ML15222B159

From kanterella
Jump to navigation Jump to search
ENT00714A - H. Cothron Et Al., EPRI, Slides: Nrc/Industry Meeting Regarding Tube-to-Tubesheet Weld and Divider Plate Cracking Report (July 30, 2015)
ML15222B159
Person / Time
Site: Indian Point  Entergy icon.png
Issue date: 07/30/2015
From:
Entergy Nuclear Operations
To:
Atomic Safety and Licensing Board Panel
SECY RAS
References
RAS 28139, ASLBP 07-858-03-LR-BD01, 50-247-LR, 50-286-LR
Download: ML15222B159 (31)


Text

ENT00714A Submitted: August 10, 2015 t==~121 1 ELECTRIC POWER

~~- RESEARCH INSTITUTE NRC/Industry Meeting Regarding Tube-to-Tubesheet Weld and Divider Plate Cracking Report Helen Cothron, Manager EPRI Steam Generator Management Program Richard Smith, Sr. Associate Structural Integrity Associates Greg Kammerdeiner First Energy

© 2015 Electric Power Research Institute, Inc. All rights reserved .

Background

  • Primary Water Stress Corrosion Cracking has been reported in the weld between the divider plate and the stub runner in French and Swedish steam generators with Alloy 600 divider plate and Alloy 82 and 182 weld material
  • SGMP completed a multi-phase project in 2010 that concluded that these cracks were not safety significant Stub Runner (SR) Tubesheet (TS) 2

© 2015 Electric Power Research Institute, Inc. All rights reserved. E1

~f21 1 ELEC TRI C POWER RESEARCH INSTITUTE

Phase I Determined the Limiting Steam Generator

  • Phase I evaluated operating experience and determined limiting model steam generator

- Cracks observed in divider plate and stub runner weld heat affected zones in foreign fleet.

- Westinghouse Model 51 steam generators determined to be the limiting case for US plants.

- Model 51 is limiting case due to thinnest divider plate and greatest vertical displacements of the centerline of the tubesheet under normal and accident conditions.

- Compared well with foreign operating experience 3 t=~121

~~-

1 ELECTRIC POWER RESEARCH IN STITUTE

© 2015 Electric Power Research Institute, Inc. All rights reserved .

Phase II Assessed Transients, Stress Reports, and Repairs

  • Analysis of multiple crack geometries

- Multiple and combined origin sites

  • Review of ASME Code stress reports to determine -if they are affected by degraded divider plate condition

~~-

1 ELECTRIC POWER RESEARCH INSTITUTE

© 2015 Electric Power Research Institute, Inc. All rights reserved.

Background I Overview

  • The design analyses for the following components did not take credit for the divider plate:

- Tubesheet

- Channel head

- The lower shell

- Tubesheet to channel head junctions

- Tubesheet to lower shell junctions

- Tube-to-tubesheet welds 5 t=~121 1 ELECTRIC POWER

© 2015 Electric Power Research Institute, Inc. All rights reserved. ~~- RESEARCH INSTITUTE

Analyses Not Affected by a Degraded Divider Plate

  • Supporting analysis and boundary conditions for lower steam generator complex
  • The performance or safety function of the steam generator and the affected loop during a postulated accident condition
  • The supporting analysis basis for tube plugs installed prior to 1989
  • C* Alternate Repair Criteria
  • H* Alternate Repair Criteria
  • Alloy 800 sleeves 6 t==~~~ ~ ELECTRIC POWER

© 2015 Electric Power Research Institute, Inc. All rights reserved. '-=1-1~ RESEARCH INSTITUTE

Analyses Sensitive to a Degraded Divider Plate but Conservative

  • Divider plate factor of 76o/o of the vertical displacement was used in analyses for
  • Laser welded and TIG welded sleeves
  • F* Alternate Repair Criteria
  • W* Alternate Repair Criteria
  • Conservative because 76% is greater than the vertical displacement associated with a fully degraded divider plate (64°/o) 7

© 2015 Electric Power Research Institute, Inc. All rights reserved. e: ~f21 1 ELECTRI C POWER RE SEAR CH IN STITUTE

Phase II Conclusions

  • No identified need for US plants to inspect for PWSCC based on program results

- Degraded divider plate is not a safety concern during operations

- Degraded divider plate is not a structural concern during operations

- Degraded divider plate does not affect existing repair criteria or repair tools

- Divider plate inspection not mandated by ASME Code 8 t=~121

~~-

1 ELECTRIC POWER RESEARCH INSTITUTE

© 2015 Electric Power Research Institute, Inc. All rights reserved.

Divider Plate Crack Propagation

  • When US plants began submitting license renewal applications to the NRC, the potential crack propagation over extended period of operation became a concern
  • SGMP began a project in 2011 to address the concerns of cracks propagating over time to pressure boundary components such as the channel head or the tube-to-tubesheet weld 9

© 2015 Electric Power Research Institute, Inc. All rights reserved. E ~f21 1 ELECTRIC POWER RESEARCH INSTITUTE

Divider Plate to Channel Head/Tubesheet Weld Configurations Cladding (Alloy 82)

Cladding Tubesheet (with alloy 182 (low AHoy Steel*)

after Z-Seam Weld)

Weld (Z-Seam t

    • . 0 h

St~~ R~nn,~r

  • , ,*4 * * * ******

(Alloy 600) h* h **

Filler Plate Weldments (Alloy 600) (Alloy 82 and 182)

-o*

oG>

Partition Plate G>.! (Alloy 600)

J:(f)

G>o C=

C< Cladding 0~

I:.o (308L SS) u...J 10 t=~121 1 ELECTRIC POWER

© 2015 Electric Power Research Institute, Inc. All rights reserved. ~~- RESEARCH INSTITUTE

Divider Plate - Tubesheet Center Configuration Tubesheet (Low Alloy Steel*)

Cladding (Alloy 182)

Alloy 82/182

  • 182 used approximately 14" in the center of the tubesheet Stub Runner (Alloy 600)

Divider Plate Alloy 82/182 (Alloy 600) t=~f21 1 11

© 201 5 Electric Power Research Institute, Inc. All rights reserved. -=*- EL ECTRI C POWER RESEAR CH IN ST ITUTE

Divider Plate Crack Propagation Scenarios Tubes Tubesheet (Alloy 690 TI) (Low Alloy Steel*)

Cladding (Alloy 182)

~*.....................................................

  • * *. -~~~ =... -~ .::~; r-::..:*: .::f*~:i.~:- . ~~~ :.. -~.

. :: ;:-:.. ~.1!9Y.)32{1 .82:*.

Tube t o / . :*=. E::.:*:":_*./-\?~*::*:: . . : ~

Tubesheet Welds PWSCC Stub Runner Crocks (Alloy 600)

Divider Plate (Alloy 600)

Scenario 1 assumes cracks initiate in the divider plate assembly and propagate to the tube-to-tubesheet weld Scenario 2 assumes cracks initiate in the tube-to-tubesheet weld or the cladding 12 t=~121

~~-

1 ELECTRIC POWER RESEARCH IN STITUTE

© 2015 Electric Power Research Institute, Inc. All rights reserved .

Divider Plate Crack Propagation Scenarios Scenario 3 assumes cracks initiate in the divider plate assembly and propagate through the stub runner, the 182 weld at the triple point and through the low allow steel Tubesheet (Low Alloy Steel*)

Cladding (AIIoy82)

WSCC Cracks Stub Runner (Alloy 600) 13

© 2015 Electric Power Research Institute, Inc. All rights reserved. E ~f21 1 ELECTR IC POWER RESEARCH INSTITUTE

Scenarios 1 and 2 Concerning the Tube-to-Tubesheet Weld 14

© 2015 Electric Power Research Institute, Inc. All rights reserved. E1

~121 l ELECTRIC POWER RESEARCH INSTITUTE

Tube-to-Tubesheet Weld Cracking

  • Industry has accepted a 24o/o chromium content as a conservative threshold for PWSCC initiation 680°F, 20 sec/kg H2, K=4S ksi-in°*5, 221 days Rates adjusted to 100%engagement and no incubation cu Q) 14.7 wt.%Cr and 20 wt.%points are estimated )\

Chromium levels down to 20o/o *-E 10 1oo E have excellent resistance to initiation based on testing and - - E

...,Q) cu operational experience 10 0::

  • Objective of this project was to .c:

determine the chromium content in ~

an autogenous weld between Alloy CJ e

690 tubing and 82 or 182 weld D ~

material u Construct a tube-to-tubesheet ~

0.1 (J mockup to investigate 16 18 20 22 24 26 28 30 Bulk Chromium Content 15

© 201 5 Electric Power Research Institute, Inc. All rights reserved. e: ~~~ ~

I-I~

ELECTRIC POW ER RES EARCH IN STITUTE

Field Material Review Alloy 690 Statistic (product FM 82 FM 182 analysis)

Mean 29.3 20.7 14.4 Maximum 29.7 21.5 16.5 Minimum 29.0 19.6 13.6

  • Review of chromium content in tube and tubesheet cladding material
  • SGMP Alloy 690TT steam generator tubing specification recommends a minimum chromium concentration of 28.5 weight percent
  • 191 heats of Alloy 690 tubing material were reviewed and the minimum concentration is 29 weight percent

- All three tubing suppliers were represented

  • 17 heats of Alloy 82 were reviewed and the minimum chromium is 19.6
  • 27 heats of Alloy 182 were reviewed and the minimum chromium is 13.6 16 t=~121 1 ElECTRIC POWER

© 2015 Electric Power Research Institute, Inc. All rights reserved. ~~- RESEARCH INSTITUTE

Fabrication of the Mockup Material used for tube-to-tubesheet mockup 182 (.187-in x 14)

.47 5.1 .001 .002 .01 .54 1.83 7.4

.34 6.67 .003 .001 _01 .49 1.39 7.27 82 (30-mm Strip) 2.96 .004 .002 .02 .85

%Cr is on the upper end for cladding but we are measuring dilution not absolute composition 17 ~~121 1 ELECTR IC POWER

© 201 5 Electric Power Research Institute, Inc. All rights reserved. ~~- RESEARCH INSTITUTE

Fabrication of the Mockup 182 Shielded Metal Arc Welding (SMAW) process SGMP developed prototypical tube-to-tubesheet mockups to analyze chromium levels in the tube-to-tubesheet welds 82 - Electroslag Welding (ESW)

Strip cladding

~f21 1 18

© 2015 Electric Power Research Institute, Inc. All rights reserved. E1 -

ELECTRIC POWER RESEARCH INSTITUTE

Analysis of the Chromium Content

  • The mockup was designed as a 3 X 4 tube-to-tubesheet matrix having 82 cladding on one side and 182 cladding on the other side
  • An Alloy 690 tube was inserted into each drilled cylinder, positioned flush with each end, and welded autogenously using the Liburdi Dimetric P-Head autogenous GTAW welder (see next slide).

19 ~~~I I ELECTRIC POWER

© 2015 Electric Power Research Institute, Inc. All rights reserved. -=*-*~ RESEARCH INST ITUTE

Analysis of the Chromium Content

  • Plugs were removed from the tubesheet, cross sectioned, and weld dilution was evaluated
  • It had been estimated that the weld dilution would be 50o/o from the tube material and 50°/o from the cladding material since the weld is autogenous GTAW 40 cross-sections were available for 82 welds and 34 for 182 welds to include in the tube-to-tubesheet weld dilution distributions 20

© 2015 Electric Power Research Institute, Inc. All rights reserved . E ~~~ ~

I-I~

ELECTR IC POWER RESEARCH INSTITUTE

Comparison of Mockups

  • SGMP has a mockup from a replacement steam generator fabricator with Alloy 82 clad and 690 tube material

- Used to compare geometry of the fabricated mockup welds and to compare dilution 21

© 2015 Electric Power Research Institute, Inc. All rights reserved . E1

~ral l ELECTRIC POWER RESEARCH INSTITUTE

Comparison of Mockups Mockup from RSG Fabricator SGMP Mockup Similarity between the etched weld cross-section in the SGMP mockup to the one fabricated for this project 22 t=~121 1 ELECTR IC POWER

© 2015 Electric Power Research Institute, Inc. All rights reserved. ~~- RESEARCH INSTITUTE

Analysis of Chromium Content

  • The EPRI SEM system was used to mark and measure areas of features on sample cross-section macrostructures

- Measurements were used to compute the weld dilution

- Dilution measurements were analyzed to a "normal distribution" 6 6

+ 0.530 5 5 ,, ' '

4 4 I I

I

' \

\

I \

I \

3 3 I \

I \

I \

2 2 I \

I 1

1

0 0 ----

0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.2 0.3 0.4 0.5 0.6 0.7 0.8 Dilution Ratio Dilution Ratio Normal distribution for the Dilution Ratio Normal distribution for the Dilution Ratio for for 690 tubes welded in 82 cladding 690 tubes welded in 182 cladding 23 ~Pf21 1

-= ELEC TRI C POWER

© 201 5 Electric Power Research Institute, Inc. All rights reserved. RESE AR CH INSTITUTE

Analysis of Chromium Content

  • The combined dataset was used to determine the mean dilution factor

- Appropriate estimation of reality for Dilution Equation (below) 18 1J=0.52 16 cr=0.08 14 12

(..)

r:::

Cl) 10

1 C"

Cl) 8 LL.

6 4

2 0 ~~~~~~-+--~~-+--~~~~

0.2 0.25 0.3 0.35 0.4 0.45 0.5 0.55 0.6 0.65 0.7 0.75 0.8 Dilution Ratio

%Crweld = %Crtube (0.52) + %Crclad(0.48)

~~~ ~

El-Ie;;;;;;;.

24 ELEC TRI C POWER

© 201 5 Electric Power Research Institute, Inc. All rights reserved. RESEAR CH IN STITUTE

Results of the Mockup Testing

  • An autogenous weld deposit for 690 tubing material and 82 cladding has sufficient chromium content to be resistant to initiation or propagation of PWSCC (exceeds 24% Cr)

- These welds constitute the majority of the tubesheet

  • The weld deposit for 690 tubing material and 182 cladding has less chromium, but the chromium is high enough to be resistant to initiation and slow propagation of PWSCC Predicted Cr Compositions

% Cr in Weld  % Cr in Weld Statistic (82 clad) (182 clad)

Mean 25.2 22.1 Maximum 25.8 23.2 Minimum 24.5 21.8 25

© 2015 Electric Power Research Institute, Inc. All rights reserved. E1

~f21 1 ELECTRIC POWER RESE ARCH IN STITUTE

Stress Analysis of the Tubesheet

  • Finite element analysis of a Westinghouse Model 51 SG channel head was used to determine the stresses during normal, upset, and accident conditions
  • Tensile loads are shown to be present in the divider plate region

- The dominant mechanical bending loads are in the plane that drives a horizontal crack and would not turn the crack vertically.

- Assuming a crack initiates, it would propagate horizontally

  • The Alloy 182 cladding in the center of the tubesheet is the most susceptible to PWSCC

- This area of concern is in compression

  • Results and conclusions were independently verified by third party t=~121 1 26

© 2015 Electric Power Research Institute, Inc. All rights reserved. -=*- ELECTR IC POWER RESEARCH INSTITUTE

Scenario 3 Concerning the Low Alloy Steel Channel Head 27

© 2015 Electric Power Research Institute, Inc. All rights reserved. E ~f21 1 ELECTR IC POWER RESEARCH IN STITUTE

Stress and Fatigue Analyses

- Design transients evaluated

- Stress paths defined through the channel head wall

- Through wall hoop and axial stresses mapped along these paths

- Fracture mechanics and fatigue crack growth evaluations performed

  • Maximum stress intensity factor less than the allowable values for all design basis plant conditions up to 75o/o of the vessel wall thickness

- Allowable flaw is 3.9" deep and 29.6" long based on flaw tolerance considerations

  • Fatigue growth evaluation performed for a postulated flaw 0.25" deep into the low allow steel

- Evaluated axial and circumferential growth 28 t=~~~ ~ ELECTR IC POWER

© 2015 Electric Power Research Institute, Inc. All rights reserved. ~~-~~ RESEARCH INSTITUTE

Conclusions

  • Critical crack depth is 75o/o of the channel head material- 3.9" deep and 29.6" long
  • The bounding crack growth results in a crack depth of 8°/o of the channel head material

- The circumferential crack would grow from 0.25" to 0.40" deep and from 1.9" to 3.07" long

  • Forty years is chosen because this is typical for fatigue analyses Results for 40 Year Crack. Growth Period Stress Case Flaw Initial Final Final Flaw lnitia Final Allowab le Pass Type Flaw Flaw DepthNessel Kmax. Kmax. K  ?

Depth Depth Thickness (ksi-..J in) (ksi- in) (ksi-\i in)

(inch)' (a) (aft}

(inch)

Path 1 Circ 0.25 0.404-0 0.080 24.632 31.048 63.2 Yes (with residua.Q Axiru 0.25 02508 0.048 7.465 7.482 63.2 Yes Path2 Circ 0.25 02 985 0.057 21.555 23.806 63.2 Yes (with residual)

Axial 0.25 02510 0.048 10.527 10.545 63.2 Yes Path 3 Circ 0.25 0.2524 0.049 9.954 9.998 63.2 Yes (with residual) Axiru 0.25 02514 0.048 9.555 9.579 63.2 Yes 29

© 2015 Electric Power Research Institute, Inc. All rights reserved . E1 ~121 l ELECTRIC POWER RES EARCH IN STITUTE

Conservatisms in the Fatigue Analysis

  • Cracks in divider plate assemblies that have been inspected for years have shown no crack growth and thus have not approached the triple point
  • It is not expected that the US fleet has divider plate cracking extending the full length of the divider plate and approaching the triple point
  • The assumed initial 0.25" through wall depth primary water stress corrosion crack in low alloy steel is not realistic; but is a conservative assumption

- Low alloy steel is not susceptible to PWSCC 30 t=~121

~~-

1 ELECTR IC POWER RESEARCH INSTITUTE

© 2015 Electric Power Research Institute, Inc. All rights reserved.

Scenarios 1, 2, and 3 Conclusions

  • Chromium levels in the tube-to-tubesheet welds are sufficient to resist PWSCC initiation
  • Orientation of the dominant stresses associated with the tubesheet to divider plate regions strongly favor horizontal crack propagation in the stub runner or the welds on top and bottom of the stub runner
  • The center of the tubesheet ts in compression thus the driving force for PWSCC initiation/propagation is not present
  • Stresses in the channel head region of the steam generator are insufficient to propagate a crack by fatigue into the channel head and thus would not compromise structural integrity of the channel head
  • Final Report 3002002850 published October 2014 E ~f21 1 31 ELECT RI C POWER

© 2015 Electric Power Research Institute, Inc. All rights reserved. .- RESEARCH INSTITUTE