ML082890206

From kanterella
Jump to navigation Jump to search

TVA Slides - EPU - Summary of October 14, 2008, Meeting with Tennessee Valley Authority Regarding Steam Dryer Portion of the Extended Power Uprate Review
ML082890206
Person / Time
Site: Browns Ferry  Tennessee Valley Authority icon.png
Issue date: 10/14/2008
From:
Tennessee Valley Authority
To:
Division of Operating Reactor Licensing
Brown, E, NRR/DORL, 415-2315
Shared Package
ML083030525 List:
References
TAC MD5262, TAC MD5263
Download: ML082890206 (25)
v • d • e


Text

I H A I TENNESSEE VALLEY AUTHORITY BROWNS FERRY NUCLEAR PLANT ExtendedPowe rate

-steam Dryers--

-7j]..

October 14, 2008

Agenda

  • Status of Unit 1 and 2 Dryer Analyses

" Decision on Acoustic Side Branches

" Plan to Address SRV Resonance

  • Changes in EIC Removal Method
  • Unit 2 Noise Removal

" Submodeling Questions

  • Review of RAI 19, 20 and 21 Responses
  • Schedule 2

Status of Unit 1 and 2 Dryer Analyses *

" TVA Decided not to Install Acoustic Side Branches (ASB)

No clear advantage

" Unit 1 and 2 Stress Reports (June 2008) Need to be Revised SR-a > 2.7 at CLTP Evaluates CLTP only Unit 2 anomalous low flow (LF) signal (19% power)

Newer strain gage data now available

" Additional Strain Gage Data Unit 1 startup August 2008 Unit 2 startup September 2008

  • Unit 1 Stress Report Being Finalized

" Unit 2 Stress Report in Progress 3

Decision on Acoustic Side Branches

  • 24-inch Quad Cities Design Chosen Governed by clearance limitations
  • Confirmation of ASB design by 1/8 Scale Model Test (SMT)

Damping effect less than expected SRV resonance still present

  • Requires Stress Analysis to Address EPU Bump-up factor 4

Decision on Acoustic Side Branches [

BFNI With ASBs 2 5 S 20 ".....MSL A Upper 20 ------------------------------------

_ M SL A Low er

_MSL B Upper MSL B Lower 15----------

MSL C Upper MSIS (1 Lower 10

- - - - ------ a 0

50 100 150 200 250 Frequency (Hz)

BFN2 With ASBs -

MSL A Upper 7

r-MSL A Lower MSL B Upper 6 _

i ------- MSL B Lower 0

MSL C Upper

-NISL C Lower 4

--I --

--e 0

0 50 100 150 200 250 Frequency (Hz) 5

IRA Plan to Address SRV Resonance

  • 1/8 SMT Performed for each Unit's Configuration Data at each strain gage location Data at CLTP and EPU Mach numbers
  • Bump-up Factors Calculated as a Function of Frequency by Equation:

BF =

PSDEPu At each frequency PSDCLTP Applied to Plant CLTP Strain Gage Data to Predict EPU Load PCLTP = CCLTP(CL TP - EICcLTP) - CLF(LF - EICLF)

PEPU = BF[CCLTP(CLTP-EICCLTP) -

CF(LF-EICLF)]

P = Steam line unsteadypressure BF = Bump-up factor for SG location C = Coherencefactor between upper and lower locations EIC = Signal taken wilh zero excitation voltage LF = Low flow signal 6

Changes in EIC Removal Method SEIC Signal Taken by Removing Strain Gage Excitation Voltage

  • Electrical Noise is Removed by Using EIC signal Mechanical Component = SG Signal - EIC
  • Additional EIC Signals on Units 1 & 2 EIC now Matched with Companion CLTP and LF Signals PCLTP = CCLTP(CL TP - EICcLTP) - CL,(LF - EICLF)

P = Steam line unsteady pressure C = Coherence factor between upper and lower locations EIC = Signal taken with zero excitation voltage LF = Low flow signal 7

Changes in EIC Removal Method EIC Signals JiV BFNI A Upper EIC

  • C @ lD%,Pm.d FO (2M)AU 0.01 0.0001 0

50 100 150 200 250 Frqpency (Hz)

BFNI B Upper EIC Ia 0

50 100 100 200 250 Fmqumney (Hr)

BFNI B Lower EIC BFNI A Lower EIC 0.1 0.01 I

01 a

0001I 00001 000001I 0000001I

--UtE @86%Povmrl4*zWIOrO)AL U*E. N %Po-4UdZW.D (5M) 8L 0.1I tfEZ:QeflflPov.m*

FHzWO0W(1Bt UECO WIk*O a01*Hz1W0{

2U) BL 0.01 0.001

..0001 wV-.

0 50 100 150 200 250 F.qmency (Hz) ffi2_LL 0

50 100 150 200 250 Frmoncy (Hz) 8

Changes in EIC Removal Method EIC Signals iRA BFNI C Upper EIC 0-1

-WQet~floVSJUFiZ*WD (TW)OU

-- U1E8 @o0Po¶.wtvo (2l8,cu 0.01 0001 0

0.0001 00l0001 0

00 100 150 200 250 F-q-n..y (H.)

BFNI D Upper EIC 0.1

-- U~lECg ewlSS t4HPz WO(736) DU 0.01 0.001 e 0.0001 0.00000_

0 50 100 150 200 250 BFN1 D Lower EIC BFNI C Lower EIC I

a

-- UIEIC

@6%P01a11t'4 H4 2W0 (1161 C

0.001 0.501

[

0001 0, 0001 i

i ____i i

I a

0 s0 100 150 200 250 50 100 150 200 250 Fmquene (Hz) 9

Changes in EIC Removal Method EIC Signals IM BFN2 A Upper EIC 0.1 0.01 0.001 0.00001 00.001 BFN2 B Upper EIC 0.1

-WEC@5%P

ýw/HzW(4018U

-WECG$3%POýnHZWO( 2)aU

-WEC@WPOd4HZW(49)BU 0.001-S 0.00001 0.~0l00 0

so 100 ISO 200 250 Fmq-.ney (Hz)

FBA quency (H.)

BFN2 A Lower EIC 0.t 001 EEG TO %Poý4536Hz'VFO (228q 0.001 0*00001-0,000001 0

50 100 10S 200 250 Frequency (Hz)

BFN2 B Lower EIC 0.1 0.01 1 0.001 ul 0.0001 0000 0.000001 0

00 100 150 200 250 Frequency (H1) 10

Changes in EIC Removal Method EIC Signals BFN2 C Upper EIC 0.1-0.01 0

0 1.0 1.020 210 BFN2 D Upper EIC 0.01 001 0.00001 100 150 200 250 Freqency (Hr)

BFN2 C Lower EIC 01

--L EC

@81%,Pov,fOe'6$HzWD (ie) c

-WEI@1,po.4S ztFD(4)O.

0.01 0.0001 0.00001 0

so 100 150 200 250 Fmq.mnmy (Hz) 11

Unit 2 Noise Removal

  • Additional Data Taken on Unit 2 to Confirm Signal Behavior Electrical noise on Unit 2 varies with recirculation pump speed (VFD frequency)

Relationship is not well understood 19% power signal originally used for noise removal was atypical Composite 19% & 30% power signals replaced

  • New LF signal at 5% Power and Companion 5% EIC Signal All strain gages -on MSL D lower damaged Substituting MSL A for MSL D due to strain gage failures CLTP signal and companion EIC signal unchanged 12

Unit 2 Noise Removal PSD Signals BFN2 A UPoer BFN2 B Uppe 1.0E-01 1.0E-01 14 0

50 100 150 200 250 Fre.n1y (.z) 50 100 Fq y(z 150 200 250 BFN2 A Lo BFN2 B Lower F1.00E-03 1*-

I.

1 0-04 I.0E-05 1.0E-01 1.0E-02 1.0E-03 01.00-04 1.00-00 1.00-06 50 100 150 200 250 Freqency (Hz)

Froequcy (Hz) 13

Unit 2 Noise Removal PSD Signals BFN2 C LpW BFN2 D Upep 0

1.00E01 10E-02 1.00-03 (0.00-04 1.00-06 50 100 150 200 250 Freq-y (Hz)

BFN2 D tow BFN2 C Loý 0

1.0E-01 1.0E-02 1.0E-03 l.E-05 I.00-00 250 so 100 150 200 250 Frequey (Hz)

FM.weoy (Ftz) 14

Submodeling Questions Is the Stress Reduction Factor (SRF) accurate and unique?

Would a different analyst get the same solution?

Limited, Specific Purpose o Avoid excess conservatism of shell model o Based on mechanistic behavior along weld line CDI Shell Model => SIA Shell Submodel o Characteristic load matches CDI stress along the weld line Drain Channel-to-skirt: Bending thru the joint - See Figure 1 Hood Stiffener-to-Hood: Membrane in stiffener - See Figure 2 SIA Shell Submodel => SIA Solid Submodel o Incorporates weld geometry o Applies characteristic loads o Accurately captures load transfer mechanism and stress distribution through weld Submodel attributes (loads & boundary conditions) are not unique, but SRF is unique & accurate. So a different analyst would get the same result.

15

Submodeling Questions IRAI Submerged Skirt - Figure 1 Node 98156, Skirt 1000 800 CL C

U) 4-600 400 200 0

50 50.1 50.2 50.3 50.4 50.5 Time [ s ]

16

Submodeling Questions Inner Hood Stiffener - Figure 2 Node 104843, Hood Support 5000

[

4 0 0 0 3000--

CD 2 0 0 0o o.....

.-U)

-.-Surface Ci 1 0 0 0

-m id d le bottom 0)0 50 50.1 50.2 50.3 50.4 50.5 Time [s ]

17

Submodeling Questions I--

ft-A-1

  • Are the submodel loads statically equivalent to the CDI model?

No - not statically equivalent, nor required Limited objective is to capture stress along weld line Simple Example:

REAL BEAM EQUIVALENT BEAM Objective: Design Connecting Weld (For FEA Model)

For Real Beam Using FEA M

,=--

M=

k-ft I

k

,M 10k-ft 10k+

~ILZ Z 0M = 10 k-ft

-21 18

Submodeling Questions Are the times used the ones which yield the largest stress intensity after application of the SRF?

Refer again to Figures 1 & 2 Alternating stress defined by either membrane or bending extrema Extrema states produce maximum strain (i.e., fatigue usage)

SRF should be based on the extrema stress state At other points in time, the product of stress intensity and SRF would have a lower value; i.e., be less conservative 19

Submodeling Questions Demonstrate that the uncertainty in calculating the SRF is small Approach produces high certainty that bounding stress of weld line is captured Solid submodel mesh sensitivity study demonstrated convergence Weld factor of 1.8 retained Low level of uncertainty subsumed by bias and uncertainty applied to overall process 20

Review of RAI 19, 20 and 21 Responses I-RA-1

  • RAI 19 EMCB.147 (Unit 2 only) o New Unit 2 stress analysis o Revised response based on revised analysis EMCB.192/150 o SRV Resonance EMCB.181 Follow-up o 0- 2 Hz mean filter EMCB.182 Follow-up 0 EIC removal (Unit 1 only)

(Unit 1 only)

EMCB.183 Follow-up (Unit 1 only) 0 SR-P values in table 21

Review of RAI 19, 20 and 21 Responses ITVA_

" RAI 19 (continued)

EMCB.181 (Unit 1) & EMCB.147 (Unit 2) Follow-up 0 PSD plot filtering EMCB.186 & EMCB.187 Follow-up (Unit 1 only) o Sub Modeling

  • RAI 20 EMCB. 194 (Unit 1 only) o 9% signal coherence EMCB.195 (Unit 1 only) 0 Fan noise 22

Review of RAI 19Y20 and 21 Responses I-R-A-1 RAI 20 (continued)

EMCB. 196 (Unit 1 only) o EIC plots EMCB.197/153 o Strain gage penetration location EMCB.154 (Unit 2 only) o 9% signal coherence EMCB.195 0 Extended frequency plots - VFD 23

Review of RAI 19 20 and 21 Responses

  • RAI 21 EMCB.198 (Unit 1 only) o EIC removal EMCB.155 (Unit 2 only) 0 EIC removal 24

I Schedule Item Date TVA response to RAI 21 on Channel Bow 10/17/08 TVA submit Unit 1 stress analysis & Unit 2 status 10/31/08 TVA submit Unit 2 stress analysis 11/14/08 Tentative ACRS meetings 2/09-3/09 Unit 2 outage begins 4/09 NRC issue EPU Amendment for Units 1, 2, and 3 4/09 Unit 2 startup at EPU 5/09 Unit 1 implement EPU 6/09 Unit 3 implement EPU Spring 2010 25

Retrieved from "https://kanterella.com/w/index.php?title=ML082890206&oldid=5324683"