Safety-Related Selected Range:

250 Ohm Accuracy:

t 0.1 %, (t 0.25 ohm)

Safety Classification :

NIA Temperature Effect:

0.5 %

for 40 -120°F Input Signal Range :

4 to 20 mAdc Exelon.

Reactor Core Thermal Power Uncertainty Calculation Unit 1 Table 2-4.

RWCU System PPC Precision Signal Resistor Unit LE-0113 Revision 0 Selected Range:

250 Ohm Accuracy:

+/- 0.1 o/Ot (+/- 0.25 ohm)

Safety Classification:

N/A Temperature Effect:

+/- 0.5 % for 40 - 120°F Input Signal Range:

4to 20 mAde 2.2.6 RWCU System Computer Point - Plant Process Computer (PPC) (Ref. 4.5.10 and Attachment 4):

The PPC calculates Core Thermal Power based in part on the measurement of Reactor Water Cleanup flow. The PPC uses an analog input card, which read the voltage drop across a precision 250 ohm resistor.

Table 2-5.

AWCU System PPC Analog Input Card Unit 1 Component 1.0.:

A1718 Location:

1Q-C603 (H12-P603)

Device Type:

PPC - Potentiometer (Analog) Input Card Manufacturer/Model No.:

AnalogiclANDS5500 Quality Classification:

N/A Accident Service:

N/A Seismic Category:

N Tech Spec Requirement:

N/A Selected Full Scale Span:

t:5VDC Calibration Span:

(-) 5 VDC to (+) 5 VDC Calibration Period:

24 months Accuracy (A3):

+/- 0.5 % of fuU scale span Input Impedance (resistance):

10 Meg Ohms Analog to Digital Converter:

Not Specified Power Supply Effect (PSE3):

N/A EMI/RFI Effect:

N/A Response time (damping):

N/A Operating Temperature Limits:

32 to 122 OF (0 to 50°C)

Humidity limits:

Not Specified Safety Classification:

Non Safety-Related Page 11 of 93

C-~celun, 2.2.7 RWCU System Local Service Environments (Ref. 4.4.2} :

2.3 REACTt?R CLEANUP SYSTEM TEMPERATURE 2.3.1 RWCU System Regenerative Heat Exchanger Inlet Temperature (Ref. 4.4.1, 4.5.6, 4.5.71, 4.5.16, 4.6.4, 4,9.5, and 4.9.7}

TE-U44-1 NL~4 Thermocouple Reactor Core Thermal Power Uncertainty Calculation Unit 1 Revision 0 Table 2-6.

RWCU System Local Service Environments PPC A7 741 Table 2-7.

RhVCU System Inlet Therm~ouple Page 1 2 of 93 Component Numbers :

TE-044-i N004 "REACTOR 1NATER CLEANUP SYSTEM REGEN HEAT EXCH INLET TEMP" Devic Type :

Type T Copper - Constantan (CU/CN }

ManufacturerlModel No. :

California Allay CalModel 117C34$5P073 Element Range :

{-}200° to (+}700 °F Calibrated Range 0° - 600°F Rated Accuracy:

t 0.75°F Output Signal:

(-} O.fi74 mV to (+}15.769 mV Safety Classification :

NIA Pipe Class Service No. :

DCC-101 "RWCU pump discharge thru Regen HXs Normal Operating Temperature 530 °F (See Note 1 }

Normal Operating Temperature Spec:

535 °F Design Temperature :

582 ~'F Maximum Operating Temperature :

582 °F Flow Transmitter Plant Process Computer Area f Room.

Area 016 Area ~B - Control Room Location 506C - Cont. H2 Recombiner Control Room (Comp. Rm. 553 }

Normal Temp. Range (°F}

65 min / 106 max l 85 norm 65 min / 78 max / norm N1A Normal Pressure

(-} 0.25 inches WG

+ 0.25 inches WG Normal Humidity {RH °/d}

50 average l 90 maximum 50 average l ~? maximum Radiation 2.50E-03 Rads/hr, 8.78E+02 TiD hUA Exelon.

Reactor Core Thermal Power Uncertainty Calculation Unit 1 LE-Q113 Revision 0 2.2.7 RWCU System Local Service Environments (Ref. 4.4.2):

Table 2-6.

RWCU System Local Service Environments Flow Transmitter Plant Process Computer Area/ Room.

Area 016 Area 008 - Control Room Location 506e - Cont. H2 Recombiner Control Room (Camp. Rm. 553)

Normal Temp. Range (OF) 65 min /106 max / 85 norm 65 min /78 max / norm N/A Normal Pressure

(..) 0.25 inches WG

+ 0.25 inches WG Normal Humidity (RH %)

50 average /90 maximum 50 average / 90 maximum Radiation 2.50E-03 Radslhr, 8.78E+02 TID NlA 2.3 REACTOR CLEANUP SYSTEM TEMPERATURE 2.3.1 RWCU System Regenerative Heat Exchanger Inlet Temperature (Ref. 4.4.1. 4.5.6, 4.5.11, 4.5.16, 4.6.4, 4.9.5, and 4.9.7)

TE-044-1 N004 PPC A1741 Thermocouple a

~

Table 2-7.

RWCU System rnlet Thermocouple Component Numbers:

TE-044-1N004 "REACTOR WATER CLEANUP SYSTEM REGEN HEAT EXCH INLET TEMP" Device Type:

Type T Copper-Constantan (CU/CN)

Manufacturer/Model No.:

Carifornia Alloy ColModel 117C3485P073 Element Range:

{..)2000 to (+)700 OF Calibrated Range 0°.. 600°F Rated Accuracy:

+/- O.75°F Output Signal:

(-) 0.674 mV to (+)15.769 mV Safety Classification:

N/A Pipe Class Service No.:

DCC-101 uAWCU pump discharge thru Regen HXs Normal Operating Temperature 530 OF (See Note 1)

Normal Operating Temperature Spec:

535°F Design Temperature:

582¢F Maximum Operating Temperature:

582 OF Page 12 of 93

Note 1 : Reactor Engineering provided normal operating temperature based on 100% power operation for both Units. Data was retrieved once per hour for one week. The Unit 2 value of 530 'F in lieu of Unit 1 528 'IF was used based on it being the most conservative.

2.3.2 RWCU System Regenerative Heat Exchanger Outlet Temperature (Ref. 4.1.1, 4.5.6, 4.5.16, 4.6.4, 4.5.11, 4.9.5, and 4.9.7)

TE-044-1 N015 Thermocouple Reactor Core Thermal Power Uncertainty Calculation Unit I Revision 0 Table 2q.

RWCU System Inlet Thermocouple PPC A1742 Table 20.

RWCU System Outlet Thermocouple Note 2 : Reactor Engineering provided normal operating temperature based on 100% power Page 13 of 93 LE-01 13 Component Numbers:

TE-044-1 N015 "REACTOR WATER CLEANUP SYSTEM REGEN HEAT EXCH OUTLET TEMP" Device Type:

Type T Copper - Constantan (CU/CN)

Manufacturer/Model No. :

California Alloy Co/Model 11 7C3485PO73 Element Range:

(-)200" to (+)700 *F Input Range 0,0 - 600"F Rated Accuracy:

t 0.75'OF Output Range
0) 0.874 mV to (+}15.789 MV Safety Classification :

N/A Pipe Class Service No. :

ECC-1 05 "RWCU Regen HX to HV-1 F042 Normal Operating Temperature based on actual plant data 440 OF (See Note 2)

Normal Operating Temperature :

438 OF Design Temperature:

434 OF Maximum Operating Temperature :

434 'IF Normal Operating Pressure :

1168 psig Design Pressure:

1290 psig Maximum Operating Pressure :

1542 psig Normal flow Rate :

360 gp Normal Operating Pressure:

1235 prig Design Pressure :

1290 prig Maximum Operating Pressure :

1542psig Normal flow Rate :

360 gpm Exelon.

Reactor Core ThermaJ Power Uncertainty Calculation Unit 1 LE-0113 Revision 0 Table 2-7.

RWCU System Inlet Thermocouple Normal Operating Pressure:

1235 psig Design Pressure:

1290 psig Maximum Operating Pressure:

1542 psig Normal flow Rate:

360gpm Note 1: Reactor Englneenng provided normal operating temperature based on 1000/0 power operation for both Units. Data was retrieved once per hour for one week. The Unit 2 value of 530 of in lieu of Unit 1 528 of was used based on it being the most conservative.

2.3.2 RWCU System Regenerative Heat Exchanger Outlet Temperature (Ref. 4.1.1, 4.5.6, 4.5.16, 4.6.4, 4.5.11,4.9.5. and 4.9.7)

TE-044-1 N015 PPC A1742 Thermocouple Table 2-8.

RWCU System Outlet Thermocouple Note 2: Reactor Eng,neerlng provided normal operating temperature based on 100% power

~.---

Component Numbers:

TE-044-1 N015 "REACTOR WATER CLEANUP SYSTEM REGEN HEAT EXCH OUTLET TEMP" Device Type:

Type T Copper - Constantan (CU/CN)

Manufacturer/Model No.:

California Alloy ColModeJ 117C348SP073 Element Range:

(-)200° to (+)700 of Input Range 0° - 600°F Rated Accuracy:

I: O.75°F Output Range

(-) 0.674 mV to (+)15.769 mV Safety Classification:

N/A Pipe Class Service No.:

ECC-10S "RWCU Regen HX to HV-1F042 Normal Operating Temperature 440 of (See Note 2) based on actual plant data Normal Operating Temperature:

438 of Design Temperature:

434 of Maximum Operating Temperature:

434 of Normal Operating Pressure:

1168 psig Design Pressure:

1290 psig

[MaXimumOperating Pressure:

~._~

1542 psig

--~------

Normal flow Rate:

360 gpm n+~","""",,,"""'H"""N**'

.. ?................,......"....H.

-~-,-,=-,---<<~<<~.~.. _-

Page 13 of 93

Exele)

Reactor Core Thermal Power Uncertainty Calculation Unit 1 Revision 0 operation for both Units. Data was retrieved once per hour for one week.

2.3.3 RWCU System Plant Process Computer (PPC) (Ref. 4.5.6, 4.5.16, and Attachment 4) :

Table 2-9.

RWCU System Thermocouple PPC Analog Input Card Unit 1 2.3.4 RWCU System Local Service Environments (Ref. 4.4.2):

Table 2-10.

RWCU System Thermocouple Local Service Environments Page 1 4 of 93 LE-01 1 3 Component I.D. :

Al 718 and Al 742 Location :

10-C603 (H12-P603)

Device Type :

PPC - Potentiometer (Analog) Input Card Manufacturer/Model No. :

Analogic/ANDS5500 Quality Classification :

N/A Accident Service :

N/A Seismic Category:

N Tech Spec Requirement:

N/A Selected Range :

Upper : -25 mV to + 25mV Calibration Span:

(-) 0.674 mV to (+) 15.769 mV Calibration Period :

24 months Accuracy (A2) :

t 0.5 % of full scale span Input Impedance (resistance) :

10 Meg Ohms Analog to Digital Converter :

Not Specified Signal Source Resistance :

2800 0 maximum Power Supply Effect (PSE2) :

N/A EMI/RFI Effect :

N/A Response time (damping) :

NIA Operating Temperature Limits :

32 to 122 °F (0 to 50 °C)

Humidity limits:

Not Specified Safety Classification :

Non Safety-Related Thermocouple Plant Process Computer Area 1 Room.

Area 016 Area 008 - Control Room Location 506C -- Cant. H2 Recombiner Control Room (Comp. Rm. 553)

Normal Temp. Range (F) 65 min 1112 max / 104 norm 65 min / 78 max / norm NIA Normal Pressure

(-) 0.25 inches WG

+ 0.25 inches WG Exel(tn.

Reactor Core Thermal Power Uncertainty Calculation Unit 1 LE-0113 Revision 0 operation for both Units. Data was retrieved once per hour for one week.

2.3.3 RWCU System Plant Process Computer (PPC) (Ref. 4.5.6, 4.5.16, and Attachment 4):

Table 2..9.

RWCU System Thermocouple PPC Analog Input Card Unit 1 Component 1.0.:

A1718 and A1742 Location:

10-C603 (H12-P603)

Device Type:

PPC - Potentiometer (Analog) Input Card Manufacturer/Model No.:

AnalogiclANDS5500 Quality Classification:

NJA Accident Service:

N/A Seismic Category:

N Tech Spec Requirement:

N/A Selected Range:

Upper: -25 mV to + 25mV Calibration Span:

(-) 0.674 mV to (+)15.769 mV Calibration Period:

24 months Accuracy (A2):

+/- 0.5 % of full scale span Input Impedance (resistance):

10 Meg Ohms Analog to Digital Converter.

Not Specified Signal Source Resistance:

2800 n maximum Power Supply Effect (PSE2):

N/A EMI/RFJ Effect:

N/A Response time (damping):

N/A Operating Temperature Limits:

32 to 122 of (0 to 50°C)

Humidity limits:

Not Specified Safety Classification:

Non Safety-Related 2.3.4 RWCU System Local Service Environments (Ref. 4.4.2):

E Table 2-10.

I L IS RWCU S t

Th ysem ermocouple oca ervlce nVlronments Thermocouple Plant Process Computer i--,

Areal Room.

Area 016 Area 008 - Control Room Location 506e - Cont. H2 Recombiner Control Room (Camp. Rm. 553)

-~_.

Normal Temp. Range (OF) 65 min /112 max /104 norm 65 min I 78 max I norm NIA Normal Pressure

(-) 0.25 inches WG

+ 0.25 inches WG k....,=,,~"

  • "",,"if=~m

~.."_".,,,,,~,,.",.*,.,,

.,.._..~

>>-"~,-,.,.,..

Page 14 of 93

Exelon.

Normal Humidity (RH %)

50 average / 90 maximum 50 average / 90 maximum NIA Radiation 2.4 CRD FLOW RATE UNCERTAINTY Flow element :

FE-046-1 N003 Reactor Core Thermal Power Uncertainty Calculation Unit 1 Revision 0 2.50E-03 Rads/hr, 8.78E+02 TID 2.4.1 CRD Hydraulic System Flow Loop Diagram Flow Transmitter FT-0046-1 N004 Input Resistor 8 ohm Comp. Point A1711 LE-0113 Each analyzed instrument loop consists of a flow element supplying a differential pressure to a pressure transmitter, and a PPC input/output (1/0) module with a precision resistor (8 0) across the input. The loop is shown as follows :

The loop components evaluated in this document (and the applicable performance specification and process parameter data) :

2.4.2 CRD Hydraulic System Flow Element (Ref. 4.4.1, 4.5.8, 4.5.12, 4.6.2, 4.6.3, 4.8.5, and 4.9.5)

Table 2-11.

CRD Hydraulic System Flow Element Unit 1 Page 1 5 of 93 Component I.D. :

FE-046-1 N003 "CRD HYDRAULIC SYS DRIVE WTR FLOW CONT" Device Type :

Flow Nozzle Manufacturer/Model No. :

GE - Vickery Simms Inc./ 158B7077AP016 Reference Accuracy (A1) :

t 1.0 % flow Design Temperature :

150°F Design Pressure :

2000 psig Pipe Size:

2 inch schedule 80 Material

..Stainless Steel Maximum Flow 100 y.

gpm Pipe Class Service No. :

DCD-112, "Control Rod Drive Hyd. from DBD-I 08 to Hydraulic Control Units Normal Operating Temperature :

100 °F Design Temperature :

150 °F Maximum Operating Temperature :

150 OF Normal Operating Pressure :

1448 psig Design Pressure :

1750 psig Maximum Pressure :

~..

Operating

~.,,.

A......_

.....r 1750 psig

,TH~...m._.__~ ,.

.,M.._c....

Exelon.

Reactor Core Thermal Power Uncertainty Calculation Unit 1 LE-0113 Revision 0 Normal Humidity (RH %)

50 average I 90 maximum 50 average / 90 maximum Radiation 2.50E-03 Radslhr. 8.78E+02 TID NlA 2.4 CRD FLOW RATE UNCERTAINTY 2.4.1 CRD Hydraulic System Flow loop Diagram Each analyzed instrument loop consists of a flow element supplying a differential pressure to a pressure transmitter. and a PPC input/output (I/O) module with a precision resistor (80) across the input.

The loop is shown as forrows:

Flow element:

Flow Input Compo Point FE-046-1 NOO3 f---t Transmitter FT* r---t Resistor

---+

A1711 0046-1NOO4 80hm The loop components evaluated in this document (and the applicable performance specification and process parameter data):

2.4.2 CRD Hydraulic System Flow Element (Ref. 4.4.1, 4.5.8, 4.5.12, 4.6.2, 4.6.3, 4.8.5, and 4.9.5)

Table 2-11.

CRD Hydraulic System Flow Element -- Unit 1 Component 1.0.:

FE-04f3..1 NOO3 "CRD HYDRAULIC SYS DRIVE WTR FLOWCONT" Device Type:

Flow Nozzle Manufacturer/Model No.:

GE - Vickery Simms Inc./15887077AP016 Reference Accuracy (A1):

+/- 1.0 Ok flow Design Temperature:

150°F Design Pressure:

2000 psig Pipe Size:

2 inch schedule 80 Material Stainless Steel Maximum Flow 100 gpm Pipe C'ass Service No.:

DCD-112, "Control Rod Drive Hyd. from DBD-I 08 to Hydraulic Control Units I Normal Operating Temperature:

100 OF

~--

Design Temperature:

150 OF Maximum Operating Temperature:

150 OF Normal Operating Pressure:

1448 psig

_n.

Design Pressure:

1750 psig f--~'

Maximum Operating Pressure:

1750 psig N.*',........""'*,*,,

,....;.'W'w"..,.,..

."rN""

  • """""",,-,~rr Page 15 of 93

Exel6n*

Reactor Core Thermal Power Uncertainty Calculation Unit 1 Revision 0 Table 2-11.

CRD Hydraulic System Flow Element - Unit 1 Normal flow Rate :

AP Max. Flow Rate :

200 inches H20 0 100 gpm 105 gpm @ 1515.8 prig 2.4.3 CRD Hydraulic System Flow Transmitter (Ref. 4.5.8, 4.5.12, 4.5.2, 4.6.3, 4.7.4 and 4.9.5)

Table 2-12.

CRD Hydraulic System Differential Pressure Transmitter Page 1 6 of 93 LE-0113 Component I.D. :

FT-046-1 N004 "CRD HYDRAULIC SYS DRIVE WTR FLOW CONT" Location (AREA / EVEL / RM) :

015 / 253'/ 402 Device Type:

Differential Pressure Transmitter Manufacturer/Model No. :

Rosemount/ 1151 DP5D22PB Quality Assurance Classification :

N Accident Service & Seismic Category.

NIA Tech Spec Requirement :

N/A Upper Range Limit :

750 inches H2O Lower Range Limit:

0-125 inches H2O Calibrated Span:

0 to 197.5 inches H2O - Static pressure corrected Operating Span :

0 to 2(30 inches H2O at 100 gpm Output Signal :

4 - 20 mA, Corresponding to 0 -100 gpm Calibration Period:

24 months Accuracy (A2):

t 0.25 % calibrated span Calibration Accuracy:

+/- 0.5 Stability (Drift, D2):

f 0.25 °lo of URL for 6 months [2a]

Temperature Effect (DTE1), per 100"F t 1.5% of URL per 100 "F t 2.5 % for low range (URL/6)

This taken to be equal to 2.4 %

at 197.5 inches H2O span Temperature Normal Operating Limits :

(-)40 to 150 °F (Amplifier)

Overpressure Effect:

Zero shift of less than t 2.0 Static Pressure Zero Effect:

t 0.5 % of upper range limit for 2000 psi Static Pressure Span Effect (SPNE2) :

(-) 0.5 % t 0.1% input reading per 1,000 psi. This is a systematic error which can be calibrated out for a particular pressure before installation.

Exelon.

Reactor Core Thermal Power Uncertainty Calculation Unit 1 Table 2-11.

CAD Hydraulic System Flow Element - Unit 1 LE-0113 Revision 0 Normal flow Rate:

105 gpm @ 1515.8 psig L\\P @ Max. Flow Rate:

200 inches H20 @ 100 gpm 2.4.3 CAD Hydraulic System Flow Transmitter (Ref. 4.5.8, 4.5.12, 4.6.2, 4.6.3, 4.7.4 and 4.9.5)

Table 2-12.

CRD Hydraulic System Differential Pressure Transmitter Component 1.0.:

FT-046-1 NOO4 "CAD HYDRAULIC SYS DRIVE WTR FLOW CaNT" Location (AREA 1EVEL 1AM):

015/253' 1402 Device Type:

Differential Pressure Transmitter ManufacturerlModel No.:

Rosemount/1151DP5D22PB Quality Assurance Classification:

N Accident Service & Seismic Category:

N1A Tech Spec Requirement:

N/A Upper Range Limit:

750 inches H2O Lower Range Limit:

0-125 inches H2O CaJibrated Span:

oto 197.5 inches H20 - static pressure corrected Operating Span:

oto 200 inches H20 at 100 gpm Output Signal:

4 - 20 rnA, Corresponding to 0 - 100 gpm Calibration Period:

24 months Accuracy (A2):

+/- 0.25 % calibrated span Calibration Accuracy:

+/-0.5%

Stability (Drift. 02):

+/- 0.25 %

of URL for 6 months [20']

Temperature Effect (DTE1), per 100°F

+/- 1.5% of URl per 100 OF

+/- 2.5 % for low range (URL/6)

This taken to be equal to 2.4 % at 197.5 inches H2O span Temperature Normal Operating Limits:

(-}40 to 150 of (Amplifier)

Overpressure Effect:

Zero shift of less than +/- 2.00/0 Static Pressure Zero Effect:

+/- 0.5 % of upper range limit for 2000 psi Static Pressure Span Effect (SPNE2):

(-) 0.5 % +/- 0.1 % input reading per 1,000 psi. This is a systematic error which can be calibrated out for a particular pressure before installation.

Page 16 of 93

&elonl.

Reactor Core Thermal Power Uncertainty Calculation Unit 1 Revision 0 Table 2-12.

CAD Hydraulic System Differential Pressure Transmitter 2.4.4 CAD Hydraulic System Flow Signal Resistor Unit (Ref. 4.1.1, 4.5.12, Attachment 12, and 4.7.3) :

Table 2-13.

CAD Hydraulic System PPC Precision Signal Resistor Unit 2.4.5 CAD Hydraulic System Flow PPC 1/0 (Refs. 4.5.8, 4.5.12, and Attachment 4) :

Table 2-14.

CAD Hydraulic System PPG Analog Input Card Unit 1 Page 1 7 of 93 LE-01 13 Component I.D. :

A1711 Location :

10-C603 (H12-P603)

Device Type :

PPC - Potentiometer (Analog) Input Card Manufacturer/Model No. :

Analogic/ANDS5500 Quality Classification :

N/A Accident Service & Seismic Category : --N/A Dwg. Designation :

8 D Device Type :

Precision signal resistor unit (wire-wound)

Manufacturer/Model No. :

Bailey Type 766 Selected Range:

8 Ohm Accuracy :

  • 0.008 ohm (+/- 0.1 %)

Safety Classification :

N/A Temperature Effect :

0.5 % for 40 -120°F Input Signal Range :

4 to, 20 mAdc Seismic (vibration) Effect (SEIS2) :

+/- 0.05 % of URL per g at 200 Hz in any axis.

Power Supply Effect (PSE2) :

< 0.005 % of calibrated span per volt.

Mounting Position Effect:

No span effect. Zero-shift can be calibrated out.

EMI/RFI Effect :

Not Specified Response time (damping) :

Not Specified Harsh temperature effect (HTE2) :

Not Applicable Humidity limits :

0 to 100 % RH Safety Classification :

Non-safety related Radiation Effect (e2R) :

Not Specified Exelon.

Reactor Core Thermal Power Uncertainty Calculation Unit 1 LE-Q113 Revision 0 Table 2*12.

CAD Hydraulic System Differential Pressure Transmitter Seismic (vibration) Effect (SEIS2):

+/- 0.05 % of URL per 9 at 200 Hz in any axis.

Power Supply Effect (PSE2):

< 0.005 % of calibrated span per volt.

Mounting Position Effect:

No span effect. Zero-shift can be calibrated out.

EMIIRFI Effect:

Not Specified Response time (damping):

Not Specified Harsh temperature effect (HTE2):

Not Applicable Humidity limits:

Oto 100 % RH Safety Classification:

Non-safety related Radiation Effect (e2R):

Not Specified 2.4.4 CAD Hydrau'ic System Flow Signal Resistor Unit (Ref. 4.1.1, 4.5.12, Attachment 12, and 4.7.3):

Table 2-13.

CAD Hydraulic System PPC Precision Signal Resistor Unit Dwg. Designation:

an Device Type:

Precision signal resistor unit (wire-wound)

Manufacturer/Model No.:

Bailey Type 766 Selected Range:

aOhm Accuracy:

+/- 0.008 ohm (+/- 0.1 0/0)

Safety Classification:

N/A Temperature Effect:

+/- 0.5 % for 40 - 120°F Input Signal Range:

4 to 20 mAdc 2.4.5 CAD Hydraulic System Flow PPC I/O (Refs. 4.5.8, 4.5.12, and Attachment 4):

Table 2-14.

CAD Hydraulic System PPC Analog Input Card Unit 1 f Component I.D.:

A1711 r---------

Location:

1o-C603 (H12-P603)

,.~~

Device Type:

PPC - Potentiometer (Analog) Input Card

.~~~._~

Manufacturer/Model No.:

Analogic/ANDS5500

..m...._

..' **~_ _

I Quality CJassification:

N/A I

[~~~~~~,~~_"~ervice & Seismic Category:

--~-

N/A

, ~--<<.~-

_-,.......w~;-_-..,_*_*_*,..N,_~_*,_*_*.*_*,?,.,.''_*'N-"''''''.*."',,*,.

Page 17 of 93

Exel6n, Reactor Core Thermal Power Uncertainty Calculation Unit 1 Revision 0 Table 2-14.

CRD Hydraulic System PPC Analog Input Card Unit 1 2.4.6 CRD Hydraulic System Flow Process Parameters (Refs. 4.3.2 and 4.8.5):

Process Temp Maximum :

1501 "F Process Temp Minimum 4511E The minimum water temperature is based on the CST being outside and exposed to winter elements. The maximum temperature is based on 140°F of the condensate and condenser system plus 10"F nominal heat addition from the CRD Water pump. (Ref. 4.3.2) 2.4.7 CRD Hydraulic System Local Service Environments (Ref. 4:4.2) :

Table 2-15.

CRD Hydraulic System Local Service Environments Page 1 8 of 93 LE-0113 Tech Spec Requirement :

N/A Selected Full Scale Span :

t 160 mV Calibration Span :

0 160 mV to (+) 160 mV Calibration Period :

24 months Accuracy (A3) :

t 0.5 °lo of full scale span Calibration Accuracy:

N/A Input Impedance (resistance):

10 Meg Ohms Analog to Digital Converter:

Not Specified Power Supply Effect (PSE3) :

N/A EMI/RFI Effect:

N/A Operating Temperature Limits :

32 to 122 °F (0 to 50 "C)

Humidity limits:

Not Specified Safety Classification :

Non Safety-Related Flow Transmitter Plant Process Computer Area / Room.

Area 015 / Room 402 Area 008 - Control Room Location Room 402 Control Room (Camp. Rm. 553)

Normal Temp. Range (°F) 65 min / 106 max / 90 norm 65 min / 78 max / norm N/A Normal Pressure

(-) 0.25 inches WG

+ 0.25 inches WG Normal Humidity (RH °to) 50 average / 90 maximum 50 average / 90 maximum l

Radiation 1.00E-01 Rads/hr, 3.51 E+04 TID I N/A Exelon.

Reactor Core Thermal Power Uncertainty Calculation Unit 1 Table 2*14.

CAD Hydraulic System PPC Analog Input Card Unit 1 LE-0113 Revision 0 Tech Spec Requirement:

N/A Selected Full Scale Span:

+/- 160 mV Calibration Span:

(-) 160 mV to (+) 160 mV Calibration Period:

24 months Accuracy (A3):

+/- 0.5 %

of fuJI scafe span Calibration Accuracy:

N/A Input Impedance (resistance):

10 Meg Ohms Analog to Digital Converter:

Not Specified Power Supply Effect (PSE3):

N/A EMI/RFI Effect:

N/A Operating Temperature Limits:

32 to 122 of (0 to 50°C)

Humidity limits:

Not Specified Safety Classification:

Non Safety-Related 2.4.6 CAD Hydraulic System Frow Process Parameters (Refs. 4.3.2 and 4.8.5):

Process Temp Maximum:

Process Temp Minimum The minimum water temperature is based on the CST being outside and exposed to winter elements. The maximum temperature is based on 140°F of the condensate and condenser system plus 10°F nominal heat addition from the CRD Water pump. (Ref. 4.3.2) 2.4.7 CAD Hydraulic System Local Service Environments (Ref. 4.4.2):

Table 2-15.

CRD Hydraulic System Local Service Environments

-~

Flow Transmitter Plant Process Computer Area I Room.

Area 015 I Room 402 Area 008 - Control Room Location Room 402 Control Room {Compo Rm. 553}

l1----"

Normal Temp. Range (OF) 65 min /106 max /90 norm 65 min / 78 max I norm N/A Normal Pressure

(-) 0.25 inches WG

+ 0.25 inches WG f------

Normar Humidity (RH 0/0) 50 average / 90 maximum 50 average I 90 maximum f.*"f'N'_vN-"*.*",.,....N"..Nf,..'WUN.""'>N"""ff....*.~...'.*,...

Radiation 1.00E-01 Radslhr, 3.51 E+04 TID! NlA Page 18 of 93

&elon.

2.5 RECIRCULATION PUMP MOTOR UNCERTAINTY 2.5.1 Recirculation Pump Motor Loop Diagram Each analyzed instrument loop consists of a Watt Transducer, and a PPC input/output (I/O) module.

The uncertainty magnitude of the CT and PT is negligible for this calculation.

PT Potential Transformer Watt Transducer Reactor Core Thermal Power LE-0113 Uncertainty Calculation Unit 1 Revision 0 CT Current Transformer Table 2-16.

Recirculation Pump Motor Watt Transducer Page 1 9 of 93 PPC Analog Input Card Computer Points - Al 725 & Al 726 2.5.2 Recirculation Pump Motor Watt Transducer (Ref. 4.5.13 to 4.5.15, Attachment 9, 4.9.5, and 4.9.7)

Component I.D. :

MT1 A and MT1 B Location :

10-C603 (H 12-P603)

Device Type:

Watt Transducer Manufacturer/Model No. :

Ametek Power Systems 1 XL3-1 K5A2-25 Quality Classification :

N/A Accident Service & Seismic Category :

N/A Tech Spec Requirement :

N/A Rated Output (RO) 10.5 MW Current Input (Current Transformer) :

0 -- 5 Amps (1500/5)

Voltage Input (Potential Transformer) :

0 - 120 V (4160/120)

Output Range:

0 -1 mAdc Calibration Period :

24 months Accuracy (A1) :

t (0.2% Reading + 0.01 % Rated Output) at 0-200% Rated Output Stability (Drift, D1) per year :

t 0.1 % RO, Non-cumulative Temperature Effect:

1 0.005 %

/ ° C PF Effect on Accuracy t 0.1% VA (maximum)

EMI/RFI Effect:

N/Au.W.

Exelon.

Reactor Core Thermal Power Uncertainty Calculation Unit 1 LE-0113 Revision 0 2.5 RECIRCULATION PUMP MOTOR UNCERTAINTY 2.5.1 Recirculation Pump Motor Loop Diagram Each analyzed instrument loop consists of a Watt Transducer, and a PPC input/output (I/O) module.

The uncertainty magnitude of the CT and PT is negligibre for this calculation.

PT CT Potential Current Transformer Transfonner

~Ir Watt Transducer PPC Analog Input Card Computer Points - A1725 & A1726 2.5.2 Recirculation Pump Motor Watt Transducer (Ref. 4.5.13 to 4.5.15, Attachment 9,4.9.5, and 4.9.7)

Table 2-16.

Recirculation Pump Motor Watt Transducer Component J.D.:

MT1 A and MT1 B Location:

10-C603 (H12-P603)

Device Type:

Watt Transducer Manufacturer/Model No.:

Ametek Power Systems I XL3-1 K5A2-25 Quality Classification:

N/A Accident Service & Seismic Category:

N/A Tech Spec Requirement:

N/A Rated Output (RO) 10.5 MW Current Input (Current Transformer):

0-5 Amps (1500/5)

Voltage Input (Potentiar Transformer):

0- 120 V (4160/120)

Output Range:

0-1 mAde

.....~,~.....

Calibration Period:

24 months Accuracy (A1):

+/- (0.2°/0 Reading + 0.01 % Rated Output) at 0-200°10 Rated Output Stability (Drift, 01) per year:

+/- 0.1% RO, Non-cumulative Temperature Effect:

+/- 0.005 % I 0 C

~,--,-

l;~;FE~:~~curacy

+/- 0.1% VA (maximum)

,.......,.,.,.".wc..w *.".'"",,*.w.*'va''''',N,.,.*

N/A

.""""#'-h"W"*""_""'"'''.~~_~'F:'C''''

.'<'"-.:.,~,,,,,,,,,,,,""

Page 19 of 93

Exel6n, Reactor Core Thermal Power LE-0113 Uncertainty Calculation Unit 1 Revision 0 Table 2-16.

Recirculation Pump Motor Watt Transducer 2.5.3 Recirculation Pump Motor Watt Meter Transducer Precision Signal Resistor (Ref. 4.5.13 to 4.5.15, and 4.9.7) :

Table 2-17.

Recirculation Pump Motor Watt Transducer PPC Precision Signal Resistor Unit 2.5.4 Recirculation Pump Motor Watt Transducer PPC Analog Input Card (Ref. Attachment 4) :

Table 2-18.

Recirculation Pump Motor Watt Transducer PPG Analog Input Card Unit 1 Page 20 of 93 Component I.D. :

Al 725 and Al 726 Location :

10-C603 (H12-P603)

Device Type:

PPC - Potentiometer (Analog) Input Card Manufacturer/Motel No. :

Analogic/ANDS5500 Quality Classification :

N/A Accident Service & Seism ic Category :

N/A Tech Spec Requirement :

N/A Selected Fu l l Scale Span:

t 160 mV Calibration Span:

(-) 160 mV to (+) 160 mV Calibration Period :

24 months Accuracy (A2) :

....._~~_..___~...

t 0.5 % of full scale span Dwg. Designation :

R3A & R3B Device Type :

Precision signal resistor unit (wire-wound)

Manufacturer/Model No. :

CE / Type H R41 13513 Selected Rating :

160 Ohm Accuracy :

t 0.1 °lfl of input signal range, 0.1 watt Safety Classification :

N/A Temperature Effect :

NIA Input Signal Range :

0 - 1 mAdc Operating Temperature Limits :

(-)4°F to 158 'IF (-200 C to x-70° C)

Operating Humidity:

0 to 95 °lo RH non condensing Safety Classification :

Non Safety-Related Exelon.

Reactor Core Thermal Power Uncertainty Calcufation Unit 1 Table 2-16.

Recirculation Pump Motor Watt Transducer LE..0113 Revision 0 Operating Temperature Limits:

(-)4°F to 158 of (-20° C to +70 0 C)

Operating Humidity:

oto 95 % RH non condensing Safety Classification:

Non Safety-Related 2.5.3 Recirculation Pump Motor Watt Meter Transducer Precision Signal Resistor (Ref. 4.5.13 to 4.5.15, and 4.9.7):

Table 2-17.

Recirculation Pump Motor Watt Transducer PPC Precision Signal Resistor Unit Dwg. Designation:

R3A & R3S Device Type:

Precision signal resistor unit (Wire-wound)

Manufacturer/ModeJ No.:

GE/Type HR41D5B Selected Rating:

1600hm Accuracy:

+/- 0.1 Ok of input signal range, 0.1 watt I Safety Classification:

N/A Temperature Effect:

N/A Input Signal Range:

0-1 mAdc w_

2.5.4 Recirculation Pump Motor Watt Transducer PPC Analog Input Card (Ref. Attachment 4):

Table 2-18.

Recirculation Pump Motor Watt Transducer PPC Analog Input Card Unit 1 Component 1.0.:

A1725 and A1726 Location:

10-COO3 (H12-P603)

-~

Device Type:

PPC - Potentiometer (Analog) Input Card Manufacturer/Model No.:

Analogic/ANDS5500 Quality Crassification:

N/A Accident Service &Seismic Category:

N/A Tech Spec Requirement:

N/A f---

Selected Full Scale Span:

+/- 160 mV Calibration Span:

(-) 160 mV to (+) 160 mV Calibration Period:

24 months

-_.~.~~

Accuracy (A2):

+/- 0.5 %

of full scale span

"'~<c"',...,c~<<,_-,~-.,

Page 20 of 93

Exel6n.

Reactor Core Thermal Power Uncertainty Calculation Unit I Revision 0 Me 218.

Recirculation Pump Motor Watt Transducer PPG Analog Input Card Unit 1 Page 21 of 93 LE-01 13 Input Impedance (resistance) :

1 0 Meg Ohms Analog to Digital Converter:

Not Specified Power Supply Effect (PSE2) :

N/A EMI/RFI Effect:

N/A Operating Temperature Limits :

32 to 122 "F (0 to 50 OC) limits:

L Humidity Not Specified I Safety Classification :

- Non Safety-Related Exelon.

Reactor Core Thermal Power Uncertainty Calculation Unit 1 LE-0113 Revision 0 Table 2..18.

Recirculation Pump Motor Watt Transducer PPC Analog Input Card Unit 1 Input Impedance (resistance):

10 Meg Ohms Analog to Digital Converter:

Not Specified Power Supply Effect (PSE2):

N/A EMI/RFI Effect:

N/A Operating Temperature Limits:

32 to 122 of (0 to 50°C)

Humidity limits:

Not Specified Safety Classification:

Non Safety-Related Page 21 of 93

Facelun.

3.0 ASSUMPTIONS AND LIMITATIONS Reactor Core Thermal Power LE-0113 Uncertainty Calculation Unit 1 Revision 0 3.1 Standard practice is to specify calibration uncertainty in calculations equal to the uncertainty associated with the instruments under test (Ref. 4.1.1).

3.2 Instrumentation uncertainty caused by the operating environment's temperature, humidity, and pressure variations are evaluated when these error sources are specified by the instrument's vendor. If the instrument's operating environment specifications bound the in-service environmental conditions where the equipment is located and separate temperature, humidity, and pressure uncertainty terms are not specified for the instrument, then these uncertainties are assumed to be included in the manufacturer's reference accuracy specification.

3.4 Published instrument vendor specifications are considered to be based on sufficiently large samples so that the probability and confidence level meets the 2a criteria, unless stated otherwise by the vendor.

3.5 Seismic effects are considered negligible or capable of being calibrated out unless the instrumentation is required to operate during and following a seismic event.

3.6 The insulation resistance error is considered negligible unless the instrumentation is required to operate in an abnormal or harsh environment.

3.7 Regulated instrument power supplies are assumed to function within specified voltage limits ;

therefore, power supply error is considered negligible with respect to other error terms unless the vendor specifically specifies a power supply effect.

3.8 Measurement of CRD hydraulic system purge water temperature is found to be accurate within t 0.7

°F per Attachment 1. The enthalpy of water at the CRD normal operating pressure of 1448 psig and normal operating temperature of 100 OF as listed in P-300 (reference 4.4.1) are used to determine uncertainty of the CRD hydraulic system enthalpy because this is more conservative than using the higher temperatures normally found during operation.

3.9 The CRD system uses a Rosemount 1151 differential pressure transmitter (Section 2.4.3), which is mounted in a radiation exposure area. A radiation exposure effect is not specified for the Model 1151 transmitter; therefore the radiation effect applicable to this transmitter is assumed to be 10 % of span. This assumption is considered conservative based on three factors : (1) periodic surveillance is performed on this transmitter and it is required to operate within 0.5 % of span or maintenance activities must be performed. Existing calibration records did not indicate anything unusual occurring with the calibration of these transmitters in this service. (2) A Rosemount Model 1153 series B transmitter is rated for radiation exposure and may be expected to have a radiation effect of t 4 % of upper range limit (UPL) during and after exposure to 2.2 x 107 rad (Section 7.3.1.16). However, this exposure is over a 1000 times the expected exposure for the CRD system flow transmitter s during normal service. The estimated TO during normal service for the CRD flow transmitter is 3.51 x 104 rad (Section 4.4.2). The 10 % span estimated effect is more than an order of magnitude greater than the threshold for maintenance activity and of the same order of magnitude of the effect on a similar transmitter with 1000 times the exposure.

3.10 Interim results are rounded to the level of significance of the input data to avoid implying that a higher level of precision exists in the calculated values. For example, uncertainty may be specified by a supplier to one significant figure (e.g., 0.5 %). This value says that the level of significance associated with this uncertainty is one part in two hundred. The results are rounded when the numeric value of a result implies a higher level of significance than what the input data suggests.

Page 22 of 93 Exelon.

Reactor Core Thermal Power Uncertainty Calculation Unit 1 LE-0113 Revision 0 3.0 ASSUMPTIONS AND LIMITATIONS 3.1 Standard practice is to specify calibration uncertainty in calculations equal to the uncertainty associated with the instruments under test (Ref. 4.1.1).

3.2 Instrumentation uncertainty caused by the operating environment's temperature, humidity, and pressure variations are evaluated when these error sources are specified by the instrument's vendor. If the instrument's operating environment specifications bound the in-service environmental conditions where the equipment is located and separate temperature, humidity, and pressure uncertainty terms are not specified for the instrument, then these uncertainties are assumed to be included in the manufacturers reference accuracy specification.

3.4 Published instrument vendor specifications are considered to be based on sufficiently large samples so that the probability and confidence revel meets the 20 criteria, unless stated otherwise by the vendor.

3.5 Seismic effects are considered negligible or capable of being calibrated out unless the instrumentation is required to operate during and following a seismic event.

3.6 The insulation resistance error is considered negligible unress the instrumentation is required to operate in an abnormal or harsh environment.

3.7 Regulated instrument power supplies are assumed to function within specified voltage limits; therefore, power supply error is considered negligible with respect to other error terms unless the vendor specifically specifies a power supply effect.

3.8 Measurement of CRD hydraulic system purge water temperature is found to be accurate within +/- 0.7 OF per Attachment 1. The enthalpy of water at the CAD normal operating pressure of 1448 psig and normal operating temperature of 100 OF as listed in P-300 (reference 4.4.1) are used to determine uncertainty of the CRD hydraulic system enthalpy because this is more conservative than using the higher temperatures normally found during operation.

3.9 The CRD system uses a Rosemount 1151 differential pressure transmitter (Section 2.4.3). which is mounted in a radiation exposure area. A radiation exposure effect is not specified for the Model 1151 transmitter; therefore the radiation effect applicable to this transmitter is assumed to be 10 % of span. This assumption is considered conservative based on three factors: (1) periodic surveillance is perlormed on this transmitter and it is required to operate within 0.5 % of span or maintenance activities must be performed. Existing caJibration records did not indicate anything unusual occurring with the calibration of these transmitters in this service. (2) A Rosemount Model 1153 series B transmitter is rated for radiation exposure and may be expected to have a radiation effect of +/- 4 % of upper range limit (URL) during and after exposure to 2.2 x 107 rad (Section 7.3.1.16). However, this exposure ;s over a 1000 times the expected exposure for the CAD system flow transmitter s during normal service. The estimated TID during normal service for the CRD flow transmitter is 3.51 x 10" rad (Section 4.4.2). The look span estimated effect is more than an order of magnitude greater than the threshold for maintenance activity and of the same order of magnitude of the effect on a similar transmitter with 1000 times the exposure.

3.10 Interim results are rounded to the level of significance of the input data to avoid implying that a higher fevel of precision exists in the calculated values. For example, uncertainty may be specified by a supplier to one significant figure (e.g., 0.5 Ok). This value says that the level of significance associated with this uncertainty is one part in two hundred. The results are rounded when the numeric value of a result implies a higher level of significance than what the input data suggests.

Page 22 of 93

Exel6n Reactor Core Thermal Power LE-4113 Uncertainty Calculation Unit 1 Revision 0 3.11 An uncertainty of t 10 psig is assumed for the feedwater pressure to cover the variations in the actual steam dome pressure. The variation in pressure has a negligible effect on the enthalpy of the feedwater.

3.12 An uncertainty of 10 °lo is assumed for the RPV thermal radiation heat loss term, ORAD, based on a review of calculation LM-0553 (Reference 4.8.3). LM-0553 determined the RPV heat loss by calculating the actual heat load in the drywell, subtracting the heat load attributed to any operating equipment in the drywell, and proportioning the heat load based on the shared chilled water system design flows assigned to Unit 1 drywell and Unit 2 drywell on a percentage basis. LM-0553 assumes Unit 1 and Unit 2 are operating at 100 percent power and the drywell air cooling fans are aligned as designed.

3.13 Steam table excerpts have been provided for convenience as Attachment 5, National Institute for Standards and Technology (NIST) Saturated Properties of Water, and Attachment 6, NIST Isobaric and Isothermal Properties of Water, as extracted from the NIST fluid properties WebBook (Reference 4.9.3). For conservatism, factors of one-half the least significant figure in the tables are used for the interpolation error. The factors are 0.05 BtuAbm for vapor and 0.005 Btullbm for liquid.

Page 23 of 93 Exelon.

Reactor Core Thermal Power Uncertainty Calculation Unit 1 LE-0113 Revision 0 3.11 An uncertainty of +/- 10 psig is assumed for the feedwater pressure to cover the variations in the actual steam dome pressure. The variation in pressure has a negligible effect on the enthalpy of the feedwater.

3.12 An uncertainty of 10 % is assumed for the RPV thermal radiation heat loss term, ORAD, based on a review of calculation LM-0553 (Reference 4.8.3). LM-DS53 determined the RPV heat loss by calculating the actual heat road in the drywell, subtracting the heat load attributed to any operating equipment in the drywell, and proportioning the heat load based on the shared chilled water system design flows assigned to Unit 1 drywell and Unit 2 drywell on a percentage basis. lM*0553 assumes Unit 1 and Unit 2 are operating at 100 percent power and the drywall air cooling fans are aligned as designed.

3.13 Steam table excerpts have been provided for convenience as Attachment 5, National Institute for Standards and Technology (NIST) Saturated Properties of Water, and Attachment 6, NIST Isobaric and Isothermal Properties of Water, as extracted from the NIST fluid properties WebBook (Reference 4.9.3). For conservatism, factors of one-half the least significant figure in the tables are used for the interpolation error. The factors are 0.05 Btullbm for vapor and 0.005 Btu/Ibm for liqUid.

Page 23 of 93

Exelon.

4.0 REFERENCES

4.1 METHODOLOGY Reactor Core Thermal Power Uncertainty Calculation Unit 1 Revision 0 4.1.1 CC-MA-103-2001, Rev 0, "Setpoint Methodology for Peach Bottom Atomic Power Station and Limerick Generating Station 4.1.2 IC-11-00001, Rev. 4, Calibration of Plant Instrumentation and Equipment 4.2 PROCEDURES 4.2.1 ST-2-044-400-1, Rev. 23, Reactor Water Cleanup High Differential Flow Isolation Calibration 4.2.2 IC-C-11-00307, Rev. 5, Calibration of Rosemount Model 1153 and 1154 Transmitters 4.3 DESIGN BASIS DOCUMENTS 4.3.1 L-S-11, Rev. 15, DBD Feedwater System 4.3.2 L-S-15, Rev. 10, DBD Control Rod Drive System 4.3.3 L-S-19, Rev. 10, DBD Recirculation System 4.3.4 L-S-36, Rev. 10, DBD Reactor Water Makeup System 4.3.5 L-S-42, Rev. 09, DBD Nuclear Boiler System 4.4 SPECIFICATIONS, CODES, & STANDARDS 4.4.1 P-300, Rev. 45, Specification "Piping Materials and Instrument Piping Standards 4.4.2 M-171, Rev. 16, Specification for Environmental Service Condition LGS Units 1 & 2 4.5 LIMERICK STATION DRAWINGS 4.5.1 M-06 Sheet 3, Rev. 58, P&ID Feedwater 4.5.2 M-23 Sheet 4, Rev. 33, P&ID Process Sampling 4.5.3 M-41 Sheets 112, Rev. 46/62, P&ID Nuclear Boiler 4.5.4 M-42 Sheets 112, Rev. 41134, P&ID Nuclear Boiler Vessel Instrumentation 4.5.5 M-43 Sheets 112, Rev. 48139, P&ID Reactor Recirculation Pump 4.5.6 M-44 Sheets 112, Rev. 56147, P&ID Reactor Water Clean-up 4.5.7 M-45 Sheet 1, Rev. 30, P&ID Cleanup Filter and Demineralizer 4.5.8 M-46 Sheet 1, Rev. 51, P&ID Control Rod Drive Hydraulic-Part A 4.5.9 M-47 Sheet 1, Rev. 45, P&ID Control Rod Drive Hydraulic-Part B 4.5.10 B21-1050-E-008, Rev. 13, Elem. Diagram Steam Leak Detection Schematic 4.5.11 G31-N011-C-003, Rev 002, Purchased PT. ORF. PLT. SH 1 4.5.12 C11-1060-E-002, Rev. 23, Elem. Diagram CRD Hydraulic System, LGS U1 4.5.13 B32-1030-E-050, Sheet 18, Rev. 5, Elem. Diag. Reactor "A" Recirc Pump an MG Set, Unit 1 Page 24 of 93 LE-0113 Exelon.


~--~---

Reactor Core Thermal Power Uncertainty Calculation Unit 1 LE..o113 Revision 0 4.0 4.1 4.1.1 4.1.2 4.2 4.2.1 4.2.2 4.3 4.3.1 4.3.2 4.3.3 4.3.4 4.3.5 4.4 4.4.1 4.4.2 4.5 4.5.1 4.5.2 4.5.3 4.5.4 4.5.5 4.5.6 4.5.7 4.5.8 4.5.9 4.5.10 4.5.11 4.5.12 4.5.13 REFERENCES METHODOLOGY CC-MA-1 03-2001, Rev 0, "Setpoint Methodology for Peach Bottom Atomic Power Station and Limerick Generating Station IC-11-00001, Rev. 4, Calibration of Plant Instrumentation and Equipment PROCEDURES ST-2-044-40D-1. Rev. 23, Reactor Water Cleanup High Differential Flow Isolation Calibration IC-C-11-00307, Rev. 5, Calibration of Rosemount Model 1153 and 1154 Transmitters DESIGN BASIS DOCUMENTS L-S-11, Rev. 15, DBD Feedwater System L-S..15, Rev. 10, DBD Control Rod Drive System L-S-19, Rev. 10, DBD Recirculation System L-S..36, Rev. 10, DBD Reactor Water Makeup System L-S-42, Rev. 09, DBD Nuclear BoHer System SPECIFICATIONS, CODES, & STANDARDS P-300, Rev. 45, Specification "Piping Materials and Instrument Piping Standards" M-171, Rev. 16, Specification for Environmental Service Condition LGS Units 1 & 2 LIMERICK STATION DRAWINGS M-06 Sheet 3, Rev. 58, P&ID Feedwater M-23 Sheet 4, Rev. 33, P&ID Process Sampling M-41 Sheets 1/2, Rev. 46/62, P&ID Nuclear Boiler M-42 Sheets 1/2, Rev. 41/34, P&ID Nuclear Boiler Vesse' Instrumentation M-43 Sheets 1/2, Rev. 48/39, P&ID Reactor Recirculation Pump M-44 Sheets 1/2, Rev. 56/47, P&ID Reactor Water Clean-up M-45 Sheet 1, Rev. 30, P&ID Cleanup Filter and Demineralizer M-46 Sheet 1, Rev. 51, P&fD Control Rod Drive Hydraulic-Part A M-47 Sheet 1, Rev. 45, P&JD Control Rod Drive Hydraulic-Part B 821-1050-E-008, Rev. 13, Elem. Diagram Steam Leak Detection Schematic G31-N011-C-003, Rev 002, Purchased PT. ORF. PLT. SH 1 C11-1060-E-002, Rev. 23, Elem. Diagram CRD Hydraulic System, LGS U1 832-1030-E-050, Sheet 18, Rev. 5, Elem. Diag. Reactor "A" Recirc Pump an MG Set, Unit 1 Page 24 of 93

Exel6n.

4.5.14 B32-1030-E-050, Sheet 19, Rev. 4, Elem. Diag. Reactor "B" Recirc Pump an MG Set, Unit 1 4.5.15 B32-1030-E-050, Sheet 3, Rev. 7, Elem. Diag. Reactor "A"& "B" Recirc Pump an MG Set, Parts List; Unit I 4.5.16 G31-1040-E-003, Rev. 28, Elementary Diag. Reactor Water Cleanup System, Unit 1 4.5.17 C32-1020-E-003, Rev. 33, Elem. Diag. Feedwater Control System, Unit 1 4.5.18 B21-1040-E-003, Rev. 21, Elem. Diag. Nuclear Boiler Process 4.5. 19 ET701, Sheets 6016, Rev. 9/8, Schematic & Connection Diagram NSSS/BOP Computer Analog Inputs, Unit 1 4.5.20 G31-1030-G-001, Rev. 20, Process Diagram Reactor Water Clean-Up System (High Pressure) 4.5.21 B32-C-001-J-023, Rev. 1, Recirculation Pump Curve (Attachment 10) 4.6 GENERAL ELECTRIC (GE) DOCUMENTS 4.6.1 C11-4010-H-004, Rev. 13, Control Rod Drive System 4.6.2 C11-N003-C-041 Rev 002 PPD-Flow Nozzle 4.6.3 C19 -3050-H-001, Rev. 14, CRD Instrumentation System 4.6.4 G31-3050-H-001, Rev. 26, Reactor Water Clean-Up System 4.6.5 B21-3050-H-001, Rev. 27, Nuclear Boiler System 4.6.6 B32-3050-H-001, Rev. 6, Reactor Recirculation System 4.7 VENDOR INFORMATION 4.7.1 M-1-832-0001-K7, Recirculation Pump Vendor Manual Reactor Core Thermal Power Uncertainty Calculation Unit 1 Revision 0 4.7.2 Rosemount Product Data Sheet 00809-0100-4302, Rev BA, January 2008, Rosemount 1153 Series B AlphalineO Nuclear Pressure Transmitter 4.7.3 A41-8010-K-018.6 -Bailey Signal Resistor Unit (SRU) Type 766 (Attachment 12) 4.7.4 Rosemount Inc., Instruction Manual 4259, Model 1151 Alphaline@ N[AP Flow Transmitter, 1977 (Attachment 11) 4.7.5 C95-0000-K-002(1 }.2 - ANDS481 0 - Data Acquisition System Instruction Manual 4.7.6 Rosemount Specification Drawing 01153-2734, "NO039 Option -Combination N0016 & N0037" 4.7.7 Rosemount Product Data Sheet 00813-0100-2655, Rev. AA June 1999 "N-Options for Use with the Model 1153 &11154 AlphalineO Nuclear Pressure Transmitters" 4.7.8 Fluke 8050A - Digital Multimeter Measurement P/N 530907 Rev 2 1984, Instruction Manual 4.7.9 Ametek Power Instruments, Digital & Analog Transducers, Power Measurement Catalog, Scientific Columbus Exceltronic AC Waft Transducer Specification (Attachment 9) 4.8 CALCULATIONS AND ENGINEERING ANALYSIS 4.8.1 LM-547, Rev. 2, Reactor Core Thermal Power Calculation Correction for Unaccounted Flow to Reactor Vessel 4.8.2 LM-552, Rev. 7, Reactor Heat Balance Calculation for Limerick Units 1 & 2 Page 25 of 93 LE-01 13 Exelon.

Reactor Core Thermal Power Uncertainty Calculation Unit 1 LE-0113 Revision 0 4.5.14 4.5.15 4.5.16 4.5.17 4.5.18 4.5.19 4.5.20 4.5.21 4.6 4.6.1 4.6.2 4.6.3 4.6.4 4.6.5 4.6.6 4.7 4.7.1 4.7.2 4.7.3 4.7.4 4.7.5 4.7.6 4.7.7 4.7.8 4.7.9 4.8 4.8.1 4.8.2 B32-1030-E-050i Sheet 19i Rev. 4, Elem. Diag. Reactor "B" Recirc Pump an MG Set, Unit 1 832-1030-E..Q50, Sheet 3, Bev. 7, Elem. Diag. Reactor "An & nB H Recirc Pump an MG Set, Parts Ust, Unit 1 G31-1040-E-003, Rev. 28, Elementary Diag. Reactor Water Cleanup System, Unit 1 C32-102D-E..Q03, Rev. 33, Elem. Diag. Feedwater Control System, Unit 1 821

  • 1040-E-003, Rev. 21, Elem. Diag. Nuclear Boiler Process E-Q701, Sheets 6/16, Rev. 9/8, Schematic & Connection Diagram NSSS/BOP Computer Analog Inputs, Unit 1 G31-103Q-G-001, Rev. 20, Process Diagram Reactor Water Clean-Up System (High Pressure)

B32-C-001-J..Q23, Rev. 1, Recirculation Pump Curve (Attachment 10)

GENERAL ELECTRIC (GE) DOCUMENTS C11-4010-H-004, Rev. 13. Control Rod Drive System C11-NOO3-C-001 Rev 002 PPO-Flow Nozzle C11-3050-H..OO1, Rev. 14, CRD Instrumentation System G31-305Q-H-001, Rev. 26, Reactor Water Clean-Up System B21-30SD-H-001, Rev. 27, Nuclear Boiler System B32-3050-H-001 t Rev. 6, Reactor Recirculation System VENDOR INFORMATION M-1-B32-C001-K7. Recirculation Pump Vendor Manual Rosemount Product Data Sheet 00809-0100-4302 t Rev BA, January 2008, Rosemount 1153 Series B Alphaline Nuclear Pressure Transmitter A41-801 Q-K-018.6 - Bailey Signal Resistor Unit (SRU) Type 766 (Attachment 12)

Rosemount Inc.* Instruction Manual 4259, Model 1151 Alphaline liP Flow Transmitter, 1977 (Attachment 11)

C95-0000-K-002(1 ).2 - ANOS481 0 - Data Acquisition System Instruction Manua' Rosemount Specification Drawing 01153-2734, "N0039 Option - Combination N0016 &NOO31' Rosemount Product Data Sheet 00813-0100-2655, Rev. AA June 1999 "N-options for Use with the Model 1153 &11154 Alphaline Nuclear Pressure Transmitters" Fluke 8050A - Digital Multimeter Measurement PIN 530907 Rev 2 1984, Instruction Manual Ametek Power Instruments. Digital &Analog Transducers, Power Measurement Catalog, Scientific Columbus Exceltronic AC Watt Transducer Specification (Attachment 9)

CALCULATIONS AND ENGINEERING ANALYSIS LM-547t Rev. 2, Reactor Core Thermal Power Calculation Correction for Unaccounted Flow to Reactor Vessel LM-552, Rev. 7, Reactor Heat Balance Calculation for Limerick Units 1 & 2 Page 25 of 93

Exelon, 4.8.3 LM-553, Rev. 0, Determination of the Reactor Pressure Vessel (RPV) Heat Loss 4.8.4 EE-94LGS, Rev. 16, Proper Calibration of Feedwater Elements FE-006-1(2)N001A, B, C (GE SIL 452) 4.8.5 LM-562, Rev. 2, CRD Flow Rates and System Pressures Reactor Core Thermal Power LE-0113 Uncertainty Calculation Unit 1 Revision 0 4.8.6 LEAF-MUR-0001, Rev. 0, Bounding Uncertainty Analysis for Thermal Power determination at Limerick Unit 1 Using the LEFM v+ System 4.8.7 LEAE-MUR-0002, Rev. 0, Pre-Commissioning Uncertainty Analysis for Thermal Power determination at Limerick Unit 1 Using the LEFM,/+ System 4.8.8 LE-0116, Rev. 0, Reactor Dome Narrow Range Pressure Measurement Uncertainty 4.8.9 Limerick PEPSETM MUR PU and EPU Heat Balances - Eval 2009-03880 RU 4.9 OTHER REFERENCES 4.9.1 ASME, "Fluid Meters Their Theory and Application" Sixth Edition, 1971.

4.9.2 Limerick Updated Final Safety Analysis Report, R14 4.9.3 Lemmon, E.W., McLinden, M.O., and Friend, D.G., "Thermophysical Properties of Fluid Systems",

NIST Chemistry WebBook, NIST Standard Reference Database Number 69, Eds. P.J. Linstrom and W.G. Mallard, National Institute of Standards and Technology, Gaithersburg MD, 20899, http://webbook.nist.gov, (retrieved September 30, 2009) 4.9.4 NUREG/CR-3659, Dated January 1985, NRC Guidance, A Mathematical Model for Assessing the Uncertainties of Instrumentation Measurements for Power and Flow of PWR Reactors 4.9.5 Plant Information Management System (PIMS) Data 4.9.6 TODI Tracking No : SEAG #09-000167, Plant Process Computer Data of various plant parameters, to support Core Thermal Power Uncertainty Calculations, Station/Unit(s) U1/U2, 9/3/09 4.9.7 IISCP (Improved Instrument Setpoint Control Program) Datasheets) Version 7.5 4.9.8 Edwards, Jerry L. Rosemount Nuclear Instruments letter in reference to "Grand Guff Nuclear Station message on INPO plant reports, subject Rosemount Instrument Setpoint Methodology, dated March 9, 2000". Letter dated 04/04/2000. (Attachment 3) 4.9.9 TODI A169446-80, subject "Steam Carryover Fraction on Process Computer Heat Balance Calculations" (Attachment 2)

Page 26 of 93 Exelon.

Reactor Core Thermal Power Uncertainty Calculation Unit 1 LE-G113 Revision 0 4.8.3 LM-553, Rev. 0, Determination of the Reactor Pressure Vessel (RPV) Heat Loss 4.8.4 EE-94LGS, Rev. 16, Proper Calibration of Feedwater Elements FE*OO6*1 (2)NOOl A, B, C (GE SIL 452) 4.8.5 LM*562, Rev. 2, CRO Flow Rates and System Pressures 4.8.6 LEAE-MUR-0001, Rev. 0, Bounding Uncertainty Analysis for Thermal Power determination at Limerick Unit 1 Using the LEFM,/+ System 4.8.7 LEAE-MUR-0002, Rev. 0, Pre-Commissioning Uncertainty Analysis for Thermal Power determination at Limerick Unit 1 Using the LEFM../+ System 4.8.8 LE-0116, Rev. 0, Reactor Dome Narrow Range Pressure Measurement Uncertainty 4.8.9 Limerick PEPSETM MUR PU and EPU Heat Balances - EvaI2009*03880 RO 4.9 OTHER REFERENCES 4.9.1 ASME, "Fluid Meters Their Theory and Application" Sixth Edition, 1971.

4.9.2 Limerick Updated Final Safety Analysis Report, R14 4.9.3 Lemmon, E.W., Mclinden, M.D., and Friend, D.G., uThermophysical Properties of Fluid Systems*,

NIST Chemistry WebBook, NIST Standard Reference Database Number 69, Eds. P.J. Linstrom and W.G. Mallard, National Institute of Standards and Technology, Gaithersburg MD, 20899, http;/Iwebbook.nist.gov, (retrieved September 30,2009) 4.9.4 NUREG/CR-3659, Dated January 1985, NRC Guidance, A Mathematical Model for Assessing the Uncertainties of Instrumentation Measurements for Power and Flow of PWR Reactors 4.9.5 Plant Information Management System (PIMS) Data 4.9.6 TOOl Tracking No: SEAG #09-000167, Plant Process Computer Data of various prant parameters, to support Core Thermal Power Uncertainty Calculations, Station/Unit(s) U1/U2, 913/09 4.9.7 f1SCP (Improved Instrument Setpoint Control Program) Datasheets) Version 7.5 4.9.8 Edwards, Jerry L Rosemount Nuclear Instruments letter in reference to "Grand Gulf Nuclear Station message on INPO plant reports, subject Rosemount Instrument Setpoint Methodology, dated March 9,2000". Letter dated 04/0412000. (Attachment 3) 4.9.9 TOOl A169446-80. subject IISteam Carryover Fraction on Process Computer Heat Balance Calculations lt (Attachment 2)

Page 26 of 93

5.0 IDENTIFICATION OF COMPUTER PRQGRAMS Reactor Core Thermal Power LE-0113 Uncertainty Calculation Unit 1 Revision 0 The results of calculations by special computer programs were not directly used in this design analysis. MicrosoftO Office Excel 2003 SP 3 was used to confirm the arithmetic results.

6.0 METHOD OF ANALYSIS 6.1 METHODOLOGY The methodology used to calculate Section 6 is based on CC-MA-103-2001, "Setpoint Methodology for Peach Bottom Power Station and Limerick Generating Station" (Ref. 4.1.1).

These are non-safety-related indication loops, but the indication is used to calculate Core Thermal Power, which is a licensing limit. This analysis will use the Square Root of the Sum of the Squares (SRSS) methodology for combining the random and independent uncertainties. The dependent uncertainties will be combined according to their dependency relationships and biases will be algebraically summed in accordance with the Reference 4.1.1. The level of confidence for each uncertainty will be normalized to a 2a confidence level.

6.2 CORE THERMAL POWER (CTP) CALCULATION :

The process computer provides a calculation of the CTP based on a system heat balance, where CTP is the difference between the energy leaving the system and the energy input into the system from energy sources external to the core. The process computer steady state reactor heat balance equation is based on a summation of all heat sources raising the inlet feedwater and other cold water to steam exiting the pressure vessel. Figure 5-1 shows the Limerick heat balance control volume.

Page 27 of 93 Exelon.

Reactor Core Thermal Power Uncertainty Calculation Unit 1 LE-0113 Revision 0 5.0 IDENTIFICATION OF COMPUTER PROGRAMS The results of calculations by special computer programs were not directly used in this design analysis. Microsottefl) Office Excel 2003 SP 3 was used to confirm the arithmetic results.

6.0 METHOD OF ANALYSIS 6.1 METHODOLOGY The methodology used to calculate Section 6 is based on CC*MA-1 03-2001, ICSetpo;nt Methodology for Peach Bottom Power Station and Limerick Generating Station" (Ref. 4.1.1).

These are non-safety-rerated indication loops, but the indication is used to calculate Core Thermal Power, which is a licensing limit. This analysis wiU use the Square Root of the Sum of the Squares (SASS) methodology for combining the random and independent uncertainties. The dependent uncertainties will be combined according to their dependency relationships and biases will be algebraicarJy summed in accordance with the Reference 4.1.1. The level of confidence for each uncertainty wilf be normalized to a 20' confidence level.

6.2 CORE THERMAL POWER (CTP) CALCULATION:

The process computer provides a calculation of the CTP based on a system heat balance, where CTP is the difference between the energy leaving the system and the energy input into the system from energy sources external to the core. The process computer steady state reactor heat balance equation is based on a summation of all heat sources raising the inlet feedwater and other cold water to steam exiting the pressure vessel. Figure 5-1 shows the Limerick heat balance control vo'ume.

Page 27 of 93

Figure 5-1, limerick Heat Balance Control Volume Diagram QFW-IN QCRO4N QR Heat added by the recirculation pumps Qs-F.,,

Energy of steam from feedwater supply Reactor Core Thermal Power Uncertainty Calculation Unit 1 Revision Page 28 of 93 CTP = Energy out - Energy in (Equation 1)

Energy in = QFVV-1N + QCRD-N + QP (Equation 2)

Energy out = Qs.Fw + QCRa-OUT + QFiAD + ORWCU (Equation 3)

Where, CTP Core Thermal Power generated by nuclear fuel Energy of feedwater required to raise inlet FW to Steam Energy of CRD purge water and recirculation pump seal purge water going to feedwater Exel~n.

Reactor Core Thermal Power Uncertainty Calculation Unit 1


.Os..fW + QCFlQ.OUT te II QFWJn o

aCRO-IN

'-------fo--.---

cu Filter QAWCU Demo RWCU Heat Exchangers LE..0113 Revision 0 Figure 5-1, Limerick Heat Balance Control Volume Diagram CTP =Energy out - Energy in Energy in = OFW.JN + aeRO-IN + Qp Energy out = QS.FW + QCR()..()tJT+ ORAD +aRWCU

Where, (Equation 1)

(Equation 2)

(Equation 3)

CTP OFW.IN OCR().IN Op QS.FW Core Thermal Power generated by nuclear fuer Energy of feedwater required to raise inlet FW to Steam Energy of CRD purge water and recirculation pump seal purge water going to feedwater Heat added by the recirculation pumps Energy of steam from feedwater suppry Page 28 of 93 Exel~n.

Reactor Core Thermal Power Uncertainty Calculation Unit 1


.Os..fW + QCFlQ.OUT te II QFWJn o

aCRO-IN

'-------fo--.---

cu Filter QAWCU Demo RWCU Heat Exchangers LE..0113 Revision 0 Figure 5-1, Limerick Heat Balance Control Volume Diagram CTP =Energy out - Energy in Energy in = OFW.JN + aeRO-IN + Qp Energy out = QS.FW + QCR()..()tJT+ ORAD +aRWCU

Where, (Equation 1)

(Equation 2)

(Equation 3)

CTP OFW.IN OCR().IN Op QS.FW Core Thermal Power generated by nuclear fuer Energy of feedwater required to raise inlet FW to Steam Energy of CRD purge water and recirculation pump seal purge water going to feedwater Heat added by the recirculation pumps Energy of steam from feedwater suppry Page 28 of 93

QRAD QRWCU 6.2.1 Energy In ACRD-OUT 6.2.2 Energy out xelun.

6.2.2.1 Energy of Steam Qs-FW = WFw " hG(Ps) 6.2.2.2 Energy of CRD Purge Water Reactor Core Thermal Power LE-0113 Uncertainty Calculation Unit 1 Revision 0 Energy of CRD purge water and recirculation pump seal purge water going to steam Radiative heat losses from the reactor pressure vessel Heat removed by the RWCU system regenerative heat exchangers (includes both a heat removal term and a heat additions term)

Each of the above heat contributors are individually evaluated as follows :

6.2.1.1 Energy of feedwater (QFW.IN) is equal to the feedwater mass flow rate (wFw) multiplied by the enthalpy of the water at the bulk temperature of the feedwater (hFW(Tr-W)) entering the reactor.

Changes to the bulk temperature of the feedwater due to the influx of recirculation water and RWCU water are ignored because these mass flows represent less than 1 % of the total mass flow and the temperature change caused by their influx negligible.

QFw.w = wFw " hF(TF:W)

(Equation

4) 6.2.1.2 Energy of Control Rod Drive purge water and recirculation pump seal purge water (QcRD-,N) is taken to be fixed at the enthalpy of water for a given temperature and pressure.

QCFM-IN = WORD - hF(TcRo)

(Equation 5) 6.2.1.3 Energy of recirculation pumps (Qp)

Energy of recirculation pumps (Qp) is taken as the number of pump motors (n) multiplied by the efficiency of the pump motors (n,) multiplied by the power of the pump motors (WE). This is a conservative value because the combined net energy of the two recirculation pump motors contributes to the energy of the recirculation water. This estimate is fixed relative to CTP because relatively large random variations in Op will be negligible compared to QFw.

Op = n " rlm " WE (Equation 6)

Energy of steam from feedwater (Qs.Fw) is equal to the feedwater mass flow rate (WW) multiplied by the enthalpy (hG(Ps)) of the steam (hG) at the steam dome pressure (Ps). The moisture carryover mass fraction is conservatively set to 0 (See Attachment 2).

Page 29 of 93 (Equation 7)

In a Boiling Water Reactor (BWR), the energy of CRD purge water and recirculation pump seal purge water going to steam (QCAD-OUT) is equal to the mass flow rate of control rod drive and recirculation pump seal purge water (WCAD) multiplied by its enthalpy. However, the Feedwater mass flow rate will makes up greater than 99 % of the steam mass flow rate ; therefore, WORD will be quantified as a fixed number because relatively large random variations in WCRD will be negligible in determining Qs.

QcRD-OUT = wcRo - hG(Ps)

(Equation 8)

Exelon.

Reactor Core Thermal Power Uncertainty Calculation Unit 1 LE-Q113 Revision 0 Energy of CRD purge water and recirculation pump seal purge water going to steam Radiative heat losses from the reactor pressure vessel Heat removed by the RWCU system regenerative heat exchangers (incfudes both a heat removal term and a heat additions term) 6.2.1 6.2.1.1 6.2.1.2 6.2.1.3 6.2.2 6.2.2.1 6.2.2.2 Energy In Each of the above heat contributors are individually evaluated as follows:

Energy of feedwater (aFW.IN) is equal to the feedwater mass flow rate (WFW) multiplied by the enthalpy of the water at the buJk temperature of the feedwater (hFW(TFW)) entering the reactor.

Changes to the bulk temperature of the feedwater due to the influx of recirculation water and RWCU water are ignored because these mass flows represent less than 1 % of the totar mass flow and the temperature change caused by their influx negligible.

QFW.fN = WFW

  • hF(TFW)

(Equation 4)

Energy of Control Rod Drive purge water and recirculation pump seal purge water (OCRO-IN) is taken to be fixed at the enthatpy of water for a given temperature and pressure.

aCRO.IN =WCRO

  • hF(TCRO)

(Equation 5)

Energy of recirculation pumps (Op)

Energy of recirculation pumps (Op) is taken as the number of pump motors (n) multiplied by the efficiency of the pump motors (Flm) multiplied by the power of the pump motors (We). This is a conservative value because the combined net energy of the two recirculation pump motors contributes to the energy of the recirculation water. This estimate is fixed relative to CTP because relatively large random variations in Qp will be negligible compared to OFW.

Qp=n

  • 11m *We (Equation 6)

Energy out Energy of Steam Energy of steam from feedwater (as-FW) is equal to the feedwater mass flow rate \\'NFW) multiplied by the enthalpy (hG{Ps)) of the steam (ha) at the steam dome pressure (Ps). The moisture carryover mass fraction is conservatively set to 0 (See Attachment 2).

QS-FW = WFW

  • hG(Ps)

(Equation 7)

Energy of CRD Purge Water In a Boiling Water Reactor (BWR), the energy of CRD purge water and recirculation pump seal purge water going to steam (OCRo-QUr) is equal to the mass flow rate of control rod drive and recirculation pump seal purge water (WCRO) multiplied by its enthalpy. However, the Feedwater mass flow rate will makes up greater than 99 % of the steam mass flow rate; therefore, WCRO will be quantified as a fixed number because relatively farge random variations in WCAD will be negHgibfe in determining Os.

OeRo.ouT ::: WCRO

  • hG(Ps)

(Equation 8)

Page 29 of 93

6.2.2.3 Energy of Reactor Pressure Vessel Radiative Meat Loss Energy of reactor pressure vessel radiative heat loss (QRAD) includes both heat loss due to thermal radiation as well as heat loss through convection. This value is fixed relative to CTP because relatively large random variations in QRAD will be negligible in determining Qs.

6.2.2.4 Energy of Reactor Water Clean-up Energy of reactor water clean up (QRwcu) is based on the net heat removed by the non-regenerative heat exchanges from the recirculated water stream bypassed from Recirculation Loop B to the RWCU. The actual contribution to the heat removed from the reactor pressure vessel is negligible because the non-regenerative heat exchangers cool the stream going to the cleanup filters and demineralizers and the regenerative heat exchangers use the incoming stream to reheat the RWCU flow back up to the feedwater temperature. The mass flow rate is equal to the nominal pump flow rate the higher of Pump A or the combination of Pump B & C. The pump flow rate is fixed relative to CTP because relatively large random variations in QRwcu will be negligible in determining Qs. The net heat removed is equal to the mass flow rate (WRwcu) multiplied by difference of enthalpies across the RWCU.

QRWCU = WRWCU - [hF(TRPV) - hF(TFw)1 (Equation 9) 6.2.3 Neat Balance Equation The reactor heat balance is based on the principle that heat input to the reactor water equals heat out. Substituting Equations 2 and 3 into Equation 1 yields :

(QS.FW + ACRD-our + QRAD + QRwcu) = CTP + (QFW -IN + QDRD4N + QP)

(Equation 10)

Solving for CTP yields, CTP = (QS.FW + ACRD-OUT + QRAD + RWCU) - (QFw-IN + ACRD-IN -!- QP)

[WFW - hG(Ps) -r-WCRD - hG(Ps) + QRAD + WRWCU - [hF(TRPV) - hF(TFW)j

- [WFw - hF(TFW) + wcRD - hF(TCRD) + n. nm - W jj (Equation 11)

Combining like terms, Reactor Core Thermal Power LE-0113 Uncertainty Calculation Unit 1 Revision 0 CTP = ((WFw " hc(PS) - WFw - hF(TFW)l + (wcRD " ho(Ps) - wcRD " hF(Tcno)J

+ QRAD + WRwcu " [hr(TRPv),- hF(TFw)) - n - nm - WE)

(Equation 12)

CTP = wFw - [hG(PS) - hF(TFw)J + wcRD - [hc(ps) - hF(TcRD)1

+ QRAo + WRwcu " (hF(TRPv) - hF(TFw)J - n " nm - WE (Equation 13)

Equation 13 is the form of the equation used by the Plant Process Computer (PPC) to calculate CTP (Reference 4.8.2). The CTP can now be expressed as a function of WFW, hc(PS), QCRD.aur, QRAD, QRwcu, hF(TFw), wcRD-IN, QP.

Page 30 of 93 Exelon.

Reactor Core Thermal Power Uncertainty Calculation Unit 1 LE-0113 Revision 0 6.2.2.3 Energy of Reactor Pressure Vessel Radiative Heat Loss Energy of reactor pressure vessel radiative heat loss (ORAO) includes both heat loss due to thermal radiation as well as heat Joss through convection. This value is fixed relative to CTP because relatively large random variations in ORAD will be negligible in determining as.

6.2.2.4 Energy of Reactor Water Clean-up Energy of reactor water clean up (QRwCU) is based on the net heat removed by the non-regenerative heat exchanges from the recirculated water stream bypassed from Recirculation Loop B to the AWCU. The actual contribution to the heat removed from the reactor pressure vessel is negligible because the non-regenerative heat exchangers cool the stream going to the cleanup filters and demineralizers and the regenerative heat exchangers use the incoming stream to reheat the RWCU trow back up to the feedwater temperature. The mass flow rate is equal to the nominal pump flow rate the higher of Pump Aor the combination of Pump B& C. The pump flow rate is fixed relative to CTP because relatively farge random variations in QRWCU will be negligible in determining Os. The net heat removed is equal to the mass flow rate (WRWCU) multiplied by difference of enthalpies across the RWCU.

QRWCU = WRWCU' [hF(TAPV) - hF(TFW))

(Equation 9) 6.2.3 Heat Balance Equation The reactor heat balance is based on the principle that heat input to the reactor water equals heat out. Substituting Equations 2 and 3 into Equation 1 yields:

(QS.FW + QCRD-OUr+ ORAD + QRWCU) = CTP + (QfW.IN + OCRD..fN + Qp)

(Equation 10)

Solving for CTP yields, CTP =(QS.fW + OCRD.QUT+ QRAD + QRWCU) * (QFW.IN + OCRO-IN + Qp)

= [WFW

  • ha(Ps ) + WCRO' hG(Ps ) + ORAD + WRWCU' [hF(TRPV) - hF(TFW)]

- [WFW

  • hF(TFW) + WCRD
  • hF(TcRo) + n* 11m' WE]

Combining like terms, eTP =[(WFW

  • ho(Ps ) - WFW
  • hF(TFW)J + [WCRD' hG(Ps) - WCRD* hF(TcRD)]

+ ORAD + WRWCU * [hF(TRPV)' - hF(TFW)] - n

  • 11m *WE]

CTP = WFW * [hG{Ps ) - hF(TFW)] + WCRO *lhG(Ps ) - hF(TCRD)]

+ ORAD + WRWCU * [hF(TRPV) - hdTFW)]

  • n* 11m' WE (Equation 11)

(Equation 12)

(Equation 13)

Equation 13 is the form of the equation used by the Pfant Process Computer (PPC) to calculate CTP (Reference 4.8.2). The CTP can now be expressed as a function of WFW, hG(Ps), OCRO-OUT, ORAD.

QAWCU, hF(TFW), aCRo-lN, Qp.

Page 30 of 93

bcel6n.

Further simplification of Equation 14 can be made by setting variables with negligible input into the uncertainty of the CTP to constant values. The variables that have negligible contribution to the determination of the uncertainty of the heat from CTP are QCRD.OUT, QRAD, QRwcu, QCRD-IN, and Op as discussed above (Sections 6.2.2.2, 6.2.2.3, 6.2.2.4, 6.2.1.2, and 6.2.1.3).

Where, CTP = f (wFw, hc(Ps), ACRD-our, QRAo, Qpwcu, hF(TFw), QCRO-IN, QP)

CTP = CwFw

  • hG(P5) + QCRD-out + QRAD + Qawcul -- [wFw
  • hF(Trw) + QCRD-!N + QP)

WFW * [hG(PS) - hF(TFw)) + (QcRD-OUT - ACRD-IN) + QRna + QRWCU - QP 6.2.4 Uncertainty Determination From NURECICR-3659 (Ref. 4.9.4), the standard uncertainty (uc) of a function (y) containing multiple statistically independent terms may be expressed as follows:

Y=Axt,x2,

...,XN)

N N

Uc =

~ZICAXIY E

uJ'(Y}

1 (Equation 16) c, = TK and u,(y) = I q I u(y4 )

(Equation 17)

The standard uncertainty, uc, is multiplied by a coverage factor, k, which is equivalent to the number of standard deviations for a given confidence level to arrive at the measurement uncertainty U and the expected value of y, Y, is taken as y plus or minus the measurement uncertainty.

U = kuC(y) and Y = y :t U (Equation 18) 6.2.4.1 Standard Uncertainty for CTP The standard uncertainty for CTP can be determined by taking the square root of the sum of the squares of the partial derivatives of each subcomponent of CTP multiplied by the square of the uncertainties of the subcomponents as shown in Equation 19 (Reference 4.9.4).

CTP k

(V FW

)2 DCTP wFw

(

+

ah0(Pg)

~ ahctPsa

)

CTP

(

ahF (TFw )

`6hF(TFw )

2 aCTP aQ

~ ~60c~

 irr CAD _

. IN IaCTP r

aQRWGU

,~aawc1) u Reactor Core Thermal Power Uncertainty Calculation Unit 1 Revision 0 acTP I)Z CRD _ OUT

  • (010CRO. arr ~

t The partial derivative terms can be solved for by recalling CTP from Equation 15.

Page 31 of 93 (Equation 14)

(Equation 15)

(Equation 19)

LE-0113 Exelon.

Reactor Core Thermal Power Uncertainty Calculation Unit 1 LE-0113 Revision 0 (Equation 14)

Further simplification of Equation 14 can be made by setting variables with negligible input into the uncertainty of the CTP to constant values. The variables that have negligible contribution to the determination of the uncertainty of the heat from CTP are aCRO.OUTt ORAO, aRWcu, aCRO-IN, and Op as discussed above (Sections 6.2.2.2, 6.2.2.3, 6.2.2.4, 6.2.1.2, and 6.2.1.3).

CTP =[WFW

  • ho(Ps) + QCRO-QUT + ORAD + QRWCU] - [WFW
  • hF(TFW) + OCFIO-IN + Qp]

(Equation 15) 6.2.4 Uncertainty Determination From NUREG/CR..3659 (Ref. 4.9.4), the standard uncertainty (uc) of a function (y) containing multiple statistically independent terms may be expressed as foUows:

y =~X1t X2, ***, XN)

(Equation 16)

Where, at cl =-

a~ and Ul(y) = I Cj f u('<4)

(Equation 17)

The standard uncertainty, uc, is multiplied by a coverage factor, k, which is equivalent to the number of standard deviations for a given confidence level to arrive at the measurement uncertainty U and the expected varue of y, V, is taken as y plus or minus the measurement uncertainty.

U =kuc(y) and Y = Y+/- U (Equation 18) 6.2.4.1 Standard Uncertainty for CTP The standard uncertainty for CTP can be determined by taking the square root of the sum of the squares of the partial derivatives of each subcomponent of CTP multiplied by the square of the uncertainties of the subcomponents as shown in Equation 19 (Reference 4.9.4).

nsr*(uwp) ]+[( d~~~~)r*(uhG(Ps,f]+

[(Idh~~;:J*(UhF(TPN)f ]+[( dQ:~::UTr*hCAO~~f]+

[( ~~::INr.(UOCAO~'.f]+[(~r,(uOAAQf]+

[( ~~::ur*(UORWCJ]+[(~r.(uo.f]

The partial derivative terms can be solved for by recalling CTP from Equation 15.

Page 31 of 93 (Equation 19)

Exelon.

Reactor Core Thermal Power Uncertainty Calculation Unit 1 LE-0113 Revision 0 (Equation 14)

Further simplification of Equation 14 can be made by setting variables with negligible input into the uncertainty of the CTP to constant values. The variables that have negligible contribution to the determination of the uncertainty of the heat from CTP are aCRO.OUTt ORAO, aRWcu, aCRO-IN, and Op as discussed above (Sections 6.2.2.2, 6.2.2.3, 6.2.2.4, 6.2.1.2, and 6.2.1.3).

CTP =[WFW

  • ho(Ps) + QCRO-QUT + ORAD + QRWCU] - [WFW
  • hF(TFW) + OCFIO-IN + Qp]

(Equation 15) 6.2.4 Uncertainty Determination From NUREG/CR..3659 (Ref. 4.9.4), the standard uncertainty (uc) of a function (y) containing multiple statistically independent terms may be expressed as foUows:

y =~X1t X2, ***, XN)

(Equation 16)

Where, at cl =-

a~ and Ul(y) = I Cj f u('<4)

(Equation 17)

The standard uncertainty, uc, is multiplied by a coverage factor, k, which is equivalent to the number of standard deviations for a given confidence level to arrive at the measurement uncertainty U and the expected varue of y, V, is taken as y plus or minus the measurement uncertainty.

U =kuc(y) and Y = Y+/- U (Equation 18) 6.2.4.1 Standard Uncertainty for CTP The standard uncertainty for CTP can be determined by taking the square root of the sum of the squares of the partial derivatives of each subcomponent of CTP multiplied by the square of the uncertainties of the subcomponents as shown in Equation 19 (Reference 4.9.4).

nsr*(Uwp) ]+[( d~~~~)r*(uhG(Ps,f]+

[(Idh~~;:J*(UhF(TPN)f ]+[( dQ:~::UTr*hCAO~~f]+

[( ~~::INr.(UOCAO~'.f]+[(~r,(uOAAQf]+

[( ~~::ur*(UORWCJ]+[(~r.(uo.f]

The partial derivative terms can be solved for by recalling CTP from Equation 15.

Page 31 of 93 (Equation 19)

CTP = wFw - (hG(PS) - hF(TFw)) + AFO-OUT - 0CF04N) + ORAD + QRWCU - QP The partial derivatives are determined from Equation 15 and shown below, remembering that the terms 0CRD-0VT I QCAD.1N I OnAD I Qlqwcu I and Qp are fixed relative to the determination of the uncertainty of CTP.

'ICTP (hG (PS) - hF (TFW))

aWFW a CTP a ha(PS)

- wow a CTP DhATIV)

DCTP wn FW a OCRD,OUT (Equation 23) aCTP a QCRD-1N a CTP a () RAO (Equation 25)

- aCTP a QRWcU (Equation 26) a CTP a 0p The Feedwater mass flowrate uncertainty, oWFw, is equal to the measurement uncertainty as shown in Equation 28.

The steam enthalpy uncertainty, aho, is determined by Equation 29.

0 he (Ps ITS 1 10)

=

a ha

, 6T H +

a T Substituting finite differences for the partial derivatives :

ahG 2.

CFP 2 + (ah (3 )2 7-P) a la

(_G~h,, ) 2 2

(.dh,3 )

2 2 + (AhG

)2 crio 2

AT UT + AP CrP

-El -

0 The interpolation error for steam enthalpy, a lo = +/- 0.05 Btu/lbm.

Reactor Core Thermal Power LE-01 13 Uncertainty Calculation Unit I Revision 0 Page 32 of 93

. alo 9 (Equation 20)

(Equation 21)

(Equation 22)

(Equation 24)

(Equation 27)

(Equation 28)

(Equation 29)

(Equation 30)

Noting that for saturated steam, pressure determines the temperature of the steam ; thus, CFT =0 In addition, the steam tables used were derived from NEST Chemistry WebBook (Reference 19.3).

Exelc!)n.

Reactor Core Therma' Power Uncertainty Calculation Unit 1 LE..0113 Revision 0 (Equation 20)

CTP = WFW * (hG(Ps)

  • hF(TFW)) + (OeRO-ouT - QCfID.IN) + Q RAD + QRWCU - Qp The partial derivatives are determined from Equation 15 and shown below, remembering that the terms QCRD-otJr, OCRO-IN, ORAD, QAWCU, and Qp are fixed relative to the determination of the uncertajnty of CTP.

acTP

-- =(hG (ps) - hF (TFW ))

dwFW aCTP


w

() ho(Ps) -

FW aCTP ahF(TFW ) := -WFW aCTP

=1 aQCRO"OUT aCTP

=-1 aQCRO_'N aCTP =1 aQRAO aCTP

=1 iJQRWCU (Equation 21)

(Equation 22)

(Equation 23)

(Equation 24)

(Equation 25)

(Equation 26) aCTP =-1 aQp (Equation 27)

The Feedwater mass flowrate uncertainty, O'wf:W, is equal to the measurement uncertainty as shown in Equation 28.

(J'WFW =UFW

. W FW The steam enthalpy uncertainty, Uh(3, is determined by Equation 29.

)2

)2

()2 ahG 2

ahG 2

ahG 2

aha (Ps,Ts,I.) = (J'T"

  • aT

+(iiF'

  • ap

+

dlo

  • a~

Substituting finite differences for the partial derivatives:

(Equation 28)

(Equation 29)

(Equation 30)

Noting that for saturated steam, pressure determines the temperature of the steam; thus, 0T =0.

In addition, the steam tables used were derived from NIST Chemistry WebBook (Reference 4.9.3).

The interpolation error for steam enthalpy, 0\\ =+/- 0.05 Btu/Ibm.

Page 32 of 93

_ dho(Ps) 2 2

ahG (lo } 2 F

2 ar~(PS,10) -'

dP QP

+(

NO QF° s

a The Feedwater enthalpy uncertainty, 6hF, is determined by Equation 32.

2 2

0h,(TFWIPFW'lo)=

ahF

  • aT2 + ahF
  • CrP2 + ('IF) *a1Q 2

~(aT aP

()

atO Substituting finite differences for the partial derivatives yields The interpolation error for liquid enthalpy, ab = :t 0.005 BTUIb, and that the enthalpy of subcooled water varies with temperature and slightly with pressure ; thus, ahF (Trw,PFw,la)

Q a"O = XAAD %

- QRAD 6.2.5 Extended Instrument Drift 2

2 dhF

  • CF j

z + dhF

  • a 2 + dhF T

j P

10 AT AP (7QcP'O_ our xcAD %

" OCRU-OUT aQP T xp % - Op (Equation 38)

Reactor Care Thermal Power Uncertainty Calculation Unit i Revision 0 o T a + dhF (P)

( AP (Equation 31)

(Equation 32) 2 d

z e

h j z

ap +

F

  • aJ ° (Equation 33) d o The following uncertainty terms are relatively insignificant in the determination of CTP uncertainty; therefore, it is only necessary to quantify the uncertainty to within a reasonably conservative value.

Refinement of the uncertainty of these items after this initial determination is not required given that the uncertainty is within the tolerances shown in the sensitivity analysis (See Attachment 7).

The Control Rod Drive (CR©) outlet energy uncertainty, aoeao_our, is determined by Equation 34, where xcm is the uncertainty calculated for the GRD flaw stream.

(Equation 34)

The CRD inlet energy uncertainty, o'acRO.Inr, is determined by Equation 35, where xcaQ is the uncertainty calculated for the CRD flow stream.

Cr 0"0 - "

' xcr,c % " QcRD_JN (Equation 35)

The reactor pressure vessel heat loss uncertainty, crQpAp, is determined by Equation 36, where xFAD is the uncertainty assigned to the heat loss value calculated by LM-0553 (Reference 4.$.3).

(Equation 36)

The RWCU heat removal uncertainty, dQpwcu, is determined by Equation 37, where xPwcu is the uncertainty calculated for the RWCU heat balance terms.

LE-01 13 a°A"CU - xRwcu °la " Qawcu (Equation 37)

The Recirculation Pump heat addition uncertainty, aov, is determined by Equation 38, where xP is the uncertainty calculated far measurement of the recirculation pump motor power.

Instrument drift specifications are usually published for a defined period of time. The instrument drift for one period of time is independent from the instrument drift of any other equivalent period of time.

Therefore, the drift specification D far a period of X months can be expanded to n "X months (where n Page 33 of 93 Exelc.;n.

Reactor Core Thermal Power Uncertainty Carculation Unit 1 LE-D113 Revision 0 (P I ) _

(~hG(Ps))2*

2 (LihG(lo))2.

2 O'ha s,o -

liP O'p +

Lil 0'10 S

0 The Feedwater enthalpy uncertainty, (ThF, is determined by Equation 32.

(Equation 31 )

(Equation 32)

(Equation 33)

(

LihF)2 2

(AhF)2 2

(AhF)2 2

=

liT

  • OT

+

AP

  • O'p

+

Ala

  • 0'10 The interpolation error for liquid enthalpy, 0'10 = :t 0.005 BTU/lb, and that the enthalpy of subcooled water varies with temperature and slightly with pressure; thus,

(

Lih (T>>)2 (Lih (p>>)2 (ah )2 0'

('Tt PI) =

F

  • 0 2 +

F

.0' 2 + __F

  • 0 2

hF FW I FW, 0 AT T

liP P

ala

'0 The following uncertainty terms are relatively insignificant in the determination of CTP uncertainty; therefore, it is on'y necessary to quantify the uncertainty to within a reasonably conservative value.

Refinement of the uncertainty of these items after this initial determination is not required given that the uncertainty is within the tolerances shown in the sensitivity analysis (See Attachment 7).

The Control Rod Drive (CAD) outlet energy uncertainty, O'acRD~OUT, is determined by Equation 34, where XCAO is the uncertainty calculated for the CRD flow stream.

a QCRO_ OUT XCRD %

  • QCRO_OUT (Equation 34)

The CRD fnret energy uncertainty, (TQCRO,,)Nt Is determined by Equation 35, where XCAO is the uncertainty calculated for the CAD flow stream.

(]

aeRO << IN XCAO %

  • OCROjN (Equation 35)

The reactor pressure vessel heat loss uncertainty, O'QRAD, is determined by Equation 36, where XRAO

s the uncertainty assigned to the heat loss value calcufated by LM-0553 (Reference 4.8.3).

(J QRA(I = XRAO o~

  • ORAD (Equation 36)

The AWCU heat removal uncertainty. (TQRWCU, is determined by Equation 37, where XAWCU is the uncertainty calculated for the RWCU heat balance terms.

(JQRWCU = XAWCU %

  • ORWCU (Equation 37)

The Recirculation Pump heat addition uncertainty, (Top, is determined by Equation 38, where Xp is the uncertainty calculated for measurement of the recirculation pump motor power.

°ap = Xp %. Qp (Equation 38) 6.2.5 Extended Instrument Drift Instrument drift specifications are usually published for a defined period of time. The instrument drift for one period of time is independent from the instrument drift of any other equivalent period of time.

Therefore, the drift specification D for a period of Xmonths can be expanded to n*X months (where n Page 33 of 93 Exelc.;n.

Reactor Core Thermal Power Uncertainty Carculation Unit 1 LE-D113 Revision 0 (P I ) _

(~hG(Ps))2*

2 (LihG(lo))2.

2 O'ha s,o -

liP O'p +

Lil 0'10 S

0 The Feedwater enthalpy uncertainty, (ThF, is determined by Equation 32.

(Equation 31 )

(Equation 32)

(Equation 33)

(

LihF)2 2

(AhF)2 2

(AhF)2 2

=

liT

  • OT

+

AP

  • O'p

+

Ala

  • 0'10 The interpolation error for liquid enthalpy, 0'10 = :t 0.005 BTU/lb, and that the enthalpy of subcooled water varies with temperature and slightly with pressure; thus,

(

Lih (T>>)2 (Lih (p>>)2 (ah )2 0'

('Tt PI) =

F

  • 0 2 +

F

.0' 2 + __F

  • 0 2

hF FW I FW, 0 AT T

liP P

ala 10 The following uncertainty terms are relatively insignificant in the determination of CTP uncertainty; therefore, it is on'y necessary to quantify the uncertainty to within a reasonably conservative value.

Refinement of the uncertainty of these items after this initial determination is not required given that the uncertainty is within the tolerances shown in the sensitivity analysis (See Attachment 7).

The Control Rod Drive (CAD) outlet energy uncertainty, O'acRD~OUT, is determined by Equation 34, where XCAO is the uncertainty calculated for the CRD flow stream.

a QCRO_ OUT XCRD %

  • QCRO_OUT (Equation 34)

The CRD fnret energy uncertainty, (TQCRO,,)Nt Is determined by Equation 35, where XCAO is the uncertainty calculated for the CAD flow stream.

(]

aeRO << IN XCAO %

  • OCROjN (Equation 35)

The reactor pressure vessel heat loss uncertainty, O'QRAD, is determined by Equation 36, where XRAO

s the uncertainty assigned to the heat loss value calcufated by LM-0553 (Reference 4.8.3).

(J QRA(I = XRAO o~

  • ORAD (Equation 36)

The AWCU heat removal uncertainty. (TQRWCU, is determined by Equation 37, where XAWCU is the uncertainty calculated for the RWCU heat balance terms.

(JQRWCU = XAWCU %

  • ORWCU (Equation 37)

The Recirculation Pump heat addition uncertainty, (Top, is determined by Equation 38, where Xp is the uncertainty calculated for measurement of the recirculation pump motor power.

°ap = Xp %. Qp (Equation 38) 6.2.5 Extended Instrument Drift Instrument drift specifications are usually published for a defined period of time. The instrument drift for one period of time is independent from the instrument drift of any other equivalent period of time.

Therefore, the drift specification D for a period of Xmonths can be expanded to n*X months (where n Page 33 of 93

Reactor Core Thermal Power LE-0113 Uncertainty Calculation Unit 1 Revision 0 is the station surveillance interval divided by the vendor drift interval and n is an integer greater than zero) by the SRSS method. Instrument drift for surveillance intervals exceeding the instrument suppliers' specified drift Interval is calculated using Equation 39 (Section 4.1.1 of reference 4.1.1,).

Dr, = [n " (D.)2]112 (Equation 39)

Page 34 of 93 Exelon.

Reactor Core Thermal Power Uncertainty Calculation Unit 1 LE-Q113 Revision 0 is the station surveiJtance interval divided by the vendor drift interval and n Is an integer greater than zero) by the SASS method. Instrument drift for surveillance intervals exceeding the instrument suppliers' specified drift interval is calculated using Equation 39 (Section 4.1.1 of reference 4.1.1 ~).

Dn = (n. (Dx)2fl2 (Equation 39)

Page 34 of 93

Exelon, 7.0 NUMERIC ANALYSIS 7.1 FEEDWATER FLOW UNCERTAINTY The uncertainty of the feedwater mass flow rate measurement for the Caldon(b Ultrasonics Leading Edge Flow Meter Check Plus (LEFMV+) system is taken from Reference 4.8.6, Section 2.0, using Equation 28.

Or WFW - UFw measurement

  • w Fw = 0.32 %
  • w Fw

= 0.0032

  • 15,090,000 Ibmlhr 48,288 lbmlhr Reactor Core Thermal Power LE-0113 Uncertainty Calculation Unit 1 Revision 0 7.2 STEAM DOME PRESSURE MEASUREMENT UNCERTAINTY The uncertainty of the steam dome pressure measurement is taken from Reference 4.8.8.

csns W t 20 psig 7.3 REACTOR WATER CLEAN-UP (RWCU) FLOW LOOP UNCERTAINTY 7.3.1 RWCU Flow Loop Accuracy (LAR,rcu_Fjow) 7.3.1.1 RWCU Flow Element Reference Accuracy (A1)

Reference Accuracy is specified as f 1.50% of actual rate of flow (Section 2.2.3).

A1 2, t 1.50% Flow 7.3.1.2 RWCU Flow Element Installation Effect (IE1)

The flow elements meet the installation requirements of Ref. 4.6.5. Therefore, IE1

=

  • 0.50% Flow 7.3.1.3 RWCU Flow Element Temperature Effect on Flow Element Expansion (TN1)

Per section 2.2.3, the maximum temperature of the water passing through the flow element is 582

°F and the normal temperature is 539 °F with the flow elements located just upstream of the RWCU Recirculation Pumps. Since the system temperature operation band is small, there is a minor change in the flow element expansion factor. The change is in order of 0.003 inches or less for the temperature range of 515 '"F to 560 OF. Therefore the temperature effect on flow element expansion can be neglected.

TN I

= t 0% Flow 7.3.1.4 RWCU Flow Element Temperature Effect on Density (TD1)

During normal operations the temperature band for the fluid passing through the flow element at the inlet of the RWCU system is approximately 530 °F. The effect temperature has on density will be evaluated at t 4.37 aF (Section 7.4.6) from the base condition of 530 OF at 1060 prig at the flow element (Section 2.2.3). Density is relatively constant at 1060 psig ; however, for conservatism a variation of

  • 10 psig is used to show that the pressure effect is negligible even at four times the nominally specified transmitter reference accuracy of 0.25 %. The uncertainty in the density, ap(T,P), as a function of temperature and pressure will follow Equation 33.

Page 35 of 93 Reactor Core Thermal Power Uncertainty Calculation Unit 1 LE-0113 Revision 0 7.0 NUMERIC ANALYSIS 7.1 FEEDWATER FLOW UNCERTAINTY The uncertainty of the feedwater mass flow rate measurement for the Caldon Ultrasonics Leading Edge Flow Meter Check Plus (LEFM.,I+) system is taken from Reference 4.8.6, Section 2.0, using Equation 28.

O'Wr:w =UFWmeasurement

  • WFW =0.320/0
  • Wr:w

= 0.0032 *15j 090,OOO IbmJhr

=: 48,288 Ibmlhr 7.2 STEAM DOME PRESSURE MEASUREMENT UNCERTAINTY The uncertainty of the steam dome pressure measurement is taken from Reference 4.8.8.

O'Ps = +/- 20psig 7.3 7.3.1 7.3.1.1 7.3.1.2 7.3.1.3 7.3.1.4 REACTOR WATER CLEAN-UP (RWCU) FLOW LOOP UNCERTAINTY RWCU Flow Loop Accuracy (LARWCU_FJow)

RWCU Flow Element Reference Accuracy (A1)

Reference Accuracy is specified as +/- 1.50% of actual rate of flow (Section 2.2.3).

A12<1

=

+/- 1.50% Flow RWCU Flow Element fnstanatlon Effect (IE1)

The flow elements meet the installation requirements of Ref. 4.6.5. Therefore j IE1

=:t: 0.50% Flow RWCU Flow Element Temperature Effect on Flow Element Expansion (TN1)

Per section 2.2.3, the maximum temperature of the water passing through the flow element is 582 of and the normal temperature is 539 of with the flow elements located just upstream of the RWCU Recirculation Pumps. Since the system temperature operation band is small. there is a minor change in the flow element expansion factor. The change is in order of 0.003 inches or less for the temperature range of 515 of to 560 OF. Therefore the temperature effect on flow element expansion can ba neglected.

TN1

=+/-ook Flow RWCU Flow Element Temperature Effect on Density (TD1)

During normal operations the temperature band for the fluid passing through the flow element at the inlat of the RWCU system ;s approximately 530 of. The effect temperature has on density wUl be evaluated at :t 4.37 of (Section 7.4.6) from the base condition of 530 (tF at 1060 psig at the flow element (Section 2.2.3). Density is relatively constant at 1060 psig; however, for conservatism a variation of :t 10 psig is used to show that the pressure effect is negligible even at four times the nominally specified transmitter reference accuracy of 0.25 0/0. The uncertainty in the density, O'p(TIP), as a function of temperature and pressure wUl follow Equation 33.

Page 35 of 93

ap CTRWCU cs, - 2

  • a, (p(534.4) - p(525.6)1z 534.4-- 525.6 r p(1070) - p(1050) 12 i

11070070--105 1050 psi 47.046 -- 47.612 534.4 - 525.6 0.26307 Ibm Reactor Core Thermal Power LE-01 1 3 Uncertainty Calculation Unit 1 Revision 0 z

( Ibm ft 3

°F

°F Ibrn 47.339-47.326 z 7 * (10 si 2 1070 --1050 psi p

Page 36 of 93

  • (4.4 `F)'
  • (10 psi ),

Ibm ft3

  • (4.4 OF)' +

fr Divide by the base density to convert to a percentage value c',' tT xtVcU, PI o 0.283 @ 530V= 0.00598 =0.598 %

~

47.333 Ibm V psi

  • (10 psi Y The percent change in density, a,,, as a function of the percent change in flow, a,, is given by the following equation (Reference Attachment 8, Equation A8-10) :

Exelon.

Reactor Core Thermal Power Uncertainty Calculation Unit 1

[

lbm]2

(

P(534.4) - P(525.6>>)2 jt3

  • (4.4 "FY +

534.4-525.6 of

[

Ibm]2

(

P(1070) - P0050))2 jt'

  • (10 psi)l 1070 - 1050 psi LE-Q113 Revision 0

[

lbm]2

(

47.046 - 47.612)2 jt3

  • (4.4 0FY +

534.4 - 525.6 0 F

=

[

lbm]2

( 47.339-47.326)2 jt'

  • (10 psiY 1070 - 1050 psi

[

ibm]2

[lhm]2

= (-0.57)2fi3"

  • (4.40Fy+(O.Ol)"jt'.
  • (lOpsi)"

8.8 of 20 pSI

= 0.08009 (Ibm)2 +0.00004 (ibm)2 ft3 ft3 Ibm 0'jJ (TFlV ) ::: 0.28307--r

/t Dividebythe basedensity toconvert to a percentage value O'p(TRWCU,~o)::: 0.283 @530°F::: 0.00598 = 0.598 %

47.333 The percent change in density, O'x, as a function of the percent change in flow, 01, is given by the following equation (Reference Attachment 8, Equation A8-10):

Page 36 of 93 Exelon.

Reactor Core Thermal Power Uncertainty Calculation Unit 1

[

lbm]2

(

P(534.4) - P(525.6>>)2 ft3

  • (4.4 "FY +

534.4-525.6 of

[

Ibm]2

(

P(1070) - P0050))2 ft'

  • (10 psi)l 1070 - 1050 psi LE-Q113 Revision 0

[

lbm]2

(

47.046 - 47.612)2 ft3

  • (4.4 0FY +

534.4 - 525.6 0 F

=

[

lbm]2

(

47.339-47.326)2 ft'

  • (10 psiY 1070 - 1050 psi

[

ibm]2

[lhm]2

= (-0.57)2fi3"

  • (4.40Fy+(O.Ol)"ft'.
  • (lOpsi)"

8.8 of 20 pSI

= 0.08009 (Ibm)2 +0.OO4 (ibm)2 ft3 ft3 Ibm U jJ (TFlV ) ::: 0.28307--r It Dividebythe basedensity toconvert to a percentage value Up (TRWCU,~o)::: 0.283 @530°F::: 0.00598 = 0.598 %

47.333 The percent change in density, O'x, as a function of the percent change in flow, 01, is given by the following equation (Reference Attachment 8, Equation A8-10):

Page 36 of 93

Exele)

J Rearranging Equation A8-10 and substituting oa for a, to solve for the unknown flow uncertainty, cs, :

TDl %

TD1

m 7.3.1.5 RWCU Flow Element Humidity Error (e1H)

The flow element is a mechanical device installed within the process. Therefore, humidity effects are not applicable.

e1H

=0 7.3.1.6 RWCU Flow Element Radiation Error (e1 R)

The flow element is a mechanical device installed within the process. Therefore, radiation effects are not applicable.

e1 R

=

0 e1S

=

0

. 0.598%

2

.0.299 %

0.3 % of flow 7.3.1.7 RWCU Flow Element Seismic Error (e1 S) 7.3.1.8 RWCU Flow Element Static Pressure Offset Error (el SP)

Reactor Core Thermal Power Uncertainty Calculation Unit 1 Revision 0 For normal error analysis, normal vibrations and seismic effects are considered negligible or capable of being calibrated out. Therefore, there is no seismic error for normal operating conditions (Section 3.5).

The flow element is a mechanical device installed within the process. Therefore, static pressure effects are not applicable.

e1SP =

0 7.3.1.9 RWCU Flow Element Ambient Pressure Error (e1 P)

The flow element is a mechanical device installed within the process. Therefore, therefore the flow element is not subject to ambient pressure variations.

e1P

=

0 7.3.1.10 RWCU Flow Element Process Error (e1 Pr)

Any process errors have been accounted for as errors associated with Temperature Effect on Density. Therefore, e1 Pr

=

0 7.3.1.11 RWCU Flow Element Temperature Error (e1 T)

Temperature error Is considered to be a random variable for the flow elements and is addressed under Temperature Effect on Flow Element Expansion (Section 7.3.1.3). Therefore, e1T 0

Page 37 of 93 LE-0113 Exelon.

Reactor Core Thermal Power Uncertainty Calculation Unit 1 LE-0113 Revision 0 Rearranging Equation A8-10 and substituting O'p forO'x to solve for the unknown flow uncertainty, 0'1:

TD]~.f&tw I

= _. TDl~dmrftr-2

~.!..O.598 %

2

0.299 %

=

0.3 % of flow 7.3.1.5 RWCU Flow Element Humidity Error (elH)

The flow erement Is a mechanical device installed within the process. Therefore, humidity effects are not applicable.

elH

=0 7.3.1.6 RWCU Flow Element Radiation Error (e1 R)

The flow element is a mechanical device installed within the process. Therefore, radiation effects are not applicable.

alA

=

0 7.3.1.7 RWCU Flow Element Seismic Error (e1S)

For normal error analysis, normal vibrations and seismic effects are considered negligible or capable of befng calibrated out. Therefore, there is no seismic error for normal operating conditions (Section 3.5).

alB

=

0 7.3.1.8 AWCU Flow Element Static Pressure Offset Error (e1 SP)

The flow element is a mechanical device installed within the process. Therefore, static pressure effects are not applicable.

e1SP

=

0 7.3.1.9 RWCU Flow Element Ambient Pressure Error (e1 P)

The flow element is a mechanicar device installed within the process. Therefore, therefore the flow element is not sublect to ambient pressure variations.

alP

=

0 7.3.1.10 RWCU Flow Element Process Error (el Pr)

Any process errors have been accounted for as errors associated with Temperature Effect on Density. Therefore, elPr

=

0 7.3.1.11 RWCU Flow Element Temperature Error (e1T)

Temperature error Is considered to be a random variable for the flow elements and is addressed under Temperature Effect on Flow Element Expansion (Section 7.3.1.3). Therefore, elT

I 0

Page 37 of 93 Exelon.

Reactor Core Thermal Power Uncertainty Calculation Unit 1 LE-0113 Revision 0 Rearranging Equation A8-10 and substituting O'p forO'x to solve for the unknown flow uncertainty, 0'1:

TD]~.f&tw I

= _. TDl~dmrftr-2

~.!..O.598 %

2

0.299 %

=

0.3 % of flow 7.3.1.5 RWCU Flow Element Humidity Error (elH)

The flow erement Is a mechanical device installed within the process. Therefore, humidity effects are not applicable.

elH

=0 7.3.1.6 RWCU Flow Element Radiation Error (e1 R)

The flow element is a mechanical device installed within the process. Therefore, radiation effects are not applicable.

alA

=

0 7.3.1.7 RWCU Flow Element Seismic Error (e1S)

For normal error analysis, normal vibrations and seismic effects are considered negligible or capable of befng calibrated out. Therefore, there is no seismic error for normal operating conditions (Section 3.5).

alB

=

0 7.3.1.8 AWCU Flow Element Static Pressure Offset Error (e1 SP)

The flow element is a mechanical device installed within the process. Therefore, static pressure effects are not applicable.

e1SP

=

0 7.3.1.9 RWCU Flow Element Ambient Pressure Error (e1 P)

The flow element is a mechanicar device installed within the process. Therefore, therefore the flow element is not sublect to ambient pressure variations.

alP

=

0 7.3.1.10 RWCU Flow Element Process Error (el Pr)

Any process errors have been accounted for as errors associated with Temperature Effect on Density. Therefore, elPr

=

0 7.3.1.11 RWCU Flow Element Temperature Error (e1T)

Temperature error Is considered to be a random variable for the flow elements and is addressed under Temperature Effect on Flow Element Expansion (Section 7.3.1.3). Therefore, elT

I 0

Page 37 of 93

Exelon LE-0113 Uncertainty Calculation Unit 1 Revision 0 Reactor Core Thermal Power 7.3.1.12 RWCU Flow Transmitter Reference Accuracy (A2)

Reference Accuracy Is specified as +/- 0.25 °Ia of span considered to be a 3a value (Section 2.2.4).

The reference accuracy is set to the calibration accuracy per plant procedure (Reference 4.1.1).

A23,,

t 0.50 % (3a]

Converting to a 2a value A22a

+/- 0.50 °la

  • 2 / 3 A22a t 0.3333 a!a of span Converting °Ja span to °!a flow Rearranging Equation A8-10, app = 2
  • aFi ow, to solve for flow (Reference Attachment 8) :

A22a%Fbvv A2%span / 2 1 0.3333 % of span / 2 t 0.1667 % of Flow A22a%Flow t 0.1667 °la of Flow 7.3.1.13 RWCU Flow Transmitter Power Supply Effects (a2PS2)

Power supply effects are considered to be negligible (Section 3.7). Therefore, a2PS2 =

t 0 7.3.1.14 RWCU Flow Transmitter Ambient Temperature Error (a2T)

The temperature effect is +/- (0.75 °la URL + 0.5 % span)/100°F (3a] (Section 2.2.4). The maximum temperature at the transmitter location is 106 °F, and minimum temperature during calibration could be 65 IF, so the maximum difference = 106 - 65 OF = 41 IF (Section 2.3.4) a2T~ =

+/- [(0.0075

  • 750 INWC + 0.005
  • 220 INWC)
  • 41 IF/ 100 OF]

=

+/- [(5.625 INWC + 1.1 INWC)

  • 0.41 ]

a2T3,, =

t 2.76 INWC [3a], rounded to level of significance Converting to a 2a value a2T2a = t 2.76 INWC

  • 2 / 3 o2T2a

=

+/- 1.8382 I NW C Converting to % span a2T2a =

t 1.8382 INWC / 220 INWC

-=

t 0.008355 = 0.8355 %

Converting °la span to a/n flow Rearranging Equation A8-10, ajp = 2

  • an_ow, to solve for flow (Reference Attachment 8) :

o-2T2a%Fjow a2T2o%pow Page 38 of 93

=

a2T2a%span / 2

+/- 0.8355 % / 2 t 0.4178 %

=

+/- 0.418 °la Exelc;n.

Reactor Core Thermal Power Uncertainty Calculation Unit 1 LE*0113 Revision 0 7.3.1.12 RWCU Flow Transmitter Reference Accuracy (A2)

Reference Accuracy is specified as +/- 0.25 % of span considered to be a 30' value (Section 2.2.4).

The reference accuracy is set to the calibration accuracy per plant procedure (Reference 4.1.1).

A230

=

I: 0.50 % [30']

Converting to a 20' value A220

=

+/- 0.50 %

  • 2 /3 A220

=

+/- 0.3333 % of span Converting % span to % flow Rearranging Equation A8-10, a/1p =2

  • O'FLOW, to solve for flow (Reference Attachment 8):

A22o%FIow

=

A2%Span / 2

=

I: 0.3333 % of span /2

=

+/- 0.1667 % of Flow A22o%Flow

=

+/- 0.1667 % of Flow 7.3.1.13 RWCU Flow Transmitter Power Supply Effects (a2PS2)

Power supply effects are considered to be negligible (Section 3.7). Therefore, a2PS2 =

1:0 7.3.1.14 RWCU Flow Transmitter Ambient Temperature Error (a2T)

The temperature effect is +/- (0.75 % UAL + 0.5 % span)/100°F [30'] (Section 2.2.4). The maximum temperature at the transmitter location is 106 OF, and minimum temperature during calibration could be 65 bF, so the maximum difference =106.... 65 \\f>F = 41 l)F (Section 2.3.4) o2T30 =

+/- [(0.0075

  • 750 INWC + 0.005
  • 220 INWC)
  • 41 ~F/1 00 oF]

=

+/- [(5.625 fNWC + 1.1 INWC)

  • 0.41]

o2T3cr =

+/- 2.76 INWC [3c1], rounded to level of significance Converting to a 20 value a2T2o

=

+/- 2.76 INWC

  • 2/3 a2T20 =

+/- 1.8382 INWC Converting to Ok span a2T20 =

+/- 1.8382 fNWC 1220 INWC

=

+/- 0.008355 =0.8355 Ok Converting % span to % flow Rearranging Equation A8-10, a/1P =2* anow, to solve for flow (Reference Attachment 8):

a2T2o%Flow

=

0'2T20'%Sp8n 12

=

+/- 0.8355 % I 2

=

+/- 0.4178 0/0 02T2o%Ffow =

+/- 0.418 %

Page 38 of 93

Exeloln.

7.3.1.15 RWCU Plow Transmitter Humidity Error (e2H)

The manufacturer specifies the transmitter operating humidity limits between 0 and 100 % RH (Section 2.2.4). The transmitter is located in Containment H2 Recombiner Room 506C, Area 16 where humidity may vary from 50 to 90% RH (Reference 4.4.2). Humidity errors are set to zero because they are considered to be included within the reference accuracy specification under these conditions. (Section 3.2) e2H

=

0 7.3.1.16 RWCU Flow Transmitter Radiation Error (e2R)

The manufacturer specifies the transmitter operating radiation effect during and after exposure to 2.2 x 107 rads TO (Section 2.2.4). The transmitter is located in Containment H2 Recombiner Room 5060, Area 16, where total integrated dose (TID) could be as high as 8.78 x 102 rad (Reference Section 4.4.2). Therefore, e2R e2R Converting to % span e2R%S

=

t 1.20 x 10-31NWC / 220 INWC

= +/-5

.44x 1O=5.44x 10,4 %

Converting % span to % flow Rearranging Equation A8-10, crap = 2

  • arLo ,, to solve for flow (Reference Attachment 8) :

e2R e2R

, / 2 t5.44x 10'° %/2 t 2.72 x 10'4

= t 2.72 x 10,4%, which is negligible e2R %F,,

=

0 7.3.1.17 RWCU Flow Transmitter Seismic Error (e2S)

Reactor Core Thermal Power LE-01'13 Uncertainty Calculation Unit 1 Revision 0 The transmitter's accuracy is within t 0.5% of URL (upper range limit) during and after a seismic disturbance defined by a required response spectrum with a ZPA of 4 g's. A seismic event defines a particular type of accident condition. Therefore, there is no seismic error for normal operating conditions (Section 3.5) e2S

=

0 7.3.1.18 RWCU Flow Transmitter Ambient Pressure Error (e2P)

The flow transmitter is an electrical device and therefore not affected by ambient pressure.

e2P

=

0 7.3.1.19 RWCU Flow Transmitter Temperature Error (e2T)

Temperature error is considered to be a random variable for a Rosemount transmitter. Therefore e2T 0

Page 39 of 93 t (4.0 % of URL)

  • dose / rated dose t (0.04.750 INWC)
  • 8.78 x 102 / 2.2 x 107 t30INWC*3.99x10`5 t 1.20 x 10"3 INWC Exelon.

Reactor Core Thermal Power Uncertainty Calculation Unit 1 LE-0113 Revision 0 7.3.1.15 RWCU Flow Transmitter Humidity Error (e2H)

The manufacturer specifies the transmitter operating humidity limits between 0 and 100 % RH (Section 2.2.4). The transmitter is located in Containment H2 Recombiner Room 506C, Area 16 where humidity may vary from 50 to 90% RH (Reference 4.4.2). Humidity errors are set to zero because they are considered to be included within the reference accuracy specification under these conditions. (Section 3.2) e2H

=

0 7.3.1.16 RWCU Ffow Transmitter Radiation Error (e2R)

The manufacturer specifies the transmitter operating radiation effect during and after exposure to 2.2 x 107 rads TID (Section 2.2.4). The transmitter is located in Containment H2 Recombiner Room 506C, Area 16, where total integrated dose (TID) could be as high as 8.78 x 102 rad (Reference Section 4.4.2). Therefore, e2R

=

+/- (4.0 % of URL)

  • dose I rated dose

=

+/- (0.04 *750 INWC)

  • 8.78 x 102 /2.2 x 107

=

+/- 30 INWC

  • 3.99 x 10-5 e2R

=

t: 1.20 x 10*3 INWC Converting to % span e2Rcx,Span

=

+/- 1.20 x 10~3fNWC / 220 INWC

=

+/- 5.44 x 10-6 = 5.44 X 10-4 0/0 Converting % span to % flow Rearranging Equation A8-1 0, O'AP = 2

  • O'FlOW, to solve for flow (Reference Attachment 8):

e2R %Aow

=

e2R %Span I 2

=

+/- 5.44 x 10.4 °/012

+/- 2.72 x 10.4 %

=

+/- 2.72 x 10.4 %, which is negligible e2R%FIow 0

7.3.1.17 AWCU Flow Transmitter Seismic Error (e28)

The transmitter's accuracy is within +/- 0.50/0 of UAL (upper range limit) during and after a seismic disturbance defined by a required response spectrum with a ZPA of 4 g's. A seismic event defines a particular type of accident condition. Therefore, there is no seismic error for normal operating conditions (Section 3.5) e2S

=

0 7.3.1.18 AWCU Flow Transmitter Ambient Pressure Error (e2P)

The flow transmitter is an electr;cal device and therefore not affected by ambient pressure.

e2P

=

0 7.3.1.19 RWCU Flow Transmitter Temperature Error (e2T)

Temperature error;s considered to be a random variable for a Rosemount transmitter. Therefore e2T

=

0 Page 39 of 93 Exelon.

Reactor Core Thermal Power Uncertainty Calculation Unit 1 LE-0113 Revision 0 7.3.1.15 RWCU Flow Transmitter Humidity Error (e2H)

The manufacturer specifies the transmitter operating humidity limits between 0 and 100 % RH (Section 2.2.4). The transmitter is located in Containment H2 Recombiner Room 506C, Area 16 where humidity may vary from 50 to 90% RH (Reference 4.4.2). Humidity errors are set to zero because they are considered to be included within the reference accuracy specification under these conditions. (Section 3.2) e2H

=

0 7.3.1.16 RWCU Ffow Transmitter Radiation Error (e2R)

The manufacturer specifies the transmitter operating radiation effect during and after exposure to 2.2 x 107 rads TID (Section 2.2.4). The transmitter is located in Containment H2 Recombiner Room 506C, Area 16, where total integrated dose (TID) could be as high as 8.78 x 102 rad (Reference Section 4.4.2). Therefore, e2R

=

+/- (4.0 % of URL)

  • dose I rated dose

=

+/- (0.04 *750 INWC)

  • 8.78 x 102 /2.2 x 107

=

+/- 30 INWC

  • 3.99 x 10-5 e2R

=

t: 1.20 x 10*3 INWC Converting to % span e2Rcx,Span

=

+/- 1.20 x 10~3fNWC / 220 INWC

=

+/- 5.44 x 10-6 = 5.44 X 10-4 0/0 Converting % span to % flow Rearranging Equation A8-1 0, O'AP = 2

  • O'FlOW, to solve for flow (Reference Attachment 8):

e2R %Aow

=

e2R %Span I 2

=

+/- 5.44 x 10.4 °/012

+/- 2.72 x 10.4 %

=

+/- 2.72 x 10.4 %, which is negligible e2R%FIow 0

7.3.1.17 AWCU Flow Transmitter Seismic Error (e28)

The transmitter's accuracy is within +/- 0.50/0 of UAL (upper range limit) during and after a seismic disturbance defined by a required response spectrum with a ZPA of 4 g's. A seismic event defines a particular type of accident condition. Therefore, there is no seismic error for normal operating conditions (Section 3.5) e2S

=

0 7.3.1.18 AWCU Flow Transmitter Ambient Pressure Error (e2P)

The flow transmitter is an electr;cal device and therefore not affected by ambient pressure.

e2P

=

0 7.3.1.19 RWCU Flow Transmitter Temperature Error (e2T)

Temperature error;s considered to be a random variable for a Rosemount transmitter. Therefore e2T

=

0 Page 39 of 93

Exel6n, 7.3.1.20 RWCU PPC I/0 Module and SRU Reference Accuracy (A3)

Reference accuracy of the computer input is taken to be the SRSS of the reference accuracies of the SRU and the I/0 module. The reference accuracy of the SRU is 0.1 % of span (Section 2.2.5).

Reference Accuracy for the i/0 module is specified as t 0.25 % of span (Section 2.2.6).

A32,,

= 0.12 + 0.5 2 = 0.5099 = 0.51 %span Converting % span to % flow Rearranging Equation A8-10, o'ap = 2

  • GROW, to solve for flow (Reference Attachment 8) :

A3%Flc, =

A3%s a  / 2 e3H

=

0

+/-0.51%/2 A3%Fk,

+/-0.255 %

7.3.1.21 RWCU PPC 1/0 Module Humidity Error (e3H)

The manufacturer specifies the UO module operating humidity limits between 0 and 95 % RH (Section 2.2.6). The I/0 module is located in the Control Room 533, where humidity may vary from 50 to 90% RH (Reference 4.4.2). Humidity errors are set to zero because they are considered to be included within the reference accuracy specification under these conditions (Section 3.2).

7.3.1.22 RWCU PPC !/0 Module Radiation Error (e3R)

Reactor Core Thermal Power LE-0113 Uncertainty Calculation Unit 1 Revision 0 No radiation errors are specified in the manufacturer's specifications. The instrument is located in the Control Room 533, a mild environment (Section 4.4.2). Therefore, it is reasonable to consider the normal radiation effect as being included in the reference accuracy.

e3R

= 0 7.3.1.23 RWCU PPC l/0 Module Seismic Error (e3S)

No seismic effect errors are specified in the manufacturer's specifications. A seismic event defines a particular type of accident condition. Therefore, there is no seismic error for normal operating conditions (Section 3.5).

e3s

=

0 7.3.1.24 RWCU PPC 1/0 Module Static Pressure Offset Error (e3SP)

The 1/0 module is an electrical device and therefore not affected by static pressure.

e3SP =

0 7.3.1.25 RWCU PPC l/0 Module Ambient Pressure Error (e3P)

The I/0 module is an electrical device and therefore not affected by ambient pressure.

e3P 0

7.3.1.26 RWCU PPC I/0 Module Process Error (e3Pr)

The 1/0 module receives an analog current input from the flow transmitter proportional to the pressure sensed. Any process errors associated with the conversion of pressure to a current signal have been accounted for as errors associated with Flow Element. Therefore, e3Pr

=

0 Page 40 of 93 Exelan.

Reactor Core Thermal Power Uncertainty Calculation Unit 1 LE-0113 Revision 0 7.3.1.20 RWCU PPC I/O Module and SRU Reference Accuracy (A3)

Reference accuracy of the computer input is taken to be the SRSS of the reference accuracies of the SRU and the I/O module. The reference accuracy of the SRU is 0.1 % of span (Section 2.2.5).

Reference Accuracy for the 1/0 module is specified as +/- 0.25 % of span (Section 2.2.6).

A~(J

= JO.1 2 + 0.52 =0.5099 = 0.51 0,10 span Converting % span to % ffow Rearranging Equation A8-10, GAP = 2 It O'FLOW, to solve for flow (Reference Attachment 8):

A~FIow :::

A3%Span / 2

+/-O.51%/2 A3%Flow

=

+/- 0.255 %

7.3.1.21 RWCU PPC va Module Humidity Error (e3H)

The manufacturer specifies the I/O module operating humidity limits between aand 95 % RH (Section 2.2.6). The I/O module is located In the Control Room 533, where humidity may vary from 50 to 90%

AH (Reference 4.4.2). Humidity errors are set to zero because they are considered to be included within the reference accuracy specification under these conditions (Section 3.2).

e3H

=

0 7.3.1.22 AWCU PPC I/O Module Radiation Error (e3R)

No radiation errors are specified in the manufacturer's specifications. The instrument is located in the Control Room 533, a mild environment (Section 4.4.2). Therefore, it is reasonable to consfder the normal radiation effect as being included in the reference accuracy.

e3A

=

a 7.3.1.23 AWCU PPC flO Module Seismic Error (e3S)

No seismic effect errors are specified in the manufacturer's specifications. A seismic event defines a particular type of accident condition. Therefore, there is no seismic error for normal operating conditions (Section 3.5).

eSS

=

0 7.3.1.24 RWCU PPC 1/0 Module Static Pressure Offset Error (e3SP)

The flO module is an electrical device and therefore not affected by static pressure.

e3SP

=

0 7.3.1.25 RWCU PPC I/O Module Ambient Pressure Error (e3P)

The 1/0 module is an electricar device and therefore not affected by ambient pressure.

e3P

=

a 7.3.1.26 RWCU PPC I/O Module Process Error (e3Pr)

The 110 module receives an analog current input from the trow transmitter proportional to the pressure sensed. Any process errors associated with the conversion of pressure to a current signal have been accounted for as errors associated with Flow Element. Therefore, e3Pr

=

0 Page 40 of 93

7.3.1.27 RWCU Flow Loop Accuracy (LARwcu_Fkyw)

LAAWCU rr,, _ +/- [(A1)2 + (IE1)'+ (TN1)2 + (TD1)2 + (el H)2 + (e1 R)2 + (e1 S)z + (e1 SPI2 + (e1 P~2 +

(ei PrI2 + (e1T)2 + (A2 2 + (s2PS2)2 + (s2T)2 + (e2H)2 + (e2R)2 + (e2SI + (e2P) +

(e2T) + (A3) 2 + (e3H) + (e3R)2 + (e3S)2 + (e3SP)2 + (e3P)2 + (e3Pr) ]'12 D3

=

t 0

+/- [(1.5)2 + (0.5)2 + (0)2 +

2 + (0)2 + (0)2 + (0)2 + (0)2 + (0)a + (0)2 + (0)2 +

=

(0.

(0.3) 1667)2 + (0)2 + (0.41 ) + (0)2 + (0)2 + (0)2 + (0)2 + (0)2 + (0.255)2 + (0)2 + (0)2 +

(0)2 + (0)2 + (0)2 + (0)a],

=t[2.25+0.25+0+0.09+0+0+0+0+0+0+0+0.027789+0+0.174724+

0+0+0+0+0+0.065025+0+0+0+0+0+0]"2

= t [2.85754] 1"2 t 1.690425 °gyp LAFwcug. = t 1.7 %

7.3.2 RWCU Flow Loop Drift (LDRwcu_now) 7.3.2.1 RWCU Flow Element Drift Error (D1)

The flow element is a mechanical device ; drift error is not applicable for the flow elements.

Therefore, D1

= t 0.00% Flow 7.3.2.2 RWCU Flow Transmitter Drift Error (D2)

Drift error for the transmitter is t 0.2% of URL / 30 months, taken as a random 2cs value (Section 2.2.4). The calibration frequency is 2 years, with a late factor of 6 months.

D22a

=

t (0.2%

  • URL)

Drift is applied to the surveillance interval (SI) using Equation 39 (Section 6.2.5) as follows D22.

t [n(D.)2]1r2, t ([(24 months + 6 months) / 30 months] " [0.002 URL] 2)112

+/- ([(30 / 30)] " [1.5 1NWCh1/2 D22a

+/- 1.50 I NW C Converting to % span D2%sw

=

t 1.5 INWC / 219.8 INWC

=

t 0.00682439 = 0.682439 %

Converting % span to % flow 7.3.2.3 RWCU Flow PPC 1/O Module Drift Error (D3)

Reactor Core Thermal Power LE-0113 Uncertainty Calculation Unit 1 Revision 0 Rearranging Equation A8-10, aAp = 2

  • 6FLow, to solve for flow (Reference Attachment 8) :

D2%now =

D2q sw / 2 t 0.682439% / 2 0.341219 %

D2%Fbw =

+/-0.34 %

The vendor does not specify a drift error for the 1/O module. Therefore, per Ref. 4.1.1 Section I, it is considered to be included in the reference accuracy.

Page 4 1 of 93 Exelon.

Reactor Core Thermal Power Uncertainty Calculation Unit 1 LE-G113 Revision 0 7.3.2.2 7.3.2 7.3.2.1 7.3.1.27 RWCU Flow loop Accuracy (lAAWCU_FIoW)

LARWCU Flow =+/- [(A1}2 + (IE1)2 + (TN1)2 + (T01)2 + (e1H)2 + (e1R)2 + (e15)2 + (e1Spt + (e1pt +

(e1prt + (e1n2 + (A2~2 + (S2PS2)2 + (s2T)2 + (e2H)2 + (e2R)2 + (e2S~ + (e2P) +

(e2T) + (A3)2 + (e3H) + (e3R)2 + (e38)2 + (eSSp)2 + (e3p)2 + (e3Pr) ]112

== +/- [(1.5)2 + (0.5}2 + (0)2 + ~0.3)2 + (0)2 + (0)2 + (0)2 + (0)2 + (0)2 + (0)2 + (0)2 +

(0.1667)2 + (0)2 + (0.418l + (0)2 + (0)2 + (0)2 + (0)2 + (0)2 + (0.255)2 + (0)2 + (0}2 +

(0)2 + (0)2 + (0)2 + (0)2]1/1

== +/- [2.25 + 0.25 + 0 + 0.09 + 0 + 0 + 0 + 0 + 0 + 0 + 0 + 0.02n89 + 0 + 0.174724 +

o+ 0 + 0 + 0 + 0 + 0.065025 + 0 + 0 + 0 + 0 + 0 + 0]112

=+/- [2.85754]112

=::t: 1.690425 0/0 LARWCU.,.Aow== +/- 1.7 %

AWCU Flow Loop Drift (lDRWcu_Flow)

RWCU Flow Element Drift Error (01)

The flow element is a mechanical device; drift error is not applicable for the flow elements.

Therefore, 01

== +/- 0.00% Flow RWCU Flow Transmitter Drift Error (02)

Drift error for the transmitter is +/- 0.2°10 of URL 130 months, taken as a random 20' value (Section 2.2.4). The calibration frequency is 2 years, with a late factor of 6 months.

0220

=

+/- (0.2%

  • UAL)

Drift is appUed to the surveiUance interval (51) using Equation 39 (Section 6.2.5) as follows 0220

=

+/- [nO(Dx)2fl2,

=

+/- ([(24 months + 6 months) 130 months] * [0.002 URL] 2)112

=

+/- ([(30 I 30)] *[1.5 INWCf)112 022<1

=

t: 1.50 fNWC Converting to % span D2%$pan=

+/- 1.5INWC 1219.81NWC

=

+/- 0.00682439 =0.682439 0/0 Converting % span to % flow Rearranging Equation A8-l0, (j~p== 2 It O'FLOW, to solve for flow (Reference Attachment 8):

D2%Aow =

D2,,.span I 2

=

+/- 0.682439% 12

=

+/- 0.341219 0/0 D2%FIow =

+/- 0.34 0/0 7.3.2.3 RWCU Flow ppe flO Module Drift Error (03)

The vendor does not specify a drift error for the JlO module. Therefore, per Ref. 4.1.1 Section I, it is considered to be included in the reference accuracy.

D3

=

+/-O Page 41 of 93 Exelon.

Reactor Core Thermal Power Uncertainty Calculation Unit 1 LE-G113 Revision 0 7.3.2.2 7.3.2 7.3.2.1 7.3.1.27 RWCU Flow loop Accuracy (lAAWCU_FIoW)

LARWCU Flow =+/- [(A1}2 + (IE1)2 + (TN1)2 + (T01)2 + (e1H)2 + (e1R)2 + (e15)2 + (e1Spt + (e1pt +

(e1prt + (e1n2 + (A2~2 + (S2PS2)2 + (s2T)2 + (e2H)2 + (e2R)2 + (e2S~ + (e2P) +

(e2T) + (A3)2 + (e3H) + (e3R)2 + (e38)2 + (eSSp)2 + (e3p)2 + (e3Pr) ]112

== +/- [(1.5)2 + (0.5}2 + (0)2 + ~0.3)2 + (0)2 + (0)2 + (0)2 + (0)2 + (0)2 + (0)2 + (0)2 +

(0.1667)2 + (0)2 + (0.418l + (0)2 + (0)2 + (0)2 + (0)2 + (0)2 + (0.255)2 + (0)2 + (0}2 +

(0)2 + (0)2 + (0)2 + (0)2]1/1

== +/- [2.25 + 0.25 + 0 + 0.09 + 0 + 0 + 0 + 0 + 0 + 0 + 0 + 0.02n89 + 0 + 0.174724 +

o+ 0 + 0 + 0 + 0 + 0.065025 + 0 + 0 + 0 + 0 + 0 + 0]112

=+/- [2.85754]112

=::t: 1.690425 0/0 LARWCU.,.Aow== +/- 1.7 %

AWCU Flow Loop Drift (lDRWcu_Flow)

RWCU Flow Element Drift Error (01)

The flow element is a mechanical device; drift error is not applicable for the flow elements.

Therefore, 01

== +/- 0.00% Flow RWCU Flow Transmitter Drift Error (02)

Drift error for the transmitter is +/- 0.2°10 of URL 130 months, taken as a random 20' value (Section 2.2.4). The calibration frequency is 2 years, with a late factor of 6 months.

0220

=

+/- (0.2%

  • UAL)

Drift is appUed to the surveiUance interval (51) using Equation 39 (Section 6.2.5) as follows 0220

=

+/- [nO(Dx)2fl2,

=

+/- ([(24 months + 6 months) 130 months] * [0.002 URL] 2)112

=

+/- ([(30 I 30)] *[1.5 INWCf)112 022<1

=

t: 1.50 fNWC Converting to % span D2%$pan=

+/- 1.5INWC 1219.81NWC

=

+/- 0.00682439 =0.682439 0/0 Converting % span to % flow Rearranging Equation A8-l0, (j~p== 2 It O'FLOW, to solve for flow (Reference Attachment 8):

D2%Aow =

D2,,.span I 2

=

+/- 0.682439% 12

=

+/- 0.341219 0/0 D2%FIow =

+/- 0.34 0/0 7.3.2.3 RWCU Flow ppe flO Module Drift Error (03)

The vendor does not specify a drift error for the JlO module. Therefore, per Ref. 4.1.1 Section I, it is considered to be included in the reference accuracy.

D3

=

+/-O Page 41 of 93

Exelon, PMAnwcu_Fww = 0 PEARwcu_Flow = 0 LE-0113 Uncertainty Calculation Unit 1 Revision 0 Reactor Core Thermal Power 7.3.2.4 RWCU Flow Loop Drift (LDRvCU.u Frow)

LDRwcu_now = [(D1) 2 + (D2)2 +

(D3)2]ir2

= [(0)2 + (0.34)2 + (0)2]ll

= t [0 + 0.116431 + 0]"

  • [0.1164311"2 0.341219 LDRwcu-Raw =
  • 0.34 7.3.3 RWCU Flow Loop Process Measurement Accuracy (PMAR,cu_Fj'W)

No additional PMA effects beyond the effects specified in the calculation of loop accuracy.

7.3.4 RWCU Flow Loop Primary Element Accuracy (PEARwcu_Fj'w)

No additional PEM effects beyond the effects specified in the calculation of loop accuracy.

7.3.5 RWCU Flow Loop Calibration Accuracy (CAswcu-now)

Standard practice is to specify calibration uncertainty in calculations equal to the uncertainty associated with the instruments under test (Reference 4.1.1).

Therefore, CAawc., F~ow =

[(A1)2 + (A2)2 + (A3)2]"2

[(1.5 %)2 + (0.1667 %)2+ (0.255 %)2

]"2

[2.3428 %2 ]f2 1.5306 CARwcu_r1ow =

1.5 7.3.6 Total Uncertainty RWCU Flow Loop (TURwcu_Fl.)

TURrcu.tl = [(LA)2 + (LD)2 + (PMA)2 + (PEA)2 + (C A) a] "2

  • [(1.7)2 + (0.34)2 + (0)2 + (0)2 + (1.5)2]'

=t[2.89+0.1156+0+0+2.25]"

=

  • 15.2556]'t2

= t 2.29251 TURwcu,Ftow ' :t 2.3 °fo 7.4 RWCU TEMPERATURE LOOP UNCERTAINTY 7.4.1 RWCU Temperature Loop Accuracy (LARwcu-r) 7.4.1.1 RWCU Temperature Element Reference Accuracy (A1)

RWCU Temperature Element Reference Accuracy is provided in Section 2.3.1.

A 1

=

  • 0.75 OF Page 42 of 93 Exelon.

Reactor Core Thermal Power Uncertainty Calculation Unit 1 LE-6113 Revision 0 7.3.2.4 RWCU Flow Loop Drift (LDRWCU,~FIow)

LDRWcu_Flow ::: [(01)2 + {D2)2 + (03)2]112

=[(0)2 + (0.34)2 + (0)2]112

+/- [0 + 0.116431 + 0]1/2

=::1: (0.116431]1/2

+/- 0.3412190/0

[(A1)2 + {A2)2 + (A3)2Jll2

[(1.5 0/0)2 + (0.1667 0/0)2 + (0.255 0/0)2]112

[2.3428 0/02]112 1.5306 0./0

=

=

LDRWcu_FloW ::: +/- 0.34 %

RWCU Flow Loop Process Measurement Accuracy (PMARWCU_Aow)

No additional PMA effects beyond the effects specified in the calculation of loop accuracy.

PMARwcu",Flow::: 0 RWCU Flow Loop Primary Element Accuracy (P~CU_FIow)

No additional PEM effects beyond the effects specified in the calculation of loop accuracy.

PEARwcu_Flow ::: 0 RWCU Flow Loop Calibration Accuracy (CARWCU_FIow)

Standard practice is to specify calibration uncertajnty in calculations equal to the uncertainty associated with the instruments under test (Reference 4.1.1).

Therefore, CARWCU=FIOW :::

7.3.3 7.3.4 7.3.5 7.3.6 CARWCU_Flow =

1.5 %

Total Uncertainty RWCU Flow Loop (TURWCUJlow)

TURWcuYJow ::: +/- [(lA)2 + (LD)2 + (PMA)2 + (PEA)2 + (CA)2J1f2

+/- [(1.7)2 + (0.34)2 + (0)2 + (0)2 + (1.5)2]112

=+/- [2.89 + 0.1156 + 0 + 0 + 2.25]112

= +/- [5.2556]112

=+/-2.29251%

TURWCU,.FIow = +/- 2.3 %

7.4 RWCU TEMPERATURE LOOP UNCERTAINTY 7.4.1 7.4.1.1 RWCU Temperature loop Accuracy (lARWCUJ)

RWCU Temperature Element Reference Accuracy (A1)

RWCU Temperature Element Reference Accuracy is provided in Section 2.3.1.

A1

=:to.75°F Page 42 of 93 Exelon.

Reactor Core Thermal Power Uncertainty Calculation Unit 1 LE-6113 Revision 0 7.3.2.4 RWCU Flow Loop Drift (LDRWCU,~FIow)

LDRWcu_Flow ::: [(01)2 + {D2)2 + (03)2]112

=[(0)2 + (0.34)2 + (0)2]112

+/- [0 + 0.116431 + 0]1/2

=::1: (0.116431]1/2

+/- 0.3412190/0

[(A1)2 + {A2)2 + (A3)2Jll2

[(1.5 0/0)2 + (0.1667 0/0)2 + (0.255 0/0)2]112

[2.3428 0/02]112 1.5306 0./0

=

=

LDRWcu_FloW ::: +/- 0.34 %

RWCU Flow Loop Process Measurement Accuracy (PMARWCU_AOW)

No additional PMA effects beyond the effects specified in the calculation of loop accuracy.

PMARwcu",Flow::: 0 RWCU Flow Loop Primary Element Accuracy (P~CU_FIow)

No additional PEM effects beyond the effects specified in the calculation of loop accuracy.

PEARwcu_Flow ::: 0 RWCU Flow Loop Calibration Accuracy (CARWCU_FIow)

Standard practice is to specify calibration uncertajnty in calculations equal to the uncertainty associated with the instruments under test (Reference 4.1.1).

Therefore, CARWCU=FIOW :::

7.3.3 7.3.4 7.3.5 7.3.6 CARWCU_Flow =

1.5 %

Total Uncertainty RWCU Flow Loop (TURWCUJlow)

TURWcuYJow ::: +/- [(lA)2 + (LD)2 + (PMA)2 + (PEA)2 + (CA)2J1f2

+/- [(1.7)2 + (0.34)2 + (0)2 + (0)2 + (1.5)2]112

=+/- [2.89 + 0.1156 + 0 + 0 + 2.25]112

= +/- [5.2556]112

=+/-2.29251%

TURWCU,.FIow = +/- 2.3 %

7.4 RWCU TEMPERATURE LOOP UNCERTAINTY 7.4.1 7.4.1.1 RWCU Temperature loop Accuracy (lARWCUJ)

RWCU Temperature Element Reference Accuracy (A1)

RWCU Temperature Element Reference Accuracy is provided in Section 2.3.1.

A1

=:to.75°F Page 42 of 93

Exelun.

7.4.1.2 RWCU Temperature Element Power Supply Effects (al PS)

Power supply effects are considered to be negligible (Section 3.7). Therefore, al PS =

0 7.4.1.3 RWCU Temperature Element Ambient Temperature Error (a1T)

All thermocouple extension wire junctions are on adjacent terminals and are assumed to be at the same temperature. Therefore, for the thermocouple, 7.4.1.4 RWCU Temperature Element Humidity Error (e1 H)

The manufacturer does not specify the thermocouple operating humidity limits (Section 2.3.1). The thermocouple is located In the RWCU System Room 506, where humidity may vary from 50 to 90%

RH (Reference 4.4.2). Humidity errors are set to zero because they are considered to be included within the reference accuracy specification under these conditions (Section 3.2).

e1H 0

7.4.1.5 RWCU Temperature Element Radiation Error (e1 R)

No radiation errors are specified in the manufacturer's specifications. The instrument is located in the Control Room 533, a mild environment (Reference 4.4.2). Therefore, it is reasonable to consider the normal radiation effect as being included in the reference accuracy. Therefore, e1 R

=

0 7.4.1.6 RWCU Temperature Element Seismic Error (e1 S)

No seismic effect errors are specified in the manufacturer's specifications. A seismic event defines a particular type of accident condition. Therefore, there is no seismic error for normal operating conditions (Section 3.5).

e1S

=

0 7.4.1.7 RWCU Temperature Element Vibration Effect (e1 V)

Reactor Core Thermal Power Uncertainty Calculation Unit 1 Revision 0 The error due to vibration is considered to be negligible because it is small and unaffected by vibrations in the system.

e1V

= 0 7.4.1.8 RWCU Temperature Element Static Pressure Error (e1 SP)

The thermocouple output is not subject to pressure variations.

e1 SP,, =

0 7.4.1.9 RWCU Temperature Element Ambient Pressure Error (e1 P)

The thermocouple is an electrical device and therefore not affected by ambient pressure.

e1P

=

0 7.4.1.10 RWCU Temperature Element Temperature Error (e1 T)

The temperature error is assumed to be included in the reference accuracy (Reference 4.1.1).

Therefore, e1T 0

7.4.1.11 RWCU Temperature Loop PPC i/0 Module Reference Accuracy (A2)

Reference Accuracy is specified as +/- 3.0 IF (Section 2.3.3) and considered to be a 20 value (Section 3.4)

Page 43 of 93 LE-0113 ExelC)n.

Reactor Core Thermal Power Uncertainty Calculation Unit 1 LE-0113 Revision 0 7.4.1.2 RWCU Temperature Element Power Supply Effects (<f1 PS)

Power supply effects are considered to be negligible (Section 3.7). Therefore, 0'1PS

=

0 7.4.1.3 AWCU Temperature Element Ambient Temperature Error (o1T)

All thermocouple extension wire junctions are on adjacent terminals and are assumed to be at the same temperature. Therefore, for the thermocouple, 0'1T10= 0 7.4.1.4 RWCU Temperature Element Humidity Error (e1 H)

The manufacturer does not specify the thermocouple operating humidity limits (Section 2.3.1). The thermocouple is located in the RWCU System Room 506, where humidity may vary from 50 to 900/0 RH (Reference 4.4.2). Humidity errors are set to zero because they are considered to be included within the reference accuracy specification under these conditions (Section 3.2).

e1H

=

0 7.4.1.5 RWCU Temperature Element Radiation Error (e1 A)

No radiation errors are specified in the manufacturer's specifications. The instrument is located in the Control Room 533, a mild environment (Reference 4.4.2). Therefore, it is reasonable to consider the normal radiation effect as being included in the reference accuracy. Therefore, e1R

=

0 7.4.1.6 RWCU Temperature Element Seismic Error (e1 S)

No seismic effect errors are specified in the manufacturer's specifications. A seismic event defines a particular type of accident condition. Therefore, there is no seismic error for normal operating conditions (Section 3.5).

e1S

=

0 7.4.1.7 RWCU Temperature Element Vibration Effect (elV)

The error due to vibration is considered to be negligible because it is small and unaffected by vibrations in the system.

elV

=

0 7.4.1.8 RWCU Temperature Element Static Pressure Error (el SP)

The thermocouple output is not subject to pressure variations.

el SP1G =

0 7.4.1.9 RWCU Temperature Element Ambient Pressure Error (e1 P)

The thermocouple is an electrical device and therefore not affected by ambient pressure.

alP

=

0 7.4.1.10 RWCU Temperature Element Temperature Error (e1n The temperature error is assumed to be included in the reference accuracy (Reference 4.1.1).

Therefore, elT

=

0 7.4.1.11 RWCU Temperature Loop PPC 1/0 Module Reference Accuracy (A2)

Reference Accuracy;s specified as +/- 3.0 ~F (Section 2.3.3) and considered to be a 20' value (Section 3.4)

Page 43 of 93

A22a t 3.0 *F 7.4.1.12 RWCU Temperature Loop PPC I/O Module Humidity Error (e2H)

The manufacturer specifies the I/O module operating humidity limits between 0 and 95 % RH (Section 2.3.3). The I/O module is located in the Control Room 533, where humidity may vary from 50 to 90% RH (Reference Section 4.4.2). Humidity errors are set to zero because they are considered to be included within the reference accuracy specification under these conditions.

(Section 3.2) e2H 0

7.4.1.13 RWCU Temperature Loop PPG 1/O Module Radiation Error (e2R)

Reactor Core Thermal Power LE-0113 Uncertainty Calculation Unit 1 Revision 0 No radiation errors are specified in the manufacturer's specifications. The instrument is located in the Control Room 533, a mild environment [Section 4.4.2]. Therefore, it is reasonable to consider the normal radiation effect as being included in the reference accuracy. Therefore, e2R

= 0 7.4.1.14 RWCU Temperature Loop PPC I/O Module Seismic Error (e2S)

No seismic effect errors are specified in the manufacturer's specifications. A seismic event defines a particular type of accident condition. Therefore, there is no seismic error for normal operating conditions (Section 2.8).

e2S 0

7.4.1.15 RWCU Temperature Loop PPC I/O Module Static Pressure Offset Error (e2SP)

The t/O module is an electrical device installed in the control room and is not subject to pressure effects.

e2SP 0

7.4.1.15 RWCU Temperature Loop PPG I/O Module Ambient Pressure Error (e2P)

The I/O module Is an electrical device and therefore not affected by ambient pressure.

e2P

=

0 7.4.1.17 RWCU Temperature Loop Accuracy (LARwcu_n.)

LARwcuj 7.4.2 RWCU Temperature Loop Drift t [(Al )2 + (a1 PS)2 + (a1T)2 + (e1 H)2 + (el R)2 + (e1 S~2 + (e1 V 2 + (e1 SP)2 + (e1 P)2

+ (e1T) 2 + (A2)2 + (e2H)2 + (e2R)2 + (e2S)2 + (e2SP) + (e2P) ]'/2 t

2+(0)2+(0) 2 +40) 1 + (0)z + (0)2 + (0)2 + (0)2 + (0)2 +

[ 0.75)

40)

(0)2+(3. 0)2+(0)2+

(0) +(0) 2 +(0) 2 +(0) 2]1

=t[0.5525+0+0+0+0+0+0+0+0+0+9.0+0+0+0+0+0]1r2

= [9.5625 °F2]112 LARwcuj = t 3.09 °F 7.4.2.1 RWCU Temperature Element Drift Error (D1)

The error associated with the thermocouple is already included in the reference accuracy (Section 3.2). Therefore, for the thermocouple, D1 t0 Page 44 of 93 Exel~n.

Reactor Core Thermal Power Uncertainty Carcuration Unit 1 LE-0113 Revision 0 A22c

=

+/- 3.0 of 7.4.1.12 RWCU Temperature Loop PPC I/O Module Humidity Error (e2H)

The manufacturer specifies the I/O module operating humidity limits between 0 and 95 ok RH (Sectjon 2.3.3). The 1/0 module is focated in the Control Room 533, where humidity may vary from 50 to 900/0 RH (Reference Section 4.4.2). Humidity errors are set to zero because they are considered to be included within the reference accuracy specification under these conditions.

(Section 3.2) e2H

=

0 7.4.1.13 RWCU Temperature Loop PPC I/O Module Radiation Error (e2R)

No radiation errors are specified in the manufacturer's specifications. The instrument is located in the Control Room 533, a mild environment [Section 4.4.2J. Therefore, it is reasonable to consider the normal radiation effect as being included in the reference accuracy. Therefore, e2R

=

0 7.4.1.14 RWCU Temperature Loop PPC I/O Module Seismic Error (e2S)

No seismic effect errors are specified in the manufacturer's specifications. A seismic event defines a particular type of accident condition. Therefore, there is no seismic error for normal operating conditions (Section 2.8).

e2S

=

0 7.4.1.15 RWCU Temperature Loop PPC I/O Module Static Pressure Offset Error (e2SP)

The riO module is an erectrical device installed in the control room and is not subject to pressure effects.

e2SP =

0 7.4.1.16 RWCU Temperature Loop ppe I/O Module Ambient Pressure Error (e2P)

The 1/0 module is an electrical device and therefore not affected by ambient pressure.

a2P

=

0 7.4.1.17 AWCU Temperature Loop Accuracy (LARWCU_FlOW)

LARWCU..T =+/- [(A1)2 + (01 PS)2 + (0'1T)2 + (el H)2 + (el R)2 + (elSi2 + {e1vt + (el SP)2 + (el p)2

+ (e1T)2 + (A2)2 + (e2H)2 + (e2R)2 + (e2S)2 + (e2SP) + (e2P) ]112

+/- [~o.75)2 + (0)2 + (0)2 +jO)2 + (0)2 + (0)2 + (0)2 + (0)2 + (0)2 + (0)2 + (3.0)2 + (0)2 +

CO) + (0)2 + (0)2 + (0)2]1

=+/- [0.5625 + 0 + 0 + 0 + 0 + 0 + 0 + 0 + 0 + 0 + 9.0 + 0 + 0 + 0 + 0 + 0]112

+/- [9.5625 °F2]1f2 LARwcu-T

=+/- 3.09 OF 7.4.2 7.4.2.1 RWCU Temperature Loop Drift RWCU Temperature Element Drift Error (01)

The error associated with the thermocouple is already included in the reference accuracy (Section 3.2). Therefore, for the thermocouple, 01

=

+/-O Page 44 of 93

x.

rw P-NrnAeft

^qw1on.

7.4.2.2 RWCU Temperature Loop PPC 1/0 Module Drift Error (D2)

The vendors do not specify drift errors for the SRU and 1/0 module. Therefore, per Section 3.2, it is considered to be included in the reference accuracy.

D2

+/- 0 7.4.2.3 RWCU Temperature Loop Drift (LDRWCU-T)

LDRWCU-T

= [(DI)2 + (D2)2]1/2

= [(0)2 + (0)2]1,2

[0 + 0]""2

[0/2

+/-0.0 %

LDFWCU-T

+ 0 %

7.4.3 RWCU Temperature Loop (PMARwCU,,,T)

No additional PMA effects beyond the effects specified in the calculation of loop accuracy.

PMApwcu_T = +/- 0 7.4.4 RWCU Temperature Loop (PEARwcu_T)

No additional PMA effects beyond the effects specified in the calculation of loop accuracy.

PEARWCU-T = 0 7.4.5 RWCU Temperature Loop Calibration Accuracy (CARwcu - T)

Standard practice is to specify calibration uncertainty in calculations equal to the uncertainty associated with the instruments under test (Reference 4. 1. 1 Therefore, CAFiwcu,:r

[(Al

)2+ (A2)2 + (A3)21"2

[(0.75 -F)2 + (3.09 -F)21"2

[0-5625 + 9]"

[9.5625]"

CARWCU-T 3.09 'F 7.4.6 Total Uncertainty RWCU Temperature Loop (TURwcuj)

TURWCU_T

= :t [(LA)2 + (LD)' + (PMA)2 + (PEA)2 + (CA)']""2

=+/- [(3-09)' + (0)2 + (0)' + (0)2 + (3.09)']"2

=t[9.5625+0+0+0+9.5625] 2

[19.125]"

4.373 *F TURwcu-T

4

.37 *F 7.5 CRD FLOW RATE UNCERTAINTY 7.5.1 CRD Flow Rate Loop Accuracy (LAcRD_F1,w)

Reactor Core Thermal Power LE-01 13 Uncertainty Calculation Unit I Revision 0 Page 45 of 93 Exelon.

Reactor Core Thermal Power Uncertainty Calculation Unit 1 LE-0113 Revision 0 7.4.2.2 RWCU Temperature Loop PPC 1/0 Module Drift Error (D2)

The vendors do not specify drift errors for the SAU and va module. Therefore, per Section 3.2, it is considered to be included in the reference accuracy.

D2

=

+/-O 7.4.2.3 RWCU Temperature Loop Drift (LDRWCU_T)

LDRWcu_T

= [(Dl)2 + (02)2]112

= [(0)2 + (0)2]1/2

= :t (0 + 0]1/2

= +/- [ot12

= +/-O.O 0/0 LDRWcu_:T

=+/- 0 %

7.4.3 RWCU Temperature Loop (PMARWCU",T)

No additional PMA effects beyond the effects specified in the calculation of loop accuracy_

PMARWCU..T = :t 0 7.4.4 RWCU Temperature loop (PEARWCU_r)

No additional PMA effects beyond the effects specified in the calculation of loop accuracy.

PEARWCU~T =0 7.4.5 AWCU Temperature Loop Calibration Accuracy (CARWCU_r)

Standard practice is to specify caJibration uncertainty in calculations equal to the uncertainty associated with the instruments under test (Reference 4.1.1 ).

Therefore, CARWCU.",T

=

[{A1)2 + (A2)2 + {A3)2J112

=

[(0.75 °F)2 + (3.09 °F)2]112

=

[0.5625 + 9]1/2

=

[9.5625]112 CARWCU_T

=

3.09 OF 7.4.6 Total Uncertainty RWCU Temperature Loop (TURWCU3)

TURWCU_T

=:t [(LA)2 + (lD)2 + (PMA)2 + {PEA)2 + (CA)2]1/2

=+/- [(3.09)2 + (0)2 + (0)2 + {O)2 + (3.09)2)112

= +/- [9.5625 + 0 + 0 + 0 + 9.5625]1I2

= +/- [19.125]112

=+/-4.373 OF TURWcu_T

=:t 4.37 OF 7.5 CRD FLOW RATE UNCERTAINTY 7.5.1 CRD Flow Rate Loop Accuracy (LACRO_Row)

Page 45 of 93

Converting to % span 1-5-0113 Reactor Core Thermal Power Uncertainty Calculation Unit I Revision 0 7.5.1.1 CRD Flow Element Reference Accuracy (Al)

The accuracy of the flow element is +/- 1% of actual rate of flow (Section 2.4.2).

Therefore, Al I % flow 7.5.1.2 CRD Flow Element Humidity, Radiation, Pressure, and Temperature Errors (elH, el R, el P, e1T)

The flow element is a mechanical device mounted in the process and its output is not subject to environmental or vibration effects. Therefore ;

e1H = el R = el P = e1T = 0 7.5.1.3 CRD Flow Element Seismic Error (e1 S)

A seismic event is an abnormal operating condition and is not addressed by this calculation (Section 3.5). Therefore ;

els 0

7.5.1.4 CRD Flow Element Static Pressure Error (e1 SP}

The flow element is constructed of stainless steel and is not affected by process pressure.

Therefore, e1SP =

0 7.5.1.5 CRD Flow Transmitter Reference Accuracy (A2)

Reference Accuracy is

  • 0.25 % of span (Section 2.4.3). The reference accuracy is set to the calibration accuracy per plant procedure (Reference 4.1.1).

A2

t 0.50 %

Converting % span to % flow Rearranging Equation A8-10, (

AF = 2
  • CROW, to solve for flow (Reference Attachment 8) :

A2%Flow A2%Span / 2

+/- 0.50 % of span / 2

+/- 0.25 % of Flow A2%Flow

+/- 0.25 % of Flow 7.5.1.6 CRO Flow Transmitter Power Supply Effects (a2PS)

Power supply effects are considered to be negligible (Section 3.7). Therefore, a2PS =

L- 0 7.5.1.7 CRD Flow Transmitter Ambient Temperature Error (a2T)

The temperature effect is

  • 2.4 % of span per 100 OF at 197.5 INWG span (Section 2.4.3). The calibrated span is 197.5 INWC. The maximum temperature at the transmitter location is 106 C

-F, and minimum temperature during calibration could be 65 OF, so the maximum difference = 106 - 65 OF = 41 "F (Section 2.3.4).

a2T

  • t(O.024.197.5 INWC) / 100 "F].41 '-F

[3a]

+/- [(4.74 1 NWC)/100 *F] - 41 *F

+/- 1.943 INWC Page 46 of 93 Exelon.

Reactor Core Thermal Power Uncertainty Calcuration Unit 1 LE*0113 Revision 0 7.5.1.1 CRD Flow Element Reference Accuracy (A1)

The accuracy of the flow element is +/- 1°k of actual rate of flow (Section 2.4.2).

Therefore.

A1

=

+/- 1 % now 7.5.1.2 CAD Flow Element Humidity, Radiation, Pressure, and Temperature Errors (e1 H, e1 A, 91 P, e1T)

The flow element is a mechanical device mounted in the process and its output is not SUbject to environmental or vibration effects. Therefore; el H = e1 R = el P=e1T = 0 7.5.1.3 CAD Flow Element Seismic Error (e18)

A seismic event is an abnormal operating condition and is not addressed by this calculation (Section 3.5). Therefore; elS

=

0 7.5.1.4 CAD Flow Element Static Pressure Error (el SP)

The flow element is constructed of stainless steel and is not affected by process pressure.

Therefore, e18P

=

0 7.5.1.5 CAD Flow Transmitter Reference Accuracy (A2)

Reference Accuracy;s +/- 0.25 Ok of span (Section 2.4.3). The reference accuracy is set to the calibration accuracy per plant procedure (Reference 4.1.1 ).

A2

=

+/- 0.50 0/0 Converting % span to % flow Rearranging Equation A8-10, craP = 2

  • O'FLOW, to solve for trow (Reference Attachment 8):

A2%FIoW ;:

A2%Span / 2

=

+/- 0.50 % of span / 2

=

+/- 0.25 % of Flow A2%FIOW =

I:: 0.25 % of Flow 7.5.1.6 CAD Flow Transmitter Power Supply Effects (cr2PS)

Power supply effects are considered to be negligible (Section 3.7). Therefore, cr2P8

=

+/- 0 7.5.1.7 CRD Ffow Transmitter Ambient Temperature Error (02T)

The temperature effect is +/- 2.4 % of span per 100 OF at 197.5 rNWC span (Section 2.4.3). The calibrated span is 197.5 INWC. The maximum temperature at the transmitter location is 106 tlF, and minimum temperature during calibration could be 65 OF. so the maximum difference = 106 - 65 t>F = 41 GF (Section 2.3.4).

cr2T

=
I:: [(0.024* 197.5INWC) /100 oF]. 41 "*F (30']

=

+/- [(4.74 INWC)1100 #F]

  • 41°F
=

+/- 1.943 fNWC Converting to % span Page 46 of 93

E~celc'm.

a2T2a =

t 1.943 INWC / 197.5 INWC t 0.0098 Converting % span to % flow Rearranging Equation A8-10, aep = 2

  • aF.ow, to solve for flow (Reference Attachment 8) :

a2T2,%Fiow -

a2T2,,%S,,, / 2

+/-0.98 %/2

+/-0.49 a2T2,%F,,,

t 0.49 7.5.1.8 CRD Flow Transmitter Humidity Error (e2H)

The manufacturer specifies the transmitter operating humidity limits between 0 and 100 % RH (Section 2.4.3). The transmitter is located in the CRD Equipment Area, Room 402, where humidity may vary from 50 to 90% RH (Reference 4.4.2). Humidity errors are set to zero because they are considered to be included within the reference accuracy specification under these conditions.

(Section 3.2) e2H 0

7.5.1.9 CRD Flow Transmitter Radiation Error (e2R)

Radiation error Is assumed to be 10 % of span (Section 3.9).

e2R 10 % of Span 1 0 °!o " 197.5 INWC e2R

=

19.751NWC Converting to % span o2 R t 19.75 INWC / 197.5 INWC 10.10 = 10 %

Reactor Core Thermal Power LE-01 13 Uncertainty Calculation Unit 1 Revision 0 Converting % span to % flow Rearranging Equation A8-10, aap = 2 '` aFLow, to solve for flow (Reference Attachment 8) :

cr2R%F[Ow e2V

=

0 a2R%F,,,w e2S

=

0 7.5.1.10 CRD Flow Transmitter Seismic Error (e2S)

No seismic effect errors are specified In the manufacturer's specifications. A seismic event defines a particular type of accident condition. Therefore, there is no seismic error for normal operating conditions (Section 3.5) 7.5.1.11 CRD Flow Transmitter Vibration Effect (e2V)

The error due to vibration is considered to be negligible because it is small and unaffected by vibrations in the system.

Page 47 of 93 a2R2^sW / 2

t 10.0% / 2 t 5.00%

t 5.0%

Exelon.

Aeactor Core Thermal Power Uncertainty Calculation Unit 1 LE-0113 Revision 0 02T20

=

+/- 1.943INWC/ 197.5 rNWC

=

+/- 0.0098 Converting % span to % flow Rearranging Equation A8-1 0, <1M' =2

  • O'FlOW, to solve for trow (Reference Attachment 8):

<12T2G%FIow

=

a2T2o%Span I 2

=

t 0.98 %

/ 2

=

+/-0.49%

<12T2o%Flow

=

+/- 0.49 o/Q 7.5.1.8 CAD Flow Transmitter Humidity Error (e2H)

The manufacturer specifies the transmitter operating humidity limits between 0 and 100 % RH (Section 2.4.3). The transmitter is located in the CRD Equipment Area, Room 402, where humidity may vary from 50 to 900/0 RH (Reference 4.4.2). Humidity errors are set to zero because they are considered to be included within the reference accuracy specification under these conditions.

(Section 3.2) e2H

=:

0 7.5.1.9 CAD Flow Transmitter Radiation Error (e2R)

Radiation error is assumed to be 10 % of span (Section 3.9).

e2R

=

10 % of Span

=

10 %. 197.5 rNWC e2R

=:

19.751NWC Converting to % span cr2R

=:

+/- 19.75 INWC / 197.5 INWC

=

+/-0.10 =: 10 Ok Converting % span to % flow Rearranging Equation A8-10, crAP = 2 11 <1FlOW, to solve for flow (Reference Attachment 8):

0'2R%FIow

=:

cr2R2o%Span I 2

=:

+/- 10.0%/2

=

t 5.000k cr2R%FIOW 7.5.1.10 CAD Flow Transmitter Seismic Error (e2S)

No seismic effect errors are specified in the manufacturer's specifications. A seismic event defines a particu'ar type of accident condition. Therefore, there is no seismic error for normal operating conditions (Section 3.5) e2S

=

0 7.5.1.11 CRD Ffow Transmitter Vibration Effect (e2V)

The error due to vibration is considered to be negligible because it is smalf and unaffected by vibrations in the system.

e2V

=

0 Page 47 of 93

7.5.1.12 CRD Flow Transmitter Static Pressure Zero Error (e2SP)

The transmitter has a Zero Error of :t 0.5 % of URL for 2000 psi (Section 2.4.3). The calibrated range shown in Table 2-12 shows the static pressure adjustment made to account for the span error effect. Therefore, the total Static Pressure Error is e2SP =

+/- 0.5 % of URL for 2000 psi The normal operating pressure (Table 2-11) is 1448 prig. Therefore, e2SP

+/-0.5%.7501NWC-1448prig /2000psi

+/- 2.715 INWC e2SP =

+/- 2.72 INWC, rounded Converting to % span cr2SP =

t 2.72 INWC / 200 INWC X0

.01316=1

.32%

Converting % span to % flow Rearranging Equation A8-1 0, (Y,&p 2

  • GROW, to solve for flow (Reference Attachment 8) :

a2SP%O,,

QSP,aso. 2 11.32%12 0.66 %

a2SP%Fl.

+/- 0.66 %

7.5.1.13 CRD Flow Transmitter Ambient Pressure Error (e2P)

The flow transmitter is an electrical device and therefore not affected by ambient pressure.

e2P

= 0 7.5.1.14 CRD Flow Transmitter Temperature Error (e2T)

Temperature error is considered to be a random variable for a Rosemount transmitter. Therefore e2T

=

0 7.5.1.15 CRD Flow Loop PPC 1/0 Module Reference Accuracy (A3)

Reference accuracy of the computer input is taken to be the SRSS of the reference accuracies of the SRU and the 1/0 module. The reference accuracy of the SRU is 0.1 % of span. Reference Accuracy for the 1/0 module is specified as +/- 0.5 % of span (Sections 2.4.4 and 2.4.5).

At.

y4yoj + 002

.12

.5 = 0.5099 = 0.51 % span Converting % span to % flow Rearranging Equation AS-10, aAp = 2

  • aFLOW, to solve for flow (Reference Attachment 8) :

A3%n. =

A3%spa. / 2 A3%Ff.

=

+/- 0

.5i%/2

=

+/- 0.255 %

7.5.1.16 CRD Flow Loop PPC 1/0 Module Humidity Error (e3H)

Reactor Core Thermal Power LE-01 13 Uncertainty Calculation Unit I Revision 0 The manufacturer specifies the 1/0 module operating humidity limits between 0 and 95 % RH (Section 2.4.4). The 1/0 module is located in the Control Room 533, where humidity may vary from Page 48 of 93 Exelon.

Reactor Core Thermal Power Uncertainty Calculation Unit 1 LE-G113 Revision 0 7.5.1.12 CRD Flow Transmitter Static Pressure Zero Error (e2SP)

The transmitter has a Zero Error of:t 0.5 % of UAL for 2000 psi (Section 2.4.3). The calibrated range shown in Table 2*12 shows the static pressure adjustment made to account for the span error effect. Therefore, the total Static Pressure Error is e2SP

+/- 0.5 % of URL for 2000 psi The normal operating pressure (Table 2-11) is 1448 psig. Therefore, e2SP

+/- 0.5 %

  • 750 INWC. 1448 psig / 2000 psi
i:2.715INWC e2SP

+/- 2.72 INWC, rounded Converting to % span a2SP

+/-2.72INWCI200INWC

=

t 0.01316 = 1.32 ok Converting % span to % ffow Rearranging Equation A8-10, O'M>::: 2
  • O'FlOW, to solve for flow (Reference Attachment 8):

0'2SP%FIOW

=

a2SP2~Span /2

=

+/- 1.32 % I 2

=

+/-0.66 %

0'2SP%Flow

+/- 0.66 %

7.5.1.13 CAD Flow Transmitter Ambient Pressure Error (e2P)

The flow transmitter is an electrical device and therefore not affected by ambient pressure.

e2P

=

0 7.5.1.14 CAD Flow Transmitter Temperature Error (e2T)

Temperature error Is considered to be a random variable for a Rosemount transmitter. Therefore e2T

=

0 7.5.1.15 CAD Flow Loop PPC I/O Module Reference Accuracy (A3)

Reference accuracy of the computer input is taken to be the SRSS of the reference accuracies of the SRU and the 1/0 module. The reference accuracy of the SRU is 0.1 % of span. Reference Accuracy tor the 1/0 module is specified as +/- 0.5 °k of span (Sections 2.4.4 and 2.4.5).

A~(J

= ~O.12 +0.52 ::: 0.5099::: 0.51 % span Converting % span to % flow Rearranging Equation A8-10, O'AP =2

  • O'FLOW, to solve for flow (Reference Attachment 8):

A3%Flow =

A~span / 2

=

t 0.51°,10 /2

=

+/-0.255 %

7.5.1.16 CAD Flow Loop PPC I/O Module Humidity Error (a3H)

The manufacturer specifies the 1/0 module operating humidity limits between 0 and 95 0,10 RH (Section 2.4.4). The I/O module is located in the Control Room 533, where humidity may vary from Page 48 of 93

50 to 90 % RH (Reference 4.4.2). Humidity errors are set to zero because they are considered to be included within the reference accuracy specification under these conditions. (Section 3.2) e3H 0

7.5.1.17 CRD Flow Loop PPC 1/O Module Radiation Error (e3R)

Reactor Core Thermal Power LE-0113 Uncertainty Calculation Unit 1 Revision 0 No radiation errors are specified in the manufacturer's specifications. The instrument is located in the Control Room 533, a mild environment [Reference 4.4.2]. Therefore, it is reasonable to consider the normal radiation effect as being included in the reference accuracy. Therefore, e3R

=

0 7.5.1.18 CRD Flow Loop PPC I/0 Module Seismic Error (e3S)

No seismic effect errors are specified in the manufacturer's specifications. A seismic event defines a particular type of accident condition. Therefore, there Is no seismic error for normal operating conditions (Section 3.5).

e3S 0

7.5.1.19 CRD Flow Loop PPC I/C? Module Static Pressure Offset Error (e3SP)

The l/O module is an electrical device and therefore not affected by static pressure.

e3SP =

0 7.5.1.20 CRD Flow Loop PPC i/O Module Ambient Pressure Error (e3P)

The t/O module is an electrical device and therefore not affected by ambient pressure.

e3P 0

7.5.1.21 CRD Flow Loop PPC I/O Module Process Error (e3Pr)

The I/O module receives an analog current input from the flow transmitter proportional to the pressure sensed. Any process errors associated with the conversion of pressure to a current signal have been accounted for as errors associated with flow transmitter. Therefore, e3Pr

=

0 7.5.1.22 CRD Flow Loop Accuracy (LAcRD-Fl1,)

LAM-Flow Therefore, D 1

= :t 0.00°lam Flow

[(A1)2 + (e1 H)2 + (e1 R)2 + (el P)2 + (e1T)2 + (e1 S)2 + (ell SP)2 + (A2) 2 + (a2PS2)2 +

(o2T)2 + (e2H)2 + (e2R)2 + (e2S)2 + (e2V)2 + (e2SP)2 +(e2P) 2 + (e2T)2 + (A3)2 +

(e3H)2 + (e3R)2 + (e3S)2 + (e3SP) 2 + (e3P)2 + (e3Pr)2]1

[(1) 2 +(0) 2 +(0) 2 +(0) 2 + (0)2 + (0)2 +(0) 2 +(0.25) 2 +

2 +

2 +(0) 2 +(5.2

0) 0.49)
0) +

(0)2+(0)2+ (0.66)2+ (0)2+ (0)2 + (0.255)2 + (0)2 + (0), + (0) + (0)2 + (0)2+(0)1]11

=t[1 +0+0+0+0+0+0+0.015625+0+0.2401 +0+25.0+0+0+0.4356+

0+0+0.065025+0+0+0+0+0+0]112 t [26.80323]'

=15.177183 %

LACRD_Fbw = 5.2 7.5.2 CRD Flow Loop Drift (LDCRO-Row) 7.5.2.1 CRD Flow Element Drift Error (D1)

The flow element is a mechanical device; drift error is not applicable for the flow elements.

Page 49 of 93 Exelon.

Reactor Core Thermal Power Uncertainty Calculation Unit 1 LE-G113 Revision 0 7.5.2 7.5.2.1 50 to 90 % AH (Reference 4.4.2). Humidity errors are set to zero because they are considered to be included within the reference accuracy specification under these conditions. (Section 3.2) e3H

=

0 7.5.1.17 CRD Flow Loop PPC 1/0 Module Radiation Error (e3R)

No radiation errors are specified in the manufacturer's specifications. The instrument is located in the Control Aoom 533, a mild environment [Reference 4.4.2]. Therefore, it Is reasonable to consider the normal radiation effect as being included in the reference accuracy. Therefore, e3R

=

0 7.5.1.18 CAD Flow Loop PPC 1/0 Module Seismic Error (e3S)

No seismic effect errors are specified in the manufacturers specifications. A seismic event defines a partiCUlar type of accident condition. Therefore, there Is no seismic error for normal operating conditions (Section 3.5).

e3S

=:

0 7.5.1.19 CAD Flow Loop PPC 1/0 Module Static Pressure Offset Error (e38P)

The 1/0 module is an electrical device and therefore not affected by static pressure.

e3SP

=

0 7.5.1.20 CRD Flow Loop PPC I/O Modure Ambient Pressure Error (a3P)

The JlO module is an electrical device and therefore not affected by ambient pressure.

e3P

=

0 7.5.1.21 CRD Flow Loop PPC I/O Module Process Error (e3Pr)

The 1/0 module receives an analog current input from the flow transmitter proportional to the pressure sensed. Any process errors associated with the conversion of pressure to a current signal have been accounted for as errors associated with flow transmitter. Therefore, e3Pr

=

0 7.5.1.22 CAD Flow Loop Accuracy (LAcRD_FlOW)

LAcAO_FIow =[{A1)2 + (e1 H)2 + (e1 A)2 + (e1 p)2 + (e1T)2 + (e1 8)2 + (e1 SP)2 + (A2)2 + (0'2P82)2 +

(0'2T)2 + (e2H)2 + (e2R)2 + (e28)2 + (e2V)2 + (e28p)2 +J92P)2 + (e2T)2 + (A3)2 +

(e3H)2 + (e3R)2 + (e3S)2 + (e3Sp)2 + (e3p)2 + (e3Pr)21'

=[(1)2 + (0)2 + (0)2 + (0)2 + (0)2 + (0)2 + (0)2 + (0.25)2 + ~0)2 + ~0.49)2 + (0)2 + (5.0).2 +

(0)2 + (0)2 + (0.66)2 + (0)2 + (0)2 + (0.255)2 + (0)2 + (0) + (0) + (0)2 + (0)2 + (0) ]'12

=+/- [1 + 0 + 0 + 0 + 0 + 0 + 0 + 0.015625+ 0 + 0.2401 + 0 + 25.0 + 0 + 0 + 0.4356 +

0+0 + 0.065025 + 0 + 0 + 0 + 0 + 0 + 0]112

= +/- [26.80323]112

=+/- 5.1n183 0/0 LAcRD,..FIoW =+/- 5.2 %

CRD Flow Loop Drift (LDcRD_Row)

CAD Flow Element Drift Error (D1)

The flow element is a mechanical device; drift error is not applicable for the flow elements.

Therefore, D1

=+/- 0.00%

Flow Page 49 of 93

7.5.2.2 CRD Flow Loop Transmitter Drift Error (D2) 7.5.2.3 CRD Flow Loop Drift Error (D2)

Drift error for the transmitter is t 0.25 % of URL / 6 months, taken as a random 2a value (Section 3.4). The calibration frequency is 2 years, with a late factor of 6 months.

D22Q t (0.25 %

  • URL)

Drift is applied to the surveillance interval as follows (Section 6.2.5) :

+/- ([(24 months + 6 months) / 6 months] * [0.0025 URL]

  • 030/6]'[0.0025
  • 750 INWC]2)"

t [17.578125]"

t 4.192627458 1 NWC t 4.191NWC, rounded to level of significance Reactor Core Thermal Power Converting to % span D2

 =

t 4.19 INWC / 197.5 INWC

+/-0.02121-2.121%

Converting % span to % flow Rearranging Equation A8-10, asp = 2

  • aFLrnar, to solve for flow (Reference Attachment 8) :

D2%Fbw D2%s, / 2

+/-2.121 %/2

+/- 1.0608 D2%n,. =

+/- 1.0 7.5.2.4 CRD Flow Loop PPC I/0 Module Drift Error (D3)

The vendor does not specify a drift error for the 1/0 module. Therefore, per Ref. 4.1.1 Section I, it is considered to be included in the reference accuracy.

LE-0113 Uncertainty Calculation Unit 1 Revision 0 D3

+/- 0 7.5.2.5 CRD Flow Loop Drift (LDRwcu-Fk,,)

LDCRD-n,+ = [(D1)2 + (D2)2 + (D3)2]1r2

= [(0)2 + (1.0)1 + (0)2]112

~t[0+ 1.0+0]"2 t [11 1/2 LDCRO_Ft. = t 1.0 7.5.3 CRD Flow Loop Process Measurement Accuracy (PMAcRO_Frow)

No additional PMA effects beyond the effects specified in the calculation of loop accuracy.

PMARwcu_m" = 0 7.5.4 CRD Flow Loop Primary Element Accuracy (PEACRD-Ft

.)

No additional PEM effects beyond the effects specified in the calculation of loop accuracy.

PEARwcu_jqo,r = 0 7.5.5 CAD Flow Loop Calibration Accuracy (CAcRO_Row)

Page 50 of 93 Exelon.

Reactor Core Thermal Power Uncertainty Calculation Unit 1 LE..0113 Revision 0 7.5.2.2 7.5.2.3 CAD Flow Loop Transmitter Drift Error (02)

CAD Flow Loop Drift Error (D2)

Drift error for the transmitter is +/- 0.25 % of UAL /6 months, taken as a random 20' value (Section 3.4). The calibration frequency is 2 years, with a late factor of 6 months.

D220

=

+/- (0.25 %

  • URL)

Drift is applied to the surveillance interval as follows (Section 6.2.5):

D22c

=

+/- {[(24 months + 6 months) 16 months]... [0.0025 URL] 2)1/2

=

+/- ((30/6]"[0.0025... 750 INWCJ2)1/2

=

I: [17.578125]1/2

=

l:4.192627458INWC 0220'

=

+/- 4.19 INWC, rounded to Jevel of significance Converting to Ok span D2%Span=

+/-4.19INWC/197.5INWC

=

+/-0.02121=2.121°k Converting % span to % flow Rearranging Equation AS-1 0, (fAP =2

  • O'FlOW, to solve for flow (Reference Attachment 8):

02%FIow =

D2%Span I 2

=

+/-2.121 %/2

=

+/- 1.06080/0 D2%fJow =

+/- 1.0 0/0 7.5.2.4 CAD Flow Loop PPC I/O Module Drift Error (03)

The vendor does not specify a drift error for the VO module. Therefore, per Ref. 4.1.1 Section I, it is considered to be included in the reference accuracy.

03

=

to 7.5.2.5 CAO Flow Loop Drift (LORWCU_Aow)

LOCRO_AOW

= [(01)2 + (02)2 + (03)2]112

=[(0)2 + (1.0)2 + (0)2]112

=:i: [0 + 1.0 + OJ1/2

=+/- [1]1/2 LDcRo_Flow

= +/- 1.0 %

7.5.3 CAO Flow Loop Process Measurement Accuracy (PMAcRO_Aow)

No additional PMA effects beyond the effects specified in the calcuJation of loop accuracy.

PMAAWcu_Flow = 0 7.5.4 CAD Flow Loop Primary Element Accuracy (PEAcRD_FIow)

No additional PEM effects beyond the effects specified in the calculation of loop accuracy.

PEARWcu_Aow = 0 7.5.5 CAD Flow Loop Calibration Accuracy (CACRO_FIOW)

Page 50 of 93

Standard practice is to specify calibration uncertainty in calculations equal to the uncertainty associated with the instruments under test (Section 4.1.1).

Therefore, CARwcu_Frow =

[(Al )2 + (A2)2 +

(A3)2]1r2 w

[(1 %)2 +(0.25 %)2 +(0.255 %)211r2

[1.25 %2]"2 1.12753 CARwcu-Frnw =

1.06185 % = rounded to 1.1 7.5.6 Total Uncertainty CRD Flow Loop (TUCRD_Flow)

TUDRD_Fl. = [(LA)2 + (LD)2 + (PMA)2 + (PEA)2 + (CA)2]1"2

=f[27.04+1+0+0+1.21]1

-- t [29.25]1 5.4083 TUc.Ro.

= 5.4 7.6 RECIRCULATION PUMP HEAT UNCERTAINTY 7.6.1.1 Recirculation Pump Motor Power (QP)

The recirculation pump system consists of two parallel pumps that maintain forced circulation flow loops in the reactor core. The water originates In the core and returns to the core at a higher pressure. The work performed by the recirculation pumps includes the specific work of the pump plus the pump inefficiency. This energy can be estimated by measuring the power consumed by the pump motor and multiplying the pump motor power by the motor efficiency. The maximum design output of the recirculation motor M-G set is 7700 HP, which is monitored by the watts transducer.

The 7700 HP is conservatively used in lieu of the motor 7500 HP rated in determining the recirculation pump heat uncertainty.

= [(5.2)2 + (1.0)2 +

(0)2

+ (0)2 + (1.1)2]112 Motor WA

_ WA

~m -

WE, motor efficiency ;

pump efficiency

)7P-WA Reactor Core Thermal Power LE-0113 Uncertainty Calculation Unit 1 Revision 0 Brake HP where WA ' -- '?m ' WE, motor output to pump (brake horsepower (HP))

W, = ideal work energy input to fluid system (fluid HP)

Page 51 of 93 Fluid HP OPP Exelon.

Reactor Core Thermal Power Uncertainty Calculation Unit 1 LE-0113 Revision 0 Standard practice is to specify calibration uncertainty in calculations equal to the uncertainty associated with the instruments under test (Section 4.1.1).

Therefore, CARWCU,~FIOW =

[(A1)2 + (A2)2 + (A3)2J'12

=

[(1 %)2 + (0.25 %)2 + (0.255 0/0)2)"2

=

[1.25 %2]'12

=

1.12753 Ok CAAWCU_Flow =

1.06185 % ::: rounded to 1.1 7.5.6 Total Uncertainty CRD Flow Loop (TUCRD_Row)

TUCRO_Flow

[(LA)2 + (LD)2 + (PMA)2 + (PEA)2 + (CA)2]'12

=[(5.2)2 + (1.0)2 + (0)2 + (0)2 + (1.1 )2]'12

= +/- [27.04 + 1 + 0 + 0 + 1.21J'12

=+/- [29.25]112

=+/-5.4083 %

TUCAOMf1o'N

= +/- 5.4 %

7.6 7.6.1.1 RECIRCULATION PUMP HEAT UNCERTAINTY Recirculation Pump Motor Power (Qp)

The recirculation pump system consists of two parallel pumps that maintain forced circulation flow loops in the reactor core. The water originates in the core and returns to the core at a higher pressure. The work performed by the recirculation pumps includes the specific work of the pump plus the pump inefficiency. This energy can be estimated by measuring the power consumed by the pump motor and multiplying the pump motor power by the motor efficiency. The maximum design output of the recirculation motor M-G set is 7700 HP, which is monitored by the watts transducer.

The noo HP is conservatively used in lieu of the motor 7500 HP rated in determining the recircuration pump heat uncertainty.

17m Motor Brake HP Fruid HP W

'l __

A m-WE 1 motor efficiency; where WA := 'lm.We, motor output to pump (brake horsepower (HP>>

'lP = WI 1 pump efficiency WA WI =ideal work energy input to fluid system (fluid HP)

Page 51 of 93

Exelon'.

Heat Added by Pump =

(

(1-77 ~00- WA Reactor Core Thermal Power LE-0113 Uncertainty Calculation Unit 1 Revision 0 WE = electric power input to motor (measured power, Watts)

The difference between the input (actual pump power) and the output (ideal) pump power is the power lost to friction in the pump. The heat added to the pump is due to inefficiency.

The calculation of the total recirculation pump heat input, QP, is taken from Equation 6 (Section 6.2.1.3).

QP

=

2. qr". We Motor Efficiency, qm, is 94.8% at maximum speed of 1690 rpm (Reference 4.8.2) 2

  • 0.948
  • 5.74 MW (based on 7,700 Hp motor (Table 2-1) 10.8866 MW Multiply by 3,412,000 Btu/hr per MW to convert to Btu/hr

=

10.8866 MW

  • 3,412,000 Btu/hr/MW 37,145,079 Btu/hr QP 37,145,000 Btu/hr, rounded to level of significance Where, standard conversion factors are :

1 HP = 550 ft-IbVs 1 ft 3 = 7.48 gal 1 W = 3.412 Btu/hr 1 HP = 0.7457 kW 1 HP = 2544.43 Btu/hr 7.6.2 Recirculation Pump Motor Watt Transducer Loop Uncertainty 7.6.2.1 Recirculation Pump Motor Watt Transducer Reference Accuracy (A1)

The accuracy of the transducer is t 0.2 °la of Reading + 0.01 % Rated Output at 0 to 200% of the Rated Output (reference Section 2.5.2).

Maximum error is when the reading Is equal to the recirculation pump motor power rating in MW, i.e.,

7700 HP (5.74 MW, Reference 4.3.3).

A1 2a=

t(0.2%*5.74MW+0.01%*10.5MW)

t 0.01253 MW A12a

t 0.013 MW, rounded to level of significance 7.6.2.2 Recirculation Pump Motor Watt Transducer Power Supply Effects (a1 PS)

Power supply effects are considered to be negligible (Section 3.7). Therefore, ai PS =

0 7.6.2.3 Recirculation Pump Motor Watt Transducer Ambient Temperature Error (a1T)

The Watt Transducer is located in the auxiliary electrical room. This is a controlled environment ;

therefore, temperature error can be neglected.

Page 52 of 93 (Equation 40)

Exelon.

Reactor Core Thermal Power Uncertainty Calculation Unit 1 LE-0113 Revision 0 7.6.2.2 7.6.2 7.6.2.1 WE = electric power input to motor (measured power, Watts)

The difference between the input (actual pump power) and the output (ideal) pump power is the power lost to friction in the pump. The heat added to the pump is due to inefficiency.

Heal Added by Pump = ((1-17j{00). WA The calculation of the total recirculation pump heat input, Qp, is taken from Equation 6 (Section 6.2.1.3).

Qp 2* 11m' We (Equation 40)

Motor Efficiency, 'lm, is 94.8%

at maximum speed of 1690 rpm (Reference 4.8.2)

=

2

  • 0.948
  • 5.74 MW (based on 7,700 Hp motor (Table 2-1) 10.8866 MW Multiply by 3,412,000 Btulhr per MW to convert to Btuthr

=

10.8866 MW

  • 3,412,000 Btulhr/MW

=

37,145,079 Btuthr Qp 37,145,000 Btulhr, rounded to level of significance Where, standard conversion factors are:

1 HP::: 550 ft-Ib,ls 1 fe = 7.48 gal 1 W ::: 3.412 Btulhr 1 HP::: 0.7457 kW 1HP =2544.43 Btulhr Recirculation Pump Motor Watt Transducer Loop Uncertainty Recirculation Pump Motor Watt Transducer Reference Accuracy (A1)

The accuracy of the transducer is:t: 0.2 % of Reading +0.01 % Rated Output at 0 to 200% of the Rated Output (reference Section 2.5.2).

Maximum error is when the reading is equal to the recirculation pump motor power rating in MW, i.e.,

7700 HP (5.74 MW, Reference 4.3.3).

A1 20':::

t: (0.20/0
  • 5.74 MW + 0.010/0'" 10.5 MW)

+/- 0.01253 MW A120' :::

+/- 0.013 MW, rounded to revel of significance Recirculation Pump Motor Watt Transducer Power Supply Effects (0'1 PS)

Power supply effects are considered to be negligible (Section 3.7). Therefore, 0'1 PS =

0 7.6.2.3 Recirculation Pump Motor Watt Transducer Ambient Temperature Error (0'1T)

The Watt Transducer is located in the auxiliary electrical room. This is a controlled environment; therefore, temperature error can be neglected.

Page 52 of 93

e1 SP 61T

=

0 7.6.3 Watt Transducer Humidity, Radiation, Pressure, and Temperature Errors (el H, e1 R, e1 P, e1T)

The watt transducer is an electrical device and the output is not subject to environmental or vibration effects. Therefore ;

elH=elR=elP=e1T=0 7.6.3.1 Recirculation Pump Motor Watt Transducer Seismic Error (e1 S)

A seismic event is an abnormal operating condition and is not addressed by this calculation.

Therefore ;

e 1 S 0

7.6.3.2 Recirculation Pump Motor Watt Transducer Static Pressure Error (e1 SP)

The watt transducer is an electrical device not affected by process pressure. Therefore ;

0 Reactor Core Thermal Power Uncertainty Calculation Unit 1 Revision 0 7.6.3.3 Recirculation Pump Motor Watt Transducer PPC I/O Module Reference Accuracy (A2)

Reference accuracy of the computer input is taken to be the SRSS of the reference accuracies of the SRU and the I/0 module. The reference accuracy of the SRU is 0.1 % of span as given in Section 2.5.3. Reference Accuracy for the 1/0 module is specified as +/- 0.5 % of span (Section 2.5.4) and considered to be a 2a value (Section 3.4).

A22d

=

0.12 + 0..52 = 0.509901951= 0.51

A22, t 0.51 %
  • range A22a t 0.0051
  • 10.5 MW = 0.05355 MW 7.6.3.4 Recirculation Pump Motor Watt Transducer PPC I/0 Module Humidity Error (e2H)

The manufacturer specifies the I/0 module operating humidity limits between 0 and 95 % RH LE-0113 (Section 2.4.4). The I/0 module is located in the Control Room 533, where humidity may vary from 50 to 90 % RH (Reference 4.4.2). Humidity errors are set to zero because they are considered to be included within the reference accuracy specification under these conditions. (Section 3.2) e2H 0

7.6.3.5 Recirculation Pump Motor Watt Transducer PPC 1/0 Module Radiation Error (e2R)

No radiation errors are specified in the manufacturer's specifications. The instrument is located in the Control Room 533, a mild environment (Section 4.4.2). Therefore, it is reasonable to consider the normal radiation effect as being included in the reference accuracy. Therefore, e2R

= 0 7.6.3.6 Recirculation Pump Motor Watt Transducer PPC I/0 Module Seismic Error (e2S)

No seismic effect errors are specified in the manufacturers specifications. A seismic event defines a particular type of accident condition. Therefore, there is no seismic error for normal operating conditions (Section 3.5).

e2S

=

0 7.6.3.7 Recirculation Pump Motor Watt Transducer PPC I/0 Module Static Pressure Offset Error (e2SP)

The 1/0 module is an electrical device and therefore not affected by static pressure.

e2SP =

0 Page 53 of 93 Exelon.

Reactor Core Thermal Power Uncertainty Calculation Unit 1 LE-0113 Revision 0 0'11

=

0 7.6.3 Watt Transducer Humidity, Radiation, Pressure, and Temperature Errors (e1 H, e1 A, e1 P, e1T)

The watt transducer is an electrical device and the output is not subject to environmental or vibration effects. Therefore; e1 H =e1 R = e1 P =elT = 0 7.6.3.1 Recirculation Pump Motor Watt Transducer Seismic Error (e1 S)

A seismic event is an abnormal operating condition and is not addressed by this calculation.

Therefore; e1S

=

0 7.6.3.2 Recirculation Pump Motor Watt Transducer Static Pressure Error (e1 SP)

The watt transducer is an electrical device not affected by process pressure. Therefore; e1SP =

0 7.6.3.3 Recirculation Pump Motor Watt Transducer PPC I/O Module Reference Accuracy (A2)

Reference accuracy of the computer input is taken to be the SASS of the reference accuracies of the SRU and the 1/0 module. The reference accuracy of the SRU is 0.1 % of span as given in Section 2.5.3. Reference Accuracy for the I/O module is specified as +/- 0.5 % of span (Section 2.5.4) and considered to be a 20' value (Section 3.4).

= ~O.12 +0.5 2 =0.509901951= 0.51 0/0 A22C1

=

+/- 0.51 %

.. range A22G

=

+/- 0.0051

  • 10.5 MW =0.05355 MW 7.6.3.4 Recirculation Pump Motor Watt Transducer PPC 1/0 Module Humidity Error (e2H)

The manufacturer specifies the 1/0 module operating humidity limits between 0 and 95 % RH (Section 2.4.4). The I/O module is located in the Control Room 533, where humidity may vary from 50 to 90 Ok RH (Reference 4.4.2). Humidity errors are set to zero because they are considered to be included within the reference accuracy specification under these conditions. (Section 3.2) e2H

=

0 7.6.3.5 Recirculation Pump Motor Watt Transducer PPC I/O Module Radiation Error (e2A)

No radiation errors are specified in the manufacturer's specifications. The instrument is located in the Control Room 533, a mild environment (Section 4.4.2). Therefore, it is reasonable to consider the normal radiation effect as being included in the reference accuracy. Therefore, e2R

=

0 7.6.3.6 Recirculation Pump Motor Watt Transducer PPC 110 Module Seismic Error (e28)

No seismic effect errors are specified in the manufacturer's specifications. A seismic event defines a partiCUlar type of accident condition. Therefore, there is no seismic error for normal operating conditions (Section 3.5).

e2S

=

0 7.6.3.7 Recirculation Pump Motor Watt Transducer PPC I/O Module Static Pressure Offset Error (e2SP)

The I/O module is an electrical device and therefore not affected by static pressure.

e2SP =

0 Page 53 of 93

C-xeiurt.

7.6.3.8 Recirculation Pump Motor Watt Transducer PPC I/0 Module Ambient Pressure Error (e2P)

The I/(3 module is an electrical device and therefore not affected by ambient pressure.

e2P 0

7.6.3.9 Recirculation Pump Motor Watt Transducer PPC 1/0 Module Process Error (e2P)r Any process errors associated with the conversion of pressure to a current signal have been accounted for as errors associated with Watt Transducer. Therefore, e2Pr 0

7.6.3.10 Recirculation Pump Motor Watt Transducer Loop Accuracy (LAp)

LAP LAP 7.6.4 Recirculation Pump Motor Watt Transducer Loop Drift (LDP) 7.6.4.1 Recirculation Pump Motor Watt Transducer Drift Error (D1)

Reactor Core Thermal Power Uncertainty Calculation Unit 1 Revision 0 The drift will be considered random and independent over time. The surveillance frequency Is 30 months ; therefore, drift error follows Equation 39 (Section 6.2.5 and reference 2.5.2).

D1 t 0.1 % of RO per year

[(30!12) - (+/- 0.1 % RO)2]t12

[2.5 - 0.00011025 MW2]1/2 0.016601958 MW D1 0.017 MW, rounded to level of significance 7.6.4.2 Recirculation Pump Motor Watt Transducer PPC 1/0 Module Drift Error (D2)

The vendor does not specify a drift error for the I/0 module. Therefore, per Ref. 4.1.1 Section I, it is considered to be Included in the reference accuracy.

D2

=

t 0 7.6.4.3 Recirculation Pump Motor Watt Transducer Loop Drift (LDRWCU_FjO')

LDp LDP 7.6.5 Recirculation Pump Motor Watt Transducer Process Measurement Accuracy (PMAP)

No additional PMA effects beyond the effects specified in the calculation of loop accuracy.

PMAP =

0 7.6.6 Recirculation Pump Motor Watt Transducer Primary Element Accuracy (PEAp)

Page 54 of 93 LE-0113

[(Al

+

+

+

+

+

+

2 (A2)2

)2 (a1 PS)2 (alT)2 (e1 H)2 (e1 R)2 (e1 (e1T)2 + (e1 S) + (e1 SP)2 +

+ (e2H)2 + (e2R)2 + (e2S)2 + (e2SP)2 + (e2P)

P)2

+ (e2Pr)2]1/ 2 t [(0.013)2 + (0) 2 +~(0)2 +(0) 2 + (0)2 + (0)2 + (0)2 + (0.05355)2 + (0)2 + (0)2 + (0) 2 +

(0)2+

(0) 2 + (0)] '

t[0.000169+0+0+0+0+0+0+0+0+0.002867603+0+0+0+0+0+01 1

~2 t [0.003036603 MW']

  • 0.055105376 MW t 0.055 MW

=

t [(D1)2 + (D2)2]'r2

=

t [(0.017)2 + (0)2],r2 t 0.017 MW Exelon.

Reactor Core Thermal Power Uncertainty Calculation Unit 1 LE-G113 Revision 0 7.6.3.8 Recirculation Pump Motor Watt Transducer PPC I/O Module Ambient Pressure Error (e2P)

The I/O module is an electrical device and therefore not affected by ambient pressure.

e2P

=

0 7.6.3.9 Recirculation Pump Motor Watt Transducer PPC I/O Module Process Error (e2P)r Any process errors associated with the conversion of pressure to a current signal have been accounted for as errors associated with Watt Transducer. Therefore.

e2Pr

=

0 7.6.3.10 Recirculation Pump Motor Watt Transducer Loop Accuracy (LAp)

LAp

=

t [(A1)2 + (<11 PS)2 + (<11T}2 + (el H)2 + (el R}2 + (el Pt2 + (e1T)2 + (e18)2 + (e1 SP)2 +

(A2)2 + (e2H)2 + (e2R)2 + (e2S)2 + (e28p)2 + (e2P) + (e2Pr)2]112

=

+/- [(0.013f + (0)2 +,JO)2 + (0)2 +(of + (0)2 + (0)2 + (0.05355)2 + (0)2 + (0)2 + (0)2 +

(0)2 + (0)2 + (O>>)'

=

+/- [0.000169 + 0 + 0 + 0 + 0 + 0 + 0 + 0 + 0 + 0.002867603 + 0 + 0 + 0 + 0 + 0 + 0]1/2 at

+/- [0.003036603 MW2)112

=

+/- 0.055105376 MW LAp

=

+/- 0.055 MW 7.6.4 7.6.4.1 7.6.4.2 7.6.4.3 7.6.5 7.6.6 Recirculation Pump Motor Watt Transducer Loop Drift (LOp)

Recircu'ation Pump Motor Watt Transducer Drift Error (01)

The drift will be considered random and independent over time. The surveillance frequency;s 30 months; therefore, drift error follows Equation 39 (Section 6.2.5 and reference 2.5.2).

D1

=

+/- 0.1 °16 of RO per year

=

[(30/12) *{+/- 0.1 Ok RO)2]112

=

[2.5. 0.00011025 MW2]1/2

=

0.016601958 MW 01

=

0.017 MW, rounded to level of significance Recirculation Pump Motor Watt Transducer PPC I/O Module Drift Error (02)

The vendor does not specify a drift error for the I/O module. Therefore, per Ref. 4.1.1 Section I, it is considered to be included in the reference accuracy.

02

=

+/-O Recirculation Pump Motor Watt Transducer Loop Drift (lORWCtLAow)

LOp

=

+/- [(Dl)2 + (02)2]112

=

+/- [(0.017)2 + (0)2fl2 LOp

=

+/- 0.017 MW Recirculation Pump Motor Watt Transducer Process Measurement Accuracy (PMAp)

No additional PMA effects beyond the effects specifjed in the calculation of roop accuracy.

PMAp =

0 Recirculation Pump Motor Watt Transducer Primary Element Accuracy (PEAp)

Page 54 of 93

Exelon.

No additional PEM effects beyond the effects specified in the calculation of loop accuracy.

PEAP =

0 7.6.7 Recirculation Pump Motor Watt Transducer Loop Calibration Accuracy (CAP)

Standard practice is to specify calibration uncertainty in calculations equal to the uncertainty associated with the instruments under test.

Therefore, CAp t [(A1)2 + (A2)2]'"2

=

t [(0.013)2 + (0.05355)2]u2 t [0.000169 + 0.002667603]"2

+/- [0.003036603]'*

t 0.055105376 CAp

+/- 0.055 MW 7.6.8 Total Uncertainty Recirculation Pump Motor Watt Transducer Loop (TUP)

TUp Converting to °lo motor power TUP

=

t 0.08 MW l 5.74 MW t 0.0 13937282 t1.4%

7.7 DETERMINATION OF CTP UNCERTAINTY 7.7.1 Numerical Solutions for the Partial Derivative Terms Solutions are found in two parts by first substituting values into Equations 20 through 27, as shown.

a CTP = (ho (Ps) - hF (TFW ))

a WFw

= (ho (1043 psig) - hF (430.8 O F)) Btullbm

= (1191.1 - 409.71) Btulibm 781.39 Btullbrn Reactor Core Thermal Power Uncertainty Calculation Unit 1 Revision 0 dCTP = WFw = 15,090,000 Ibmlhr (Reference Equation 21) aho (Ps )

Page 55 of 93 (Reference Equation 20)

LE-0113 t [(LA)2 + (LD)2 + (PMA)2 + (PEA)2 + (CA)2]'"z

=

t [(0.055)2 + (0.017)2 + (0)a + (0)2 + (0.055)2]'"

t [0.003025+ 0.000289 + 0 + 0 + 0.003025]1"2

=

t [0.006339]'"

t 0.079617837 MW Exelon.

Reactor Core Thermal Power Uncertainty Calculation Unit 1 LE-G113 Revision 0 No addnional PEM effects beyond the effects specified in the calculation of roop accuracy.

PEAp =

0 7.6.7 Recirculation Pump Motor Watt Transducer Loop Calibration Accuracy (CAp)

Standard practice is to specify calibration uncertainty in calculations equal to the uncertainty associated with the instruments under test.

Therefore, CAp

=

+/- [(A1)2 + (A2)2]112

=

+/- [(0.013)2 + (0.05355)2]'/2

=

+/- [0.000169 + 0.002867603]112

=

+/- [0.003036603j1fl

=

+/- 0.055105376 CAp

=

+/- 0.055 MW 7.6.8 Tatar Uncertainty Recirculation Pump Motor Watt Transducer Loop (TUp)

TUp

=

t: [(LA)2 + (LD)2 + (PMAf + (PEA)2 + {CA)2]112

=

+/- [(0.055)2 + (0.017)2 + (0)2 + (0)2 + (0.055)2]'12

=

+/- [0.003025+ 0.000289 + 0 + 0 + 0.003025J1/2

=

+/- [0.006339]1/2

=

+/- 0.079617837 MW Converting to Ok motor power TUp

=

to.08 MW /5.74 MW

=

+/- 0.013937282

=

+/-1.4 %

7.7 DETERMINATION OF CTP UNCERTAINTY 7.7.1 Numerical Solutions tor the Partial Derivative Terms Solutions are found in two parts by first substituting values into Equations 20 through 27. as shown.

=(hG (1043 psig) - hF (430.8 <>F)) Btullbm (Reference Equation 20)

= (1191.1 409.71) Btullbm

= 781.39 Btu/Ibm

_dC_T_P_ = WFW = 15,090,000 Ibmlhr dha(Ps)

Page 55 of 93 (Reference Equation 21)

Reactor Core Thermal Power Uncertainty Calculation Unit 1 Revision 0 7.7.2 Component Uncertainty Terms The component uncertainty terms, uc, are found by substituting values into Equations 31, 33 through 37, as shown.

7.7.2.1 Peedwater Mass Plowrate Uncertainty or -Fw `` 48,2881bm/hr (Section 7.1) 7.7.2.2 Steam Enthalpy Uncertainty ah,(Ps.lo) x

( A

~ " +

h ;s (lo )

APs

~

Al, Aho (Ps)

(Reference Equation 31)

LE-01 1 3 The nominal steam dome pressure is evaluated at 1043 psig (Reference 4.8.8) with an uncertainty of t 20 psig (Section 7.2). Steam table interpolation error is taken to be one-half of the least significant figure shown, i.e., t 0.05 Btu/Ibm.

Page 56 of 93 aCTP = -VVFw = --

,15090000

,Ibm/h r

(Reference Equation 22) ahr (TFW )

a CTP (Reference Equation 23) a QCRO OUT

_CTP

_1 (Reference Equation 24) a Qcna-Ire a CTP (Reference Equation 25) a Q wAo a CTP1

=

{Reference Equation 26) a QRWCU a CTP =-1 (Reference Equation 27) a (~ p Exelon.

Reactor Core Thermal Power Uncertainty Calculation Unit 1 LE-0113 Revision 0 aCTP

--- = - WFW

== -15,090,000 Ibm/hr ohF(TFW )

aCTP


1 dQCRO_OUT aCTP

=-1 aOCROIN aCTP =1 iJQRAO aCTP

=1 a°AWCU (Reference Equation 22)

(Reference Equation 23)

(Reference Equation 24)

(Reference Equation 25)

(Reference Equation 26)

(Reference Equation 27) aCTP =-1 oap 7.7.2 Component Uncertainty Terms The component uncertainty terms, uc, are found by sUbstituting values into Equations 31, 33 through 37, as shown.

7.7.2.1 Feedwater Mass FJowrate Uncertainty q w

= 48,288 Jbmlhr (8 t'

7 1)

FW ec Ion 7.7.2.2 Steam Enthalpy Uncertainty a

(P I);::: (. ~hG(PS>>)2

  • a 2+(l1hG(Jo >>)2.... a. 2 ha

$, 0

~Ps.

p..

~Io rill (Reference Equation 31)

The nomina' steam dome pressure is evaluated at 1043 psig (Reference 4.8.8) with an uncertainty of +/- 20 psig (Section 7.2). Steam table interpolation error is taken to be one-half of the least significant figure shown. i.e.* +/- 0.05 Btu/Ibm.

Page 56 of 93

Exelon, a4" (PS 11") =

=.10

0.80156 Btu Ibm QN(TFnr,PFW "lo)

he (1063) -- he (1023)

Btu / lbm 20 psi ' +

1063-1023 psi

([h, (1190.3) + 0.05] - [hc (1191.9) -- 0.05])'

1190.3 -- 1191.9

)2

( Btu / Ibm

  • (20 si ' +

1063-1023 psi p

l Btu / Ibm (1191.2 ---1191.112 0-05+0.05

-1.60 2 Btu I Ibm 2

40

  • (20 psi2 )+

) (

psi

)

0.1 2 0.1

()

7.7.2.3 Feedwater Enthalpy Uncertainty 0.05- (-- 0.05)

Btu 2 +0.0025 Btu Ibm

!bm 2

Reactor Core Thermal Power Uncertainty Calculation Unit 1 Revision U Btu hr 0.05 Btu 2

hr Btu / Ibm Btu hr 6h,, (Ps,lo ) = 0.802 bm

, rounded to the inherent uncertainty of the steam table.

2 0.05 Btu hr )

Ahr(T) 2

, 0T2 + ~hF (P

(

)

2

  • ap2 + AhF 2 _a,.2 (Reference Equation 33)

AT AP )

©lo )

From Table 2-1, feedwater pressure is given as 1155 t 10 psig. From Table 2-1, feedwater temperature is given as 430.8 t 0.57 °F. Enthalpy of water is from Attachment 6. Steam table interpolation error is taken to be one-half of the least significant figure shown, i.e., +/- 0.005 Btu/Ibm.

Page 57 of 93 Exel~n.

Reactor Core Thermal Power Uncertainty Calculation Unit 1 LE-0113 Revision 0

(

hG (1063) - hG (1023>>)2 (Btu/lbmJ2 * (20 psiY +

1063 - 1023 psi

(

[hG (1190.3) +0.05] - rho (1191.9) - 0.05lJ2[Btu!Ibm]' *(0.05 Btu)2 0.05 - (- 0.05)

Btu hr hr

(

1190.3 -1191.9 J2(BfUllbmJ2 * (20 pSi)2 +

1063 - 1023 psi

= CI~'~~:~~:;'lr[BtrtmJ *(O.05~;r

-1.60)2(StulI~m)2 11 (20pSi2)+

40 PSI

= (o.1)2[StU/lbm]2-(0.05 BfU)2 0.1 Btu hr hr

=

0.064 Btu 2 +0.0025 Btu 2

fbm2 Ibm2

= 0.80156 Btu Ibm O'h (Ps I ):: 0.802 Btu, rounded to the inherent uncertainty of the steam table.

G

' 0 Ibm 7.7.2.3 Feedwater EnthaJpy Uncertainty o

('t P

J) = (AhF(T>>)2.C1 2 +(AhF(p))2 * (] 2 +(ilhF)2 -0 2 (Reference Equation 33) hy:

FW' FW, 0 liT T

,ap Pilla 10 From Table 2-1, feedwater pressure is given as 1155 +/- 10 ps;g. From Table 2-1, feedwater temperature is given as 430.8 +/- 0.57 OF. Enthalpy of water is from Attachment 6. Steam table interpolation error is taken to be one-half of the least significant figure shown, i.e., +/- 0.005 Btu/Ibm.

Page 57 of 93

Exelon'.

47hF (TF, PFw, la ) =

= 0.62504000 BTU ffh, (TFyy, PFw, Io ) = 0.625 BTU Reactor Core Thermal Power Uncertainty Calculation Unit 1 Revision 0 h F (431.37) -- hF (430.23)

)2( Btu l lbm )' * (0.57 -F)2 +

431.37-430.23

°F hF (1165) - hF (1145) )2 Btu/lbm

)2

' (10 Psi)' +

1165-1145 psi (1h F (430.8) + 0.005] - (h, (430.8) - 0.005] z 0.005- (:0.005) 410.33 - 409.08 )2( Btu 1 Ibm 431.37 - 430.23 nF

"` (0.57 F)

)

409.71-- 409.70

)2 Btu /Ibm

)2

" (10 psi)2 +

1165 -1145 psi 409.72-409.71 )2 Btu / Ibm 0.005 13tu 0.005 +0.005 Btu Ibm lbrn 1.25 2 Btu l lbm )2

  • (0.57 OF) 2 +

1.14 OF 0.01 2 Btu / !bm )2, (10 psi)2 +

20 psi Btu /lbrn Btu

!bm Ibm 2

0.005 lBbm 2

2 2

0.39063 Btu 2 +0.00002 Btu 2 +0.00002 Btu Ibm Ibm Ibm 2

Ibm, rounded to the inherent uncertainty of the steam table.

The accuracy of the temperature measurement determines the feedwater enthalpy uncertainty.

Variations in pressure and steam table interpolation are negligible.

Page 58 of 93 Btu / Ibm Btu lbm (0.005 Btu Ibm)

LE-0113 Exelon.

Reactor Core Thermal Power Uncertainty Calculation Unit 1 LE-0113 Revision 0 hF(431.37) - hF(430.23>>)2 (Btu/lbrn)2

(

)2

  • 0.57'1=

+

431.37 - 430.23 of q~ (TFW,PFw,lo ) =

(hF(\\\\6~=~~~145)r(Blu:~~mr*(10ps1)2 +

(

[hF(430.8)+ 0.005]-". (hF(430.8)~ 0.005])2[8tU/ Ibm]2 *(0.005 BIU)2 0.005- (-0.005)

Btu Ibm Ibm 410.33 - 409.08)2(Btu/lbrn)2 * (O.57~)2 +

431.37 - 430.23 of

= (409.71-409.70)2 (Btu/lbrn)2 * (10 pSi)2 +

1165 -1145 psi

( 409.72 - 409.71)2[StU/lbm]2 * (0.005 BIU).2 0.005 +0.005 Btu Ibm Ibm 1.25J2(Btu/lbrn)2 * (O.57Of)2 +

1.14 of

(0.01)2 (Btu/lbrn)2 * (10 psi)2 +

20 psi

(

...0.01)2[BtU/lbm]2 '(0.005 BtU)~

ani Bfu bm Ibm 0.39063 Btu 2

+0.00002 Btu 2

+ 0.00002 Btu 2

Ibm2 rbm 2 Ibrn2

0.62504000 BTU Ibm BTU O'h (TFw,PFW,lo );;;; 0.625--

F Ibm,rounded to the inherent uncertainty of the steam table.

The accuracy of the temperature measurement determines the feedwater enthalpy uncertainty.

Variations in pressure and steam table interpolation are negligible.

Page 58 of 93

xelu~

7.7.2.4 Control Rod Drive Outlet Energy Uncertainty The control rod drive outlet energy uncertainty is dominated by the uncertainty of the mass flow rate. The steam table interpolation error is negligible and is not used in the following computation.

CRD flow rate uncertainty, XCFID, is 5.4 % of flow (Section 7.5.6).

The variation in hf is shown below.

Converting oaf to % of hf

= :t 0.70 BtuAbm / 71.93 Btu/Ibm

=

  • 0.00973 = 0.97 %

Reactor Core Thermal Power LE-01 13 Uncertainty Calculation Unit I Revision 0 hf (100. 7) - hf (99.3 )'( Btu / Ibm (0.70F Y 100.7-99.3 OF a

11f V CRD ) =

(hf (1458) - hf (1438 2 Btu / 1bm (10 psi)'

1 1458 --1438 0F I (

)()

(72-63)-(71.24 2 Btu l Ibm 2

  • (0.7-Fy 100.7-99.3 OF

+

(71,96)7

)

2 (11u l lbrn ) * (1 ()pSi)2

( 14,58 -1438 OF (1.39 2 (Btu/Ibm 2

(0-7-Fy 140)

OF

)

+ ((0.05, SW IM

(_)

(1 op$1 )2 20 OF VO.483025 130 / Ibn? + 0.000676 Btu' / Ibm'

= 40.483740113tU 2 /IbM2

= 0.69549 Btuftm, rounded 0.70 Btu/lbrn The SRSS method is used to combine the flow ate and enthalpy uncertainty terms.

XCW.OUT % = ~,5.42 -

+ o.972 = 30.1 0(79 =: 5.486429 % = 6 %, conservatively rounded up COCIO-1111T

= XCFID-OUT %

- QCA0,,0UT (Reference Equation 34)

= 6 %

  • hG(1 043 psig)
  • WCRD

= 6 %

  • 1131.1 Btu/Ibm
  • 105 gal/min
  • 60 m in/hr
  • 62.265 Ibm/fe / 7.48 gaVft3

= 3,747,851.883 Btu/hr

= 3,748,000 Btu/hr, rounded to level of significance Page 59 of 93 Exelon.

Reactor Core Thermal Power Uncertainty CaJculation Unit 1 LE-0113 Revision 0 7.7.2.4 Control Rod Drive Outlet Energy Uncertainty The control rod drive outlet energy uncertainty is dominated by the uncertainty of the mass flow rate. The steam table interpolation error is negligible and is not used in the folfowing computation.

CRD flow rate uncertaintYf XCRO, is 5.4 % of flow (Section 7.5.6).

Tha variation in hf is shown below.

(

hl (100.7) - hl (99.3.)l( BIU o/lbm)2 *(O.7f>F yz too.7 - 99.3 F

+(hi (1458) - hi(1438)2( Btu I Ibm ) *(10 siY 1458 -1438 of P

(72.63) -(71.24)2(BtU:lbm)2,. (o.7ot=f 100.7-99.3 F

+(71.96) - (71.91)2(Btu/lbrn) *(1 OpSi)2 1458 -1438 of (l~:n Btu~bmr*(0.70f'f

+(l~~5 nBtU~bm)- (1 Opsl)2

~0.483025 Bft? IIbm2 +O.000676Btif IIbrif
J0.48374018tu2 /1bm2

=0.69549 Btullbm, rounded 0.70 Btu/Ibm Converting Chf to % of h,

t
0.70 Btullbm /71.93 Btul1bm

=:t: 0.00973 ;:: 0.97 %

The SASS method is used to combine the flow rate and enthalpy uncertainty terms.

XCROc,OUT %

~5.4 2 + 0.972

=:.J30.1 009

5.486429 % =6 0/0, conservatively rounded up (Reference Equation 34)

=6 %

1r ha(1043 pslg) 11 WeRD

6 %
  • 1191.1 Btu/Ibm
  • 105 gaVmin,. 60 minlhr" 62.265Ibm/ft3 /7.48 garJft3
3,747,851.883 Btu/hr
3,748,000 Btulhr, rounded to level of significance Page 59 of 93

on.

7.7.2.5 Control Rod Drive Inlet Energy Uncertainty Reactor Core Thermal Power LE-0113 Uncertainty Calculation Unit 1 Revision 0 The control rod drive inlet energy uncertainty is dominated by the uncertainty of the mass flow rate.

CRD flow rate uncertainty, xcFO, is 6 %, which was conservatively rounded up from 5.5%, of flow (Section 7.7.2.4). CRD normal operating temperature and pressure is 100 "F and 1448 psig (Section 4.4.1).

aQCFO-. W

= XCRO-ire %

' QCRO -IN (Reference Equation 35)

= 6 %

  • h t(100 °F at 1448 psig)
  • WCRD 6 %
  • 71.93 Btullbm
  • 105 gal/min
  • 60 min/hr
  • 62.265 Ibmlft3 I 7.48 gal/ W

= 226,331.111 Btu/hr

= 226,000 Btu/hr, rounded to level of significance 7.7.2.6 Reactor Pressure Vessel Heat Loss Uncertainty The determination of reactor heat loss was done in Reference 4.8.3 for both Unit 1 and Unit 2. The Unit 1 heat loss was calculated to be 0.89 MWt. The Unit 2 heat loss was calculated to be 1.04 MWt. A conservative variance of 10 % of the heat loss rate is used to estimate the heat loss uncertainty.

GORao = XRAO %

" ORAD TORWCU W

RWCU 10 %

  • 0.89 MWt
  • 3,412,000 Btu/hr l MWt

= 303,668 Btu/hr 7.7.2.7 RWCU Heat Removal Uncertainty xrWcu is set 2.3 % based on the flow loop uncertainty.

aoRw

= XRWCU %

" QRWCU Remembering that Q'awcu = WRWCU

  • thF(TRd<<j,) -- hf(TFw)j (See Equation 9)

= 2.3 %. wnwcu - [hf (TnEcjRc_gN) - hf (TFW )l Page 60 of 93 (Reference Equation 36)

The RWCU flow loop uncertainty is t 2.3 % of flow (Section 7.3.6). The RWCU temperature measurement uncertainty is t 4.37 "F (Section 7.4.6). The variation in enthalpy over the range of temperatures in this loop of about 0.5 % is negligible compared to the 2.3 % flow variation. RWCU normal operating temperatures and pressures are 530 OF and 1060 psig (RWCU suction) and 438

  • F and 1168 psig (RWCU discharge) (Reference 4.4.1). Temperature values of 530 'IF (suction) and 440'OF (discharge) are used to be more in line with actual measurements.

(Reference Equation 37)

=154,000 Ihm (Pump A) and 133,000 I h m (Pump B and C) r (Section 2.2.1)

Maximum heat removal occurs with Pump A running, therefore, WRwcu = 154,000 Ibmlhr.

Exelon.

Reactor Core Thermal Power Uncertainty Calculation Unit 1 LE*0113 Revision 0 7.7.2.5 Control Rod Drive Inlet Energy Uncertainty The control rod drive inlet energy uncertainty is dominated by the uncertainty of the mass flow rate.

CRD flow rata uncertainty, XCAO, is 6 %, which was conservatively rounded up from 5.5%, of flow (Section 7.7.2.4). CRD normal operating temperature and pressure is 100 ~F and 1448 psig (Section 4.4.1).

(Reference Equation 35)

(Reference Equation 36)

= 6 %

  • ht(100 of at 1448 psig)
  • WCRD

= 6 %

1/ 71.93 Btu/Ibm

  • 105 gaVmin.. 60 min/hr" 62.265Ibm/fe /7.48 gaV te

=226,331.111 Btulhr

=226,000 Btu/hr, rounded to level of significance 7.7.2.6 Reactor Pressure Vessel Heat Loss Uncertainty The determination of reactor heat loss was done in Reference 4.8.3 for both Unit 1 and Unit 2. The Unit 1 heat loss was calculated to be 0.89 MWt. The Unit 2 heat loss was calculated to be 1.04 MWt. A conservative variance of 10 % of the heat loss rate is used to estimate the heat loss uncertainty.

(J ORAD

XRAO %
  • ORAD

= 10 %

  • 0.89 MWt* 3,412,000 Btulhr / MWt

=303,668 Btulhr 7.7.2.7 RWCU Heat Removal Uncertainty The RWCU flow loop uncertaInty is +/- 2.3 % of flow (Section 7.3.6). The AWCU temperature measurement uncertainty is +/- 4.37 OF (Section 7.4.6). The variation in enthalpy over the range of temperatures in this loop of about 0.5 Ok is negligible compared to the 2.3 % flow variation. AWCU normal operating temperatures and pressures are 530 OF and 1060 pslg (RWCU suction) and 438 OF and 1168 psig (RWCU discharge) (Reference 4.4.1). Temperature values of 530 {)F (suction) and 440 OF (discharge) are used to be more in line with actual measurements.

XAWCU Is set 2.3 % based on the flow loop uncertainty.

XAWCU %
  • QRWCU (Reference Equation 37)

Remembering that QAWCU =WRWCU * [hdTAeclrcjn) - hF(TFW)] (See Equation 9)

O'°RWCU WRWCU Ibm Ibm

== 154,OOO-hr(PumpA) and 133,OOO-hr (Pump B and C)

(Section 2.2.1)

Maximum heat removal occurs with Pump A running, therefore, WAWCU =154,000 Ibm/hr.

Page 60 of 93

Exelon.

hf (TREciac,IH) = 524.39 Btu tat 530 O1= and 1060 psig)

Ibm hf (TFw)

= 419.83 Btu (at 440 °F and 1168 psig) or QAWCU =2.3%-154,000' h m

- (524.39 - 419.83 bm = 370,352 Btu/hr 7.7.2.8 Recirculabon Pump Neat Addition Uncertainty The power of the recirculation pumps (WE ) is measured by a watt-meter with a calculated uncertainty of 1.4 % (Section 7.6.8).

XP

= t 1.4 °la The uncertainty of the power measurement multiplied by the pump power, Qp, is the uncertainty of the pump power, a0P. Op is 37,145,000 Btu/hr (Section 7.6.1.1).

a o P

= :txp °lam. QP (See Equation 38) t 1.4%

  • 37,145,000 Btu/hr

= :t 520,030 Btu/hr CT op

= t 520,000 Btu/hr, rounded to level of significance 7.8 TOTAL CTP UNCERTAINTY CALCULATION Total CTP uncertainty, UcTP, is calculated by using Equations 19, 20 through 27, 28, 31, 33, and 34 through 37.

(Reference Equation 19)

U CTP -

CTP z

d FNr (w,. f oCTP aQGRO_._ IN Reactor Core Thermal Power Uncertainty Calculation Unit 1 Revision 0

= 3701,400 Btu/hr, rounded to level of significance z

aCTP ahF(TFW )

~ahF(7Fw)

  • (aQCRD_tN ~

CTP (4

)

~

O~rcu

~QRWCU CTP ahQ (Ps )

~

(6h. (P,)

a CTP

)

(

1aQCAD_0UT M a~c~, vuf C aCTP ) ~

r T (o'QA~

+

aQRAD (I~oPJ = ~

(aPo~

Page 61 of 93 LE-0113 Exelon.

Reactor Core Thermal Power Uncertainty Calculation Unit 1 LE-0113 Revision 0 h,(TAECIRC IN) = 524.39 Btu (at 530 "F and 1060 psig)

Ibm

419.83 Btu (at 440 of and 1168 psig)

Ibm 0'0

2.3°/0 *154,000 Ibm. (524.39 - 419.83) Btu =370,352 Btulhr AWCU hr Ibm

= 370,400 Btulhr, rounded to leve' of significance 7.7.2.8 Recirculation Pump Heat Addition Uncertainty The power of the recirculation pumps (WE) is measured by a watt-meter with a calculated uncertainty of 1.4 010 (Section 7.6.8).

Xp

= +/- 1.4 %

The uncertainty of the power measurement multiplied by the pump power, Qp, is the uncertainty of the pump power, crop. Qp is 37,145,000 Btulhr (Section 7.6.1.1).

a a p

=+/- Xp 0/0

  • Qp

=+/- 1.40/0

  • 37,145,000 Btulhr

=+/- 520,030 Btuthr crOp

+/- 520,000 Btulhr, rounded to level of significance (See Equation 38) 7,,8 TOTAL CTP UNCERTAINTY CALCULATION Total CTP uncertainty, UCTP, is calculated by using Equations 19,20 through 27, 28, 31, 33, and 34 through 37.

(Reference Equation 19)

UCTP =

Page 61 of 93

DCTP = (ha (Ps) - hF MA aWFW a CTP r wFw a CTP ahF (TF  )

ra CTP a OCaa OUT a CTP UCTP UCTP =

a QCRO..IN aCTP ~1 a 0 RAO a CTP ~1 (Reference Equation 26) a OFiWCU a CTP ---1 (Reference Equation 27) a 0p Substituting the previously determined values for the variables shown gives :

-15,090,000 lbrn 2 = 0.625 Btu 2

+

1

  • 3,748,000 Btu 2 hr 1)

Ibm hr lOly Reactor Core Thermal Power Uncertainty Calculation Unit 1 Revision 0 (Reference Equation 22) 1 (Reference Equation 23) 2 1781.39 Btu

" 48,288 Ibm 1 h m

+

)

2 15,090,000 1 h m

  • 0802 Btu bm 2

2 2 226,000 Btu

+ j 1 If

  • 303,668 Btu hr hr

)2) +(l 2

370,400 Btu 1

  • 520,000 Btu hr hr 610,570 Bt 2
  • 2.3317 x 109 !bm 2 + 2.2771 x 1014 lbm 2
  • 0.6432 Btu 2 +

lbm hr

-h-r2 Ibm2 )

~ 2.27710 x 1014 lbm 2

  • 0.3906 Btu 2

1+ 1 1

  • 1.4048 x 1013 Btu 2

hr 2

lbm2

hr 2

!r 1*5.1076x1010 Btu 2 + ( 1*9.2210x101 Btu 2 1 +

hr' hr 2 1

  • 1.3700 x 1011 Btu 2 )+

1

  • 2.7040 x

(

101, Btu 2 hr hr2 Page 62 of 93 (Reference Equation 20)

(Reference Equation 21)

(Reference Equation 24)

(Reference Equation 25)

LE-01 13 Exelon.

Reactor Core Thermal Power Uncertainty Calculation Unit 1 LE-0113 Revision 0 aCTP

-- = (hG (ps) - hF (TFW ))

aWFW aCTP

--=WFW dhG(Ps) aCTP

---=-WFW dhF(TFW )

aCTP

=1 aOeRO_OUT aCTP

=-1 aOCAOJN (Reference Equation 20)

(Reference Equation 21)

(Reference Equation 22)

(Reference Equation 23)

(Reference Equation 24)

(Reference Equation 26)

(Reference Equation 25) aCTP =1 dOAAO aCTP

=1 aoRWCU aCTP =-1 (Reference Equation 27) aop Substituting the previously determined values for the variables shown gives:

UCTP =.

( f81.39 ~~IT *(48,288 I~~r) +((115,090,000 I~~ I)" *(0.802 ~~n+

((/-15,090,000 I~~Ir *(0.6251~~n

+( ~ l/f *(3,748.000 ~t~n

+

(~-llr *(226,000 ~~n+(~llf *(303'668~~rr +

( ~ llf *(370,400 ~t~n

+(~ llf *(520,000 ~t~n 610,570 Btu 2 *2.3317x109 fbm 2

]+(2.2771Xl014 Ibm 2.0.6432 Btu 2). +

Ibm2 hr2 hr2 Ibm2

( 2.27710x1014 Ibm 2.0.3906 Btu 2 )+(1*1.4048 X l013 Btu 2

..)+

hr2 fbm2 hr2

( 1*S.1076Xl010 Btu2J+(1*9.2210X1010 Btu 2

J+

hr2 hr2

( 1*1.3700X1011 Btu2 ]+(1*2.7040X1011 Btu 2 J hr2 hr2 Page 62 of 93

UCTP 2

2 1.4237 x 10,5 hr Btu ) + ( 1.4546 hr x 101 Bt2

+

)

8.8940 x 1013 Btu 2 hr 2 ucTP = 1.6740 x 1015 Btu2 hr 2 UCTP -

= 40,914,572 Btu 40,910,000 Btu 3,412,000 hr Btu MWt ucTP -

11.99 MWt.100%

3,458 MWt UCTP = 0.347 Reactor Gore Thermal Power LE-0113 Uncertainty Calculation Unit 1 Revision 0 1.4048 x 1013 Btu hr 2 2

21 5.1080 x 1 Oio Bt

+ 9.2210 x i 01° Btu hr hr 1

.3700x1011 Btu

+ 2.7040 x 1011 Btu 2

hr

) (

hr

)

ucTP =40,910,000 Btu

, rounded to level of significance Divide UcTp by 3,412,000 Btu/hr to convert units to megawatts thermal (MWt) ucTP = 11.99138 MWt = 11.99 MWt, rounded Divide UcTp by 3,458 Btu/hr to convert to percent of rated reactor thermal power.

The determination of total CTP uncertainty is sensitive to two measured parameters, feedwater mass flowrate measurement uncertainty and feedwater temperature measurement uncertainty.

Page 63 of 93 Exel~n.

Reactor Core Thermal Power Uncertainty Calculation Unit 1 LE-0113 Revision 0 UCTP =

1.4237x1015 Bt~ 2)+( 1.4646 x1014 Bt~ 2)+

hr hr

( 8.8940x10'3 ~;~2)+( 1.4048x10'3 ~:~2)+

(5.1080X1010 ~:~ 2)+( 9.2210x10'O ~:~ 2)+

( 1.3700x10 f

' ~t~22 )+( 2.7040x10'1 ~t~22 )

UCTP =

1.6740x1015 Btu2 2

hr

40,914,572 Btu hr UCTP
;;; 40,91 0,000 ~t~, rounded to level of significance Divide UCTP by 3,412,000 Btu/hr to convert units to megawatts thermal (MWt) 40910000 Btu I

hr UCTP -

Btu 3412000-l:!!..-

t MWt UCTP =11.99138 MWt =11.99 MWt, rounded Divide UCTP by 3,458 Btulhr to convert to percent of rated reactor thermal power.

u

= 11.99 MWt.1000/0 CTP 3,458MWt UCTP=0.347 %

The determination of totar CTP uncertainty is sensitive to two measured parameters, feedwater mass flowrate measurement uncertainty and feedwater temperature measurement uncertainty.

Page 63 of 93

8.0 CONCLUSION

S Page 64 of 93 LE-0113 Reactor Core Thermal Power Uncertainty Calculation Unit 1 Revision 0 The total uncertainty associated with reactor core thermal power calculation performed by the Plant Process Computer is 0.347 % of rated reactor thermal power (2a).

Measurement Uncertainty Recapture (MUR) is based on the Nuclear Regulatory Commission (NRC) amended 10CFR50, Appendix K "Emergency Core Cooling System Evaluation Models" on June 1, 2000. The original regulation did not require the power measurement uncertainty be demonstrated, but rather mandated a 2% margin. The new rule allows licensees to justify a smaller margin for power measurement uncertainty based on the installed highly accurate feedwater flow measurement instrumentation.

Therefore, the Core Thermal Power (CTP) uncertainty of 0.347% allows the original 2% margin to be reduced to 1.653 % (2.0 % - 0.347 % = 1.653 %), which is conservatively rounded down to 1.65%.

Exelon.

8.0 CONCLUSrONS Reactor Core Thermal Power Uncertainty Calculation Unit 1 LE-0113 Revision 0 The total uncertainty associated with reactor core thermaJ power calculation performed by the Plant Process Computer is 0.347 °/0 of rated reactor thermal power (20).

Measurement Uncertainty Recapture (MUR) is based on the Nuclear Regulatory Commission (NRC) amended 10CFR50, Appendix K"Emergency Core Cooling System Evaluation Models lt on June 1, 2000. The original regulation did not require the power measurement uncertainty be demonstrated.

but rather mandated a 2% margin. The new rule allows licensees to justify a smaller margin for power measurement uncertainty based on the installed highly accurate feedwater flow measurement instrumentation.

Therefore, the Core Thermal Power (CTP) uncertainty of 0.3470/0 allows the original20k margin to be reduced to 1.653 % (2.0 °/0 - 0.347 % = 1.6530/0), which is conservatively rounded down to 1.650/0.

Page 64 of 93

.Exe 'on LE-0113 Reactor Core Thermal Power Uncertainty Calculation Unit 1 Revision 0 9.0 ATTAgHMENTS Telecon URS and TemTex Accuracy of RTDs for TE-046-103............................................... 66 TODI Al695446-801-Steam Carryover Fraction Design Input (1 page)................................. 67 Rosemount Nuclear Instruments customer letter to Grand Gulf (2 pages).............................. 68 Analogic Analog Input Card ANDS5500 (4 pages).................................................................. 70 KIST Saturated Properties of Water........................................................................................ 74 NIST Isobaric and Isothermal Properties of Water...................................................................75 CTP Calculation Results Sensitivity Analysis..................................................... . ,.,., 76 Derivation of the relationship between flow, AP, and density.................................................. 77 Ametek Scientific Columbus Exceltronic AC Watt Transducer Specification (4 pages).......... 80 0 B32-C-001-J-023, Rev. 1, Recirculation Pump Curve (1 pages)........................................... 84 1 Rosemount Inc., Instruction Manual 4259, Model 1151 Alphaline dP Flow Transmitter, 1977 (PAGES 1, 6, 14, 24, AND 29)....................................................................................................... 85 2 Bailey Signal Resistor Unit Type 766 (2 pages)..................................................................... 88 3 Rosemount Specification Drawing 01153-2734, N0039 Option - Combination N0016 &

N0037 (2 Pages)......................................................................................................................................90 4 Rosemount Product Data Sheet 00813-0100-2655, Rev. AA June 1999 "N-Options for Use with the Model 1153 & 1154 AlphallneS Nuclear Pressure Transmitters" (2 Pages)............................. 92 Page 65 of 93 Exelon.

9.0 ATTACHMENTS Reactor Core Thermal Power Uncertainty Calculation Unit 1 LE-Q113 Revision 0 Taleeon URS and TemTex Accuracy of RTDs for TE-046*103 66 TOOl A1695446-801 - Steam Carryover Fraction Design Input (1 page) 67 Rosemount Nuclear Instruments customer letter to Grand Gulf (2 pages) 68 Analoglc Analog Input Card ANDS5500 (4 pages) 70 NIST Saturated Properties of Water 74 NIST Isobaric and Isothermal Properties of Water 75 CTP Calculation Results Sensitivity Analysis 76 Derivation of the relationship between flow,.1P, and density 77 Ametek Scientific Columbus Exceltronic AC Watt Transducer Specification (4 pages) 80 0 B32-C-001-J-023, Rev. 1t Recirculation Pump Curve (1 pages) 84 1 Rosemount Inc., Instruction Manual 4259, Model 1151 Alphaltne<!>.j;;:p Flow Transmitter, 19n (PAGES 1,6, 14,24, AND 29) 85 2 Bailey Signal Resistor Unit Type 766 (2 pages) 88 3 Rosemount Specification Drawing 01153-2734, N0039 Option - Combination N0016 &

NOO37 (2 Pages) 90 4 Rosemount Product Data Sheet 00813-0100-2655, Rev. AA June 1999 "N.Qptions for Use with the Model 1153 & 1154 Alphaline Nuclear Pressure Transmitters" (2 Pages) 92 Page 65 of 93

Shah, Pravin TemTex Temperature Systen-ts, Inc.

700 E. Houston St.

Shermaan, TX 75090 Phone: (903) 813-ISM John E. Pehush, P. E washirigton Group URS Supewiskr9 Discipike Engkseer - I&C (609) 720-2274 w (609) 216-13922 c John. Pehueh4V"Int. com Reactor Core Thermal Power Uncertainty Calculation Unit 1 Revision 0 Telecon URS and TemTex Accuracy of RTDs far TE-046-103 Frorn:

Pshushjohn sent:

Frkfay. Sopterrsber 25, 2000 9:58 AM To :

Kimura. SW%w Shah, Pravin Cc:

Clconnor, John: Law. Richard su b CRD Orift walw als&aw" z r

Page 66 of 93 Pravin:

CRD Drive Water Discharge Temperature Is measured by TE-OW103 tuns 2 TE-M-2o3), whirr, is supplied Per telecom between John Pehush and TemTex application engheering on 09124109, the RTDs supplied to Limerick Unit 1 are 100 ohm platinum. The model number 10457-8785 specified in PIMS its the TemTex drawing number. The RTD was manufactured to the ft" standard IEC-751, which mans the accuracy is

+ f- 0.1296 (of ice) at 0*C, commonly knows as Class 8. Thwefone the RTD will

`provide an accuracy of

+ f 0.3'C at 0*C (+I- 0.54'F at 32*F). The "Temperature Coefficient of Resistance (TCR), as so called the ALPHA, Is the average increase in resWance per degree increase. The TCR of a platinum RTD Is 0.00385 T'he range of TE-046-103 Is 0 to 200`F (-17.78 to 93.J3*C). The overall accuracy is conservatively calculated at 93-33-C (2007), which is +l- 0.359'C (+/- 0.65'F), Therefore, temperature accuracy of TE-046-103 used is rounded to +/- 0.7'F.

LE-0113 Exelon.

Reactor Core Thermal Power Uncertainty Calcufation Unit 1 LE-0113 Revision 0 Telecon URS and TemTex Accuracy of RTDs for TE-046-103 Shah, Pravfn From:

Sent:

To:

Co:

SubjHt:

P....ush.John Friday. 5eptltmber 25. 200Q 9;58 AM Kimura. Stephen: Shah. Pr.avin Ocomot. JoM; Low, Richard eRO 0riYe Watat~ Temperature Pravin:

eRn Drive Water Discharge Temperature is measured by TE..Q46..103 (Unit 2 TE-046-203)~ which is auppUed by; TemTex Temperature S~ Inc.

700 ~ HousIon St.

Sherman, TX 75090 Phone: (903) 813-1500 Per telecom betWeen John Pehush and TemTex application engineering on 09/24109. the RTDs supplied to Limerick Unit 1 are 100 ohm pIatWIum. The modet number 10457-8185 specined in PIMS is the TemTex drawing 1'KJITlt)er. The RTD was manufactured to the indusCry standard IEe-7S1, which means the accuracy is

+1- 0.12% (of resistance) at O*C, convnonly knows 88 CJasa B, Therefore the RTD will provide., accuracy of

+I. 0.3*C at O*C (+1-O.54*F at 32*F).. The lrfemperature Coefficient of Resistance" (TeR), as so called 1Ile ALPHA, Is the average Increase in reslatance per degree increase, The TeR of 8 piamum RTD is 0.00385 OJOIC*,

The I'8f'I98 ofTE..Q46-103 is 0 Co 20£r'F (-17.78 to 93.33*C). The ovet'8JIacCtl"8cy is conservatively caJculated at 93.33*C (2OO-F), which ia +/- O.359*C (+'- 0.65*f)~ Therefore, tempef8111re acctncyofTE..()46..103 used is rtlUnded to +/- O.7"F.

John E. Pehush. P,E.

washington Group URS SUperviIing Discipline Engineer - f&C (609) 720-2274 w (609) 2160-1392 c John.pehtJ8hOWgfnt.com Page 66 of 93 TODI A1695446-801-Steam Carryover Fraction Design Input (1 page)

Page 67 of 93 LE-0113 Reactor Core Thermal Power Uncertainty Calculation Unit 1 Revision 0 EXELON TRANSMITTAL OF DESIGN INFORMATION dSAFETY-RELATED Originating Organization Tracking No :

R

~~elon

~REGULATO RELATED (Other (specify)

A 1695446-84 StationAJnit(s) U 2rigk_ Units i & 2 Page t of 1 To : John Pehush WGI - Mid Atlantic Subject : Transmittal of information.

I h4o 2eve Drm&h Preparer P parers Signature Date Chris, Vie ie tnd Z

Appto er Signatur :

ate MUR Project Reviewer MUR i_s ject Signature D to For Quality/Completeness Status of Information : SApproved for Use OUnverified Description of Information-The following information was requested by URS Washington Division (WGI) for input to Limerick's core thermal power (CTP) uncertainty calculations (units 1 & 2).

The steam carryover fraction that should be used as design input to the calculations is zero.

Purpose of Issuance and Limitations an Use: This information is being supplied solely for (lie uw as design input for the Link-rielc CTP uncertainty calculations (unit; 1 & 2).

Source Documents :

G.E. Document "Impact of Steam Carryover Fraction on Process Computer Heat Balance Calculations",

September 2001.

Distribution:

Original: Limerick file CC : Ray George, Electrical Design Manager, Limerick John Pehush, I&C Lead, WGl - Mid Atlantic Exelon.

Reactor Core Thermal Power Uncertainty Calculation Unit 1 LE-G113 Revision 0 TOOl A1695446-801 - Steam Carryover Fraction Design Input (1 page)

EXELON TRANSMllTAL OF DESIGN INFORMATION DSAFETY-RELATEO

~NON-SAFETY*AELATED DREGULATOAY RELATED Originating Organization I8lExelon OOther (specify)

Tracking No:

A169544&80 StationlUnit(s) Umerick Units 1& 2

Subject:

Transmittal of information.

MUR Project Reviewer For QuatitytCompfeteness Status of Information: f8JApproved for Use OUnverified Page 1 of 1 To: John Pehush WGI

  • Mid Atlantic Description of Information:

The following information was requested by URS Washington Division (WGI) for input to Limerick's core thermal power (CTP) uncertainty carculatkms (units 1 & 2).

The steam carryover fraction that should be used as design Input to the calculations is.<<£2.

Purpose of Issuance and Limitations on Use: This information is being supplied solely for lhe use as design input for the Limerick CfP uncertainty calculations (units I & 2).

Source Documents:

G.E. Document "Impact of Steam Carryover Fraction on Process Computer Heat Balance CalcuJations",

September 2001.

Distribution:

Original: Umerick fire CC: Ray George, Electrical Design Manager, Umerick John Pehush, I&C Lead. WGI - Mid Atlantic Page 67 of 93 Rosemount Nuclear Instruments customer letter to Grand Gulf (2 pages)

Rosemount Nuclear Instruments 4 April 2000}

Ref Grand Gulf Nuclear Station message on INPO plant reports, subject Rosemount Instrument Setpoint Methodology, dated March 9, 2000

Dear Customer:

`this letter is intended to eliminate any confusion that may have arisen as a result of the reference message from Grand Gulf. The massage was concerned with statistical variation associated with published performance variables and how the variation relates to the published specifications in Rosemount Nuclear Instruments, Ine.(RNII) pressure transmitter models 1152, 1153 Series E3, 1153 Series 0, 1154 and 1154 Series H. According ten our understntnding, the performance variables of primary concern are those discussed in GE Instrument Setpoint Methodology document NEDC 31336, namely I.

Reference Accuracy 2.

Ambient Temperature Effect 3.

Overpressure Effect 4 :

Static Pressure Effects 5.

Power Supply Effect It is RNII's understanding that GE and (lie NRC have accepted the methodology of using transmitter testing to insure specifications are met as a basis for confirming specificatiorns are

_+3o< "13se conclusions we draw regarding specifications being +3o arc based on manufacturing testing and screening, final assembly acceptance testing, periodic (e.g., every 3 months) audit testing oftransmitter samples and limited statistical atnalysis. Please note that all performance specifications are based ots zero-based ranges urxie r refcrcnce conditions. Finally, we wish to make clear that no inferences are made wills respect to confidence levels associated with any specification.

1.

Referenee Accuracy_

2.

Ambient Temperature Effect Reactor Core Thermal Power Uncertainty Calculation Unit 1 Revision 0 Ro

ftlBleu im t ?0¢ r i admo"

(}, ne "lien Priw`a MN S-44 WA I "I Efot2)82ihWs?

F,x t ltt$2) 828 6280 All (100*,) RNII transafttets, including models 11 52, 1153 Series B, 1153 Series D, 1154 and 1154 Series H, are tested to verify accuracy to +0.25% of span at W*. 20'%, 40%, 60'10, 80% and 100% of span, Therefore, tlse reference accuracy published in our specifications is considered +3Q.

All(1t)0%)

ausplI ier boards are tested for compliance with their temperature effect specifications prior to final assembly. All sensor modules, with the exception of model 1154, are temperature compensated to assure compliance with their temperature effect specifications. All (100010) model 1154, model 1154 Series H and model 1153 gages and absolute pressure transmitters are tested following final assembly to verify compliance with specification. Additionally, a review of audit test data performed on final assemblies of model 1152 and model 1153 transmitters not tested following final assembly indicate Page 68 of 93 LE-0113 Exelon.

Reactor Core Thermal Power Uncertainty Calculation Unit 1 LE..0113 Revision 0 Rosemount Nuclear Instruments customer letter to Grand Gulf (2 pages)


~-----,-_..,

.,.~-------*.... ---

Rosemount Nuclear Instruments 4 April 2000 RCltealOUnt Jfy¢~.. lnstrumtnls. Inc-1700I 1Cd"'O\\o9V I}.lYe E,~n Pr"",. 1,4/0: l>S,1U USA I(/1 I (Gl t} S211*8;?S~

FiliI 1(IlZ1 8:21) 1f200 Ref: Grand Gulf Nuclear Station message on INPO plant reports, subje<:t Rosemount Instrument Setpoint Methodology, dated Marcb 9, 2000

Dear Customer:

This fetter is intended to eliminate any confusion that may have arisen as a result ofthe reference message (rom Orand Gulf. The message was concerned with statistical variation associated with published performance variables and how the variation relates to the published specifications in Rosemount Nuclear Instruments, Inc.(RNU) pressure transmitter models 1152. 11 S3 Series D, 11 SJ Series D, t 154 and II S4 Serie~ H. According to our understanding, the performance variables ofprimary concern are those discussed in OJ:: Instrument Setpoint Methodology document NEDC 31336, namely 1.

Reference Accuracy 2.

Ambient Temperature Effect 3

Overpressure Effect 4.

Static Pressure Effects 5.

Power Supply Effect It is RNU's understandi.lB that GE and the NRC have accepted the methodology of using transmitter testing to insure specifications arc met as a basis for confirming speeificatiofls are

.:3<1. 'niC conclusions we draw regarding spccification.'i being,!3a are based on manufacturing testing and screening, final assembly acceptance testing, periodic (e.g., every 3 months) audit testing oftransmitter samples and limited statistical analysis. Please note that aU perfonnance specifications are based on zero-based ranges under reference conditions. Finally, we wish to m.tke clear that no inferences are made wilh r\\-'Spect to confidence levels associated with any specification.

I.

RercrcMe Aecuracy,

All (tOO-A) RNII rransmilt~s, including models "i52, liS) Series at 1153 Series D, 1154 and 1154 Series H, are tested to verify acxuracy to +0.25% ofspan at 00/0. 20%, 40%, 6~/o.

80% and 100% ofspan. Therefore, the reference accuracy published in our specifications is considered ~J(1.

2.

Ambknt T~mpcraturcEfred All (100%) ampllfeer boards are tesled for compliance with their temperature effect specifications prior to final assembly. All sensor modules, with [he exception ofmodel J154, are temperature compensated to assure compliance with their temperature effect specifications. All (J00%) model t 154. model 1154 Series H and model 1153 gage and absolute pressure transmitters are tested following final assembly to verify compliance with specification. AdditionaUy, a review ofaudit test data performed on final assemblies or model 1152 and model I J53 transmitters not tested following final assembly indicate Page 68 of 93

meloi cOnfortttartee to specification. Therefore, the ambient temperature effect published in our specifications is considered *3a, 3.

O,v,~frprcssure Effect Testing of this variable is done at the module stage. All (100%) range 3 through 8 sensor modules are tested for compliance to specifications. We do not test range 9 or 10 modules for overpressure for safety reasons. However, design similarity permits us to conclude that statements made for ranges 3 through $ would also apply to ranges 9 and 10. Therefore, the overpressure effect published in our specifications is consi lered +3a.

4.

Static Pressure Effects Reactor Core Thermal Power Uncertainty Calculation Unit 1 Revision 0 All (100%) differential pressure sensor modules are tested for compliance with static pressure zero errors. Additionally, Models 1153 and 1134 Ranges 7, 6,7 and 8 arc 100% tested after final assembly for added assurance of specification compliance. Audit testing performed on ranges 4 and 5 have shown compliance to the specification. Therefore, static pressure effects publisehed in our specifications are considered }3Q,.

5.

Power 8 Oct Testitr8, har conformance to this specification is performed on all trnnsinittcrs undergoing arnpic (audit) testing. `this variable has historically exhibited extremely small performance errors ntul small standard deviation (es +:ntially a mean error of zero with a standard deviation typically less than I Or% of ilea specifica(ion), Al i transmitters tested were found in compliawc with the specification

. 'Iltereforu, power supply effect published in our speciticati icatiorm ii considered -+3a.

511()411(1 ycata lutvc any further gtwstiotty, please contact Jerry Ldwards at (612) 828-395 1,

%inusruly, Jerry L. rdwards Manager, Sales, Marketing and Contracts Rosemount Nuclear Instntments, lnc.

FOB-A IT Page 69 of 93 Exelc)n.

Reactor Core Thermal Power Uncertainty Calculation Unit 1 LE-0113 Revision 0 conformanoo to specification. Therefore. the ambient temperature effect published in our s~dflCatiot1s is considered.:3<1.

Te${iflg of this variable is done at the module stage. All (100%) range) through 8 sensor modules are tC5ted for compliance 10 specifications. We do not test range 9 or 10 modules (Of' overpressure for safety reasOllS, However, design similarity permits us to conclude that statements made for ranges J through 8 would also opply to ranges 9 and IG. Therefore, the overpressure effect published in our specifications is considered!,3<J.

4.

Stade Pressure Effects All (100%) differential pressure sensor modules are tested for compliance with static pressure 2ero crrol"$. Additioonlty. Models 1153 and 1154 Rongcs 3,6,7 and 8 arc 100% tested aftel' linnlflssembly for added assurance ofspecificalion compliance. Audit testing performed on ra..gcs 4 and Shave shown compliance to the speciftcation. Therefore, static pressure effects published in our specifications are considered ~3<J, TcstlllG, fOr conformance to this specification is pcrlbrmed on aillmnsmiltcrs undergoing

~amplc (audit) lesting, "n.is vadnblc Ims historically exhibited extremely smaU performance cm~ nnd small siandard t1cviatioll (cssc"tially" mcan error afzero with a siandard deviation lypicnlly lo,;s than Io-A! oftile specil1cntion). All transmitters tested were f()tlnd in eompltno<< with the specification, 'l111,,'fc!Orc. power supply COcci published in our spccificatiQll$ i~ considered.+30.

SIWtlid you have any further questions. plCOlSC contact Jt..'fTY IldwanJs at (612) 828-395 L Sinccrely.

Jerry L. Edwards Man.1gcr. Sales. Marketing and Contracts Rosemount Nuclear Instntments. rne.

FISIIR'IGSEMIIlJIJ Page 69 of 93

Reactor Core Thermal Power Uncertainty Calculation Unit 1 Revision 0 Analogic Analog Input Card ANDS5500 (4 pages)

PEABODY, MA 01860 POTENTIOMETER INPUT CARD ANDS5500 SPECIFICATION 2-15227 REV 01 First Used On : ANDS5500 Page 1 of 8 Code Ident:

1 BMOO File Name : 2-15227revOl_doa

_Printed :_. April 4, 2003 REVISION HISTORY Approvals for Release:

Richard Lane 2113103 Originator Date Richard Lane 3131103 Engineer Date Page 70 of 93 LE-0113 REV DESCRIPTION DWN APVD DATE 01 SEE E.C.0 K.Q RWA 08125183 Exelon.

Reactor Core Thermal Power Uncertainty Calculation Unit 1 Analogic Analog Input Card ANDS5500 (4 pages)

ANALaC51~.

PEABODY, MA 01960 LE-0113 Revision 0 POTENTIOMETER INPUT CARD ANDS5500 SPECIFICATION 2-15227 REV 01 First Used On; ANOS5500 Page 1 of a Code Ident: 1BMOO File Name: 2-15227rev01.doo Printed:

April 4, 2003 REVISION HISTORY REV OESCRIPnON DWN APVD DATE 01 SEEE.C.O K.Q RWA 08125183 Approvals for Release:

Richard Lane Originator Richard Lane Engineer Page 70 of 93 2113/03 Date 3/31/03 Date

ANDS5500 POTENTIOMETER INPUT CARD RELATED DOCUMENTS :

Schematic D5-8814 Theory of Operation A2-5568 1.

GENERAL DESCRIPTION Reactor Core Thermal Power LE-0113 Uncertainty Calculation Unit 1 Revision 0 SPECIEICATI0N 2-15227 The Potentiometer Card is a user card for the ANDS 5500 Data Acquisition System. It consists of a power supply and voltage reference, an output multiplexer and four identical signal' conditioning channels. There are two 44-pin edge-card connectors (male) on the PC board, one is connected internally to the system and the other is for user connection.

Each signal channel consists of an excitation output section and a signal input section. The latter can be preset to accept a variety of voltage ranges and types via jumpers. It provides a DC voltage to, and permits low frequency and DC measurement of the signals from, appropriate external devices such as piezoelectric accelerometers.

11.

SPECIFICATIONS 1. GENERAL Number of Channels :

4 Size and Shape :

Approximately 7 W x 4 '/z" x '/2" (similar to Analogic D4-7443)

Operating Temperature :

0 - 50 Degrees C.

Storage Temperature : 85 Degrees C.

Input & Output Connection :

EDGECARD, gold plated 2. ELECTRICAL a) Power Requirement:

+5v (+/-5%) at 100mA

+15v (+1-5%) at 1OOmA

-15v (+l-5%) at 1OOmA Page 7 1 of 93 Exelon.

Reactor Core Thermal Power Uncertainty Calculation Unit 1 LE-0113 Revision 0 ANDS5500 POTENTIOMETER INPUT CARD SPECIFICATION 2-15227 RELATED DOCUMENTS:

Schematic Theory of Operation 05-8814 A2-5568 I.

GENERAL DESCRIPTION The Potentiometer Card is a user card for the ANDS 5500 Data Acquisition System~ It consists of a power suppfy and voltage reference, an output multiplexer and four identical signal conditioning channels. There are two 44-pin edge-card connectors (male) on the PC board, one is connected internally to the system and the other is for user connection.

Each signal channel consists of an excitation output section and a signal input section. The latter can be preset to accept a variety of valtage ranges and types via jumpers. It provides a DC voltage to, and permits low frequency and DC measurement of the signals from, appropriate external devices such as piezoelectric accelerometers.

II.

SPECIFICATIONS

1. GENERAL Number of Channels:

Size and Shape:

Operating Temperature:

Storage Temperature:

Input & Output Connection:

2. ELECTRICAL a) Power Requirement:

Page 71 of 93 4

ApprOXimately 7 ~" X 4 %" X ~" (similar to Analogic 04-7443) o-50 Degrees C.

..40 - 85 Degrees C.

EDGECARD, gold plated

+5v (+/-5%) at 100mA

+15v (+1-5%) at 100mA

..15v (+/-5%) at 100mA

Reactor Core Thermal Power Uncertainty Calculation Unit 1 Revision 0 ANALOGIC - PEABODY, MA 01960 2-15227 REV 01 POTENTIOMETER INPUT CARD ANOSS500 Page 4 of 9 Code Ident, 18M00 File Name. 2-15227revOl.doc Printed :

April 4, 2003 b) Excitation Voltage Output (1 per channel)

Voltage Lever

+5v and -5v Voltage Accuracy:

+!- 0.2% over operating temperature range Output Current:

10mA max.

Output Impedance :

< 0.1 ohm Protection:

Tolerate direct short circuit or short to

-30vdc protected up to 250v by a thermistor c) Signal Input (1 per channel) rwontrolled Document. Source Unknown Input Impedence 10Meg ohm Input Signal Flanges :

+f-5v, +/-2.5v, +M.25v, +1-500mv,

+/-250mv, +/-I 25mv Jumper selected (see Table 1) and fine adjustment by trimpot Primary Frequency Content:

DC to 100Hz Signal Source Resistance :

2800 ohm max.

Overall Accurancy : (includes

+l- 0.5% over operating excitation accurancy) temperature range Protection :

Input protected up to 250v continuous AmplIfler Upper Bandwidth :

10Hz, 45Hz and 1 OOHz (2-pole filter)

Jumper selected (see Table 2)

Input Coupling Jumper selected for either, AC OR DC.

Lower AC bandwidth is 0.5Hz (see Table 2).

Page 72 of 93 INPUT RANGE (F.S)

JUMPERg JUMPER

+/-5v 2

6to7 9 to 10

+I-Z 5v 4

5 to 7 9 to 10

+/-1.5v 8

4 to 8 9 to 10 Exelon.

Reactor Core Thermal Power Uncertainty Calculation Unit 1 Uncontrolled Document. Source Unknown LE..0113 Revision 0 ANALOGIC

  • PEABODY, MA 01960 POTENTIOMETER INPUT CARD ANOS5500 Code Ident 1BMOO File Name: 2~15221rev01.doc b) Excitation Voltage Output (1 per channel) 2-15227 REV 01 P8084 019 Printed:

April 4 2003 Voltage Level:

Voltage Accuracy:

Output Current:

Output Impedance:

Protection:

c) Signal Input (1 per channel)

Input Impedence:

'nput Signal Ranges:

Primary Frequency Content:

Signal Source Resistance:

Overalf Accurancy: (includes excitation accurancy)

Protection:

Amplifier Upper Bandwidth:

Input Coupling

+5v and -5v

+/~ 0.2% over operating temperature range 10mAmax.

< 0.1 ohm Tolerate direct short circuit or short to

+1-30vdc protected up to 250v by a thermistor 10Meg ohm

+1-5v, +1-2,5v, +1-1,25v, +1-500mv,

+1-250mv. +/-125mv Jumper selected (see Table 1) and fine adjustment by trimpot DC to 100Hz 2800 ohm max.

+1- 0.5% over operating temperature range Input protected up to 250v continuous 10Hz, 45Hz and 100Hz (2-pote filter)

Jumper selected (see Table 2)

Jumper selected for either, AC OR DC.

Lower AC bandwidth is O.5Hz (see Table 2).

'NPVI RANGE fE.S)

+1-5v

+1-2,5v 2

4 8

Page 72 of 93 JUMPER 2 6 to 7 5to 7 4 to 6 JUMPER 3 9 to 10 9 to 10 9 to 10

Exelon.

LE-01 1 3 Reactor Core Thermal Power Uncertainty Calculation Unit 1 Revision 0 Unoontrofled Deauwtent. source unknown INPUT RANGE (CHANNEL GAIN) VS. JUMPER POSITION TABLE 2.

CHANNEL BANDWIDTH VS. JUMPER POSITION d) Analog Signal Output to Bus Analog Output Signal:

Ch. 1, Ch.2, Ch.3, Ch.4 or Hiz (high impedance state),

digitally selected Output Offset:

+!- 5mv over operating temperature Range, adjustable to zero by trimpot Full Scale Output Range :

+l-10v (+10.5a1o) with full scale input Maximum Output Voltage in Hiz state :

+/- 15v e) Digital Signal Input From Bus :

Channel select (see Table 3)

Page 73 of 93 BANDWIDTH JUMPER,1 JUMPE

. Y=5 DC to 10Hz 2 to 3 12 to 13 16 to 17 DC to 45Hz 2 to 3 13 to 14 11 to 12 DC to100Hz 2to3 17 to 16 15 to 16 0.5 to 10Hz 1 tot 12 to 13 16 to 17 0.5 to 45Hz 1 to 2 13 to 14 11 to 12 0.5 to 100Hz 1 to 2 17 to 18 15 to 16 ANALOGIC - PEASODY, MA 01964 POTENTIOMETER INPUT CARD ANDS550 Case Ident: 1 BMD0 File Name : 2 15227revOl.doc___

2-15227 REV 01 Page 5 of 9 Printed:

April 4. 2003

+f-500mv 20 6 to 7 8 to 9

+- 250mv 40 5 to 7 8 to 9

+ - 125mv 80 4 to 6 8 to 9 TABLE 1.

Exelon.

Reactor Core Thermal Power Uncertainty Calculation Unit 1 LE-0113 Revision 0 Uncontrotfed Ooctr'Aenl SocJrce Unknown ANALOGIC.. PEABODY, MA 01960 POTENTIOMETER INPUT CARD ANDS5500 Code Ident: 1BMOO rile Name: 2-15227rev01.doe 2..15221 REV 01 Page 5 019 Printed:

Aprl14.2oo3

+1.. 500mv 20 6to 7 8to 9

+1-250mv 40 5 to 7 8to 9 8to 9 4t06 80 TABLE 1.

INPUT RANGE (CHANNEL GAIN) VS. JUMPER POSITION

+1-125mv BANDWIDTH JUMPER 1 JUMPEB4 JUMPERS DC to 10Hz 2 to 3 12 to 13 16 to 17 DC to 45Hz 2 to 3 13 to 14 11 to 12 DC to 100Hz 2 to 3 17 to 18 15 to 16 0.5 to 10Hz 1 to 2 12 to 13 16 to 17 0.5 to 45Hz 1 to 2 13 to 14 11 to 12 0.5 to 100Hz 1 to 2 17to18 15 to 16 TABLE 2.

CHANNEL BANDWIDTH VS. JUMPER POSITION d) Analog Signal Output to Bus Analog Output Signal:

Ch.l, Ch.2, Ch.3, Ch.4 or Hiz (high impedance state),

digitally selected Output Offset:

+/- 5mv over o~rating temperature Range. adjustable to zero by trimpot Full Scale Output Range:

Maximum Output Voltage in Hiz state:

+1-10v (+f-O.5%) with full scale input

+/- 15v e) Digital Signal rnput From Bus:

Channel select (see Table 3)

Page 73 of 93

Exelon.

Reactor Core Thermal Power Uncertainty Calculation Unit I Revision 0 NIST Saturated Properties of Water Thermodynamic Properties of Water

~Lemmon, E.W,, McLinder, M

.O and Friend, D.G,"Thermophysioal Properties of Fluid Systems",

NIST Chemistry WebBook, NIST Standard Reference Database Number 69, Eds. Rj, Linstrom and Wa Mallard, National Institute of Standards and Technology; Gaithersburg MD, 20899, hftp, IMebbook

. nist, gov, (retrieved September 30, 2009) tpsig = psia, 14.696, rounded Page 74 of 93 LE-01 13 Saturated PM M ies - Pressure Pressure UK Temperature I

(F)

Enthalpy

_ILMAK)

Density (y, lb Enthalpy (1, BtuAbmI, Density Ibm/ft) 1023.0 549.16 1191.9 2.337 548.77 46.008 1032.7 55031 1191.6 2.361 550.23 45234 1033.0 550.3 11191.5 2.362 560.27 45.931 1033.5 550.4 1191 s 2.364 56135 445.927 11734.4 550-6 1191.5 2.366 550.48 46.921 10303 550.6 1191.5 2.368 550.61 45.914 1036.0 550.7 1191.4 2.370 550.72 45-909 1036.1 550.7 1191.4.

2.370 550.74 45.908 1037.0 550.8 1191.4 ~~Mv#T; 45.901 1037.8 550.9 1191.4 IMOMMEN9 MR M W

- - Q3 45,895 1038.0 550.9 1191.3 ~Qk~

Mfg 45.893 1038.7 551,0 1191.3 MIMARMI 45,8 1039.5 551.1 1191.3 2.379 551-25 45.882 1040.4 MM 1191.2 2.381 551.38 45.875 1041.3 ~~~

I 1191.2 2.383 551.51 45.869

1042.1 F

551A 1191.2 Z385 551.64 45.862 104&0 051.5 1191.1

1387

.8S56 10413.8 561.6 1191.1 2390 1.89 46519 1044.7 WA 91.1 2.392 552.02 45-842 1045.6 551.8 1191.0 Z3964 45.836 1046.4 551.9 1191.0 2.396 552.28 45.829 1047.3 552.0 1191.0 2.398 552.41 45.823 1048.0 552.1 1190.9 2.400 552.52 45.817 1048.1 552.1 1190.9 2.401 552.54 45.81 1049.

MMMMM 1190.9 2.403 552.67 45.810 1049.9 MMMGXJ~--

1190.9 2.405 552.80 41843 1050.0 MMMMM~ 1190.9 2.405 552.81 W02 O

1052.5 55160 11918 2A12 553.18 45.783 1053.0 552-66 11907 2.413 553.26 45.779 10533 552.70 1190.7 2.414 F 553.31 41777 1

.0 55181 119Q3 2.439 9

55174 41704 Exelon.

Reactor Core Thermal Power Uncertainty Calculation Unit 1 NIST Saturated Properties of Water Thermodynamic Properties of WaterE LE-0113 Revision 0 Saturated Properties* Pressure Pressure Temperature Enthalpy Density Enthalpy Density (PSia)t CFl tv. Btu/Ibm)

{v lbmlft1

{I BtuJlbml o.lbm/fti 1023.0 549.16 1191.9 2.337 548.n 46.008 1032.7 550.3 1191.6 2.361 550.23 45.934 1033.0 550.3 1191.5 2.362 550.27 45.931 1033.5 550.4 1191.5 2.364 550.35 45.927 1034.4 550.5 1191.5 2.366 550.48 45.921 1035.3 550.6 1191.5 2.368 550.61 45.914 1036.0 550.7 1191.4 2.310 550.72 45.909 1036.1 550.7 1191.4 2.370 550.74 45.908 1037.0 550.8 1191.4 2.372 550.87 45.901 1037.8 550.9 1191.4 2.374 551.00 45.895 1038.0 550.9 1191.3 2.375 551.02 45.893 1038.7 551.0 1191.3 2.3n 551.12 45.888 1039.5 551.1 1191.3 2.379 551.25 45.882 1040.4 551.2 1191.2 2.381 551.38 45.875 1041.3 551.3 1191.2 2.383 551.51 45.869 1042.1 551.4 1191.2 2.385 551.64 45.862 1043.0 551.5 1191.1 2.387 551.77 45.856 1043.8 551.6 1191.1 2.390 551.89 45.849 1044.7 551.7 1191.1 2.392 552.02 45.842 1045.6 551.8 1191.0 2.394 552.15 45.836 1046.4 551.9 1191.0 2.396 552.28 45.829 1047.3 552.0 1191.0 2.398 552.41 45.823 1048.0 552.1 1190.9 2.400 552.52 45.817 1048.1 552.1 1190.9 2.401 552.54 45.816 1049.0 552.2 1190.9 2.403 552.67 45.810 1049.9 552.3 1190.9 2.405 552.80 45.803 1050.0 552.3 1190.9 2.405 552.81 45.802 1052.5 552.60 1190.8 2.412 553.18 45.783 1053.0 552.66 1190.7 2.413 553.26 45.779 1053.3 552.70 1190.7 2.414 553.31 45.m 1063.0 553.81 1190.3 2.439 554.74 45.704

'lemmon, E.W., Mclinden, M.O., and Friend, D.G., 'Thelmophysical Properties of Fluid Systems",

NIST Chemistry WebBook, NIST Standard Reference Database Number 69, Eds. P.J. Linstrom and W.G. Mallard, National Institute of Standards and Technology, Gaithersburg MO. 20899, hftpJlwebbook.nistgov, (retrieved September 30,2009) tpsig:= psia 14.696, rounded.

Page 74 of 93

'Prig = psfe " 14,696, rounded Reactor Core Thermal Power Uncertainty Calculation Unit 1 Revision 0 NIST Isobaric and Isothermal Properties of Water Thermodynamic Properties of Water'i kwimm, L.W., mctindtn, wo., srwd Friend, txG., "Themrophysicw Properties ot Fluid systems".

NIST Chemistry WebBook, MIST Stsrdard Refererae Datebase hkimbef 69, Eds. P.T Unstrom and W.G. Mallerd, Nationsi WORM of SWWarct end Tedvwlagy, GaMerstxrg MD, 20895.

h0plAvebbook.rrast.gov, (retrieved SWrxrrber 30, 2009)

Page 75 of 93 L E-!0113 isobaric P Temperature F

Pressurer iM Enthalpy 6tuAbm Density ibmltt3 Volume twbm Phase Feedwater In 428.8 1155.0 407.52 52.729 0.018865 liquid 429,8 1155.0 408.61 52.685 0.018981 li Id 430.2 1 155.0 409.05 52.667 0.018987 lq.

430.23 1169.7 409.08 52.688 0.0189 li y1d 430.8 1155.0 409.71

$2.640 0.018997 liquid 431..37 1169,7 410.33 52.615 0.019006 liquid 431 :4 1155,0 410.36 52.613 0.019007 liquid 431 ;8 1156,0 410.80 52.595 0.019013 li id 432.8 1155.0 411.513 52.551 0,01 W29 liquid RWCU Suction 525.6 1060.0 -._

518.94 47.612 0.021003 liquid x30.0 1060.0 524.39 47.333 0.021127 liquid 534.4 1060.0 529.88 47.046 0.1121256 uid RWCU Dischar e 4M.0 1168.0 417.62 52.32 0.019113 li.

440.0 1168.0 419.83 52.229 0.019147 442.0 1168.0 422.04 52.137 0.01918

!i u~f Recirculation 532.0 1280.0 526.57 47.352 0.0211119 liquid 533.0 1280.0 527.81 47.288 0.021147 liquid 534.0 1280,0 529.06 47.223 0.021176 fi id 5,35.0 1280.0 530.30 47.159 0.021205 liquid 536.0 1280.0 531,55 47.094 0.021234 liquid CRD 99.3 1448.0 71.24 62.274 0.016055 H Uid 100.0 144&0 71.93 62,265 0.01606(}

liquid 100.7 1448.0 72.63 62.256 0.010163 liquid Isothermal Pro

'es Feedwater In 430.8 1145.0 409.70 52.636 0.018958 liquid 430.8 1155.0 409.71 52640 0.018897 liquid 430.8 1165.0 409.71 52644 0.018996 1'

id RWCU Suction 530.0 1050.0 524.40 47.326 0.021130 liquid 530.0 1060.0 524.39 47.333 0.021127 liquid 530.0 1070.0 524.37 47.339 0.021124

--liquid RWCU C.

char e 440.0 115&0 419.82 52.225 0.019148 li id 440.0 1188.0 419.83 52.229 0.019147 liquid 440.0 1178.0 419.83 52,233 0.019145 li uid Recirculation 534.0 1270.0 529.07 47.217 0.021179 liquid 534.0 1280.0

$29.06 47,223 0.021176 1i: id 534.0 1290.0 529.04 47.230 0.021173 liquid CRn 100.0 1438.0 71.91 62.263 0.01W61 liquid 100.0 1448.0 71.93 62,265 0.016060 ii uid 100.0-1 -.. _-1458.0 71.96 62.267 0.01 6060 liquid Exelon.

Reactor Core Thermal Power Uncertainty Calculation Unit 1 NJST Isobaric and Isothermal Properties of Water Thermodynamic Properties of Water' LE-0113 Revision 0 Isobaric Prooerties Ternperature Pressuret EnthaJpy Density Volume (F) losiai lBtuJIbm\\

(lbmlft31 lft3Jlbm)

Phase Feedwater In 42tH!

1155.0 407.52 52.7'2.9 0.018965 liquid 429.8 11550 408.61 52.685 0.018981 liauld 430.2 1155.0 409.05 52.661 0.018987 liauid 430.23 1169.7 409.08 52.666 0.018986 liQuid 430.8 1155.0 409.71 52JS40 0.018991 liauid 431.37 1169]

410.33 52.615 0.019006 liQuid 431.4 1155.0 410.36 52.613 0.019001 liQuid 4318 1155,0 410.80 52.595 0.019013 liauld 432.8 1155.0 411.90 52.551 0.010029 liauid RWCUSuction 525.6 1060.0 I

518.94 47.612 0.021003 liQuid 530.0 I

1060.0 I

524.39 I

47.333 0.021127 liQuid 534,,4 1060.0 I

529.88 I

47.046 0.021256 liQuid RWCU Oischarae 433.0 I

1168.0 417.62 I

5232 I 0.019113 I liouid 440.0 1168.0 419.33 52.229 I 0.019147 I

liaufd 442.0 1168.0 422.04 52.137 I

0.01918 I

liauid Recirculation 532.0 1280.0 526.57 47.352 0.021119 liQuid 533.0 1280.0 527.81 47.288 0.021147 liQuid 534.0 1200.0 529.06 47.223 0.021176 liQUid 535,0 1280.0 530.30 47.159 0.021205 liquid 536.0 1280.0 531.55 47.094 0.021234 l!<luld eRO 99.3 1448.0 I

11.24 I

62-274 I 0,016058 liouid 100,0 I

1448.0 I

71.93, 62.265 0.016060 liquid 10(),1 I

1448.0 I

72.63 62.256 I 0.016063 I liQUid Isothermal ProOOfties Feedwater In 430,6 1145,0 409.70 52.636 I 0018998 liauid 430.8 1155,0 409.71 52.640 I 0.018997 I

liauid 430,8 1165.0 409.71 5~644 I 0018996 liquid RWCUSUction 5300 1050.0 524.40 i

47.326 0,021130 liQUid 530.0 1060.0 524.39 I

47.333 0.021127 liauid 530.0 I

1070.0 524.37, 47.339 I 0.021124 liQuid RWCU Discharge 440.0 1158.0 I

41$.82 52.225 0.019148 I

llauid 440.0 1168.0 I

419.83 52.229 0.019147 I liQuid 440.0 1178,0 419.83 52.233 I 0.019145 liquid Recifculation 534,0, 1270.0 529.07 47217 I 0.021119 liquid 534,0 1280.0 529,06 47,223 I 0.021176 liouid 5340 1290.0 529.04 I

47,230 0.021173 liquid CRD 100.0 I

1438.0 I

71.91 r

62.263 0.016061 liQuid 1000 1448.0 I

71.93 I

62.265 0.016060 I liQuid 100.0 I

1458.0 I

71.96 I

62.267 0016060 I

liquid

'ltn'1f1lOn, E:'w., McUnden. M.O.* and Frlend. D.G,. "l'hermophysiclll Properties of FttJd Systems".

NIST ChemIstry Web8ook. N'ST sterodard Ref~e Datebew Number 69. Eds. P,J LInstrom and W.G. M9IIrd, N8t1onaIlnstitute of Sland4tdS end TedlnOIogy. Gaittlersbt6g MO. 20&99, hItp:/Mbbooll,nist.gov. (retfieved September 30, 2009~

'P$ig.. ps;t

  • 14696, rOllllde<l, Page 75 of 93

Exelun Reactor Core Thermal Power LE-0113 Uncertainty Calculation Unit 1 Revision 0 CTP Calculation Results Sensitivity Analysis The sensitivity of the calculation of core thermal power to variations in the energy terms is determined using estimated values for the energy out, energy in, and CTP.

Table A7-1 shows the results from Cases 1, 2 and 3. For this analysis, the error rate assumed for Qs is set equal to a predicted error for the measurement of feedwater flow.

In Case 1 and Case 2, the Qs error is kept the same ; even though, the errors for the QRAD, QRWCU, Qf=, and Qcau terms are varied from 1 °!a to 19 % (29 % for Qcrd). Case 1 and Case 2 show that CTP is relatively insensitive to the accuracy of the QRAO, QRwcu, Op, and CcRD terms, when the CTP error is rounded to level of significance.

Case 3 varied the error rate assumed for Qs. A step change in the mass flow error rate of 0.01 % (from 0.31

°lo to 0.32 %) was found necessary to change the CTP error by 0.02 °lo (a change of 1 MW from 17 MWt to 18 MWt and 0.500 % to 0.520 %).

The step change error rate was calculated to determine how fine the parameters used to calculate Qs and Qrav need to be to effect the results. Parameter changes that would result in changes in the parameter's overall error rate less than 0.02 % were found to be negligible. Small variations in the flow measurement uncertainty were found to affect the CTP uncertainty.

For example, CTP is the difference between the energy leaving the reactor and the energy put into the reactor from other sources outside of the core. The enthalpy of saturated water varies with changes in pressure. For every 1 % change in pressure, the enthalpy of saturated water vapor (and by inference the energy of the flow) between 800 and 1,300 psig will vary by an average of 0.03 %. This change in enthalpy is less than the 0.04 % found necessary to cause a significant change in the CTP error rate. Thus, CTP can be said to be tolerant of the steam dome pressure measurement error specified the pressure measurement loop is shown to be accurate to about --10 psig, which is approximately 1 % of the maximum allowable steam dome pressure.

Table A7-1. CTP Calculation Sensitivity Analysis 8ensilvity Analysis Case 3 t 18.00 Mwt

'Error rounded to 2 significant figures Page 76 of 93 Case 1 Case 2 Case 3 Energy in Assumed percent of error rate CTP Case t Predicted error as Percent of CTP Assumed error rate Case 2 Predicted error as Percent of CTP Assumed error rate Case 3 Predicted error as Percent of CTP Energy Out 09 18,030,078,635 Btuhfir 152.70%

031%

0.4730r6 031%

0.4730/6 0.32%

0.489%

QRAO 3,754,300 Btufir 0.03%

1% 0.0003%

10%

0.003%

19'0 0.000%

GOU 16,407,160 8tufir 0.14%

1 % 0.0014%

10%

0.014%

1 °70 0.001%

Q out 18,050,240,095 Btuhr 152.879'0 Energy In Caw 6,201,235,621 BtuAtir 52.520/9 0.31%

0.163%

031%

0.16^

0.32%

03.168%

OP 37,146,642 BtuAtr 0.31%

1 % 0.0031%

10% 0.031%0 1 % 0.003%

Qcrd 4,578,000 Bt Whr 0.04%

1% 0.0004%

10% 0.004%

10 l0 0.000°!0 Q

in 6,242,960,263 Stufir 52.870.6 CTP QCTP 11,807,279,832 Btwhr SRSS all error terms' 0.500%

0.5009'0 0.520%

SRSS only Os and Qfw error terms(

0.500%

0.500%

0.520010 (Error rounded to 3 significant figures Convert to MW Bturhr per W 3.412 BtLYhr per MW 3,412,000 In MWt a out 5,290 Mwt Relative errors MWt Q

in 1,830 Mwt Case 1 Case 2 CTP 3,461 MWt

t 17.31 MWt t 17.31 MWt Exelon.

Reactor Core Thermal Power Uncertainty Calculation Unit 1 LE-0113 Revision 0 CTP Calculation Results Sensitivity Analysis The sensitivity of the calculation of core thermal power to variations in the energy terms is determined using estimated values for the energy out, energy in, and CTP.

Table A7-1 shows the results from Cases 1,2 and 3. For this analysis, the error rate assumed for as Is set equal to a predicted error for the measurement of feedwater flow.

In Case 1 and Case 2, the Os error is kept the same; even though, the errors for the QRAD. ORWCU, OPt and QCRO terms are varied from 1 % to 19 % (29 % tor aero). Case 1 and Case 2 show that CTP is relatively insensitive to the accuracy of the ORAD, QRWCU, Op, and CCRO terms, when the CTP error Is rounded to revel of significance.

Case 3 varied the error rate assumed for as. A step change in the mass flow error rate of 0.01 % (from 0.31

% to 0.32 %) was found necessary to change the CTP error by 0.02 0/0 (a change of 1 MW from 17 MWt to 18 MWt and 0.500 % to 0.520 0,10).

The step change error rate was calculated to determine how fine the parameters used to calculate as and QFW need to be to effect the results. Parameter changes that would result in changes in the parameter's overall error rate ress than 0.02 % were found to be negligible. Small variations in the flow measurement uncertainty were found to affect the CTP uncertainty.

For example, CTP is the difference between the energy leaving the reactor and the energy put into the reactor from other sources outside of the core. The enthalpy of saturated water varies with changes in pressure. For every 1 % change ;n pressure, the enthalpy of saturated water vapor (and by inference the energy of the flow) between 800 and 1,300 psig wlll vary by an average of 0.03 0lc.. This change in enthalpy is less than the 0.04 % found necessary to cause a significant change in the CTP error rate. Thus. CTP can be said to be tolerant of the steam dome pressure measurement error specified the pressure measurement loop is shown to be accurate to about -10 psig, which is approximately 1 % of the maximum anowable steam dome pressure.

Tabre A7-1. CTP Calculation Sensitivity Analysis Energy 10 percent of CTP Sensilivity Analysis Case 1 Predicted Assumed erroras error rate Percent of Case 1 CTP Case 2 Predicted Assumed error as erlOr rate Percenl of Case 2 CTP Case 3 Predicted Assumed error as error rate Percent 0' Case 3 CTP Case 3

+/- 18.00 MoM 0.4'73%

0.31%

0.4730,4 0.32%

0.469%

0.0003%

10%

0.003%

1%

0.000%

0.0014%

10%

0.014%

1%

0.001%

0.163%

0.31%

0.1630,{.

0.32%

0.168%

0.0031%

10%

0.031%

1%

0.003%

0.0004%

10%

0.004%

1%

0.000%

0.500%

0.500%

0.520%

0.500%

0.500%

0.520%

Relative errors MWI Case 2 1; 17.31 MWt

+/- 17.31 MWt 0.31%

1%

1%

0.31%

1%

1%

Case 1 52.520.4 0.31%

0.04%

52.87%

152.70%

0.03%

0.14%

152.87%

QCTP 0Fw Op Qcld Oin BhihrperW Btlihr per MW Energy Out 18.030.018,635 BtuA1r 3.754.300 Btuhlr 16,407,160 BIUJflr 18.050.240.095 Btwtlr Energy In 6,201.235.621 BtOOr 37,146,642 Btu.tlr 4,576.000 Btuhl(

6.242.960,263 Btuhlr CTP 11.807,279,632 BtUlhr SRSS all error terms' SRSS only as am Qfw erlOr termst fError rounded to 3 significant figures COnvert to MW 3.412 3,412.000 InMWt Q out 5,290 MWt Q in 1,830 MWl crp 3.461 MWt

Error rounded 102 significant figures Page 76 of 93

Exelun.

A8-1 Basic Flow Equation Reactor Core Thermal Power LE-0113 Uncertainty Calculation Unit 1 Revision 0 Derivation of the relationship between flow, AP, and density Given that the basic flow equation (Reference A8.3-1) applicable to orifice plates, flow nozzles, and venturis is Q = C 4 P

' AP, where C is a constant (Equation A8-1).

Then the relationship between flow (Q) and differential pressure (AP) and density (p) can be derived.

AS-1.1.

Relation-ship between 0. AP and, For constant density, the flow can be shown to vary with the square root of AP such that, a,

I OPT 02 1AP2 (Equation A8-2)

If the variation around some nominal flow, Qo, is equal to some known uncertainty, o,, then 01 = Qo * (1-a1) and Q2 = Qo * (1 + 61)

(Equation A8-3)

Similarly, if the variation around some nominal differential pressure, APO, is equal to some unknown AP uncertainty, qX, then API = Po * (1-ax )and AP2.= Po * (1 + ax)

(Equation A8-4a)

For constant AP, if the variation around some nominal density, Pb, is equal to some unknown P uncertainty, or, then p1 = po * (1-ax ) and p2 = po * (1 + ax) 0-at}

12

= (1-6x) 1+Q1 )

(I+6X )

Equation A8-6 can be simplified by defining a new function of a, :

Page 77 of 93 (Equation A8-4b)

These values can be substituted into Equation A8-2 and the equations manipulated to solve for a, Q0 * (1-at)

P0

  • 0 - a'.J Qo
  • 1 + a1)

PO * (1 + 6-), AP (Equation A8-5a)

Qo * (1-fit) po *(1-6X1 Qo * (17-¢i) VP0*0+a.)'fo"'

(Equation A8-5b)

Crossing out like terms from the numerator and denominator, rearranging, and squaring both sides yields (Equation A8-6)

Exelon.

Reactor Core Thermal Power Uncertainty Calculation Unit 1 LE*0113 Revision 0 Derivation of the relationship between flow, ~P, and density A8-1 Basic Flow Equation Given that the basic flow equation (Reference A8.3-1) applicable to orifice plates, trow nozzles, and venturis is Q== C* ~P.~p, where C is a constant (Equation A8-1).

Then the relationship between flow (0) and differential pressure (~P) and density (p) can be derived.

A8-1.1.

Relationship between Or ~p and p For constant density, the flow can be shown to vary with the square root of ~p such that, Q 1 =J'lP 1

02 AP2 (Equation A8-2)

If the variation around some nominal flow, Qo. is equal to some known uncertainty, 0'11 then (Equation A8-3)

Similarly, if the variation around some nominal differential pressure. ~Po, is equal to some unknown LiP uncertainty, O'x. then (Equation A8-4a)

For constant LiP, if the variation around some nominal density, Po, is equal to some unknown p uncertainty, O'x. then (Equation A8-4b)

These values can be substituted into Equation A8-2 and the equations manipulated to sorve for ox<

0 0 *(1-0'1) 0 0

I- (1 + 0'1 ) =

(Equation AS-5a) 0 0 * (1 - 0'1 )

Po

  • 1- 0'x Qo * (1 + 0"1) =

Po* (1 + O")(), for p (Equation AS-Sb)

Crossing out like terms from the numerator and denominator, rearranging, and squaring both sides yields (Equation A8-6)

Equation A8-6 can be simplified by defining a new function of 0"1:

Page n of 93

Y(a1)Y0 a'1)

(1+o,1 )

Substituting Y(a1) into Equation A8-6 and rearranging, gives Y (al )2 = 0 - Qx )

1+c3'x )

A8-1.2.

The Limit of n (1+or.)-Y(61)2 =0-dx)

(Y(O'1)2 + Y(rs1 )2

- or. = (1-- t3'

.)

Salving for ax, gives the relationship between the uncertainty ax, for either AP or p, and the known uncertainty of flow, a,.

ax + Y(Q1 )

2

- o'x = (1-- Y(,a1 Y )

ax.(!+Y(a1)2)=(1-Y(a1)2)

_ 1 _ y(0,1 )2

~x

~ (1 + Y(r1)2 (Equation A8-9)

Table A8-1 shows the relationship between the uncertainties in pressure or density, ax, and the uncertainty in flow, a1, to be essentially linear with a constant of proportionality equal to 2, see A8-1.2, provided a, is small, which is taken to be less than or equal to 15 %.

Thus for small flow uncertainties, small cf 1, Equation A8-9 can be simplified as a linear function qa1), Equation A8-10, which says the uncertainty in differential pressure or the uncertainty in density is approximately equal to 2 times the flow uncertainty. The inverse is also true ; given a differential pressure or density uncertainty, the uncertainty of the flow Is one-half the differential pressure or density uncertainty.

ax, n " 61, (Equation A8-10)

The limit of the constant of proportionality, n, in Equation A8-10 as a1 approaches zero is found to confirm that n care be considered as a constant within the range of er1 less than 15 °lo to 0.

1-Y(a1)2 Lim (n = Lim ~% = Lim 1 + Y(a1)z a,-*Ot

a,-->

ff1 ) a,--4(7"1 Recalling Y(a,) and expanding Y(a1 )2 in terms of a, 1-- Y(a1)2 a1 (I + Y(cr1)2 Reactor Core Thermal Power Uncertainty Calculation Unit 1 Revision 0

- Lim r

Page 78 of 93 1-1-- 2cr1 + Q12 1 + 2a1 + ff,12 a1[1+ 1-2a,+ Q12 (Equation A8-7)

(Equation A8-8)

LE-0113 Exelon.

Reactor Core Thermal Power Uncertainty Calculation Unit 1 LE-0113 Revision 0 (Equation A8-7)

(Equation A8~8)

Substituting Y(CS1) into Equation A8~6 and rearranging, gives Y(O' )2 =(1- (1x )

1 (1+O'x)

(1 + O'x)' Y(0'1)2 = (1-o-)J (V(0'1)2 + V(0'1?.0-x)=(1- 0')( )

Solving for O'x, gives the relationship between the uncertainty csx, for either 6P or p, and the known uncertainty of flow, CS1.

O'x + Y(0'1)2.O'x =(1-Y(0'1f)

(jx. {1 + Y(0'1)2)= (1-V(0'1f)

(Equation A8-9)

Table A8-1 shows the relationship between the uncertainties in pressure or density, ax. and the uncertainty in flow, <1" to be essentialfy linear with a constant of proportionaUty equal to 2, see A8-1.2, provided <1, is small, which is taken to be less than or equal to 15 %.

Thus for small flow uncertainties, small cs" Equation A8-9 can be simplified as a linear function

((0',). Equation A8-10, which says the uncertainty in differential pressure or the uncertainty in density is approximately equal to 2 times the flow uncertainty. The inverse is also true; given a differential pressure or density uncertainty. the uncertainty of the flow is one-half the differential pressure or density uncertainty.

(Equation A8-1 0)

A8-1.2.

The Limit of n The limit of the constant of proportionality, n t in Equation A8-1 0 as <11 approaches zero is found to confinn that n can be considered as a constant within the range of 0'1 less than 15 % to O.

2 1 20'1 + 0'1 2

1+20"1 + 0"1 u{ 1+ 1-2U1 +U1:

]

1+ 20'1 + 0'1 J

Page 78 of 93

Um d'

r A8-1.3.

References 0+ 2a'1 + cxt2 )_(,_ 2rs, + d,2

= Lim 4.

or, d, 1+2di + d12 )+ 1--2d' + dj a dy (2 +2d1 z 4

Lim 2 + 2d1z

, which after substituting 0 for a, gives Lim (n) = 2 a, -40 Reactor Core Thermal Power Uncertainty Calculation Unit 1 Revision 0 1.

ASME PTC 19.5-2004, Flow Measurement, Performance Test Code, ASME International Table A8-1.

Relationship between a,, Y(a, ), a x, and n Page 79 of 93 LE-0113 Y(d~ )

O~

= 0 t

1+d,~

1-dx Y(di Y 1 + Y(dj }

n = f (dx),= dx 0.0000001%

99.9999998 %p 2.0E-09 2.0000000585 0,001%

99.998%

2.0E-05 1.9999999998 0.4%

99.2%

0.008 2.0000000004 5.0%

90.5%

0.100 1.995 10.0%

81.8%

0.198 1.980 15.0%

73.9%

0.293 1.956 20.0%

66.7%

0.385 1.923 25.0%

60.0%

0.471 1.882 30.0%

53.8%

0.550 1.835 35.0%

48.1%

0.624 1.782 40.0%

42.9%

0.690 1.724

- 45.0%

37.9%

0.748

~_

1.663 50.0%

33.3%

0.800 1.600 55.0%

29.0%

0.845 1.536 60.0% -

25.0°I° 0.882 1.471 65.0%

21.2%

0.914 1.406 70.0%

17.6%

0.940 1.342 75.0%

14.3%

0.960 1.280 80.0%

11.1 °fo 0.976 1.220 85.0%

8.1%

0.987 1.161 90.0%

6.3%

0.994 1.105 95.0%

2.6%

0.999 1.051 100.0%

0.0%

1.000 1.000 Exelon.

Reactor Core Thermal Power Uncertainty Calculation Unit 1 LE-G113 Revision 0 A8-1.3.

,~~ +
0"12)),which after sUbstnutlng 0 for cr, gives L1m(n)== 2 (11~O References 1.

ASME PTC 19.5-2004, Flow Measurement, Performance Test Code, ASME International Table A8-1.

Relationship between 0'11 Y(O'l), O'x, and n 0'1 Y(a)= (1-0'1)

O'x ;; 1-Y(0'1f n J!:::d(O'x)== (1x 1

(1 +0'1) 1+Y((11'f (11 0.0000001 0/0 99.99999980/0 2.0E-09 2.0000000585 0.001 0/0 99.998 %

2.0E-05 1.9999999998 0.4 0/0 99.2%

0.008 2.0000000004 5.0%

90.5%

0.100 1.995 10.0 %

81.8 %

0.198 1.980 15.0 ok 73.9 ok 0.293 1.956 20.0%

66.7 ok 0.385 1.923 25.0%

60.0%

0.471 1.882 30.0%

53.8 0/0 0.550 1.835 35.00/0 48.1 %

0.624 1.782 40.0%

42.9%

0.690 1.724 45.0 0/0 37.9%

0.748 1.663 50.0%

33.3%

0.800 1.600 55.0%

29.0 Ok 0.845 1.536 60.00/0 25.00/0 0.882 1.471 65.00/0 21.2 %

0.914 1.406 70.0%

17.6 %

0.940 1.342 75.0 0/0 14.3%

0.960 1.280 80.0%

11.1 %

0.976 1.220 85.00/0 8.1 0/0 0.987 1.161 90.00/0 5.3 %

0.994 1.105 95.0%

2.6%

0.999 1.051 100.0 %

0.0%

1.000 1.000 Page 79 of 93

Exeleii Reactor Core Thermal Power Uncertainty Calculation Unit 1 Ametek Scientific Columbus Exceltronic AC Watt Transducer Specification (4 pages)

Exceltmnic watt and vat transducers provide utility and industrial users with a highdcgree of accuracy for applications rc quiringprex'ise tnmsurertrents.Thm trarrsdticersprovide adc-ourputsignal proportional to input watts or vars. All models are available with a wide range of inputandoutput options.

" Accuracy to 0.2% of reading

" Substation monitoring

" 0 to t1 mAdc i Exceptional reliability

" SCADA

" 1-5 or 1-3-6 mAdc

" Excellent long-term stability f Energy-management

" 4-20 or 4-12-20 mAdc

" Self-or externally powered systems

" 10-50 or 10-30--50 mAdc

" No zero adjustment required 0 Distribution monitoring

" Most popular models are 0 Process control UL Recognized AMETEK's Power hVdiumeaft 255 Nor* Woe Street Rochateti New York 14665 Phew 1-80D-274-5=

Fein 5NA4486 42 Page 80 of 93 LE-0113 Revision 0 Exelon.

Reactor Core Thermal Power Uncertainty Calculation Unit 1 lE*0113 Revision 0 Ametek Scientific Columbus Exceltronic AC Watt Transducer Specification (4 pages)

Excelttonic watt and var transducers provide utilityand industrial u~rs witha highdegreeofaccuracy £orapplications l1..'quiringpr<..'<.:iscmcasun.mcnts.Thcsc tr<:U1Sducersprovideadc-outpUtsignal proporrional to input wattsor V'<lflj. All modelsare available wirh a wide r<lnge of inputandoutpUtuptions.

Accuracy to 0.2% of reading Exceptional reliability Excellent long.term stability Self-or externally powered No zero adjustment required Most popular models are UL Recognized Substation monitoring SCAOA Energy-management systems Distribution monitoring Process control Oto+/-l mAde 1-5 or 1-3-5 mAde 4-20 or 4-12-20 mAde 10-60 or 10-30-50 mAde AJso avaiJallfei.XlP modular, pJlIg*.in tormatfor limited-space applieatioas requiring large numbers oftransducers*

    • Two, four, or eight modu'es in one enclosure Easy to install. expand. or repair
    • Convenient front-panel access for calibration and output-current jacks available see page, n~for more information.

42 Page 80 of 93

Exeloi S~cifiefriifxns Camel i Narni"I 5 A lap" Rams" 0-10A 20 A 2% A 8.2 VA (mexkrwao) at 5 A

i.0015aA 1 C iaw% 1_" C 10.012% 1 " C 3 f
to 50

C

. -20° C to +60° C

-20° C to +50° C

'~We 2 an page 40.

10 Vde See Table 2 on page 140.

a-1000011 0.25% RO

< 0.5% RO

< 0.25% RO and to 99% [E 4M ms to 99%

< f Second to N%

f'tiwfr faCt!er. _

Any WO S

P-Option" Watts I

0 to A mAda Yarn (Watt Trwddced (Var Transducer(

120 V 0-15111V 200 V OM5 VA (maxar tum) at 1 20 V 100-130 Yet:

85-105 Vac 50-500 oz 50-500 H2 to Nominal 3 VA Nominal or 50 MAdc far rion, dependng on td autputfbrye' ret0ng ~ acs% Rai ~--

~~ba~c Rsao~ ~o.~>& got

. 0-120% RO 4t 0-120% RO 1015% VA (maxinmm)

P-Opticr" Mars (Yar Trwadfuzer) 100-130 We 66-500 Ht:

6 VA Nominal Spin (minimum) t2% of Rejig (n inimunt) 120% of Span (wutkntarn) o Point (Hrt6ttuml None Required --5%

of Zara Point (nxnimum)

(rwx3.TOx5.6'H met x 94 minx 142 own) i Case, see page 122 6e Hz a.15% of Span, 10.2% RO,

+0.24% of Spas, ortcturNai3tive Noncumulative Noncurroletive 0-95% Noscondensing complete llowoutput/Power/Case) 25M VRMS"' at 60 Hz ANSI/tEEE 037,90.1 4 [tit, 8 oz, (2 kg) 3 lb:., 5 oz. 04 kg) 4 lbs., 8 or: (2 kg)

EXCELTROMIC AC WATT 0 fix~sxi AwtiNarp Power ltcertactir.

7somilerawoEffect~r :.

GPeratfte0 Te~utriNture CeebPilaate Yelbp a

Pf Bad at Accrue staaidaw callbestfoa rretleleyt RsW swbilkht (Pat Yosr)

Op-f)""

dittr lsefsNoa _

bleloamric Withstood S

1 Wi>Il+ataad Maximan Not wsi¢m :.

APpaxintate Dimowami (excludWq ate gi plate)

Q"M" wtNt 1.lnaaity

' P-Op1l*a i&dW6i 1-Sri-3-f. 4-XV1u^?

"` Total k"t Pat to 4xcood 286% of stmw Total input eat to axcood 120% of staid

"'Riettt uic levtls as iadieated W Ul,

Reactor Care Thermal Power Uncertainty Calculation Unit 1 Revision 0 06-600 Wattviisal nt 4YWx310x4,TH 7.YWx3.TOx5.6'H j f 12 mat x 99 nua x i l9 mm) (110 mm x 94 mm x 142 mm)

Style If Case, see pigs 112 Style 1 case. 34e ps"r 122 S00-IMO Vars/Eiement 500-M VarsAaemant wror wiiW vohaga compkanca. Reduce load resistance as required.

Specifications subject to t:"a wkhortt Polka.

outPul6.

r ends wink 0 t4 ti aakdc aut:ut.

an unks whit P-0064" aatVet.

vary a w

-UL Rtcagnizad models AMffW Power hoWneft 295 NoM Union Sheet RocltasteS New Yo* 14606 P)

1-8W2M-5368 Fm 585-454?M Page 81 of 93 43 LE-0113 Exelon.

Reactor Core Thermal Power Uncertainty Calculation Unit 1 LE-G113 Revision 0 Reducelo<Jd r.slstance 3S required.

to.25'" ofSp'.,

Noncumtll.tlve 4111'.,801;. 12 kg)

<0.1$% RO

<: I Second to M P-option* Va,.

(VarTransducer)

S.. Table 2on pave 40.

7.rrw)( 3,r D)( U' H (178 mm. 9(mm.142 mm)

Sryl& 1Case. ue pate 122 6tH:

f;o.I5~VA lrnaximuml ote +/-1 mAde VI'"

(VatTransducert

12% of ReNino (minifllUlll}
tM 01 S~n (Minimum) nona RtqulteG

+/-51' of Zaro Point [minimuml Any 2500 VRMS*" at GO Hz ANSI/IEEE C31,90.1 S~n.

flU'" RO, aM Noncumulatlv.

WanS/EfMtnl 5(>>-1000 VarstElltlMnt

.. Hls., 8 ot. 12 leg' 3Ibs,. 5 01, U.5ltgl

.rrWx3.ro x5,6'H U'WxU'O xc,fH xt4 F/IlIl x142mml lJJ2Iloo )( 99 111m_ 119 mrn}

.ae, see ~alle 122 Style II Cu., see page 122 p.Optloa l.eIucMs t-i'I-J-5, 4-20/

    • Total ",,",' "' to ~JttC4 ~ of.

Tot81 inplltllqt toollCood 1m of" n1)jeltttrk Itwx 8$ i,ilwtd l/If Ulll:~(;

43 Page 81 of 93

Exelon.

ORDERING PROCEDURE Specify by Use I number and appropriate suction or option suffixes in the order shown in the following example.

EXAMPLES:

XL342KM2-S-1-RS-SM-SC Tableo 1 Base Model Nmber Setectimf Reactor Core Thermal Power Uncertainty Calculation Unit 1 Revision Auxiliary Current Voltage -RS/Freq.

Opting Option External If A2 is salected, leave this space blank: specify ext. aux. power voltage at end of model no.

Page 82 of 93 Ordering Procedure Exceitronic AC Watt or VarTransducers Table 4 Table 5 Table 6 Table 6 If other there 120 Vac

{std.L specify ext.

aux. power voltage:

69 Vac Aux.,

240 Vac Aux, 277 Vac Aux.,

or 480 Vac Aux.

3-element, 0 to it mAdc Watt Transducer: 120 Vac external auxiliary power 10 A input; 240 V Input; resistor scaling (converts current output to voltage output!; seismic brace; special calibration {example: 7200 W).

XL342KSPAN7A4-5-t-RS-SM-SC 3-element, 4.20 mAdc Watt Transducer, Internal auxiliary powor;10 A Input; 240 V input; resistor scaling !converts current output to voltage output); seismic brace; special calibration (example: 7200 W).

AMETEK" P4war Insbrunuots 255 Nadt Won Street RoclwstK fir York 14605 Pty 1-8W274-631iB Fax: 585451-7806 44 LE-0113 Table 2 fhapo Select4a Compliance Voltage/

Maximum Opeo Rout Ran" Maximum toad circuk Veluae 6 to 44 -Ad. outpsrt b I

PAID I-5 MAdc 15 Vac/3D00 i2 30 Vdc st~

rd, and ii: sprs4ified 1 PA107 011144 15 Y 0

36 Ydrr by tkrt Bass tia

.del t

PANG

v-50 rnAdc 15 Vac/300 11 30 Vdt Niuabrus' Far outputs-PANG-8 1-3-5 mAdc 15 Vdc/30012 30 Vd4 other thaa 0 to to mAdc PAN7-B 4-12-20 mAdc 15 Vdc/750 0 30 Vdc indlea" the applepriate PANS-B 10-30-50 mAdc 15 Vdc/300 11 30 Vdc P-Option in the "Output' posit;...9A.

eorapI.-

PA8 1-5 MAdc 40 Vdc18000 i1 70 Vdc

'..d.11 nuath-b PA7 4-20 m-c 40 Vdc/201X1 i1 70 vac PA8 10-50 mAdc 30 Vdc/600 0 70 Vdc PAS-t3 1-3-5 mAdc 40 Vdcj8W 11 70 Vdc PA7-8 4-12-20 mAdc 40 Vdc/2000 0 70 Vdc PA8-8 10-30-50 mAdc 30 Vdcl600 f2 70 We watt Var W

Calibration at Rated Output Ettmanil MR&I Mg.

matel N4.

cafs i5~.126V Namleat low 1

XL5C5 XLV5C5 Single Phase 500 W or Vars

" till and2ut-shu n

1 in*

XLSC51in XLVSC5ltr2 3 Phase, 3 Wire 1000 W or Vars requaria a balurced vslta 2

XL31KS XLV31K5 3 Phase, 3 Wire 1000 W or Vars 2,17" XL31K52w XLV31K52w 3 Phase, 4 Wire 1500 W or Vars 3

XL342KS XLV342KS 3 Phase, 4 Wire 1500 W or Vers Exelon.

Reactor Core Thermal Power Uncertainty Calculation Unit 1 Ordering Procedure Exceltronic AC Watt or VarTransducers LE-Q113 Revision 0 ORDERING PROCEDURE Specify bv !taSI trlfdel number and appropriate I.htetion or option wffh,es in t~ order shown in the following example.

If A2 is srNected. leave this space blank; specify ext. aux. powervoltage at end of model 00.

Ba. Mod., No.

Output r..ld Auxlllary Power 1.... 3 Currant Voltage

-RSlfreq.

Option Option Input Input Option 11 12 13 Teltlc..

T."Ie..

TlIbIe5 T.ltlttS Tabid ExtefOal Aux. Power Table 3 If other then 120 Vac (stdJ. specify ext aux. power voltag..:

69VacAux.,

240 VaeAux.,

277Vac Aux.,

or 480 Vee Aux.

EXAMPLES:

XI.J42K5A2*5,'*RS*SM-SC 3-elemarrt, 010 +/-lll1Adc Watt Transduc8I; 120 Vac external auxllla/Y power; 10 A input 240 YInput; resistor scafing Iconvel1S current Otltput to vollage output); seismic brace; special cafibr.tion (example: 1200 W).

Xl..342K5PAN7A4*5-1-RS*SM*SC 3*t'-ment. 4-20 mAde Watt TransduClr; intarnalewcillary po~r; 10 A. input; 240 Vinput; resistor scaling (celWert, eurttnt output to volttgt output); seismic brac.: specill calibration lelUlA1p": 1200WI, Watt fI1IIdt1.II" Xl5C5 XL5C51112 Xl31I<S X131K521O Xl342K5 V.r M.UtI14I.

XLV5CS XlVSC51m XlV31K5

)(LV311(52~

XlV342J<5

~

Single Ph3se 3 Phase, 3Wire 3 Phase. 3Wire 3Phase, 4Wir.

3 Phase, 4Wire C.lil,atloa at R.te" Outptt (SA. 120YNo***IIpPMlI 5OOWorVars 1000 W or Vars 1000 Wor Vars lliOOWor Vars 1500WorYer.

  • Il12*&nd2ltl-el4JlllftC'"

'....,Ilalue****ta..<

Table 2 Oatput Selectita

'.(o+/-1 mAde o~ fa It,"rd**dll it specifl...

by Ute 8 MolI,I N

hi outpIfb otIMrtfwt

  • Ie il I8AtIc.

imIicllt8 'he IfJIInJpriate P-option lit the "O..,1It" po.iti** of die COl8pktte model......

f.:Olllta PAN6 PAN1 PANS PANe*S PAN7-8 PANS*S PAS PA7 PAS PAS*8 PA1*S PAS*S

_Ur81O.'

t-SmAdc 4*21 MAde 10*50 mAde

'.3-5 mAde 4-12020 mAde 10-30-50 mAde 1-5 mAde HOmMe IQ*50mAde 1*3-5 mAde 4-12*20 mAde 10*30-50 mAde Complillce Volt.ge, MaXimum Loa_

15Vdcj3000n 15Vdc/7SfO 15Vdcl3OOU 15 Vdcl3000U 15Vdc/750U 15VdC/300U 40 Vdc/aooon 40 Vdc/2000U JOYdc/600U 40 VdC/8000 U 40 Vdcl2000 n 3tl Vdc/600 U Maximum Optll Gira.V,pI 3OVck:

30Vdc JOVde 30Vdt 30Vdc JOVdc 70Vde 10Vdc 10Vdc 70Vdc 7OVdc:

l0Vdc 44 Page 82 of 93

Exelon 1"ahte 3 Auxihry Power Supply Selection Taiite 4 Input Sekiction cmetd Table 5 _Scaling Ret:istar (AS"oency Options

-RSt Scaling Resistor

.6 400 Hz

-12 50 Nz (not UL Recognized) RSt 400 Nz and Scaling Resistor RSt 50 Hz arid Scaring Resistor Tale 6 Iffier Optim 04920 1111620110fi2e

.20 50-200% Calibration Adjustment (current outputs)

-21 50-200% Calibration Adjustment (voltage outputs)

(available only with 0 to +/-1 mAdc units)

-24 24 Vdc Loop-Powered IPA7 and PAT-B models only)

(consult factory for specifications)

-CE Analog Output Shorting Relay (available only with 0 to tl mAdc units)

-SCtt Special Calibration

-SM Seismic Brace (available with 0 to tl roAdc units)

(consult factory if you desire this option with a P-Option unit)

-2 Zero-Based Output Calibration tax-: PA7-Z m 0-20 mAdcl aavallable only with P-Option units, except PAN-8 models)

Reactor Core Thermal Power Uncertainty Calculation Unit 1 Ordering Pmcedure Exceitronic AC Watt or VarTransducers Page 83 of 93

"" For external soxillary poor rues odw fl= 120 Vac, spcly tire Y in the Iris psitiew of the comr{ose aadri amber. (ExsrspW 241 We Ass.)

DC axtaraN oaudliary patiwr ave"e; see Special Opians an page 121v 7

"" OF"* -4 "Madras a Style I case. Is" pp t22 for case dliresasl"L height of or"" WWsl is 1.4rfer soils with -a "Ves j t You Mist spoor the doelrad ouipart whop.

rer Lto t o

"Hy raw qua o io tt vdc. Loa harodenee Is t ItM}/Vdc (ice.

Fm P-omits awe-spsoffy aup boa+ e.15 Vdc (PAM medW or 4-+14W PAmotel" Load)tapdmcei2AI.9kor2atkstNdra Vim) for anito with oatprds of & 21, of 54 aAdc respectively.

?his i is riot t~ of die mold asm0or, brA rmui bo prosgfod to d1s factory when yea plow your aide(.

tt Yes nest specify the desired lopm value:

1 to tt mAde can be calibrated within Wt1O% of doll ataaCsrd-calibafiea Inpst watts or visa, f aarpfo: A 2-olearoat watt tow dscer Is atthwsd is 1010 W ssaadsrd. The -SC opdee as be added for input levels from 914 W

(

to 1110 W (11146)1 p.

nab can be cofitraNd wubdo W W% of *Wrasadard-cati6ratfea irrpai waft a were, This hhearralion is no part d Oa model fir. bet mat be pmvWsd to the foctory when you plow year order.

N you require addihioul options not shows hem see Special Optiorm an ptiga 12L' When wilarbrg my special options or mats then l option, you Moist first consult the factory for pricing and del5irsty esth-ees.

A Power lashumoaer 256 Nor* Union Street RocOesdar, New York 14MS Mtarta :1-800-274-536 Fax. 554%7M 45

!.E-0113 Revision 0 Current Range Calibration at Rated Ouipot 20110 Nominal w/ Atcwecv (14 A Nominal Inputl

-3 f A 0-2 A 100 W or VaNEWmen(

-4 2.5 A 0-5 A 250 W or VarsVElement Std.""-

5 A 0-10 A S00 W err Varaf lonamt

-11 7.5 A 0-15 A 750 W or Vers/Element

.5 IC A 0-20 A 1000 W or Vers/Elemsnt IS A 0-20 All 500 W or vary/Element

.8....

25A 0-30 A MG W or Vars/Elsment Motai14t Vokaye Rsnga wl AgMutairs+

callbratiee at Rata! out w 1120 V Nominal Inaitt)

-0 69 V 9-75 V 250 W or vers/Element Std."'"'

In v 0-116111V Mw or vwsfeawat

-1 240 V 0-300 V 1000 W or Vars/Etament

-9 277 V 0-340 V 1200 W or Vars/Ehrment

.2 480 V 0-600 V 200O W or Vars/Elament -

Matt MAdc uaisl low no kalm,om Ronnit A2--

External Auxiliary Power (120 Vac std.)

x-135 Vac 50-560 Nt 3 VA A4 Internal Auxiliary Pourer (self-powered) 70-112% of Nominal Aux. Power Voltage Equals Input Frequency 3 VA (P-optfwi usitsi A2"" (tasty bgadr)

External Auxiliary Power {120 Vac std.)

1W130 Vat 99-590 N:

5 VA A4 Internal Auxiliary Power (self-powered) 84-106% of Nominal Aux, Power Voltage Equals Input Frequency 6 VA Exelon.

Reactor Core Thermal Power Uncertainty Calculation Unit 1 LE-0113 Revision 0 Ordering Procedure Exceltronlc AC Watt or VarTransducers labre 3 Auxiliary Power SlpplySelection rable 4 InputSelection 3VA 3VA Volets' Volta.. Rail" CtUlmlti** at 0.,..

HIIIIJuj wI ACCUracy mo VN 1101ltl 69 V D-75 V 250 Wor V.rs/Eillment 121Y "lSIV

.WorVarsJEle.....

240 V 0*300 V 1000Wor VarS/Elment 217 V 00340 V 1200W crVerS/EIement 480 V 0*600 V 2000 Wor VarS/B.llHlInt -7 1*-l.M~""""'poside*IlllctkJll""'" "".1

,.pIt"p.

IRtl.II

  • 0 St***....
  • 1

-9

  • 2 Extflnal AuxllQrv Pewer (121V.c std.)85-135 V.c 50-581 It!

Iflt,rnal Auxiliary Powtr 1$.It-poweredl 76-112% of Nominal Auk. Power Volt.ge Equllllnput Frequency Cun.1trca"..

CaIlbrttioll at fill... 0UqI1t H.ltIIJul wI Warley II A"omfa.llp,l1) t A 0-2 A 100 Wor Vars/Ellm.nt 2.5 A Q-5A 250WorVars/Elemtnt 5A

.'0 A

'*W Of V.rII!IIMnt 7.5A

()-15 A 150W orV.rs/Element 10 A 0*20 A 1000War VarsjElem.nt

()'20 A1500 Wor Van/Element 25 A 6-30 A 2500 Wor VarS/Elfment DltIII

-3

-4 Stll.-*

  • 11

-5 IliA

  • S****

~

.tf+/-~llllltll

~A2**

A4 (P-o,dII1JlIks1 A2**,r..".....1 External Ailltllilry Pow., (121Veest....

,.,30 Vee

!iO-!ilII1b I VA A4 hltemal Auxiliary Power (self-powered) 84-108% of Nominal Aux. Pow.r Voltage Equals Input Fr.quencv 6 VA I**".e llIldlmy pt'IMf....... Idler !IuIll12l1 Vee. SlIftIly1ft...... the IIs1,.sitlft "1M c (u.,II: 241 Vat: AIIl.) I I DC t'ttemat auiJiary power 8pecltf o,IiouII,...,21. ~

..... 0p1I.-" reqIIlrn I ~ IcaM. (SIt pep 12l forcae dilHatIOIIL

.............of......., tcrip(s)IIurteuJliltwhIJ *...tftllJ TabJe5 Seal..Resistor (-RS)lfrequency OptiOM

.QJ!itI

-RSt

  • 6
  • 12 RSt RSt QucriptiQR ScalJng Resisfo, 400Hz 50 Hz (l1ot UL Recognized) 400 Ht and Scaling Resistor 50 Hz and Scaling R.mtor t YOll....If*iff.. dell.... 0lItpIII woINp:

Fw 1MttlJA* IIIlt1. tpeelfy.......ato :tlDYtID......

......... ll1MONlIc~

Ftr P:OptitI"speofty...".."15Vde WAN lIIOdefsl.r

...Y*(PA 1JlH~1s2OlJ. *

.,2ttkONde.

~),.

widt of 5.20. orSlm.Uc, raped"".

TW,ilIIonaadoaIsaotpuc

__.luItmnt"

........to dle factoty wtlta~,

vow......

Table 6 OUter Opti...

0atkID

  • 20
  • 21

-24

-SCtt

-SM

-z Qucrlgti.

50-200% Calibration Adjustment (current outputs) 50-200% Calibration Adjustment (voltag. outputs) lavaliable only with 0to +/-1mAde units) 24 Vdc loop-Pow.red (PA7 and PA7-B mod.,s onlyl (consult factory for specifications)

An.'og Output Shorting Relay lavailable only with 0 to +/-1 mAde units)

Special Cdbration Seismic B(ac. (available with 0to +/-1 mAdc vnits)

(consult factory If you desire this option with a P-Option unit)

Zero-Based Output Calibration lex.: PAJ*Z,. 6-20 mAdei lavaHable only with P-Optlon units, except PANoS mod4llsl tty.._sf apedfy'" lies,," 1IpIt....

It! it eu'" ClIIaII'..nwldlia..1IKoflllolr carillfadlll, C~ Az..

MorIs" to 1_W TlHI.sc optlHcn Ito addU for hlp

_weM) tll1..W(lM'll.U r:.otIiRI Mea.~widIla.......,...r...mlard-elflllta1l..

iIlpatwa. Of wars, Tllb lafaorrdoBls.... pMt01..modelllllllltle,. bit IIllISlIIa,rovfdell co tile 'ac1Oly wIIotI,.. pfKII ye. Older.

If yOI "lplirl.....1ltptiou not** hete, see Speci.1 Opticns ell PIP f21. WIle. orariq IAf $peelal optioaI, or

....thu Iblaopti'" YOll muse first conalt til. faetory for pricing... deJivefy estirIaIR 45 Page 83 of 93

e6n.

Reactor Core Thermal Power Uncertainty Calculation Unit 1 Revision 0 0 B32-C-041-J-023, Rev. 1, Recirculation Pump Curve (1 pages) fic W-30 314""* "a-F-"*

W A " na ".0 Page 84 of 93 rust-K MMOIN co

-0144 4 LE-01 13 n! TO i J q"" "" r " r"4.

~

W

^

{..

srw"s "

N

~

s>:

~ O~

.,wv"t"

, "" R"v

" `.'rtt^...'

~'

avi

" ~,~`"~.

I"f" rte

".'.""N i

~"" IIM x

1 W~ "

tV"rFS~W".w".""

" """"""" ""Y".arra4"d ~

"aa.R

-1" 0

1-1 :7 gLz

"~~~

-t11 w

"w"w

..,~, r"v+"""r. wraw

~

T PAP MEMBE BE

+~ r Ir r "

p

{w m.~ "

w J t

.1;/ -J" ~iJ1~" ~ w.j. J. ::.~.:

w """

~~."r M-C-2

- ~

r~

MIM S'I

~ t

.~~~i

~~ ~~. ~

w.

r"m 1"

r t;, ; A-I TO-V, W7 z 0 0 W

s "M

MIwN~V r.~.-_

{

M" '~s w " s Y.

11.

. ~It" 5 ~

~

'~

NV" i

~

.v. 0 X-t

" s """

=2 511 s " 1 "el "

It "

f " iz sr. - 7: 7,-. Z.

, 4 " r

~

Iwi "N _ W "

w ~. "

. " ~a:R~ "

" '11 P, it F _-F: 7 F

"" } "M j 3T.

EA1Q.

Agobr Y,

Act 1, X Exelon.

Reactor Core Thermal Power Uncertainty Calculation Unit 1 LE-Q113 Revision 0 0 832-C-001-J-023, Rev. 1, Recirculation Pump Curve (1 pages)

"C )4C.J:) ;)t(1tU.. ""..........ee.

0 * >> ~ UoI>l flllflt( OtfttD*., CD

...:t......,

  • Page 84 of 93

Reactor Core Thermal Power Uncertainty 1 Rosemount Inc., Instruction Manual 4259, Model 1151 Alphalinee AP Flow Transmitter, 1977 (PAGES 1, 6, 14, 24, AND 29)

MODEL 11510P ALPHALINE8 vAI-F FLOW TRANSMITTER CAUTION:

READ BEFORE ATTEMPTING INSTALLATION OR MAINTENANCE TO AVOID POSSOSLE WARRANTY INVALIDATION CONTENTS INSTALLATION Page 1 CaitsraVon Page 3 Span Correction for High Line Pressures Page 4 OPERATION P&ge 5 Spe6tications 1fstop Squaw Root Page I MAINTENANCE Page 8 PARTS LIST/DRAWINGS Page 11 Deaign S"cil¬cations Page 11 Parts List Page 14 Drawings and Schematics Page 15 LOPYHr6"F W)SGQ0UN1 MC;. 1974, 197i1976

'ALFHALINE' AND `6-CELi.' at* Rosemount Trademarks O'O1e41ixr ar 00. or r>>o.a of 1-0 folk->>g U-S patewr 3.0!1,413, 9 85~Q34: 3> 1ytS.02B:.rnc¬ 1.65S.

t;

`'anada Pelerrsd ?9&8; 1974 p,719nfe !6rezrc

~ t<fa. 1 1F1,8#2 0A"t"r US

. 4"'V 1prvgn filter. n " 3

":,.~ ;+ -a" t n~Frrry Page 85 of 93 INSTRUCTION MANUAL 4259 POST OFFfCE sax 35129 MINHEAPrLIS. MINKE5OTA 55435 "NE: ($412)441-5580 TWX: 910-575,3103 1FLE}C: 29 "0183 CABLE; POSEWtdtf Revised 11/77 LE-0113 Revision 0 Exelon.

Reactor Core Thermal Power Uncertainty LE-Q113 Revision 0 1 Rosemount Inc., Instruction Manual 4259, Model 1151 Alphaline<m liP Flow Transmitter, 19n (PAGES 1, 6, 14, 24, AND 29)

INSTRUCTION MANUAL 4259 MODEL 1151DP ALPHALINEe

.;filS fLOW TRAHSMlnER CAUTION:

READ BEFORE ATTEMPTINQ INSTALLATION OR MAINTENANCE TO AVOID POSSfBLE WARRANTY INVALIDATION CONTENTS

'NSTAt.!..ATION callb(ah~n Span Couectlon tor High lIn~ PreSSUfQ$

OPERATION SpeeiiicatlOM 11$tOP Squ~lIe Root MAINTENANCE PARTS LlST/O~AWINGS 08$19" $peclllcatloil$

Parts List OrawinQa lind Sf;:l\\emali<;s Page 1 Page 3 Page 4 PageS Page 7 Page (1 Page l' Pag& 11 Page 14 P3Qe 15

'ALPHAllNE' ANO -CELL' ar. Rosemount Trademarks PrOl<iJt;Url by 01'& Of mo(~ 01 (tIc.\\ fOI(f)w1119 U.S Pslimffi No 3,271.&;9: :J.J18. I~J. :USl8.3f>O; J.I5*M.!!3a..*U9:l.88~

.:1,80(1.413::l 854,009. 3.111~.02S;,lrK/ 3.859.594,

~,ma"'ll Plllttnttd 196$, tg74 P:J19ntt1 Ml!~rc~. /\\10. lHI.81t Oit:tN IJ S. ii/V 1"0'<'1 P;Jf(JMt f1M.lt'd -x ;tfi'~if""1 Rosemount Inc.

POST OFFICE. SOX 35129 MINNtAt'OlIS, MINN£SOTA 554~5 PHON£: (612}9-4L*5S60 TWX: 910*516*3103 T£lEX: Z9*0l83 CABLE; ROStMOUNT

~

R~"'IS<'e'd 11/77 Page 85 of 93

Specifications - 1'151 DP \\/A-P Flow Transmitter Functional Specifications Service Liquid, gas or vapor Ranges 0-5130 inches H20 0-2511,50 inches H20 o-126/760 inches H.0 Outputs 4-1st rnAUC. sQuars root of input Power Supply Load Limitations Sets Figure 7.

Indication External power supply required. Uts t045 VDC. Transmitter operates on 12 VDC with no toad, Optional meter with f-314" linear scale.

(k-t00%, Indication accuracy is +/-.2% of Span.

Hazardous Locations Explosion prowl. Approved by Factory Mutual for Class 1, Division 1. Groups 13',

C and 0. Class If. Division t. Groups E. F and (3 : and Class lit.

Division 1-,

Certification by Canadian Standards Association (CSR) for Class 1. Groups C and El available as an option. intrinsleatly

  • ate: FM certification optional lot Class
1. Division t. Groups B. C and fJ when used with listed barrier systems.

Spots and Zero Continuously adjustable oxternalty, Zero Elevation and Suppression Zero elevation or zero suppression up to 10% of calibrated span_

Temperature Limits

°-.20°F to i-1501F Amplifier operating.

--40* F to 42200F Sensing element Reactor Core Thermal Rower Uncertainty Dumping Fixed response time of 1/3 second.

(Corer frequency of 04 Hz)

Turn-on Time to seconds: No warmup required Performance Specifications (ZERO BASED SPANS, WEtiENCE CONOf-froNS, 376SS ISOLAMove OIAPHRAGMI APPLIES FRau 2s ro !fit,LOW).

Accuracy 1:0.25% of calibrated span for a range of 25% to 100% of flow 1814 to 100% of input pressure), Inctudes combined effete of hysteresis, repeatability and conformity of the square root function ;

Dead Sans!

None Stawttly

~,0.25*to of upper range limit fort months..

Temperature Effect The total effect amtuding zero and span errnrst t 1,5% of Lpper range limit per 100*F

(~:2<5% for low rata".)

Overpressura Effect Zero shift of less than --o.5% of upper tango limit for 2000 poi (+/-2.0% for range s7=

Static Pressure Effect Zero Error: to.S°,o of copper range limit for 2000 poi (-tt.0% for range 3).

Span Error_ ~{3.5~O.t9b of reading per 1 tOfl poi (-0.73!.tl. t'itr for range 3, L This is a systematic error which can be calibrated out for a particular pressure tsefore Installation' Vibration Effect

,+/-,o*05% of uppor tango limit pot g to 200 Hr in any axis Power Supply Effect Less than 0,005% of output span per volt-:

Page 86 of 93 Load Effect No toad effect other than the change iri power supplied to the transmitter Mounting Position Effect Zero shift of up to 1" W.0 wh ch can be calibrated out. No span offoct : No cflooa in Pfane of diaphragm.

Physical Specifications Materials of Construction t Isolating D18phragms and DrainVenl Valves:

316SS. HAST£LLOY C qr MONEL, Process Flanges and Adapters :

Cadmium Plated Carbon Sieel, 31635.

HASTELLOY C or MONEL.

wetted o-flings VITON.

Fill Fluid.

Silicone tit, B

otts`

t',admiurn Plated Carbon Steer.

Electronics Housing:

Low-copper aturnenum (Nt~MA4)

Paint:

Polyester-Epoxy, Process Connections 1f4-NPT an 2-118" centers on flanges-_

112-NPT on 2". 2-t18" or 2-114" centers with adapte(s.

Electrical Connections t/2-incti wndua with screw tvrrmrtdia:

and integral lost jacks compatible with miniature oanana plugs (Pomons 2944,.

3890 or equal)

Weight 12 pounds excluding oWions-FIGURE 7 D LIMITATIONS 4-4"* -Ant POWER SJPPCY tvol~)

tYDNt:L is a rr3f1(twork i}1 WgrRdr011a1 FYKAW Go NASYFlL Y 4.t radarusrk Ot 010 Ca,40r Canto Y1rON,% a DuPont t~4demsrk Tetrrflrnob4Y Far SA+4fA Srai:~ritrrl PnfC2r1 f t q7; i.E-0113 Revision Q operating.

-601F to r t t3(}' F Storecle, Static Pressure and Overpressure Limits LO 0 pain to 2000 p2ig on octhvr oicfo without damage to the transmitter Operates 16'.10 150.0 within specifications between static line pressures ut 112 psid alto 200 prig MV 10.000 f?sip proof pressure on the LOAD

cNrr.+s) flanges.
sot, Humidity Limits 0

()-t00% A H.

Volumetric Displacement 0.01 cub ic inchaa.

LGm than

't?pr+anat motor riot drprudV!!U for Ct ravp ~ :

Exelon.

Reactor Core Thermal Power Uncertainty LE..o113 Revision 0 Specifications - 11510P vAP flow Transmitter

,.....---------- FIGURE 1 -----------....

LOA.D LIMITATIONS POWER S.JP9t. y (VOC}

Functional Speolflcatlons S.rvlce liquid, gas or vapor Range.

0-5/30 inche, H20 0'25/' 50 incMs H2°

().1251150 lnchec H~O Outputs 4':!O mAIJC. $.quar. rool Gt Input Power Supply Externat power supply reQuited. UI> to 45 voe. TransmHler operates on 12 VDC with nQ toad Load Limitation.

See Figure 1.

Indle,tion Optional meter with 1-314'" !loUt scale.

(}. tOOlJls. Indication accuracy is !.2% of span.

Hazardous Loc"'onl Expfo.'on pr()of: Approved by Factory Mutual for Clus I. Division 1. Groups a',

C and 0; cra~s II. 0111/$;011 L Groups E. F and G;

,)n<!

Clas, lit.

Dht/slot}

1.

Certification by Canadian StandardS Association (CSA) for Class I. Group$ C and 0 available as an option. IntrtnsleaUy ute: FM certiflcatlon opUonal 'Ot Clau I. Division t. Groups B, C and 0 wilen used witl'lli$ted bar,.~r systems, Span and Zero Continuousty adjustable oxtemally, Zero Elnatton and $upprt!tS$on Zero elevatlon or zero $uppression up 10 10%0' cahofated llpan Tem..-rattmt LImit.

'20"F to i'1SO"F Amplifier op@rating.

'-40" F to

  • ~20° F Sensing elemenl operating.

-60A F to t'tfID"F StoreQe.

Static Pressu,e and OVet'prenurt Limns I) pl)iQ to 2000 p~ig on oHhcr Q;OO without damage to the traosmltter Operates within $£lacilicatlon5 between static line pft~$SUreli of 112 V::sils lifltJ 2000

~j~,

10.000 psig proof pressure on the flange'S, Humidity Limits 0-100% AH.

Vofumettlc O,spl.etmanl Lc~ than O.()1 cub!,. Inche&..

._--~-

'OptlOftal ml!tllr not Ifp(lfOV!JC! for (lroup S OiSmptng Fixed !spons. lime Qf 1/3 second, (Comer fre{jlMocy of 04 Hl)

Turn-on Tim.

10 seconds No warmup r~ulrfld Performance Specifications IlEAO OAS£O SPANS flEFEnENCE CONOf*

TTONS, Jll!SS ISOl.AflNG OIAf1HRA<Uf$-

APf>t.lES FAo.'W 2~ ro,~ flOW)

A.ccuracy JO.25~ of calibrated span for a range of 25~ to 100'\\\\ ot flow {6~ to 1~of input pressure,_ Includes comblnedeffecls ot hystel8sis. repeatability and conformIty of lh& square (oot function, Dead Band NOM Stebitltv 1,:025,. of upper range hmh lor6 monthS.

Ttmp.....tur. Effect TI\\(I IQtaA aftect Inc'tKling zerO and span errors:

t 1.5% of upper range limIt per ll)O"'P (::::2.50/. for tow raf19t'!.)

Ov.,preuu:. Effect Zero shih of

~ss lfum ::0.5% of upper range limit tor 2000 pSt (+/-2.0'l4t fQr rang&

5),

Static Pr~**ure Effect Zaro Error; to.S% of lJp~r range I(mIt for

1000 psl\\-:!:1.0% tor rango 3).

Span Error

-O.5':0.1~ 01 reading per 1000P3i (-0. 75.',0.1"IlJ for range:)'j, Thi3is a

syst~mahC error which can txt calibrated out for 3 p;trt!cular pressure l)efore rnl:UallatlOIl Vlbr.1lon Effect to,.05~ of upper (tlngo limit POI' g to 200 Hz in any 3)(ls.

Pow., Supply EfflfCt Less than O,OO5lft 0' output span per vatt 7

Page 86 of 93 Load effect No load e(fcrct other l/lao lhe change in power supplied to the t'ansmitt~r Mounting Position Erf~t Zero $hifl of up to 1" fi~O wh~ch earl be cohbrQtQO out NQ Gpon effect. No effecl jn pfane of diaphragm Physical Specifications Materi,'s of Construction t l,o**Ung Diaphragms /tAd Oratn/V(tll(

Valwtt:

3l~SS. HASTELLOY C Of MONEL, Proc..., Ffaog4's.nd Adepter.:

Cadmium Plated Carbon Sh~IH. 3163$,

HASTELlOY C or MONEL.

W.ttfd O-Rlng_

VIl"ON.

FlU Fluid:

Si1i.:one Oil, eoft~

C;Jldmlum Plated Carbon Steel.

E"clronk:. Housing:

Low~Qpper atlJfrnnurn (NCMA""}

Paint:

Palyesler*Epoxy, Proceu Conn_etlon.

1/4-NPT on 2-1/8~ ct)nlers on flanges, 112*NPT on 2~. 2-118" or 2*1/4" e~l'lters with adapt.f$.

ftfe'rlcat ConMCUon.

Ill-Wetl comJwt WillI scr~w ttmHHI<cls and integra. tC$t,ackS compatible with mintature banM3 plugs {pomona 2944, 3690 Of "qoal~

Weight 12 POllr\\dS excludiog options p.K.'HEI. 1$ d l,adft.,**,,1< ul IN,tnJI.MI/J1 NK.S,4' CiJ HASTI:HVY;.f /I ttiJd!lIVi),1f ot til')

CiJ~O(

CorJJc wrON,';$.. OUPoM ~f"l1e'm;1f*

r"tlmfll1lc!Jy P&f SA~M Stllll(1llltJPMC2iJ 1 1'}7$

PA elun.

'S LIST/ORA MOOEL 1115101)

IIslop 4

c 12 LM, US

COMPLETF0 OE5IGN SJOLCFICATION PAN"$

0" 5 to 4-34 inc 0-25 t0 0-i50 0, 125 to 0"750 In STANDARD ACCESSORIES All Modets are ship-ped with flange adapter, vent/drain calves end one instruction manual per shipment_

OPTIONAL THREE-VALVE MANIFOLDS (Packaged Separatefy) 4INGS SEC' Part NQ. 115`1-is0-1 : 3-'Valve Manifold, Carbon Meet (Anderson, Greenwood & Co,, WAVC)

Part No.

1151-1511>>2 : 3-Valve Manifold. 316SS (Anderson, Greenwood & C0_ MaAVS)

Reactor Core Thermal Power Uncertainty Design Specifications Page 87 of 93 TAGGING ALPHALINE Differential Pressure Tram milters will be tagged in accordance with customer requirements. All tags are stainless steel CALIBRATION Transmittvi are factory Calit)rated to customer's specified spurn, kt Calibration is not specified. transmitters are calibrated 211 maximum range Calibrat=on is at ambient temperature and pressure LE-0113 Revision 0 hevs ti;0 (0-127 to 0-782 mrrs Hrp)

_i___-

u+clift HIO (1'1635 to t}-3810 mw H=0}

cfas Hjo (0-311'5 to 0-19050 trim "

O)

OC. asttaerO (0011 Of IN" MATERIALS OF CONSTRUCTIO14 PLANGCS AND ARAPTERs ORAIN-VENY VALVZS ISOLATING OIARHAAGM5 Cadn++um PWsd C.3:

31Gss 31$1`15 Cadtrtium MAW C S.

HA$TELLOY C HA5Tel.LOY C-275 Cadrntusn Pletttd C.S.

MONEL MONSL 3f6ss 31fss 315'99 31 "s 31ess HASTELLOY C-276 3It3~' S'S 316""t' MOREL HASTELLOY C HASTELLOY C HASTELLOY C-2715 MONEL h4QNEt MONEL CORE OpTio" LM Linew Matte_ 0-100% scat*

M8 Optional Mounting I3ratkat rat M*Untitrg to Z<` MOO Ps Optional Mountirv 8r*Cbsi for Pane# klpuntt+rg F;3 Optional Fiat Mourttio ; 8reckel rot Maearrting to 2"' Pips n) pitta 7/nnlir}uttn. Tats 02 Skde Yltltiurofn, Bottom CE Canadla-t YitRndflrds f :9oG ttpn (CSAJ Fltploston Proof Cteltkavon for Cia3 i, Group& C and 0 ; Class It, Grrnrpa E, F and G.' Class 111; (Entl

. IV)~

INTRINSIC SAFETY APPF40VAL (All At* Apprav*d 811 Factory Mutus" CLASS 1. DIV. t, 8AARIER GROUPS AG ENCY MANUFACTURER BARRIER 1140011.

8 C

0 AI Foxbom 2AI42V-PGS, 2Al.13V-AG6 t;2 Taylor 12431134, 12451144 T~~~t 12~

S ~tS4 F3

{

WeatingtlrlusC 7aS 1

SttFC 1?

F4 Leeds & Nortrtrvp 316569, 3161747 F8

{

F*cur A *sorter 8051"1023V0l. W5tOVU01 -.. _

Ft3 ~

~

F;stutr Convot3 A<=

F7 FM Homywetf 3asa5-xxxx0lto "III/t12"F585 j

X x

Exelon.

Reactor Core Thermal Power Uncertainty LE..0113 Revision 0 PARTS LISTlORA INGS SECTION Design Specifleatlons IMOOEL AlPHALINE ~'lOW YAA..SMI'tT£R 11510P COOl RANGEl

)

().S to 0.30 incheos H~O to.t~l Ie 0-162 mm H1O)

()..;lll to 0-150 inGMt H10 (0-63$ 10 0*36'0 /nUl ~()l 0, 125 10 U*7Sa Incl'\\lU. H,rO (lhU15 10 0-19050 mll1 H.O)

CODE OUTPUT C

4-20 mAOC. lQu.r. root 0' tflPVl MA'tlR'AUl OF CON$TRVCTlOH COOl! "lANG" AHD AO"'UAS ORAIH-YtN'I'VALvet ISOlATJNG OIAPHl'tAGatlS It!

CadfTWum PI'ltd C.8, 316$$

J!$SS l~

Cadmium Plil.o C $.

HASTEI.LO'l' C HMTiLl,.OY C-21l; 14 Cadmr\\lm Pletoo C.$.

MONEl MONIR n

3tti$$

316$8 31(SS9 13 316SS 316$8 HASlalOY C-21&

24

)16Sl;i 116SS MONEL 33 HASTELLOYC HA$TEllOYC HASTeLlO.,. C*Ull MONU MONEt MONfl cooa OPT!OH&

I.M lineal Mit.", (HOO~'u'*

M8 OptfonCll MOUf'ltltl9 8raC\\01 tor Mounting to t'. l"lpe PB OptiQrlal MCIll'ltino &r.c~" for Plln~ M<luN~

F,g Oplioo.' F~I MountH'" 9fackt>t fOf Moonlit!<) to 2" Pipe 01

~Il 'J~inf"l.., TM 02 s."e '(entiO"'fl, 80ttom ce CllI\\aQl,., 8tRIl411rU) AS~Il~IIQn {eSA, E>lplO"on Proll4 CfJ!1llketloll lOt Ctus l.

Grat>p. C and 0; Cr.a" 11" Groups Iii, F.nd <k Clan Ill; (End, IV),

I..-rA'HSIC SI.F£TY APJtI\\OVl\\l. tAli At. Af1Pf0\\," a,. P"teIOf,. M.mceij CLASS I, DIY. l, BARRIE" GRou,.$

AGeNCY MANU'ACtURER 8MUUER MOOn D

c 0

ll'1 l=U F<;lJ'oo'o 2AI-IZV.tG6, QAI.I;lV*f;C9 X

It, y

F2 FM Taylor 1148'134,1%451'44 X

X 1.

124$931.12'8931

)(

X 1l4S12~. JZ4lS1264 x:

J(

FJ F,..

We1\\!i1l9'1I')l.I,e 7&5801 X

X

)(

5l'lfCl~

X X

f'4 f~

L~" & NC)l'tt1rvp 318569,3t6141 X

X

)(

F!

FM FJ$C,e, & Pt>rklr 8MHOZ3\\J01. J.JOSH017UOl X

(

ao5H()~1IJ02

)(

X

'(

HI FM FJsbet Cotltrols AC302 X

)(

F1 Ff,t

~mJyw.1f 38S4S*l(XXX,OHO

,03-10$8$

X X

)t

  • It I/H2*FS85 X

~

t f

'1l$10P 4

C 12 I.M.M8...

COMpt.E11l0 DESIGN SS'(ClflCAT'ON STANDARD ACCESSORIeS All Models are ship-ped with flange adapter, veol/drain valves and one instruction manual per shipment.

OPTIONAL THRee-VALVE MANIFOLDS (Packaged Separatefy)

Port No.

11S1~15G-1: 3Nalvo M4nlfold. Carbon Steel (Anderson. Greenwood & Co" M4AVC)

Part No.

1151-150-2: 3-Valv, Manifold.

316S8 (Anderson. Greenwood & Co" M4AVS) 11 TAGGING ALPHAlINE OiHerential Pressure To;tns--

mitters will be lagged in accordance with customer requh-ements. All tags are stainless steel CALIBRATION Transmitter'S are faCtory calibrated to customer's specified span. ff calibration is not specified. transmitters are caHbrat&d at maximum range.. Calibraton is at ambient temperature and pressure Page 87 of 93

Reactor Core Thermal Power Uncertainty 2 Bailey Signal Resistor Unit Type 766 (2 pages)

Tau OF CQNI'Bi9Ts 13733 VOLLM IV Pressure Transmitter Power Supply Conductivity Alonent Trip Calibrstios Unit Relay Switch Recorder Switch Prsssuro Switch Recorder Pressure Transmitter Signal Resistor Unit Inverter Invertor ruse-MS-125-6o Page 88 of 93 LE-0113 Revision 0 Ros"Ount model 1

l1s1Dr tmesosal Electric Models 3

OVUM, M6"88 a" 9T66T939 Balsbaugh Xode1 3

OVI-2-N/ 910. IT-IBN Rosomouat model 4

$1009-2 General Sloo trio 3'

mode 1 MA General Electric 6

model 192940 "estrosies 7

model USE Osuoral alsatria a

models SB-1. SS-0, SO-10 wad U Basksdale Model 9

BIT-M1288-BB Loeda wad Northrop 10 Speed*w: model M Rosemount Model il 11S10P Alphaline Bailey lype 766 12 Topaz $000 Sorios 13 Topaz model 14 Exelon.

Reactor Core Thermal Power Uncertainty 2 Bailey Signal Resistor Unit Type 766 (2 pages)

LE-0113 Revision 0 G11-1513S VOLUB IV TA8LI OP CONTENTS SQQIPgNt '9'PSATIIII DtI!OIJllmlL NVMIII 1M Pt***ar. Tr....itt.r 101..-0_'.....1 1

1151D' Powu hpplJ' O***~.l l1.otrio 104.1.

1 n"n", mm** aU J,"u,a, Coad.othit, 81....t

.ahba.p 1104.1 J

GYl-2"N/'10.IT-IBN Trip Cal1bratioa o.it Io.eaout.4.1

.of SI001J-2 I.la)'

O***ral BI.otria

.4.1..

SWUok G***r.l II.etrie 6

Mo.d CI2940 1.001".1'

    • ,uoDl0' 7

Model...,.

awito.

G***ral Bl.otrio Mod... S8-1. S..."

Sa-10

&A' SIll Pnuv.,kito.

8aR,4... 1104.1 BI'P-112SHB I.cor'.r 1,**4..ael Nort..r...,

10 Sp**4oaas Mod.l

  • Pt***aro Tr....itt.r 1o**lHuc. 1104.1 11 11$10' Alpha1i**

Si,.al I **i.tor Vait BaU., Tnt-166 12 Ia.nt.r Topas $000 S.ri**

13 Ia*.,t.r Topas JIoelel 14 N'Uo-Gft5-1251O Page 88 of 93

r Reactor Core Thermal Revision 0 Power Uncertainty Pap 23 RENZWAL PARTS 4576KI6-00I Page 89 of 93 WIRE-Wmm, 5 WATTS, (V64655, TYPE g95)

FIG.

Son N4.

PART NO.

RST.

DES.

SCRIPTION RESISTOR, 250 0101S, -0.17. FIXED, WIRE-WOUND, 2.5 WATTS (V91637, TYPE

"`"W-1C) (Y00213, TYPE 12005) (V12~4-0 UNITS PLRIASSY 12 4

6146K23P280 13 USISTOR, 242 OHMS,

0.1% FIXED,,

6 wlag-WOUND, 2. 5 WATTS (V91637, TYPE RS-2C) (V00213, TYPE 12048) (V12463, TYPE T-2C) 4 6148K"P013 R4 RESISTOR, 8 015, {0.1% FIXED, WIRE-WOUND, 0.66 WATT, (V17870, TYPE 11375)

S 614902P303 R1 RESISTOR, 12100

, -0.1% l1XED 6

WIRE:-WOUND, 017970, TYPE R1391-0030) ; 0.25 WATT S

6148U0P217 12 RESISM, 400 OHMS, 0.1% F171ED, 6

WIRE-WOUND, 0.66 WATT, (V17870, TYP! 1375) 6 6146K15P226 13 R11SIMR, 100 OHMS, tO.17. FIM, 12 WIRE-WOUND, 2.5 WATT, (V91637, TYP9 R3-2C) (VO0213, TYPE 12008) 012463, TYPE T-2C) 7 6146K13P271 R3 RESISTOR, 96.8 OHMS, 10.1% FIXED, 6

WME-WOM, 2.5 WATTS 7

6148K68"22 RO RESISTOR, 3.2 a1PlS, W.17. FDM, 6

WIRE-

, 0.66 WATT (VL7870, TYPE R1375 8

6146K15P277 R3 RSISTOR, 250 0WS, 10.1% FDCBD, 6

WIRE-WOUM, 2.5 WATTS, (V91637, TYPE RS-2C) (V00213, TYPE UOOS) (V12463, TYPE T-2c) 6146KISP280 R3 RESISTOR, 242 OHMS, 10.1% FDMD, 3

WIU-WOUND, 2.5 WATTS (V91637, TYPE RS-2C) (VO0213, TYPE 12003) (VI2463, TYP% T-2C) 6148K68POIS R4 RESISTOR, 8 OHM, 10.1% FDMD, 3

WIRE-SOUND, 0.6b WATT (VI7870, TYPE 11375) 6146KISP226 R1 RESISTOR, 100 0M, :0.X7. 1'17180, 6

WIRE-WOUND, 2.3 WATT 091637, TYPE RS-20 (V00213, TYPE 12046) 012463, TYPE T-20 11 6146X25P271 R1 RESISTOR, 96.8 OHNS, W.1'I FIXED, 3

WIRE-WOUND, 2.5 WATTS, 091637, TYPE RS-2C) (VO0213, TYPE 12005) (V12463, TYPE T-2C) 11 6148K6$P522 R2 RZSISTOR, 3.2 OHMS, 10.1% FIXED, 3

WM-WOjM, 0.66 WATT, (V17870, TYPE 81375) 614900007 Rl RESISTOR, 150 OWE, t5% FIXED, 6

Reactor Core Thermal Power Uncertainty LE*0113 Revision 0 61461:15'280 13 IISISTOR. 242 OHHS.

1.O.1"Z. 'IUD...

VlII-VOUllD, 2. S WATTS (V91637, TYPE as*2C) (VOO213. TYPE 1200s) (V12463,

T'Y1'I T-2C) 614a68tol' 14 RUU'tOK. a OHMS, lO.I'%. PIXED.

WIII-WOUHD, 0.66 WATT, (V17810 t

'1"tPI 1137S) 614902'303 11 llISlS'rOR, 12100 OIIHS,.to. It. 'IXED WIRI-wouMD, (V17810. TYPE 11391-0030).. 0.25 WATT 6148X60'217 U

USIS'rOIl. 400 OHMS. O.l%. rIXID, VIII-WOUND. 0.66 WAtt, (V11810, nPI 1315) 6146115'226 1.3 ltISlS'tOR. 100 O1IfS,,to. 11.. FIXED, WIU-wouNO, 2.S WAtt, (V91631, TYPI 1S-2C)

(VOO213, TYPE 12008)

(V12463, n1'I T-2C) 6146Kl,P211 R3 RESISTOR, 96.8 OHMS,

~O.l~ FIXEn, wxu-womm, 2*.5 WATTS 6148Jt68"22 14 HlStsTOl. 3.2 OHMS, to. 14 rlXlD, lim-WOUND. 0.66 WATT (V17870, TYPE 11375 6146K1SP277 13 IBSISTOR.

250 OHMS,

'0.11. FIXED.

WIR£-WOUND, l.S WATTS. (V91631. TYPE as-2C) (VOO213, tYPE l~OOS)

(V12463.

TYPI T-2C) 6146lC1SP280

&1 RESISTOR. 242 OHMS. to.l't FIXED.

WIU-wouND, 2. S WATTS (V91637, TYPI as-2Cl (VOO213, TYPI 1200s)

(V12463.

nn T-2C) 614U68POt5 14 JtlSIst'Ol. 8 O'HMS, to.l~ FIXED, wm-wGURD, 0.66 WAtt (vt1870. TYPE ttl31S) 6146ltlS'226 11 USUTOR. tOO OHMS, 'to. 17. rmD, WlU...wouND, 2.S WATT (V91631, TYPE U-2C)

(VOO213, TYPE 1200$) (V12463, mE T-2C) 6146ltlSP211 11 RESISTOR, 96.8 OIN, to. IX FIXED, WIU-wouND, 2.5 WAttS, (V91637, TYPE as-2C)

(VOO213, TYPE 1200s) (V12463, TYPB T-ZC) 6148K68P522 12 RESISTOR, 3.2 08HS,

~O.l~ FIlED, WlIE-WOUND, 0.66 WATt, (V11870, TYPE R137S) 6149K94POO7 1t1 RlSIStoI.. 150 ORMS, tS~ FIXED, W1llI-WOUttD. S WAtTS, (V446SS. tyPE 995) 614611"271

3 6

3 6

6 6

6 6

6

.3 6

3 12 3

6 12 PaseU 4576K16-001 Uf.

DES.

P.&aT NO.

3 5

4 4

6 7

8 11 11 rIO.,

DlDIX 50.

Page 89 of 93

Exelo 3 Rosemount Specification Drawing 011532734, N0039 Option - Combination N0016 & N0037 (2 Pages) 110039 I.

SCOPE Reactor Core Thermal Power Uncertainty LTR ae ft* Na APwa OA?$

A

_Original Release 626853 This specification defines a Model 1153 Series B pressure transmitter with combined options N0016 - a stainless steel 1/2 - 14 MPT pips plug installed in one of the conduit hubs, and MQ037 - a 4-20mA output signal with adjustable damping.

II.

DETAILS 1.

The pipe plug is to be assembled on the nameplate/vent valve side of the transmitter.

The plug will be sealed with thread sealant and torqued to 150 in.-Ibs.

2.

The standard "R" calibration and amplifier boards are replaced with the special damping calibration and amplifier boards.

III.

SPECIFICATIONS Maximum damping for the electronics, measured at the 63% time constant is :

IV.

APPLICABILITY AND APPROVALS Maximum Damping Range 3 not applicable Range 4 1.2 seconds Range 5-9 0.8 seconds The damping electronics are not intended for the range 3 because the slower response is not required for this transmitter pressure range.

This specification is limited to the Model 1153 Series B with "R` electronics.

Qualification with the pipe plug was addressed in Rosemount Report 108026 (see paragraph 5.3.1 )

_Qualification of the damping option was addressed in Class IE USAGE

~Da "Y

. Hildebrandt APP MASTER DRA t1'-'

DAT! TM 11/16/8 Page 90 of 93 Specification Drawing N0039 Option - Combination N0016_ & N0037 Coon 10rr. na 04274 MINNUPDLM MINNOIOTA DAAWWO MM 01153-2734 1

' 1O' 2 ram, No. SOMI, RW A LE-0113 Revision 0 Exelon.

Reactor Core Thermal Power Uncertainty LE-0113 Revision 0 3 Rosemount Specification Drawing 01153~2734f N0039 Option - Combination NOO16 & N0037 (2 Pages)

H0039 LTA DATI A

Original Release 626853 Range 3 Range 4 Range 5-9 I. ~

This specification defines a Model 1153 Series 8 pressure transmftter with combined options N0016 - a stainless steel 1/2 - 14 HPT pipe plug installed in one of the conduit hubs. and N0037 - a 4-20mA output signal with adjustable d**ping.

II.

DETAILS 1.

The pfpe plug fs to be assembled on the nameplate/vent valve side of the transmitter.

The plug will be sealed with thread sealant and torqued to 150 fn.-lbs.

2.

The standard "R M calibration and llIp] ifier boards are replaced with the special damping calibration and amplifier boards.

111.

SPECIFICATIONS Maximum damping for the electronics. measured at the 631 time constant is:

Maximum Damping not app11cab1e 1.2 seconds 0.8 seconds The damping electronics are not intended for the range 3 because the slower response is not required for this transmitter pressure range.

IV.

APPLICABILITY AND APPROVALS This specification is limited to the Model 1153 Series B with "R-electronics.

Qualification with the pipe plug was addressed 1n Rosemount Report 108026 (see paragraph 5.3.1).

Qualification of the damping option was addressed in CLASS IE USAGE Rosemount 1M.

MIHNIAPOLII. MINHQOTA OA.IV

~. Hildebrandt CJ DAft 11/16/8 111UI Specification Drawing HOO39 Option.. COIIbfnat1on "0016 &"0037 8IZI! ~ IDENT. NO DRAWING HC.

A 04274 01153-2734 ISHUT lOP 2

f'orm No. eo2t8-1, Rev. Ii Page 90 of 93

Exelun.

MASTER DRAW]

CLASS IE USAGE Reactor Core Thermal Power Uncertainty Rosemount Report 08800053. The damping option is qualified to the levels of the 1154 transmitter.

However. when being used in the Model 1153 transmitter, the qualified requirements are those for the original model 1153 transmitter.

They are not altered because of the presence of the damping electronics.

Page 91 of 93 04274 o

NO.

01153-2734 I per 2 of 2 Foen No. 629!-2. Rw. a LE-0113 Revision 0 Exelen.

Reactor Core Thermal Power Uncertainty LE-0113 Revision 0 Rosemount Report 08800053.

The damping option is qualified to the levels of the 1154 trans.ftter.

However. when being used in the Model 1153 transmitter, the qualiffed requirements are those for the original Hodel 1153 transmitter.

They are not altered because of the presence of the damping electronics.

01153-2734 A 04274 MASTER ORAWING

....--------r:=-r.:===:-:r'==::2":;;-----.

ICLASS IE USAGE Page 91 of 93

Reactor Core Thermal Power Uncertainty 4 Rosemount Product Data Sheet 00813-0100-2655, Rev. AA June 1999 "N-Options for Use with the Model 1153 & 1154 AlphalineO Nuclear Pressure Transmitters" (2 Pages)

-Options for U with the Model 1153 and Model 11 Alphalin a Nuclear f

ss Tans tt Page 92 of 93 rs R

Qu T'mismo g rim mom aww~*

00813-O s W2655 En ash June 1999 ReV, AA LE-0113 Revision 0 Exelon.

Reactor Core Thermal Power Uncertainty LE-0113 Revision 0 4 Rosemount Product Data Sheet 00813-0100-2655, Rev. AA June 1999 4N-Options for Use with the Model 1153 & 1154 Alphaline Nuclear Pressure Transmitters l

' (2 Pages) oo813..Q101).2655 EngflSh June 1999 RevAA N-Options for Use with the Model 1153 and Model 1154 Alphaline(O Nuclear Pressure Transmitters Page 92 of 93

Exel6n.

INTRODUCTION Ro,seniount`a Model 115:3 and Model 1154 Alphaline* Nuclear Pressure Tratasinitt<-r3 are designed for precise pressure measurements in nuclear applications which require reliable performance tuid safety over an extended service life. These transtuitters have been qualified per IEEE Std 3:23-1974 and IEEE Std 344-1975 as documented in the corresponding Rosemount qualification reports.

Model 1153 and Model 1154 Tnarismitiers atv available in a variety of configurations for differential, flow, gage, absolute. and level measurements. To acconimodate speck customer requirements, special N-Option features trove been developed to provide greater application flexibility.

For example, the N0010 option allows a transmitter to be calibrated up to 5 T over its standaixl tipper range limit. The N002t3 option allows xa Range Code 4 Transmitter to be calibrated up to 2101nHa0 rather than the standard Range Code 4 upper range limit of 150 in"2O.

Following is a summary of selected N.-.Options. For additional infornintion on these and other N-Options, contact Rosettwunt Nuclear Instruments. Inc.

SUMMARY

OF N--OPTIONS N0002 Specifies to reverse,acting gage pressure transmitter.

N0004 Specifies factory calibration of the transmitter at temperatures above or below ivom temperature. Transmitters may be calibrated at tenaperatures between 40 and 200 'f N0010 Allows the transmitter to be calibrated up to 5`h abo8v the standard upper range limit. For example, if the stated tipper range litn t of the transmitter is 1,000 psi, tan N0010 transinitter may be calibrated up to 1,050 psi. This caption is available on all ranges.

N0011 Allows 180' rotation of the electronics housing.

N001G Specifies it stainless steel pipe plug installed on the nameplatelvent valve side of the 1153 Series li Transmitter.

Retemmnt "Wear (astranemt, Inc.

1204* ramr y Dave Even N *I#, tit 88344 TO (612) 828^8252 Tax 4314012 Fax (612) 828-8280 0199'9 Rosem Nuc" hUuwnt5. Inc.

f&ttrllWYrt.rosetllowtcm 111111111111111111111111811111 00813-0100-2ti55 Rev. AA Reactor Core Thermal Power Uncertainty N0018 Speritles a inaximuin static pressure rating of 3,200 psi rather than the standard 3,000 psi on any high-line differential pressure transmitter.

N0022 Specifies a welded 'l4~in. Swagelok1.

Compression fitting on differential and high-line transmitters.

N0026 Allows tin 1153 Series D or 1154 Range Cede 4 transmitter to be calibrated up to

210 inH,2O rather than the standard Range Code 4 dipper range limit of 1

.i0 inll :>O. The maximum mid minimum span limits are 155 and 75 inH20, respectively.

N0029 Specifies factory calibration of the transmitter at a customer-specifieJ elevated line pressure.

N0032 Specifies.a Range Code 3 or 8 differential pressure transmitter with a maxiintim static pressure rating of 2,500 psi rather than the standard 2,000 psi. Applicable to Model 1153 Series Transmitters only.

Nt1O:13 Allows 90°clockwise rotation of the electronics housing. The terminal block lines up with the vent/drain valve side.

N0034 Specifies a Model 1153 Series D or Model 1154 Transmitter with a special mounting bracket that bas no panel mounting holes.

N0037 Specifies adjustable damping on any Model 11.5:3 or Model 1154 Transmitter N0077 Specifies a Model 1153 Series F with a SST electronics housing, SST housing covers and SST mounting bucket.

N0078 Allows 1130° rotation of the process flanges N0088 Allows 90* counterclockwise rot..ytion of the electronics housing The terminal block lines u p with the process connections.

ORDERING INFORMATION Consult the appropriate transmitter Product Data Sheet for a transmitter model number, Append the N-Option number to the end of the transmitter model number. An exrtniple of a typical model number with N-Option added is 1153DI35R.AN0010.

ROStMOtlrlt l" P03eMOU!)t logo, and Alp4abne are.,egratered rr.7dema.=xs of

+9asernoctat too 5,va0etax i$ a re9ratered "IEM.ani a'Craw+ord Fi40g Co.

Afay oe ;toteoted oy oae or ercave *fare V ou fng tl. :, pa!egt tVO.,

3.6;6,536. 3,793.886 3,800,4 13; 3,271;7 f9, R!

. 30.603. Ca,aoa0sented (8fe~ett~ :9Ta,f4TS_tSiT6, a-V 1979 yrtaend0a madei. Omit taoiga Men1S rSSU+ed aaa peldrflg CovePAofo 1753-00rAS Fisher-Rosenounl satisfies aft Obfgattvns "Mmg from fegtStatioo to harmonize trroduct requfWnnir, iq the European Vr4on 1.E-0113 Revision 0 Exelon.

Reactor Core Thermal Power Uncertainty LE-0113 Revision 0 Specifies a maximum stntic pressure rating of3,200 psi rather than the stundard 3,000 psi on any high*Jint' differential pressure trallSmitt.er.

Specifies a welded 114-in. $wngefok'U compl'ession fitting on differentinl and high-Une transmitters.

Allows no 1153 Sel'ies D OJ' 1154 Range Code 4 transmitter to be calibt'atcd up to 210 inH20 rather than the standard Range Code 4 upper l-ange limit of 150 inH,)O. The maximum and minimum spa';: limits llI'e 155 nnd 75 inHzO, respectively.

Spedfleg factory calibration of t.he transmitter at n customer-specified elevated line pressure, S~cifi~a R.'mge Code 3 or 8 differelltinl pressure traOlunitter wlth a maximum glatic pressure rating of 2,500 I)si rather than the standard 2.000 psi. Applicnble to Model 1153 Series Transmitters only.

AIIow6 90<> clockwise rotation of the electronics housing. The terminal block lilll:'S up with the vl'ntJdrain vllive side, Spedt1e.s a Model 1153 Series D Of Model 1154 Transmitter with a special mounting bracket that has no p;mel mounting holes, Specifies adjustable damping on any l\\lodel tIna or Model 1154 Transmitter_

Specifies a Model 1153 Series F with a SST electl'Onics housing, SST housing covers and SST mounting bl'8cket.

AUow$ 180" rot.'1tlon ofthe process flanges Allows 90"' counterclockwise rot.1tion of the electronics housing, The termInal block lines up with the prGee68 connections.

N0029 NOO18 N0022 N0026 NOO33 NOO32 NOO34 N0077 NOO:17 NOO78 NOOS8 Covet p/lofl)" 1153,00 tAS ORDERING INFORMATION ConsuH. the appropriate transmitter Product Data Sheet for a transmitter model numbelt Append the N-Option number to the end of the t1'ansmittet*

modelllumber. An example of a typical model number with N-Opt.ion added )$ 1153DB5RANOOlO.

ROH/l'lOvflt, lIM' F?osemoullf folIO, :lfl(J AI{)ttOtiM (Uf! regltJ.tHed triJdemi1tlu;: 1)1 RQ$emo<Jflt!IJC SI\\'~k:J/( i$ ~ Itf~i~Wtdlr.J<femiJrW. ofCriJlI-1on1 Filfing (:0, Mity oe ;Jt~crt!(1 t>y!)fIt-QI mQff! Mthe !\\:$owittg tl $, PM.'!t Ma 3.646,$J':J,1~3,8$5,J,800,H3; J,g75]19; R~" 30,403, Ci}'1a(1oJp;',fllt,'d

(~'1f!t~) :974 1975.1976,.nd 197'1 44y de;;en<t on 1I'l(I(!I!( Otl'lff 1"~lg" p;ttems!SSUe<t ~fld Pl!'fIdl!ll)

Specifif>S (l t*everse.acting gnge pressure tl*tmsmitter,.

SpE'cifi(>$ factory cnHbrntion of the transmitter at temperatures abovf: or below room temperature. Transmitters may be calibrnted nt tempernhlrel3 bet.we~n 40 and 200 'iF.

Allows the transmitter to be ca1ibr~ted up to S'-:l above the stnndnrd upper range limit, For exnmple, if the stated upper runge limit of th~ tranamitter is 1.000 p~i, nn NOO10 transmitter may be

('o}ibrated up to 1,050 psi. 'nlis option is available on aII l'noges.

Allows 180" rotntion. ofthe electronics housing.

Specifies n stniolens steel pipe plug instnllcd Oil the namepl~llelventvalve side of the 115a Series B Trnu$mitter.

SUMMARY

OF N-OPTJONS NOOO2 NOOO4 NOOlO NOOll INTRODUCTION Rosemountlfl Model 1153 and Model 1154 AlphaJine* N Ud(~ltr PI'e's,sure Transmittt*t"j ure designed for pn:.-ciac pressure me-U5urements in nuclear IIp()lkutions which require reliable perfol'nmncc lUld &:tfety over an extended selvice life. These transmitters 00'-8 been qU<llified pOl"

{EEE Std :J:23-1974 tmd £EEE Std 344-t975 t'S documented in the corroepoftd.ing Rosernount qualification reports, Model 1153 and Mod'1 1154 'rnan~mittel'S ate avnilnble in n va rieLy of coDft.gumtkms fQr differential. flow, gage, absolute. and level meO$urements. To accommodate spl'Clfic customer requirements. special N-Option featnres have heen developed to pl'Ovide grenter appliention flexibilit.y.

For example, the NOOIO option allows a t.-ansmiUer to be calibrated up to 5% over its standttJ'tl upper rtlUge limit. The NOO26opt.ion.ilUOWS n Range Code

.. Trrmstuitter to be ~llibrated up to 210 InHaO rather than the standard Runge Code 4 upper ronge limit of 150 inH20.

Following itS a summary ofsefectt"d N**Optious. For additional information 011 thf'...se nnd other N*~Options,contact Rosemount Nudeo.*

Instruments, Inc.

NOOIG RotemMlllI ","lear IDSltlll1leRlt, IIIC.

12001 T$C1l/lOlOlJYOI1\\'9

£dfn Ptaitlt. MN 5534'-

T~ (612) 828-8252 Tm4310012 Fax (612) t.:1Wm Q 1m RosemOlJllt Nuctt.3r lmttWlltnlS" Inc.

1lI11l')J'i'i\\\\'ljJ~lcom mllllllllllllllMlllllllllllll1 00813-0100-2655 Rev. AA CE:

Fisher*Rosemount sati$~$,111 ob!rgJII\\:ms coming irom legislatioo 10 harmon';;:e ",roduet req;;ifemeOI$,1) the E..,rope:m linion Page 93 of 93