Text

Nonproprietary Version of Counterflow Preheat Steam Generator Tube Expansion Rept
	ML20081A476
Person / Time
Site:	Comanche Peak
Issue date:	06/30/1983
From:	; WESTINGHOUSE ELECTRIC COMPANY, DIV OF CBS CORP.
To:	;
Shared Package
ML19268E253	List: ML20081A461; ML20081A469; ML20081A476; ML20081A484;
References
	SGPR-8301, NUDOCS 8310260284
	Download: ML20081A476 (140)
	v • d • e

.

{ '. :. :_ T

.1

-COUNTERFLOW PREHEAT STEAM GENERATOR TUBE EXPANSION REPORT SGPR-8301' June 1983

. WESTINGHOUSE ELECTRIC CORPORATION

-Nuclear Energy Systems P.O. Box 355 Pittsburgh, Pennsylvania 15230 8310260284 830826 PDR ADOCK 05000445 A

PDR

'0771c/0122c/060883:5 1-L'

r-

.r

.I t -

TABLE OF CONTENTS

1.0 INTRODUCTION

'2.0

SUMMARY

AND C0f'CLUSIONS 2.1 Expansion Process 2.2' Expanded Tube Qualification 2.3 Effect of. Tube Expansion on Vibration 2.4 Overall Conclusions 3.0 PROCESS AND EQUIPMENT DESCRIPTION 3.1 Process Description-3.2 Equipment Description 3.3 Process Qualification 4.0 EXPANSION CONSIDERATIONS 4.1 Tube Selection 4.2 Expanded Tuoe Inspection 0.0 EXPANDED TUBE QUALIFICATION 5.1 Corrosion Assessment 5.2-Structural Analysis 6.0 EFFECT OF TUBE EXPANSION ON VIBRATION 6.1 Data Acquisition and Reduction 6.2 Operating Plant Data 6.3 Scale Model Test Data 7.0 REFCRENCES

~

0771c/0122c/060883:5 2

r

1.0 INTRODUCTION

Wear has been observed on some steam generator tubes within Westinghouse Model D preheat design steam generators. ~ Such wear has been attributed to tube vibration induced by feedwater entering the preheater region of the steam Modifications to reduce tube vibration, hence wear, have focussed generator.

on reducing the incoming feedwater flowrate and modifying the tube response to the feedwater. This latter approach is' the subject of this report.

An elevation view of a counterflow preheat steam generator (Models D4, DS and E) is shown in Figure 1-1.

As depicted, the preheater region is located on The lower the cold leg side of the tube bundle and faces the feedwater inlet.

shell internals including the preheater region are shown in Figure 1-2.

Incoming feedwater enters the inlet water box and encounters the impingenent plate which. directs the water outward to fill the water box volume and downward to the preheater inlet pass located between the B and D support The water enters the tube bundle, then flows upwards around plate: (baff les).

It is in the outer rows of the tube bundle facing the the tubes and baffles.

incoming feedwater that the high tube vibration levels have been observed.

Preheater region tube vibration for a single steam generator tube is dependent on an input excithtion (feedwater) and the support conditions within the Nominally, a tube is supported within the preheater by support preheater.

baffles with clearances between the tube and the baffle of approximately 20 Under operating conditions, however, a tube may not be supported at one mils.

Vibration levels for a tube not supported or more support baffle locations.

at a baffle have been observed to be higher than those for a similar, By reducing the tube-to-baffle clearance through expansion, supported tube.

support can be provided for a previously unsupported tube and the vibration levels may be reduced.

The purpose of this report is to discuss the tube expansion process utilized by Westinghouse to increase the potential for steam generator tubes within the preheater region of Model 0 steam generators to be supported at baffle Additionally, discussion of the test and qualification program to locations.

demonstrate safe operation with expanded tubes is provided.

0771c/0122c/060883:5 3 l-1

444 f /.s

/

P

.s s

f C...

/

f

(

.=

c.

4 I

s

3. 3

-O

'4 I

ja a

(

i u.6 of AumW

~

P r..

Trus.nesses Tue. Sunes.

f I

E 2

\\

m.

"iI I"f Ill l1l

, N tswer snes --.

4

4.,

,4 a i i t iI Press ear Gess.

lll

..s r) l P emmerm ss.

s l

4 9

(

l i M's Pv.a

, s n q

p-Tues Shoes e

N

\\Prese ear Qualet Pra rv a.emn inw c

N caemn c m Figure 1-1 l

Counterflow Preheat Steam Generator

_. _. _ _.,... _ _. ~ -. _ _.. _ _. _..

e i

4

,.m #

nditADPER f

J i

Kk 4

/

e 1

i N

M PARTIT1tm \\

g n

PLATE F

G

_c

~

f t

l g

f C

4 0

N\\

I i

twinspeT iap i

r r

/ d h h

%A b

i) r s

s r

\\

s i

natsmer e

l Figure 1-2 Counterflow Prcheat Steam Generator Lower Shell Internals v--

- -, -. -, - - -. -.--e..--...-

.--.ww---

rv 4-..--..m-w-,,-,,---w_-,,.

.-.-- m

..-s,.,-.-r,-,,-,=. -,. -. - - -,,. - + + - - -. =

v-+>

2.0

SUMMARY

AND CONCLUSIONS The use of a pressure controlled expansion process for reducing flow induced vibrations in stean generator. tubes within Westinghouse counterflow preheat steam generators has been evaluated through a comprehensive test and qualification program. The process equipment has undergone verification and repeatability testing to demonstrate the controllability of the process itself and the ability to consistently acnieve tube expansions within specified tolerances.-

Expanded tube _ specimens have undergone laboratory denting and stress corrosion tests. Analyses of tubes with expanded regions h, ave been conducted in accordance with the ASME Code and an NRC tube plugging regulatory guide.

Finally, the effects of tube expansion on tube vibration have been evaluated with data obtained frm an operating plant steam generator and a laboratory full scale test model.

2.1 Expansion Process Process qualification tests have been conducted to establish the capability of the expansion equipment to consistently achieve tube expansions within specified tolerances. The qualification program utilized full scale steam generator channel head mockups and included a training program for the field i

implementation personnel. The qualification tests have shown that:

]a,b,c,e-Expansions with a minimun flat length of[

and expanded a.

tube-to-baffle diametrai gap clearance of[

] 're hnsistently achieved over the design range of process parameters. The ras.ge of process parameters included initial tube-to-baffle diametral gaps of l

0.005 to 0.045 inch and tube yield strengths of 44 to 72 ksi.

_ a.,b,c,e b.

Expansions were consistently located within of the baffle plate centerline, resulting in no tube bulges occurring outside of the baffle plate surf aces.

l l

i 0771c/0122c/060883:5 4 g

c L

, s, b,Cp Expansions of: worn tubes with wear scar depths up to

, percent c.

through-wall result in expanded tube profiles within acceptable limits for tube structural and vibration considerations, d.

The teaming process utilized 'does not affect the tube mouth weld integrity.

Detection of stress corrosion cracks in the transition regions of an e.

expanded zone is estimated to be reliable at or above 30 percent for ID originateo signals and at or about 40 percent for OD originated signals.

The EDM notches employed qualify the process for cracks whole volmes are 80 times less than the volume of a 40 percent ASME drilled hole standard ampi itude.

2.2 Expanded Tube Qualification 2.2.1 Denting Concern The effect of gap size on the extent of denting has been investigated in a i

stries of Single Tube Model Boiler tests. Tests with prepacked and I

non-prepacked crevice conditions have been conducted. This testing has shown that average dent size increases with increasing diametral gap and that the Expansion process does not result in an increased propensity for denting.

2.2.2 Stress Corrosion Cracking (SCC) Concern A combination of polythionic acid tests, controlled potential electrochemical l

tests, and magnesiun chloride tests were conducted to evaluate the effect of

(

tube expansion on residual stresses. Both nominal and off-nominal tube expansions were tested. The results from the nominal tube expansion tests in polythionic acid showed no increase in tube OD or ID residual stresses for expansions up to 0.060 inch AD (this is in excess of the maximum expected expansion AD of approximately 0.045 inch).

The results from the off-nominal tube expansion tests were initially disappointing in that the laboratory samp'es displayed cracks within 16 hours1.851852e-4 days 0.00444 hours 2.645503e-5 weeks 6.088e-6 months i

0771c/0122c/060883:5 5 z-Z

Ein a-magnesium chloride solution (possibly indicating high residual stressesi. Additional tests were then conducted comparing two heats of mill annealeu.aconel-600 tubing expanded into tubesheets and baffle plate collars.

The results from these tests showed that the baffle plate expansions resulted in lower residual stresses than did the tubesheet expansions.

2.2.3 Structural Concern Analyses were perf ormed to update a portion of the generic steam generator stress report to address the effects of the tube expansion process. The expanded -tube configuration was evaluated for Design, Normal, Upset, Emergency and Faulted conditions in accordance with ASME Code Section III, subsection NB requirements.

In addition, the tube plugging margin per NRC Regulatory Guide 1.121 has been evaluated.

The conclusions from these analyses are:

The maximum calculated usage factors for the locally expanded region of a.

the tube a' e all less than the Code allowable of 1.0.

b.

Tube f atigue limits are satisfied for wear scars in the expanded tube region.

Stress concentration f actors obtained from worn tube measurements are less than the allowable f actors for which cumulative fatigue usage would reach 1.0.

The primary stresses in the expanded region of the tube are not c.

substantially affected by the expansion, d.

The tube plugging margin is not significantly affected by the expanded region.

2.3 Effect of Tube Expansion on Vibration Vibration data have been obtained for unexpanded and expanded tubes in both an operating plant steam generator and a full scale test model.

One tube (R48C55) in steam generator No. 2 at the operating plant (Model 04) was 0771c/0122c/060883:5 6 g,y k-

~

a exp anded. Two. series of _ expanded tube. tests *(the first with three tubes and

-the second with '24' tubes) were conducted utilizing a 16* full scale water test

- model.

The results-from these tests show:

a.

Levels of both RMS displacement and impact accelerations are reduced ~ by tube expansions at the B and D plate elevations.

ecf

+ -

),m, pact g's by a b.

RMS displacements are reduced by a factor of(-

i

['an'd Ga's by a f actor of[

or'hower levels-70 factor-of percent and above, as measured in the expanded tube at the operating plant.

c.

Model data for a group of expanded tubes show RMS displacements and impact g's reduced on the average, bg a f actor of[]*dnd'h6's reduced, on the average, by a factor of

]at' $ model flow rate comparable to the 100 percent flow rate in a 2 loop,+odel 04 plant.

2.4 Overall Conclusions The overall conclusions for hydraulic expansion of mill annealed Inconel 600 tubes are:

The expansion process and associated tooling have been qualified and havt:

a, consistently been shown to achieve tube expansions within specified tolerances.

b.

- Supporting evaluations, addressing chemistry / corrosion, structural and vibration concerns, have shown the acceptability of tube expansion.

0771c/0122c/060883:5 7 g_4

9 3.0 PROCESS AND EQUIPMENT DESCRIPTION 3.1 Process Description 3.1.1 Tube Preparation g,c e 4

a 3.1.2 Baffle Plate Location

. C. {

r 3.1.3 Tube Expansion a, c, e 9

0771c/0122c/060883:5 8 3-1

A. C. O t

3.1.4 Post-expansion Inspection C, c, d I

3.2 Equipment Description 3.2.1 Cambination Gauge and Swab Probe c, c, e, $

L 0771c/0122c/060883:5 9 3

u,,

g, c, e, f 3.2.2 Reamer Tool a, c, e, S 3.2.3 Dry Hone' System The dry hone system is depicted in Figure 3-4.

3.2.4 Probe Inserter

(

a g,f

8

~

3.2.5 Combination Eddy Current and Expansion Mandrel Probe The combination eddy current and expansion mandrel probe is depicted in Figure 3-7.

G, c, f,,8 0771c/0122c/060883:5 10 3-3

cultifrequency eddy current systen in the differential mode for location of a

- baffle plate. As the. coils are drawn past a baffle plate, a trace. will be recorded on a display unit external to the steam generator. A null reading from this trace is achieved when the coils are centered within the baffle

. plate (Figure 8).

Following tube expansion, the coils will be used in an absolute mode to inspect tne expansion zone. Eddy current signatures will be recorded on a

~

strip chart and compared to signatures obtained from reference expansions of known depth and profile (Figure 3-9).. Coupled with this recording on the strip chart is a support plate recording (Figure 3-10). A comparison of these

. two recordings verifies the location of the expansion zone within the baffle

_ pl ate.

3.3 Process Qualification The qualification program utilized prototypical equipment and prototypical steam generator tube and baffle plate samples to verify the accuracy and repeatability of the tube expansion process.

3.3.1 Process Requirements Requirements for the pressure controlled tube expansion process were:

a.

An expansion lenoth within the baffle plate greater than or equal C A-r

, M. c, e to \\

~

6,having{

ydiametervariation, 3

)(e,,b,e,e.

at the point of b.

Post-expansion diametral gap of

(

maximum diametr al expansio i),

c.

No significant tube expansion outside the baffle plate surf aces, i

d.

Tube ovality of less than or equc1 to 2 percent, and A, b,C, c.

e.

Freedom for tube axi motiongreatertha(

jwithinthe baffle plate hole.

0771c/0122c/060883:5 11 3d k

Figure 3-11 depicts these requirements.

3.3.2 Baffle Plate Location i

Repeatability tests were conducted to verify the accuracy of the eddy current baffle plate location. technique. These tests showed that the eddy current coils could,be nsistently centered within the Daffle plate to an accuracy of

,c. t

('

ollowing movement of the coils to align the expansion mandrel,

- the accuracy for cent ring the mandrel was determined to be consistently m

c,<.-

within j

e accuracy of the eddy current location technique was not af[ected by stainless steel versus carbon steel support plates.

s 3.3.3 Expansion Qualification

- The expansion process qualification was performed over a range of variables beyond the limits existing in the stean generators as follows:

-~

Tube Wall Thickness

.043" +.004" Tube Outer Diameters

. 7 50 " _+ *

.Diametral Expansion

.015" to.045" Material Yield Strength 44 to 72 KSI Expansion (were made into holes having a surface roughness of(

n addition, expansions of tubes having simulated wear scars (both uniform and variable depths) and expansions of tubes either cocked or t

preloaded were made.

This value is based upon measuremants of baffle plate holes machined in the manufacturing plant with production equipment.

l l

0771c/0122c/060883:5 ~ 12 3-5 f

,-+,%-

.,m4

..ww-,-.ey---,.--.,,-

<.-,c,..ve,.,,,,,-._

-,--#me

.-v.--r ww v-

+-*---mewwnv--e - + wn t-t,i-es-e

3.3.3.1 New Tube Qualification The expansion process was qualified for mill annealed Inconel 600 tubing having initial ~ diametral gap clearances between 5 and 45 mils and varying

. combinations of. yield strengths, final gap sizes, preload conditions, hole ovality and wall thickness. For a nominal yield strength of 55 Ksi, the expanded gaps and flat lengths ranged from(

o'r 'a

. initial gap of E

. ( c, #-

ma*e e Overall, the radial gaps varied fromE i

The large gaps occurrec for' oval holes with maximtsn yield strengths under preloaded and/or tilted

,,c, e off-nominal conditions. Expanded flat lengths ranged from

+

and exceeded requirements even under tube preload conditions. Table 3-1 shows typic al results achieved for pressure controlled expansions of new tubes.

3.3.3.2 Worn Tube Qualification Both uniform and non-uniform, tapered wear scars were machined onto the surf aces of low and high yield strength,.miil annealed.Inconel 600 tubes.

r-

- A c.*

Scar depths of approximately(

j percent of nominal wall thickness were used. The tubes were then expanded into baffle plate holes having initial diametral gaps of

].

T$e results showed that the majority of the worn tube expansions were within the specified tolerances for the expansion zone. The expansions, however, were not diametrically symmetrical due to preferential expansion into the worn region having minimum wall thick ness. Table 3-2 provides typicaf results achieved for expansions of tubes having wear scars.

3.3.4 Expansion Profile Examination Tests were conducted to identify the optimum frequency and sensitivity The settings for us? in examining the expansion zone diametral profile.

process itself was used very successfully in late 1982 for measuring the expanded and hardrolled regions on approximately 3000 tubes in an operating plant.II)

Diametral gap not including the wear scar depth.

0771c/0122c/060883:5 13 3-6

)

3.3.5 Post-Exoansion Eddy Current Inspection o C 8, f g i f

f

-f I

1 l

\\

1

\\

~~

t 3-7 0771c/0122c/060883:5 14

m, C i,.4 C

i l

l l

1 l

l 0771c/0122c/060883:5 15 36 l

l 1 -.

.o Table 3-1 Typical Results for Tube / Baffle Plate Pressure Controlled Expansions

- Exp anded - Zon e Expanded Y ield Initial Final-Gap Flat length-Zone Strength

-Gap (mil)

(inch)

Ovality (ksil (Mil) 0' 90' 0'

90~

(Percent) Slidability.

b, c,

.s r

0771c/0122c/060883:5 15 3-9

TABLE 3-2 TYPICAL RESULTS FOR UNIFORM AND TAPERED, NON-UNIFORM WEAR SCAR EXPANSIONS FINAL GAP FLAT LENGTH APPROX. YIELD APPR0XIMATE WEAR NON-WORN WORN NON-WORN WORN STRENGTH INITIAL GAP DEPTH SIDE SIDE SIDE SIDE OVALITY SLIDABILITY (ksi)

(mils)

(%)

(mils)

(inch)

(%)

LINIFORM WEAR SCAR RESULTS o b '.

s i

W i

1 w

A@4 i

5 4

Figure 3-1 Typical Pressure Trace for Pressure Controlled j

Hydraulic Expansion Process l

x

-, +,, - -, ---

r,

.m I

r 4-i-

i

[

i l'

l i

I I

l Figure 3-2 Combination Gauge and Swab Probe l

/ /

=

I i

i -

I.

):

1 Figure 3-3 Reamer Tool (Hands-On) t o

y

i.,,

,.e

,-c

,.e,ey,p.

.*,,--c,,,_,,

.+.w--

2

,-c,-,,-,--,==,,-ew,-c=-.w%_

a,C t i

4 4

i.

s Figure 3-4 Dry Hone System

,- +

4.,--,r.-

- + - -3, cm,-,--,%,,, - -.., -. - +,.,,, -. - -

-,.-.m..

.... ~ -, - -

,m--,

-w

.--w-,m-

a., C C.

i

~

l l

t i

l i

l

~

I l

Figure 3-5 Probe Inserter (Hands-On)

6 0$

9 9 1 r

1 I

Figure 3-6

' s Probe Inserter (Remotely Operated)

-(

t%

~

g 4

g,

\\'w-) -

4.sg '...A..

N a

3

. -. _ =.

i

% 44

=

r i

i 4

1 Figure 3 7 Combination Eddy Current arid Expansion Mandrel Probe

....,.....,.. _ -. ~

Q.,C, e, Figure 3-8 Pictorial Demonstration of Eddy Current Probe in Baffle Plate Location Mode

et.c, e,f i

l 4

t Figure 3-9 Typical Eddy Current Signatures for Reference Expansions i

'q

-r-wm-s--

e e

-g

---w.e-,

e. e n

mpv---,

-m v--~gw y.mn,w,,-,-_,,,-r--

,-m,,

-mew---

Ae(r Q P

l l

r i.

l Figure 3-10 Typical Eddy Current Signatures Displaying Tube Expansion Profile Relative to Baffle-Plate Location l

l

o Tube g

E

-*' +-Nominal Gap [

t_

/

A

/

(

Baffle Plate ExpansionLengthb

/

//

u I

/

l

/

Final Gap I

l l

l Figure 3-11 Expansion Requirements for Pressure Controlled Process l

l

,-t

,,,,n.,

,,.-e

~w.

,..,,,,---,.,,.,--.,-.v,-,w,-,

r

.,r,,,

r,--,,,,, - -, - -

,..e,,,.

/ /

J J

i l

l l

i i

t l

i Figure 3-12 l

Cimferential Cross-Section of 8-Coil Surface Riding Probe l-l-

9 t.

s

eS 3 I

i m

Figure 3-13 I

Axial Cross-Section of 8-Coil Surface Riding Probe l

l

- -... ~,

2 a

u.

e u

2.

&_u AA,4 g

J T

Eb 0., Gi t

~

1-

)

]

f a '

4 en b

FI m 3 14 SUPPORT P W g g 9

y, g._,,

9,,;-,-,..~., -. - - - - -

--.vy-.

-,-w g-y,y

,3

.w,,v-,,,q.ny,,-,,-

o,m-m,n,,,-,&.,,-y

.v,-em w, e

..w.-3..,

,,,mm-%e.,.1--,,--e,--s v

a.,4 %

t

~

FIGURE 3-15 SUPPORT PLATE USING 550/130 Kh: MIX

s 2,4 ([

FIGURE 3-16 40% DRILL HOLE IN IGE STANDARD USING 550/130 Khz MIX

a.,4 e, $

.L' L

FIGURE 3-17 NORMAL EXPANSION USING 550/130 MIX

a, C, e,h i

i f

I i

l l

i I

i um FI@RE 3-18 NOB!AL EXPANSION WInf MG USING 550/130 MIX

4,44 1-FIGURE 3-19 NORMAL EXPANSION WITH 40%

CIRClbtFERENTIAL NUTG USING 550/130 MIX e-- + -

w e e e---

r-w-rr

-e-

, -- - - - m -

mv-ee-v-,=

w m v

- ~ - - -

,r,,m'-,

- - < -w,-m,-

,--w w--

, m-g e,w y-s m-w-em--e--.c w-,y~w a

w -~-

% C,8<[

i.

k r

}

l P

FIGURE 3-20 N2PML EXPANDED REGIOfJ WITH]N SUPPORT PLATE

LX.,C d, r

en FICURE'3-21 E).1 Ani 1:1, TUHI: W I 'l li SIMULATED CI RCOMI LPI,rJT1 Al' g,

, F.

A'l TRANS]T}t su pgg g,m

4.0 EXPANSION CONSIDERATIONS 4.1 Tube Selection Tubes were selected for expansion based on evaluation of vibration data from Krsko and steam generator test models. Details of these evaluations are included in reference 6.

The principal test models providing data for selection of tubes for expansion were the 2/3 and 16' water models. The 2/3 model is a 2/3rds scale model of the feedwater inlet area and the first pass of the preheater. Velocity and turbulent, tube excitation force distributions were measured in the model.

The 16* model is a full scale test model used to obtain tube vibration seasurements as described in Paragraph 6.3.

The plant and 16* model vibration data were evaluated using the Ga parameter [

]as NaIned from accelerometers mounted in the tubes. The Ga parameter is further described in Paragraph 6.1.2.

The vibration data base used for the design modification vibration assessment and for selection of the expanded tubes included:

a.

Krsko data - Ga values for 16 tube locations including tubes in both steam generators.

i b.

16* Model data - Ga values for about 50 tube locations in the 16* sector.

l l

c.

2/3 Model data - Turbulent tube excitation forces for about 40 tube loc ations, i

d.

Nonlinear model tube vibration analyses - These analyses utilized the 2/3 model force data to calculate tube wear at key tube locations and aided the identification of tubes requiring expansion.

0771C/0122c/060883:5 16 4-l m

1

.Ga vs.. turbulent force correlations - Measured Ga values have been e.

- correlated with turbulent force data to obtain Ga estimates at tube locations not included in the 16* model.

f.

Ga wear correlation - Ga values have been related to wear-based on measured Ga values for tubes pulled from KRSK0 to obtain direct wear measurements.

The above data base has been used to identify the high vibration level tubes requiring tube expansion at the modified flow rate.

Bot.h the Ga and the turbulent force distributions led to very similar identification of candidate tubes for expansion. A typical expanded tube ~ pattern resulting from the vibration assessment is shown in Figure 4-1.

As steam generator designs vary slightly from plant-to-plant, specific expansion patterns will be prepareo for each plant.

4.2 Expanded Tube Inspection Capability for ' inspection of the expanded tube region is provided by two independent eddy current inspection probes. The ECT probe combined with the cxpansion mandrel provides confirmation that the expansion has been completed. A newly developed ECT probe is used for detection of indications in the expanded region. These probes are discussed below:

4.2.1 Expansion Mandrel Eddy Current Probe As noted in Section 3.2.5 the expansion mandrel eddy current probe is used to characterize the axial variation of the expansion immediately after the ~

expansion process is completed. This probe naasures the average diameter of the expanded tube over the length of the expansion. Coupling the diametral signature recording on a strip chart with the support plate indication recording provides the relative elevation of the expansion diameter and the sepport plate.

l-The diametral reasurement provides confirmation that the expansion has been comp leted. The strip chart recording verifies the location of the expansion 0771c/0122c/060883:5 17

,7

zone within the baffle plate. These signatures thus characterize the expansion zone profile.

4.2.2 Post-Expansion Eddy Current Inspection As noted in Section 3.3.5, an eddy current probe and associated supporting equipment and techniques have been developed to improve resolution of indications 'in the expanded regions. This probe provides capability for detection of indications in the transition regions of expanded tubes.

The probe capability permits detection of 40 percent thru-wall indications in the expanded zone. Detection capability becomes limited in the presence of wear ~ scars for which the scar signal tends to mask other indications.

This probe is utilized to perform a post-expansion, baseline eddy current inspection of the expanded tube area.

In the unlikely event of a crack developed to 40 percent wall depth af ter the expansion, the crack would be detectable by this inspection..The-signal characterization from this inspection provides a baseline for later periodic inspections. This probe is used in. conjunction with conventional eddy current techniques for the balance of the tube to obtain a canplete tube inspection.

i F

l 0771c/0122c/060883:5 18 93

b 446 f

d 3

rl l

J t

6 Figure 4-1 Typical Model D4 Steam Generator Tube E.pansion Pattern (number of expanded tubes = - 116) i i

l -

c.---

5.0 EXPANDED TUBE QUALIFICATION 5.1 Corrosion Assessment residual Expanding the tubing into preheater baffle plates may produce:

' stresses of potential concern in the transition regions between the expanded tube section and the nominal tube diameter; and possible crevice configurations, which may differ from those in operating steam generators.

The effects of the residual stresses relative to stress corrosion cracking and the crevice confinurations on possible concentration and corrosion processes have been tested and reviewed.

Variations-in the expansion process can result in the crevice diametral gap varying between{

](I6Idb and the depth (measured from either the top or bottom face of the baffic) varying between(~

~ - -

Specific

} Ne'Iownward-facing crevice is unique to this design.

corrosion issues include a concern for denting because of the reduced gap stress corrosion dFacking in the transition region, and a size, a concern for concern for tube wastage in the crevice. Pitting is not believed to be a concern because it has only been an issue in operating plants outside of crevice regions.

5.1.1 Denting i

The effect of gap size on the extent of denting has been investigated in a Tests similar to these have been series of Single Tube Model Boiler tests.

performed in previous corrosion test programs to reliably reproduce phenomena such as denting e.1d stress corrosion cracking.

5.1.1.1 Test Specimens Test specimens were prepared from lengths of Inconel 600 steam generator d

Each of the tubing fitted with carbon steel tube support plate simulants.

first three tests used two sleulants mounted over expanded areas of the tube Two additional support to simulate expanded tube / baffle plate intersections.

plate simulants, mounted over unexpanded regions of the tube, were used as 0771c/0110c/041983:5 14 55-/

reference intersections for each test. All support plate simulants (expanded and unexpanded) were then prepacked with magnetite prior to the tube being exposed -to' the denting environment.

A fourth test utilized support plate simulants similar to those mentioned above except _ that for this test, the simulants were eccentrically mounted and the crevices were not prepacked with magnetite. For this test, magnetite and copper were injected into the model boiler to simulate the initiation of denting.

Figure 5-1 depicts the general configuration of these test specimens.

5.1.1.2 Test Environment In each test, the test specimen was exposed to a solution comprised of a feedwater chemistry (from a make-up tank) of 0.2 ppm chloride from seawater, 0.1 ppm chloride from copper chloride, 2 ppm ammonia and 101 pb hydrazine.

This chemistry is referred to as the reference denting chemistry (ROC). Under certain thermal and hydraulic (T+H) conditions it is a prove-n dent producing chemistry in model boilers. The T+H condition employed in these tests were either hot leg or cold leg conditions typical of Model 0 steam generators.

The hot' leg conditions (i.e., Tsat = 520~F, Tout = 580-600~/) were use in an accelerated test while the three long term tests were performed under cold leg conditions (i.e., Tsat = 535T, Tout = 558'F). Prior testing has shown that both of these T+H conditions cause tube denting when used

{

simultaneously with the ROC, with the dent rates being significantly higher under hot leg conditions. Figure 5-2 shows average denting rates achieved in the RDC. Figure 5-3 shows the effect of superheat on the denting rate.

TPe test environment was chosen to ensure the prototypical nature of the tests. All materials, dimensions and temperatures used in the laboratory testing were either within the design tolerances for the Operating steam

' generators or were judged to be conservative. An example of the latter is the f act that, on average, the laboratory primary temperatbres t,re 4-5*F higher l

than the range encountered in the steam generators. This higher temperature should result in.a more aggressive denting condition being realized in the 0771cIO122c/060883:5 20 6 - F-

laboratory test. Also note, that a range of temperatures were actually experienced in the laboratory testing, i.e., the uppermost tube support plate (TSP) simulant was approximately 10*F lower in temperature than was the lowest sinulant. This means that if the lowest simulant experiences a temperature of 568 F, then the uopennost simulant would experience a temperature of approximately 558'F. Each test therefore, exposed the TSP simulants to prototypical operating temperatures. Table 5-1 gives a comparison of the laboratory denting test conditions with typical steam generator conditions.

The secondary flow rates and pressure drops are not prototypical. The model boilers used for this testing are natural circulaticn loops without forced flow capability. The lack of the prototypical simulation with respect to flow and pressure drop are considered to be conservative with respect to denting studies. The denting phenomenon is produced in flow-starved regions and the lack of prototypical flow in the model boilers is believed to be an accelerating characteristic of these test vehicles.

5.1.1.3 Test Results The three tests conducted with prepacked crevices were completed and the test specimens have undergone metallographic and eddy current examinations.

Additionally, tube ID measurements were made with a diatest micrometer.

l The post-test, eddy current inspections indicated no flaws or dents on the tubes. However, the micrometer ID measurements did indicate some denting.

Table 5-2 provides results from post-test measurements. Figure 5-4 shows the relationship between average dent size and diametral crevice size. C The metallographic examinations intitcated that non-protective magnetite was found at the TSP simulants during testing. Oxide thickness measuremegts indicate a trend of[

lox'ide thickness with[

] crevice Size. Typical oxide structure is shown in Figure 5-5.

The oxide in Figure 5-Sa is from an unexpanded tube region while that shown in Figure 5-5b is from m expanded region. A protective-type oxide was also obsarved at some of the TSP simulants.

S-3

The fourth test in this series was conducted under cold leg conditions for 130 days using six tube support plate (TSP) simulators. The top and bottom TSP simulants were concentrically mounted and prepacxed with magnetite so that they might serve as references with respect to tne aggressiveness oT the denting solution. The positions adjacent to these prepacked simulants were occupigbg eccentrically mounted TSPs having nominal open crevice gaps of(

) The center two simulants were eccentrically mounted at regions of tube cxpansion which produced nominal crevice gaps ofb

]

The purpose of this test was to determine whether the reduced crevice gap caused by the tube expansion would lead to more rapid fouling and, subsequently, to more.apid dent initiation than at the unexpanded intersections. An addition of sludg[

]wY injected into the boiler daily.

The intent of this addition was to provide particulate matter which might serve to foul the TSP crevices.

Periodic inspections of the test devices showed no positive indication of denting at any of the originally open crevices although the prepacked crevices did exhibit denting. The latter phenomenon was indicated by internal cicrometer measurements shown in Table 5-3.

These measurements are also reported for the eccentric locations although the significance of the numbers is not so clear in these cases due to the wall deformations created by the set screws and, in the case of the expanded locations, by the expansion process.

A more definitive measure of the corrosion is given by the metallographic svaluation of the specimens (summarized in Table 5-4).

These data show no obvious trends and no increased concern for obtaining denting type corrosion relative to the expanded versus the unexpanded locations. The symbols NPM and PM refer to nonprotective magnetite (NPM) and protective magnetite (PM),

respectively. NPM is the type of oxide which might ultimately lead to denting.

This NPM'is shom in Figure 5-6. E sn

, a,b,c, 4 The results of the four tests performed in this series indicate that there is no increased concern relative to the propensity for obtaining denting.as a result of the tube expansion process. This is an important determination because one may infer from these tests that other corrosion processes which are primarily dependent upon chemical concentration, rather than material condition, are also no more likely as a result of the tube expansion.

5.1.2 Stress Corrosion Cracking (SCC)

A combination of polythionic acid tests, controlled potential electrochemical tests, and magnesium chloride tests were conducted to evaluate the effect of tube expansion on residual stresses.

Both nomiial and off-nominal tube expansions were tested.

5.1.2.1 Polythionic Acid Tests Tubing from two heats of mill annealed Inconel 600 was heat treated in the laboratory to cause sensitization and thus produce a microstructure susceptible to cracking in polythionic acid or sodium tetrathionate when stressed in tension.

C-ring specimens were prepared from one heat of the sensitized tubing (Figure l

5-7).

The specimens were stressed to various levels and were exposed to polythionic acid. When a specimen cracked, the exposure time was recorded and was used to plot a stress versus time-to-crack relationsnip. Other sections of the sensitized tubing were expanded into collars that had various hole diameters. These collar-tube assemblies were exposed on the 00 or 10 to polythionic acid, and the time to initiate cracking was recorded. The time to crack for each collar-tube assembly was compared to the C-ring curve to determine the maximum residual stress associated with each expanded tube.

Additional tests were conducted to determine the effect of expansion on the residual stresses of tubes expanded above the top of the tubesheet.

For these 0771c[0122c/060883:523 b6

fE tests, tubes were. expanded into collars 'such that the expansion included a portion of free standing tube length outside the collars (Figure 5-8).

These latter tube-collar assemblies were exposed from either the OD or ID to polythionic acid, and the residual stresses determined as cited previously.

The results of these tests showed no increase in 00 stress for expansions up

}a, c. e. -the maximum expansion tested (Figure 5-9).

Similarly there to(

was no increase in stresses on the ID with expansions of up to approximately

[.

] *(F ig.e

.c ure 5-10). Thus, in the rarge of expansions proposed for

_ a c, t.

the preheater modifications, the hydraulic expansion j

residual stresses on the 00 or ID are insensitive to the fina! expansion diameter.. In. addition, there is a significant margin between the maximum proposed baffle plate expansion and the magnitude of the hydraulic expansions that will cause an increase in residual stress.

5.1.2.2 Magnesium Chloride (MgCl ) Tests 3

Additional residual stress tests were conducted to evaluate the expansion of tubes at the preheater baffles. Our to the limited quantity of sensitized Inconel 600 tubing available and to simplify the proc; dure, magnesium chlorice testing of Type 304 stainless steel tubes was used in lieu of polythionic acid testing of Inconel 600 tut'es. Westinghouse experience has shown that with respect to determination of the location, orientation, and relative magnitude of residual stresses caused by processing variables, such as tube expansion, MgCl testing of austenitic stainless steel is a suitable substitute for 2

testing Inconel 600 with a corrodant such as polythionic acid.

The objective of these MgCL2 tests were tests gave similar results as the Inconel 600 a.

to confirm that the MgCl2

- polythionic acid system for the baffle plate expansions.

b.

to confirm that no unexpected residual stress conditions result from nominal hydraulic expansion of tubes at baffle plates made using prototype, manually-operated field equipment, and c.

to determine the effects of off-nominal expansion conditions.

0771c/0122c/060883:5 24 6-6

o 5.1.2.3 Testing Nominal Conditions For 'tcsting nominal tube expansions, short lengths of-Type 304 stainless steel tubing, nominally 3/4-inch OD by 0.043-inch wall thickness from two heats of steel, were expanded into carbon steel collars, 3/4-inch long. The ID's of

~

the colles were drilled to simulate the range of hole sizes and ID surf ace roughness expected in the preheater baffles of Model 04, DS, and E steam generators.

Initial testing was performed using expansions made using a constant pressure process whereas later testing was performed using the controlled pressure process. The hydraulic expansion pressures for the constant pressure process ranged from 9100 to 13,300 psi whereas an.inprocess feedback system is used with the controlled pressure process. and results in the.use of lower expansion pressures. The expansion length, exclusive of portions of the expansion transitions, was v ithin the 3/4-inch long collar (Figures 5-11 and 5-12). The elastic spring-back of the tubes was sufficient to allow the tubes to be removed from the collars without affecting' the tube r

surf aces. A schematic diagram of a-test specimen is shown in Figure 5-13.

., a., c, t-e The expanded tube specimens were exposed Q

, to a boiling _ aqueous

~

in accordance with t'he' procedures o'utlined in ASTM solution of MgC12

, ~

Recomended Practice G36. The 00 surf: ice of some specimens was exposea to th.e Each-MgCl2_ solution whereas the ID surf ace of other specimens was exposed.

specimen was remaved from the test solution periodically for dye penetrant

. inspection for the presence of cracks. When detected, the location and orientation of the cracks as well as the exposure time of the specimen to the were recorded.

MgC12 r

For the initial tests made with ' tubes expanded-by the constant pressure process, the aD range of nominal hydraulic expansions tested was fro (

'For these tests, the specimens with OD exposures did not crack in times up to(

be eas all of the specimens wit'n ID exposure cracked at theexpansiontransitionbetween{

] Ta'o e 5-5). A comparison of these times to crack with-those of the calibrati5n C-ring curve showed that the stresses at the OD surf aces were quite;10w, i.e., [

[.

t$e stresses associated with the 10 surf ace at the trsnsitions were estimated to s

0771c/0122c/060883:5 25 6-7 1

/

a, c,8-beidft'heorderof{-

3(Figure 5-14). ~ The results also showed that.

.the time toL crack, i.e., the magnitude. of pe stresses, for hydraulic Lexpansionbetweenh

[abwere-thesame. Thus, the residual stresses were not Lincreased with increased AD within the range tested. These MgCl 2 test results were completely cons.istent with the polythionic acid tests cited previously~~and. indicated that the residual stresses associated with the expansion planned within the preheater baffles would be low.

2 n.

'5T.2!4f:TestingOff-Nominal' Conditions

~

.a' Tests of stainless steel tubing specimens with various off-nominal expansion s

=

-con'ditions were conductes,. Tlibse tests. included expansions that simulated the following' off-n'ominal expansi6n cor>:itionsr

-. 4 i

n a* fsf

('

A.'

Tubes with Simulated Insbilation Scratches, t*

Te!

When tubes are inserted into the, steam generabrs, the holes in adjacent tube support plates or baffles may $e slightly out of alignment. This misalignment can _cause the tube su port plates or baffles to scratch the tubes with contact loads ranging fecm zero to an approximate]

} a,cA

^

.g

-maximum. Stainless steel tubes were' inserted into hoies in a' carbon D

- l O-steel tube support plate segment and were scratched with contact loads of

] (e, c,aFigure 5-15). These specimens were then

[

[Enidraulically expanded to nominal expansions of[

]ab.

J' f

4 M T,yjtp b,.',Allofthesescratchedspecimensweretestedwith00exposurestoMgCl2 h g]fgover 4(W hours with'ro,cfacking.

Based on these test resul.ts, 0 F stratching of the tuba.'s/ y'!#tae tube support plates should not increase L'

t L

g rs

.s; p

/, - ? the. resideal stresse's 'of th5' tUoes at the baffle. expansions.

'/

Y' f", '

c'

, -).

~

> _ f lg.

.%/

BM4s'vmetr'ic' Expansion of ~ Tubes - Tube rMfset in Hole with Axis of Tube

? 'P$rallel with Axis of Nole

,j

~

1

~/

4 g.

This condition was 'o'esIgned to simulate.a condition that may occur due to 4

v.

~

jd J/

,z slightemisalignmentJf;the baffle plates. Tube segments were inserted pc

,1 h,'-

m Ai i

$ j

' gi) b Ji-(; f a

Q122c/06080kTd2E

6-8 M-

, l ')77'1c]J y vf y

L

~ '

~..., f,i f

} d. hQ

ll ?,

~

. -.u- - -

y.;

[:

m,

' h;,

'i into simglated baffle plate holes and exptnded while a side load was imposed on the tube'(Figure 5-16).

x,S k t

y

~. C.

Assymetric Expansion of Tubes - Tube Tilted in Hole with Axis of Tube not I arallel with.;xis of Hole P

(p)

This condypion was designed to simulate a' condition that may be caused by misalignment of the baffle plates. Tube segments were inserted thrcugh

- sidiulitecMaffle plate holes and expanded while a side load was imposed on one end ofsthe tube (Figure 5-17).

p

-s D.

Tubes with Short Axial Scratches

>M k

i The purpose of Thrh, e specimens was to simulate short axial scratches that n

may be' caused by" movement between the tube and the baffle plates.

E

]A*i1s expected between these components

' Movement.ofapproximately[t

~ due to ' differential expansion of the materials during thermal transients

- plant startups and shutdowns"Tn particular.

,h

m. /$

Tubes were-inserted 'and hydraulically expanded into a hole in a baffle

< ;[j -.

g riate segment or a carbon steel plate simulant while a side load was 0

t, applied to the tube. The tube was moved cyclically f or lengths of L

(

, 3r"e'ldt$ve to the hole in the baffle segment (Figure 5-18).

. n f..J+g

.hThis movement was repeated for {

]cy'cl'es, the expected number of

\\

.a

w%

u.

.\\

cycles of th,fs type during the life of a plant. These specimens were the\\n exposedJto UgCl 2 m. s from the 00 surface only.

i 3

l

s.

'I m

\\

s

'-/

~

p

- E. < Tubes with Simulated Wear Scars

?.

h i

Simulated wear scars were machined into the OD surf ace of tube

~'

Ispecimens.. The lengths of the tubes with the simulated-wear scars were

~

m i 4 p9

.m dp then ' expanded within a steel collar.

s (w:

s uy y

I dhk

go.

~

0771c/0122c/060883:527 54 l 9 I

a. ; k.'.

y_

F.

Expanded' Tubes' with Simulated Dents -

,*st Tube specimens were -expanded fto a nominal aD -_of[

g The expanded tube ' portions 1are. then dented by ovalizing the expanded r.

e,c s length by(

j. These specimens were then tested with' both 00' and ID ' exposures to MgC1

2 r:

. G.

Tubes Expanded with Simulated Sludge in Tube-to-Baffle Plate Crevice The purpose.of-these specimens was to simulate a condition in which the steam generator has operated and corrosion product or deposits have partially 'or.completelyifilled the tube-to-baffle plate crevice prior to expansion.-l Tubes were inserted through a baffle plate crevice to form the, conditions shown in Figure 5-19.

The tubes.were then expanded using the reference expansion; process. ' Specimen tubes were exposed to boiling

~MgCl with both'ID or 00 exposures.

2 Testing of specimens _with off-nominai expansions was initiated. Unexpected t

~ accelerated cracking was detected with specimens that contained simulated wear scars prior to expansion or with simulated dents _made af ter expansion (Tables

~

5-6, 5-7,. and 5-8).- Specimenswith(

2pansions that were included in' the tests as: controls also cracked in unexpectedly short times.

Testing of 'the:off-nominal expansions was temporarily suspended while tests

' were conducted to determine the cause of the accelerated cracking and to confirm the_ validity of: the results of the' previous tests.on ncminal expansions.

- 5.1. 2 =. 5 Confirmation of Previous Results on Nominal Expansions Because only a limited quantity of the previously used heat, No. 8974, remained,: a new heat of tubing, No. 500160, was -acquired. This tubing was obtained.in the mill -annealed condition and was then stress relieved by heating to' 1600~F f or 6 minutes. This stress relief treatment was performed to. assure that the residual stresses in the tubing were uniform and unaffected 0771c/0122c/06J883:5 28 g

by material manuf actur_ing processes, such as straightening and polishing subsequent to the mill annealling process.

Specimens from this new heat of tubing were expanded to aD's ranging from

(

['i ese specimens were exposed to MgCl2 on the ID surf ace and were inspected by dye penetrant after 2, 4, 8,12 and 16 hour1.851852e-4 days 0.00444 hours 2.645503e-5 weeks 6.088e-6 months exposures and at 16 hour1.851852e-4 days 0.00444 hours 2.645503e-5 weeks 6.088e-6 months intervals thereafter to a total accumulative time of 144 hours0.00167 days 0.04 hours 2.380952e-4 weeks 5.4792e-5 months . The results of these tests are summarized in Figure 5-20.

Crack

]a e e.a6 expansion indications were detected in one specimen that had a{

at the end of 2 hours2.314815e-5 days 5.555556e-4 hours 3.306878e-6 weeks 7.61e-7 months whereas four other specimens with this same size expansion were tested 144 hours0.00167 days 0.04 hours 2.380952e-4 weeks 5.4792e-5 months without significant indications and two additional specimens were tested to 80 hours9.259259e-4 days 0.0222 hours 1.322751e-4 weeks 3.044e-5 months with no indications. All specimens with small expansions were tested 144 hours0.00167 days 0.04 hours 2.380952e-4 weeks 5.4792e-5 months with no significant dye penetrant indications.

Although there is not a one-to-one correlation between the results of the previous tests of nominal expansions conducted using heat No. 8974, the results of the tests of nominal expensions with this new heat of stainless steel do, in general, confinn the earlier results and show that no unexpectedly high stressas are expected with the nominal baffle expansions.

tests with the old heat (No. 8974) have Further, a limited number of MgCl2 been conducted on tubes with nominal expansions ranging from(

)

j

}. i$ree of these expansions with aD's of[

] *J '"

exhibited the accelerated cracking.

Four specimens with AD's less than(0.030 har'e not cracked in exposure times to 144 hours0.00167 days 0.04 hours 2.380952e-4 weeks 5.4792e-5 months . Additionally, four..e expansions in specimens from this heat with nominal aD's of(

].

.were l

tested to 64 hours7.407407e-4 days 0.0178 hours 1.058201e-4 weeks 2.4352e-5 months with no cracks detected. These results, simnarized in Figure 5-20. show a general inconsistency in tests with this heat of tubing.

I No unambiguous cause for the unexpected accelerated cracking has been established. One possible cause is that the residual stresses within the l

tubing after manufacture were not uniform.

l l

0771c/0122c/060883:5 29 S-il l

e l

5.1.2.6 Comparison of Residual Stresses in Nominal Baffle Plate and Tubesheet Expancions Three types of tests have been conducted to compare the residual stresses associated with nominal baffle plate expansions with typical tube sheet expansions. All three types of tests have shown that the residual stresses associated with baffle plate expansions are less than those associated with' tubesheet expansions. Two specimens containing tubes of the new stainless steel heat, No. 500160, were mechanically rolled into Inconel 600 collars to simulate expansion of tubes into a tube sheet. These mechanical rolled specimens and two specimens with(

]$NI e plate expansions were to exposure times of 80 tested simultaneously in the same container of MgCl2 hours. At the 80 hour9.259259e-4 days 0.0222 hours 1.322751e-4 weeks 3.044e-5 months inspection, dye penetrant indications of cracks were detected in both of the mechanical rolled specimens whereas no crack indications were detected in the two(

]ad ffle plate expansions.

A second series of tests were conducted using a heat of Inconel 600 sensitized to be susceptible to cracking in sotium tetrathionate (Na S 0 I'*

246 Sections of this sensitized tubing were mechanically rolled or hydraulically expanded at[

]$n steel collars to simulate nominal (

[besheet expansions. The expansion zone for these tubesheet expansions was contained completely within the' steel collars.

These tubesheet expansion specimens as well as nominal baffle plate expansions withaD'sof(

3 we,c, e,re exposed on the tube ID to a 0.1 a

molar aqueous solution of Na b 0. Crack indications were observed in 246 the tubes with the tubesheet expansions at the end of the first exposure period (4 hours4.62963e-5 days 0.00111 hours 6.613757e-6 weeks 1.522e-6 months ) whereas no crack indications were detected in the baffle plate expansions until the end of the third exposure period (12 hours1.388889e-4 days 0.00333 hours 1.984127e-5 weeks 4.566e-6 months ), Figure 5-21.

These results also indicate that the magnitude of the stresses in the baffle plate expansions are lower than those associated with the tubesheet expansions.

50dium tetrathionate (Na2 4 6) will react with sensitized Inconel 600 50 in a manner similar to polythionic acid. Laboratory tests have shown that the Nap 4 6 will normally result in more reproducable results than tests S0 with polythionic acid.

0771c/0122c/060883:5 30

For the third test to compare the residual stresses associated with nominal baffle plate expansions with tubesheet expansions, an electrochemical controlled potential test was used.

In this test, the mill annealed Inconel 600 spec,imens were immersed into a 10 percent aqueous solution of sodium hydroxide (NaOH) and heated to 600'F. The specimens were then polarized at

+140 millivolts and held in the polarized condition for the duration of test.

As with the polythionic acid or sodiun tetrathionate tests with sensitized Inconel 600, the specimens will crack at locations of highest. tensile stress, and the time-to-crack will provide a qualitative comparison of the magnituoe of the stresses.

In this test, the stresses associated with a tube mechanically rolled into an Inconel 600 collar, a tube hydraulically expanded to simulate a[

[d nominal baffle plate expansion, and a C-ring specimen stressed to above its yield strength were compared. After 10 days of test, the baffle plate expansion was not cracked whereas the tubesheet expansion contained short axial cracks at the roll transition and the C-ring specimen contained cracks that had penetrated approximately 75-percent through-wall. The results of this test show that the stresses associated with the baffle plate expansion are less than those of the tubesheet expansion, which in turn is significantly less than the yield strength of the material.. Again these results are completely consistent with the MgCl and Na S 0 tests that compared 2

246 the residual stresses associated with the tubesheet and nominal baffle expansions.

5.1.2.7 Tests with Corroded Baffle Plate Simulators Carbon steel collars were coated on the outside surf aces (not the ID) with an epoxy material to act as an electrical insulator.

Sensitized Inconel 600 tube sections were inserted through the collars, and the assemblies were immersed in a 1 molar aqueous solution of sodium chloride (Nacl) with a pH of 4.0 adjusted with hydrochloric acid (hcl). A 4.5 ampere direct current was impressed on the assemblies for 20 hours2.314815e-4 days 0.00556 hours 3.306878e-5 weeks 7.61e-6 months to corrode the carbon steel collars.

The corrosion product filled the crevice between the tube and collar locally.

0771c/0122c/060883:5 31 6-M L

o Subsequently, the tubes were expanded into the collars using the controlled pressure expansion technique. The expanded tubes were exposed to a 0.1 molar

. sodian tetrathionate (Na2 4 6) solution for up to 12 hours1.388889e-4 days 0.00333 hours 1.984127e-5 weeks 4.566e-6 months . For tube ID b0 exposures,.the tubes were tested with the collars in-situ whereas the collars were re" loved from the tube prior to the initiation of the 00 tests.

At the 12 hour1.388889e-4 days 0.00333 hours 1.984127e-5 weeks 4.566e-6 months inspection, dye penetrant indications were detected on the ID surfaces. No dye penetrant indications were detected on the 00 surf aces.

The ID tested tu'be-collar assemblies were sectioned and the tubes removed from the collars for profiling and for metallographic examination. The metallographic examination confirmed the presence of cracks in the ID surfaces. The location and extent of the cracking was similar to that observed during metallographic examination of nominal expanded tubes. Thus, the presence of the corrosion product in the tube-to-collar crevice did not cause any apparent increase in the stresses associated with the corroded baffle plate tube expansions as comparea with the nominal tube expansions.

5.1.2.8 Influence of Temperature on Stress Corrosion Cracking Both laboratory test data and operating plant experience confirm the inhibiting effect on stress corrosion of mill annealed Inconel 600 by operating in the temperature range to which the expanded tubes will be exposed.

A.

Secondary Side Considerations - Figure 5-22 presents laboratory data of mill annealed and thermally treated Inconel 600 C-rings, stressed to several levels, and exposed to a highly aggressive caustic solution at temperatures between 550'F and 650'F. The trend toward lower corrosion rates with decreasing temperature is observed, particularly for the highly stressed mill annealed material. The laboratory conditions were selected, of course, to accelerate any attack.

Actual operating plant experience has confirmed the benefit of lower temperatures. Figure 5-23 illustrates the experience of one operating plant which had experienced caustic attack in the hot leg tubesheet J

crevices, operating at a hot leg temperature of 590'F. Reducing the hot 0771c/0122c/060883:5 32 g /4

leg temperature to the previous. cold leg temperature of 557'F essentially arrested the corrosion. Raising the hot leg temperature to 575'F increased the rate of tube plugging, but again, reducing the temperature to 557'F resulted in a reduction in the number of tubes to be plugged.

Clearly, the effect of a reduction of maximum temperature of the -hot-leg to the value of 557'F was responsible for drastically arresting the rate of attack.

In further support of this, there has been no known instance of caustic SCC originating on the cold leg side of any Westinghouse operating plant steam generators.

B.

Primary Side Considerations - Laboratory test data and operating plant experience indicate that at cold leg temperatures the rate of " pure" water stress corrosion cracking is extremely low. The term " pure" water is generically meant to refer to primary water, containing Li and B, with or without H overpressure, and also AVT water. Laboratory test data 2

are summarized in Figure.5-24, showing what is judged to be an Arrhenius relationship of crack initiation time with temperature, i.e., inversely proportional to the. absolute temperature. Much of the available data at several temperatures has been. generated at Brookhaven National Laboratory (BNL) by D. van Rooyen,( ) utilizing severely deformed reverse U-bends (RUB) to initiate cracking at reasonable time periods. Curve A is BNL data from 0.03 w/o Carbon (C) tubing. These data are presented because 0.03 w/o C is the nominal carbon content for the Model 0 steam generator tubing.

If the BNL curve is extrapolated to 554*F, based on the Arrhenius relationship, crack initiation, for severely deformed tubing, would be predicted at about 13 years.

l l

Other data would extrapolate to significantly longer initiation times, l-assuming the same activation energy as that of the BNL curve.

Westinghouse primary water reverse U-benG data, shown as "B" would extrapolate to over 44 years.

"C" is Westinghouse data, in pure water, of reverse U-bends.

"0" are C-rings, stressed to 150 percent of yield stress (approx. 55 ksi) which have not cracked af ter 8.4 years of exposure at 630'F; an extrapolation of these data to 554'F would predict initiation times greater than 160 years. One explanation of this may be i

the lower degree of deformation of the C-rings as compared to the reverse l

0771c/0122c/060883:5 33 5-/S i

~

U-bends. While the role of deformation or stress, or strain, has not been rigorously established, one would expect that the tube expansions would behave more like the C-rings than the reverse U-bends, ir, regard to both stress and strain.

The operating plant experience is very explicit with regard to the temperature effect on " pure" water stress corrosion cracking. There has been no known primary water SCC on the cold leg sides of any Westinghouse operating plant.

5.1.3 Tube Wastace A remaining potential concern is tube wastage.

Since the transport processes and chemical environment resulting in tube wastage are currently unknown, it is difficult.to reach definitive conclusions. However, expanding the tubes into the baffles makes the geometry more similar to tubesheet joints, and there are no known instances of wastage in tubesheet crevices. Consequently, although wastage concerns cannot be precluded, there is little to suggest that expanding the tubes should aggravate the potential for tube wastage at these locations.

5.'2 Structural Analysis 5.2.1 Requirements-The steam generator tubing is a part of the primary pressure boundary and, therefore, must be analyzed in accordance with Subsection NB of Section III of U) the ASME Boiler and Pressure Vessel Code

. The analysis performed is an update of a portion of the generic steam generator stress report to address the effects of the tube expansion process. The expanded tube configuration is evaluated for Design, Normal, Upset, Emergency and Faulted conditions per Subsection NB requirements.

In addition, the tube plugging margin per Regulatory Guide 1.121 has been evaluated 0771c/0122c/060883:5 34 g_g

'5'.2.2iMethodology 5.2.2.1 Finite Element Model:

Two separate -structural models were used to' assess the effects of tube expansion both at-regions' away from the expansion-and at the local : region of the tube' expansion itself. :The global tube model-was a WECAN finitel element

- model:of-the outermost tube consisting of 3-D pipe elements (Figure 5-25).

This model was. used to evaluate tube stresses at the secondary side of the tubesheet and was' used to generate interaction loads for input into the _ local stress; analysis.

The local model was a WECAN axisynnetric, isoparametric, finite element model L-

. representation of the expanded tube. Figure-5-26-depicts this model. -The local model was developed to perform the local stress evaluation in the area of. the-tube / baffle interf ace..

-. Boundary conditions for the global -model included:

-a.

Tube temperature distribution-axial and radial

-b..

- Baffle plate displacements - radial and axial when appropriate c.

Tubesheet rotations

?

Tube analysis using -the local model consisted of performing several steps:

Step 1:

Analyze the model without regard to any local tube / baffle plate hole interference. The boundary conditions are:

4 Tube temperature distribution - The region inside the baffle a.

A plate.was considered to be at the primary fluid temperature.

step change in tube temperature distribution was assumed at the top and bottom edge of the baffle plate.

2 b.

Internal pressure. corresponding to transient conditions.

C

-c.

Axial load based on tube aP.

0771c/0122c/060883:5 35 Sr - /7 sv

+,e-,-w

,-<w-

,-w w

--,+e

-,-,r.~

,3--..w-,-

vm-,--4g-i,w.--c--re,

-rype.,~.-,,-e,,,

~ ~ -, -+

.=m

,,%-*e--w---wey.,

e----

w,-,-v-

.Steo 2:

Determine from the results of Step 1 whether or not local

^

tube! baffle plate hole interference exists.

Step 3:

Rerun those transients which resulted in tube / baffle hole interference and include the allowable displacements for those nodes that showed interference as radial displacement constraint boundary conditions (axial interaction loads from global model are also included as boundary conditions).

Step 4:

C'ombine the interaction moment ' induced stresses with the transient

" thermal" stresses calculated in steps 1 through 3.

The. interaction moment-induced stresses are calculated in one of two ways, depending on whether or not interference bitween the expanded tube and the baffle plate exists for the transient condition being considered.

5.2.2.2 Thermal Analysis The transient conditions applicable to the steam generator tubing were evaluated to-determine their characteristics. The total list of transients considered ~was subdivided into transient groupings. Within each grouping, the most severe transient was chosen as an " umbrella" transient.

The selection and grouping procedure was based on the following criteria:

a.

Secondary and primary pressure differences b.

Secondary and primary fluid temperature differences c.

Feedwater f1w rate d.

Number of occurrences (This criteria is important in the case of Normal and Upset transients because of f atigue considerations.)

Table 5-9 ~ presents the grouping arrangements and umbrella transients for the operating conditions. The number of cycles for each umbrella transient is 0771c/0122c/060883:5 36 6-/J'

equivalent to the sum of the individual transient occurrances within the grouping.

5.2.2.3 Stress Analysis Tube f atigue analyses were performed at two locations: at the expanded region and at the intersection of the tube / tube sheet secondary side interf ace. The tube at the interf ace with the secondary side of the tubesheet was evaluated via hand calculations. The expanded region of the tube was evaluated with the program WECEVAL(5),

The Code fatigue evaluation was performed for three post expansion diametral

]o,c e.

diametral. For each initial gap condition, gap sizes:

the WECEVAL runs were made for expansions at the B and D baffle plates to evaluate the f atigue usage f actor at either the B baffle or D baffle plate.

The stress analyses were performed on an elastic basis, using Code methods, and included total tube stresses foF each transient and steady state condition evaluated.

The most severe local tube / baffle plate intersection was analyzed on an elastic-plastic basis and resulted in a maximum reported total tube equivalent stress of approximately 45 ksi. This total stress included effects of:

a.

Temperature, including through-wall gradients b.

Pressure c.

Local tube / baffle radial mismatch d.

Axial interaction loads induced by tube / baffle interference 5.2.2.4 Assessment of Scar Def ects/ Stress Concentrations The purpose of this analysis was to evaluate the effect of stress risers on the fatigue analysis discussed in section 5.2.2.3.

The sensitivity of the

' f atigue analysis to scratches or wear scars in the circumferential or axial

' directions was determined through the use of theoretical stress concentration f actors, K, in either the circumferential direction, KTY, or the axial T

0771c/0122c/060883:5 37 5-/r

~

I to 2~.95 or K to 2.70, a cumulative direction, KTZ. By limiting KTY TZ f atigue usage f actor less than the ASME Code allowable of 1.0 was achieved.

The theoretical stress concentration factors allowable were then compared with stress concentration f' actors for actual wear scar geometries obtained from tubes removed from operating preheat steam generator plants.

Bounding geometries based on profilometer traces of wear scar surf aces were used. The largest stress concentration f actor was calculated to be 2.35, with 90 percent of the values being 1.85 or less. A series of f atigue tests was also performed on the pulled tube samples, showing that the ' accumulated usage during formation of the waar scar was 0.0, i.e., the only damage done to the

-tubes was of geometry, resulting in stress concentration.

5.2.2.5 Tube Plugging Analysis The tube plugging margin' evaluation was reviewed to identify those portions that coul_d be affected by a change in tube geometry due to tube expansion within baffle plates. The portions that could be affected are:

a.

Minimun wall thickness calculation l

b.

External collapse pressure calculation c.

Burst strength calculation 1) verification of leak before break l

2) margin to burst An -increase in minimum allowable wall thickness from(

l

_at the expanded regions of the tube was determined to be sufficient to meet the mininum wall thickness requirements. This[

] bickness was also adequate to satisfy external collapse pressure and burst strength criteria.

Therefore, the tube plugging margin for 04, 05 and E steam generators with l

expanded tubes decreased by approximately one percent of the wall thickness at the expanded zone.

i 077.lc/0122c/060883:5 38 6-20

e 5.2.31 Conclusions The following conclusions have been identified on the basis of~ the Model D4/05/E steam generator tube ' analyses:

The maximum calculated usage factors for the locally expanded region of a.

the tube were-all less than the Code allowable of 1.0.

. b.

Tube f atigue' limits are satisfied for wear scars in the expanded tube region.

Stress. concentration f actors obtained from worn tube measurements were less than the allowable factors for which cumulative f atigue usage would reach 1.0.

c.

The primary stresses in the expanded region of the tube were not significantly affected by the expansion.

d.

The tube plugging margin was not significantly affected by the expanded region.

0771c/0122c/060883:5 39 (T_g g

-a e

n

-e.--,-

--,,s.

~

w - -

n-m

i.

Table 5-1 Comparison of Laboratory Denting Test Conditior.s with Steam Generator Conditions Laboratory Steam Generator Tube Materia!

Inconel-600 MA Inconel-600 MA TSP

Material A-285 Carbon Steel A-285 Carbon Steel TSP Configuration Cylindrical Hole Cylindrical Hole Tube Diameter, inch 0.750 0.750-

,i c, TSP Hole Diameter, inch

^

Radial Crevice Gap After Expansion, inch ~

~

Primary Temperature, "F 555 - 568 550 - 564 Secondary Temperature, "F 535 - 540 535 - 545

" S

~

aP Across Baffle, psi

{,

i:

Tube Support Plate 3

i 4

L I

i i

t 0771c/0122c/060883:5 40 g_EZ

i Table 5-2 Test Results from Tube Support Plate (TSP)

Simulants with Prepacked Crevices Radial Average Maximum Minimum Crevice' Dent Corrosion Corrosion TSP

, Tube 00 TSP ~ID Size Size Product Product No.

(inch)

(mils)

A.

Hot leg Conditions

,b C E i

B.

Cold Leg Conditions

(*, b. C t i

-~

i

TSP-4 separated on cutting.

0771c/0122c/060883:5 41 4;.;13

+

Table 5-3 Summary of Denting Data Following Completion of Denting Test.No. 4'

. TSP

.Diametral 2.verage ' Dent '. Maximum Dent Number Packed / Unpacked Crevice' Size (mils)

Si ze '(mils)

Size (mils) hkC,6 1

Packed 2

Unpacked 3

Unpacked 4=

Unpacked 5

Unpacked 6

Packed L

4 0771c/0122c/060883:5 42

,7

. _. _. _ _... _... -. - _.,. ~ - - - _. - _

Table 5-4 Summary of Corrosion Data Following Completion of Denting Test No. 4 Diametral:

Maximum

-TSP-Packed /

Crevice Type of Penetration Number Unpack ed Size (mils) Corrosion *

(mils)

Comments

> b '2" s

1 Packed

~2-Unpacked

-3

. Unpacked.

4 Unpacked 5

..; Unpacked ~

6 Packed

_L

NPM - Nonprotective Magnetite

-PM - Protective Magnetite r

l l.

t s-25 0771c/0122c/060883:5 43 o

.. ~. -

. ~ _., _. -. _..

~:..

..a Table 5-5 Comparison of 00 and ID Cracking of Hydraulically Expanded Tubes

Expansion-Time to a0, incn Crack, hours Crack Location a,b c, c.

i 00 Surf ace

Heat F121019 e

0771c/0122c/060883:5 44 g_zg

Table 5-6

'w~-r Tests on the OD of Tubes with Machined Resu.ts of MgCl2 Wear Scars Prior to Expansion

e, h c, e.

Constant Pressure Expansion Heat No. 8974 t

4 1

5 0771c/0122c/060883:5 45 g,77

. Table 5 Tests on the ID of Tubes with Machined

Results of MgC12 Wear.Scarr Prior to Expansien*

a, b, c, c.

\\

t 1-i-

24 s.

If.'.

~

0771c/0122c/060883:5 46

r Table'5-5 Comparison-of 00 and 10-Cracking of Hydraulically Expanded Tubes *

. a, in, C, t.'

5 J

4

.i i

1 5

t i,

Heat F121019,

L i

I i

l.

, 0771c/0122c/0607,83:5. 44 g.2$

.;.,.___..._._.._..,___._.a.-.____.._.._...___.._.

9;

. Table 5-6

-Results, of MgCl Tests on theLOD.of Tubes with Machined-2 Wear Scars Prior to-Expansion

d, b. C. e

Constant Pressure Expansion'

. He'at No..8974 1

d i

~l i

t i-4 1

~

0771c/0122c/060783:5 45

- + < <

--v e r-w -4,e-r-yw-.v4+

--.9r----r s ew-s-w-+-sv--w.m--

-mm--

v,s v te.e-s-wt--wo<---p w m e n - w, w r-A -~ m 4 W r-,-

ww-w g - w~ rw -s y 4 g - w ww w-ver m y v-v9ee-w'

'a

-4

+

,4 Table 5-7'

~

Results of MgCl Tests' on the ID of' Tubes with Machined.

~-

2

-- Wear Scars Prior to Expansion *

. b.f i *

. Constant Pressure' Expansion -

Heat =No.-8974 5

p X

I

/

f i

l l

l i-

[

=

'0771c/0122c/060783:5.46 6 3I I'

r

Table 5-8 Results of MgCl2 Tests on Dented Nominal Expansions A. h. c. e.,

Constant Pressure Expansion Heat No. 8974 s

j i

l 1

m.

Dent Size = 01-D2 Qx 1,

f, '

l' D.

pg

\\

1 Mi o

i 0771c/0122c/060783:5 47 5-32

Table 5-9 Transient Groupings Umbrella Transient Transients Included i

-l'0'0 percent load Load / unload 15-100 percent Loss of load at 40 sec Small step increase Small step decrease Reactor trip Control _ rod drop Loop out of service Load / unload 0-15 percent power FW cycling Loss of power Loss of flow Inadvertant startup of inactive loop Inadvertant safety inj.

~

Loss of load i

Loss of load at 120 sec Control rod drop loss of power Inadvertant startup of inactive loop Inadvertant safety inj.

Loss of load Small step load increase Tmall step load decrease Loop out of service Load / unload 0-15 percent power FW cycling Large step load decrease Large step lead decrease Excess feedwater Excess feedwater Excess bypass feedwater Excess bypass feedwater 25 percent power 25 percent power Forward flushing Forward flushing at 32' Forward flushing at 200~

Forward flushing at 250' Heatup/cooldown Heatup/cooldown Turbine roll Inadvertant RCS depressurization Reactor trip (zero thermal stress Inadvertant RCS depressurization

+ press)

Turbine roll OBE OBE 1

0771c/0122c/060883:5 48 z,3j3

a Table 5-9 (cont.)'

Transient Groupings Umbrella Transient Transients Included Pri hydro' Primary hydro Prima ry.~ 1esk Primary leak Secondary hydro Secondary hydro Tube' leak (340 psi)

. Tube leak _840 psi Tube. leak 600 psi Tube leak 400 psi Tube leak 200 psi Secondary leak Secondary leak c

4 1

9 t

0771cIO122c/060883:5 49 di-3dj

/

-i,

~... l i

/

~.

r y-(

ri3 Liconel Frit

'j'

+

-(m=

[.

'l r

Carbon Steel TSP Simulant ~

e Inconel Tube %

/

o

/.-

l l

i E-l i

Eccentric TSP

~

'_M

,i Simulant e...-

-m j,;

4 e

f f

H Packed TSP si i

Simulant j.

s,

-=

i l

/

6 I

/(

o u

>f i

1 l

Sludge Cup fd

. d'j

/

Figure 5-I Typical. Test' Specimens used in Single Tube Model Boiler Tests

..n e.-n-,

r,.,,,,,,n,

,.,.n,

.cn.

'a~

=-

/,.. e

.c

. r,,

,, em s

Y

~

t'

./

M I

-)) E 13

!?

j.j

/, y = 0.125 a --4.25s.

O

.?

/

y on s'

12

[ RATE = 3.8 Mil.S Pt!R 30 OAYS i

3, 11,

,n

~

f' d

/

y / sf f 10 '

,' z,'

r ',

i O r,-

- o A

,,e

,n y :9 t.)

.o

~ *,,

n 2

O O.

w\\

1 a s a.

l

',4.-...

s y

I w. v-or i

- 1 r-

/e t t '!

I<* 'o O

cp O

1 ['..._g5

.ot O

o o

- '.E O

OO

\\

4 j;'

C i

t n

3 z

o h

/

2,-

\\

l

]'

ih ;

1

~ ~.

i

. 's il l'

sl l

l" l'

l-l l

l l

o-0 10 20 N'.?ic 40 50 00 70 30 90 100 110 120 DAYS AFTER ONSET OF DENTING l

t l

l L.c Figure 5-2

[

t' Average Denting Rates Versus Time l

in Reference Denting Chemistry L

/.

l l

l

4 i

i 8

y

' - 1 s

< s6

(

'1 i '\\

s 5.0 i

Make-up Tarm Reference E. 4.0 c.enico Cne=siry j

~

N O.2pom Cl from Seawater O

0,' loom C[I from CuCl2 o

o j

E 3.o 2oom w :

2 50000 N H4 bl 5

1 55254 SAT 1

2.0 (n

cn o

5 O

c 1.0

<p Q

O

'O O

-C 25 30 35 40 45 50 55 60 65 s

o Superheat *F 1

~

~.

Y s

TO Figure 5-3 t '- '

?

-4 Average Dentir g Rates Versus Superheat

<t in Reference Denting Chemistry e

e,--

e-, -,,,,,,, - ~,, -, -

,,,w-,,,.w,,---.-

,,-n--

,,,,_,,,,-_,,,,,,..,,,,,,,w,-.,----

,,,,m-

-,,---,,,-,-.,,--.,,,,,,,y

0< b5 i

l t

d E

Figure 5-4 Average Dent Size Versus Time as a Function of Initial Diametral Gap i

.+

g e-v--.-..

._.3-,

vyp.-73,,

,9-m 9 %9.

9

-,9 y-g

,,,,_g

.yg-m mcw-. _. _,_ -, * - - -.. _ _.,..,-g__._,,eew-g-w.

--w+yw. w w w w--,yi-*

e-

S b,0, C i

2 f

sw Figure S-Sa Typical Oxide Growth at Carbon Steel Tube Support Plate Simulant (Unexpanded Tube, Prepacked Crevice)

dlb&L it t

I l

l l

I 1

I l

1 Figure 3-30 Typical 0xide Growth at Carbon Steel Tube Support Plate Simulant (Expanded Tube, Prepacked Crevice) i

I 1

6 96 l

Figure 5-6a Typical Nonprotectlye Magnetite Observed at Nonexpanded Tube / Baffle Plate Intersection i

l l

Figure 5-6b Typical Nonprotective Magnetite Observed at Expanded Tube / Baffle Plate Intersection 1

RESIDUAL STRESSES ARE DETERMINED BY CORROSION TESTS IN MGCL3 (STAINLESS STEEL) OR POLYTHIONIC ACID (INCONEL-600)

'c-A)

ESTABLISH EXPERIMENTAL CALIBRATION h4 CURVE

_ o-4 PLACE CALIBRATION TEST SPECINENS IN CORROSIVE ENVIRONMENT AND LOAD TO VARIOUS KNOWN STRESS LEVELS.

bM" DETERMINE TIMES TO FAILURE.

PLOT CALIBRATION CURVE OF APPLIED STRESS VERSUS TIME TO FAILURE.

k time to faMate -+-

l B)

TEST COMPONENT EXPANDED SAMPLE) d 2

PREPARE SAMPLE AND PLACE IN CORROSIVE MSI'.t need ENVIRONMENT.

DETERMINE TIME, T',

FOR RESIDUAL STRESS TO CAUSE SCC FAILURE.

ENTER CALIBRATION CURVE AT TIME, T' TO GET RESIDUAL STRESS (R' R

p l

1 1

[

Figure 5-7 Methodology for Determining Residual Stresses in Tube Expanded Regions

~

l i

.o Bulge

\\

)

8 i

I si i

8 e

t 9

I I

I E

i I

e I

I I

e 8

I I

f l

t I

1 e

I e

8 e

t Figure 5-8 Scfiematic of Typical Specimen Expanded Outside the Collar

e-,-

g -

--..,,,,,,w.,-

.,w, w.---,

-,-,-,------,-,,-w.

.,-n---+,,

a-,-,,.

Arrows indicate that data points are maximum stresses obtained at that tube expansion.

a, b, C, f.

1 i

t Figure 5-9 Results of Polythionic Acid Tests on the 0D of Inconel 600 Hydraulically Expanded Tubes e

,. - + - - -

e.~

,,,,, - --,e,-g,---

w

i Ab0l i si 1

J

l t

il' l

Figure 5-10 Results of Polythionic Acid Tests on the ID of Inconel 600 Hydraulically Expanded Tubes l

r1 C N'I -

l

\\

Du 1

q l

A I

Note: Transition regions of expandec zone have been exaggerated to show relationship to baffle D,

plate.

~

f

(

w s-

., c, e.

Change in diameter (60)

Crevice gap width (C)

Crevice gap depth (G)

Axial misplacement of expansion Expansion pressure l

l Figure 5-11 l

Schematic of Typical Expansion used in Residual Stress Tests l

i 1l,'

e, c

d

}

i i

e bu T

0 i

0 6

leno h

c c

n n

I i

d e

a e

d b

n u

a T

px e

f E

e l

o f

2 y f

h 1 l e

a t

l 5 a B

a l

f c

I n

ei f

e r

a l

t u u D

g a l

i r

a F d 1

i y

x i

l A

lau I

tc A

fo I

hc t

e k

l S

de la l

c S

e.

a-v3 2a E ?,

BC.-

I a!i ij4 l:14 i ii '-

al;)1

I

I I

I i

l A

i B

I I

I 1 3/4" c

1 3/4" 6"

Note: Transition regions of expanded zone have been exaggerated for display purposes.

Figure 5-13 Tests Schematic of Stainless Steel Specimen Used in MgC1 2

/hf I

i i

Figure 5-14 Stress versus time to crack for 304 SS C-Rings McC1 2 solution with superimposed time to crack for hydraulically expanded tubes

Baffle Plate Simulator Tube r._ _.:

Plate Pulled Full

=

~,

Length of Tube o

I Load Figure 5-15 Schematic of Fixture Used to Scratch OD Surface of Tube

il g/ //,//j/../,i

's',

/,'/

',..////,/ /.

f,,(s', '

,y

-l

/

E

//hl '/ jj i, ' */,

/

'/

aDmax Section A-A-Cross Section of Tube and Baffle Plate Prior to Expansion A

Baffle Plate Simulator

]

Tube I

I n

o d

- p/2 A. c, e.

p/2 aD,,x i

A Figure 5-16 Schematic of Fixture Used to Restrain Tube Against Baffle Plate

- - +, - -

+---ye,,,

.----.,,.--,,-a,.-,,

I Side Load =

\\

t I

i

(

l l

l Figure 5-17 Schematic of Tube Whose Axis is not Parallel to Axis of Saffle Plate Hole l

l t

l

Note: Transition r;gions of expanded zona have ceen exaggerated to show relationsnip to baffle plate.

o Tensila Machine Cross Head h

MotionofTube:(

P

. } %*i.

a P/2 Baffle Plate Simulator Fixed b

f ' N.. ', l ' k

/

s N'-

'\\ c,

\\. \\ ' ~ Ns, s,\\

NNs

j. i s i ll t t j

' 'y' t /

s s s s s

,//;;:, ; !, i

\\

s.'

/

\\ \\

/

N'

't P/2 4

'/

\\\\

Figure 5-18 Schematic of Fixture Used to Scratch Tubes Previously Expanded e.,

~

"A 3/4-inch OD by 0.045-inch wall Type 304 stainless steel tube l

l t

t Carbon steel collar drilled l

to allow nominally a 0.040-s>A inch diametral tube expansion O

Section A-A for specimen A.

Simulated sludge to be packed uniformly around tuce prior to expansion.

i Section A-A for specimen B.

Tube to be pushed to one side of collar prior to packing.

L Crevice gap to be filled with simulated sluage.

l Section A-A for specimen C.

Tube to be pushed to one side of collar prior to packing.

Crevice gap to be only partially filled witn simulated sludge.

l 1

i Figure 5-19 Schematic of Test Specimens: Tubes Expanded in Prepacked Crevices l

i A

$C

/

a T

i Figure 5-20 Tests on the ID Surfai:es of R'esults of MgCl 2 Expanded Type 304 Stainless Steel Tubes I

r.

t-6 d4 (--

Figure 5-21 Results of Sodium Tetrathionate Tests on the ID Surfaces of Expanded Sensitized Inconel 600 Tubes

S, b & f.-

i

(

r I.

p i

I Figure 5-22 tic Effect of Temperature on SCC of Inconel-600 in Kighly Ag j

L Environment l

l:

l f

I

... ~...,

..... ~ - -

(135) v 130

110

$a E

2 90 a

e, W

G

-.E 70 b

b m.

3=

F 50

~.

a

(45)

..y 3

30 (17)

,(17.2)

~

10

%/

50 First 10 20n 30

,g 4@

Crevice Indication Hot Leg =

Lowered Hot Raised Lowered 590*F Leg to 557*F Hot Leg Hot Leg to 575'F to 557'F Effective Full Power Months Since October 1977 l

Figure 5-23 i

Effect of Hot Leg. Temperature on Number of Tubes Plugged Due to Secondary Side Corrosion J

Temoerature. OF 689 652 617 600 554 1000

/

~

/

~

/

0/

2 8

/

g 100 a

s.

C/

/

K A

/

p

/

y Leoend y

G

/

A BNL Pure Water (Reverse

/

~

U-Bend (RUB), 0.03 C.

~ jo.

f B W Primary Water (RUB), 0.03 C.

o

/

C [PureWater(RUB)

'~

[

D W AVT - C-Rings i

i i

1 1

-3 L

1.55 1.60 1.65 1.70 1.75 x10 l

Reciprocal Absolute Temperature (1/T (OK))

Figure 5-24 Effect of Temperature and Soecimen Configuration on SCC of Inconel-600 in

" Pure" Water Environments

i

]

f p-g.

y-i l

I g-

.M*

\\

. t.,

. t.

II

7" l

'K-3 y,

.g.

T' o-a c,-

I 5-A.

E 5

3-3 L

)

TmlESEEZ:"

C:13 SII:E HOT SEE l

Figure 5-25 Global Model Geometry Used in Tube Structural Analysis

2.50...

L.23.

3 16.

G5..---

L82....

LE.

LM.

L31 l

L14

.37 Y

F Y

.M f

..a.a

.=

l l

Figure 5-26

\\

i Local Model Geometry used in Tube Structural Analysis

6.0 EFFECT OF TUBE EXPANSION ON VIBRATION Vibration data have been obtained for unexpanded and expanded tubes in both an operating plant steam generator and a full scale test model. The operating plant, KRSKO, provided three phases of data collection with Model D4 steam generators. Phase I (February-May 1982) vibration data were obtained prior to feedwater system modifications; Phase II (July 1982) data were obtained af ter modifications: Phase III (December 1982) provided data at additional tube locations over those of Phase II.*

Additionally, data were obtained during Phase III for one tube (R48C55) in steam generator No. 2 that was hydraulically expanded at both the B and D support plate locations.

The full scale test model used in the counterflw steam ganerator program was a 16' water model located at Waltz Mill. This model basically includes half of the preheater inlet region, all tubes represented, with those tubes located within a 16* " slice" from the centerline being full length up to the "L" support plate location. Vibration data were obtained for unexpanded tubes and for two series of tests with expanded tubes. The first expanded tube test series involved three tubes while the second test series involved 24 tubes.

6.1 Data Acquisition and Reduction 6.1.1 Data Acquisition Vibration data were obtained from in-tube biaxial accelerometers (Figure 6-1).

The assembly includes a spring assembly to ensure contact with the tube inner wall,- hard-line tubing for instrument leads, and a forward attachment used to pull the assembly into a tube and determine the installed accelerometer orientation. The spring assembly that supports an accelerometer applies a force to the tube wall sufficient to increase the natural frequency of the mounted accelerometer assembly above the primary range of interest,

The KR5K0 feedwater system was modified in June 1982 to permit simultaneous feedwater flw of 70 percent to the main feedwater nozzle and 30. percent to the auxiliary feedwater nozzle.

0771c/0122c/060883:5 50 6-1 L

~

l O, C.,e

+

The _ hard-line tubing 3 s also supported by springs to eliminate 1

impacts that would induce spurious' signals into thb fecordeo data.

f s

Accelerometers were located at the midspan.of the inlet pass (between the B 4

and 0 support plates) and at the E support plate elevationi(Figure 6-2).

Acceleration.' data were recorded on magnetic tape during conduct of each test and later reduced..The following typical data forms were used in preparing this report:

a,C,C

(

)RMS acceleration spectra - used to identify frequen'cy a.

r,

/

content of acceleration 464*

{

)RMS displacement spectra - used to provide tube RMS-b.

displacement values.

Time histor.y visicorder traces - provides peak-to-peak acceleratior.~

c.

values and tube-to-support-plate impact information.

In addition to spectra generated using the Nicolet Fas't, Fourier Analyzer, the analyzer has the capability to provide RMS acceleration' and displacement 4

values integrated over a specified frequency jange, d

e 6.1.2 Data Parameters Used to Evaluate Tube Vibra' tion

~

c l

7 Fourmajorparameterswereusedtoevaluatetub'eviidation. The first

~

parameter is the tube response frequency or frequencies. The presence or absence of tube contact at support plates results il tube free-span lengths p

that have been related to particular response frequencies.

+

-The second parameter used to evaluate tubetvibration is a peak-to-peak acceleration value measured with the tube mounted accelerometers. A

_ peak-to-peak acceleration value is obtajned by. visual observation of 'a time history record.

In the. determination of --a peak-to-peak acceleration value, a typical impact transient was selected which was 'used to weight the nomber of impacts occurring at various magnitudes.'

0771c/0122c/060883$551 (o - t

o

[

The-third parameter used to evaluate tube vibration 'is the root mean square displ acement (RMS 6). The RMS displacement was computed from measured tube acceleration signals.

Using a Nicc'et 660 Real Time Analyzer, an RMS displacement spectrum was generated and integrated over the (

),,c,g frequency range.

The fourth parameter used to evaluate tube vibration is the product of peak-to-peak acceleration and (

[M'$isplacement(Ga).

Tube wear is proportional to the work performed by a tube in contact with and moving relative to a tube support hole under an applied force.

If the effective mass of-the tube and interacting fluid is considered to be constant, the applied force is proportional to the tube acceleration and the wear is proportional to the acceleration times the tube movement. The Ga parameter was generated to represent the product.of force and movement which has been empirically t '- correlated to wear. A Ga value for a particular tube elevhtion was com utedo' by calculating the product oY peak-to-peak acceleration and

}

displacement' for e'ch axis of a biaxial accelerometer (gx x and g 6y y),

a 6

f,

,and us ng the'following formulationu i

,j

,. t

+

r 9

(-

k /(,

3 gC g qt, i

l y-t-

Ga =

t When only one axis of a biaxial accelerometer was operational at a particular

]o,c. e elevation, the single axis g6 value was multiplied by a f actor (

to i

obtain ja total vector Ga value. This factor was determined by review of 96 valuesf at the E elevation in tubes at KRSK0 where both accelerometer axes were

. operational.

I 6.2 Operating Plant Data t

Tube 1R48C55 in KRSK0 steamcgenerator No. 2 was expanded at the B and D support plates to eva}uat? the e?/ect of this rodification upon tube vibrations.

(,6

<\\ Prior: to' expansion a diametral ga) "of approximately was measur,ed at

'both the B and D plates. After tobe expansion, gaps of

.r a,c.e Jweremeasured.

r 0771c/0122c/060883:5 52 b-3 L

F

-x

) g,j7 G: ~~

g ;-

f c

'6.2.1 Pre-expansion Data

.r-During Phase I testing at main feedwater f,1y rates of 60 percent and higher, tube R48C55 accelerometers had a (

] response. The inlet pass accelarbmeters had responses at {

a,e e.

~

,w le the E level accelerometers had (

}resdonses. At 100 percent main feedwater flow,. all ~ accelerometers displayed (

i

- a,e' Table 6-1 c

a ;! >

+.

j provides' vibratior/ data at different feedwater flow rates (and nozzle entry)

Y

% gj; A,or this tube.,l[

- 'f i;

'i, i'

.~

Bs 6.2.2

Post-expansion Data 7

~

.e iDuring Phase III testing at all feedwater flow rates, including split flow, i

~

lhoth inlet pass l accelerometers displayed only lw level responses in the frequency range *of J.,e eTable 6-2 provides 1

vibration l1 ate for the expanded tube similar to that provided in Table 6-l~ for the~unexpaddedtube.

L/.

,?

a, b, e. e

(; ~

/

"4 Comparisons of unexpanded and expanded tube responses show reductions in both accelerations and displacements for the expanded tube. Peak-to-peak

-accelerations were reduced by a factor between

}"' C$parisons of Ga' values for expanded and unexpanded conditions show expanded tube Ga values lowerbyafactorof[

[**3 *

.O.c..

[

At the inlet pass elevation, a Ga.value after expansion was(

for the 100/0 fl split, a significaat reduction of the pre-expansion Ga value of c e.

,o,

similar reduction of GA was noted for the 70/30 flow solit.

i g.

0771c/0122c/ObO883:553

&-A

JS 4

6.3 Scale Model Test' Data 6.3.1 Description of the Model The 16* water model included half of the preheater inlet and half of the tube bundle, bounded by a plane along the center of the T-slot, the center partition plate at the center of the steam generator, and the wrapper. A diagram of the model is shown in Figure 6-4.

A rectangu?ar section of the i'

tube bundle,16 tubes wide.from the T-slot.and extending the full depth of tne bundle, was composed of long tubes extending from the tubesheet up to the L support plate. The remainder of the tube bundle was composed of shorter tubes extending from the tubesheet up to the O support plate. The model inlet included half of the reverse flow limiter and half of the water box, from the centerline to the side corner.

All. dimensions in this model corresponded to the dimensions of the prototypic steam generator.. The positions of-4he B, 0, E, G, H, and K support plates were adjustable in the plane of the plate (

].,,,..

Feedwater entered the model at location I and exited at locations II, III, and IV (see Figure 6-4).

Outlet II removed flow which passed through the B p l ate.- Outlet III removed flow which passed around and through the D plate above the shorter tubes in the bundle, while Outlet IV removed flow which passed through the length of the preheater along the long tubes up past the L plate.

The first 5 rows of long tubes nearest the feedwater inlet and tubes along the T-slot were prototypic Inconel 600 material, while the remainder of the tubes were stainless steel of prototypic diameter, to simulate preheater flow resistance. Tube support plates were 405 stainless steel. The model enclosure was carbon steel with a protective coating inside to prevent oxidation.

0771c/0122c/060883:5 54 b -S

The water supply was provided by a recirculating water loop with pumping.

capacity up to 5324 gpm at 100 psig.

Bypass flows through outlets 11 and 111 were controlled by valves in each return line. All flow tests were carried out at about 100~F. All flow connections to the model incorporated flexible sections to-isolate the model from loop vibrations. Loop water was partially

-filtered during operation.

6.3.2 Identification of the Model Baseline Configuration To evaluate the effect of various modifications to reduce tube vibration, it was decided to attempt to duplicate the most pronounced vibrations seen in the operating Model 04 steam generator site test data. The most pronounced vibrations, that were observed, occurred in the KRSK0 Phases 1 and 111 test series.

By varying support plate positions, the response of the tubes in the model was varied to duplicate as closely as possible the results obtained during Phase I at KRSKO. The plate-configuration-designated SR3 was chosen as the support plate configuration which most nearly represents the KRSK0 data. Tube R49C56 was used for comparison between the model and test results (this tube exhibited the highest response in-the KRSK0 testing). Figures 6-5 to 6-10 present a spectral comparison of commonly instrumented tubes for the SR3 model configuration data and the plant data (There is no correspondence of "X" and "Y" directions of accelerometers in the 16* model and the plant). The 16' Tables 6-4 to model and plant spectra exhibit gilarresponsefrequencies.

6-6 are tabulations of s displacements, average peak-to-peak accelerations, and their

[ duct, GA.

These results show similar orders of magnitude and are considered an adequate representation of site responses.

~

The Responses of other tubes instrumented in the model are also presented.

remaining instrumented tubes showed response levels too low for evaluation of modific ations.

Following identification of the SR3 plate configuration, tests were run with the model to evaluate various design concepts.

Upon completion of these tests, the SR3 base configuration was re-established and a test to demonstrate repeatability was conducted prior to running the tube expansion tests.

0771c/0122c/060883:5 55

g. - &

These repeatability' tests (designated 'llR5) were run at flow rates comparable to 70,100 and 112 percent power.* Figures 6-11 and 6-12 present two typical spectra comparisons for the two base configurations.. The response frequencies

- and relative magnitude of the responses was nearly identical for these two sets.

The repeatability of the base case was good.

Some significant differences

- were seen in the peak-to-p,eak "g" values and can be seen in the Ga comparison. The changes seen in peak-to-peak "g" amplitudes are reasonable

.considering their sensitivity to plate movement. The repeatability of displacements and response frequencies are more indicativ'e of the similarity of the two cases.

6.3.3 Baseline.vs. Tube Expansion Af ter completion of the baseline rerun, a number of tests were performed with expanded tubes. Tubes were expanded-at the B and D plates. The first

- expanded tube test (TE-1) performed the expansion process on three tubes (R49C56, R48C55, and R48C53). The second test (TE-2) performed the expansion on 24' tubes (Rows 46 to 49 and Columns 51 to 56).

The plate positions for these tests: remained identical to the previous baseline tests.

In addition to the 70,100 and 112 percent power flow rates at which baseline tests were conducted, tests were also performed at 85 percent flow rate.

Figures 6-13 to 6-16 are typical comparisons between expanded tube and baseline data. Figures 6-17 and 6-18 are typical comparisons for non-expanded tube acceleration spectra for the base and expanded tube data sets.

4, b. S

The' 100 -percent water flow rate used in the 16* Model for these tests was l'

comparable b3 a two' loop Model D plant with a 100 percent power main feedwater 0

flow rate of 4.09 x 10 lbm/hr.

l 0771c/0122c/060883:5 56 i

6-7 g.-

-w-r1p iws-en a

~

g-4

---r-m

~ - - -

e

,e-y

,ew

-ee e

w-3

,e--

w

Q.k.*, '

~

6.3.4 Sumary Plant and full scale model data comparing the vibration of tubes before and after expansion at the B and 0 plate elevations show significant reduction of tube vibration levels. Levels of both RMS displacement and impact accelerations are reduced in both plant and model data.

,t t

e RMS displacements are reduced by a factor of j, impact g's by a factor of (

['adGa'sbyafactorof(

70'r $ower levels 70 percent and above in KRSK0 based on the-expansion of a single tube.

Model' data obtained for a group of expanded tubes show RMS displacements and s e*e peak-to-peak accelerations reduced, on the average, by a f actor of[ }1n'd gs

,c. e reduced, on the average, by a f actor of

]a*t a model flow rate comparable to the 100 percent flow rate in a 2 loop, t[odel D4 plant.

0771c/0122c/060883:5 57

4. - 8

a Table 6-1 Vibration Data for Unexpanded Tube R48C55 in KRSK0 Steam Generator No. 2(I)

Q,be C 6 i

=

4 l;

~

b -1

. Table 6-2 j~

^

- Vibration Data for Expanded' Tube R48C55 III l..

-in KRSK0 Steam Generator No.12

. 6. s. e J

i 1 -.

I-t 4

t i

(

4 l -:

l

0771c/0122c/060883: 5 59 6 - (O

u.-....

'i A

Table 6-3

Ga Comparison for Unexpanded and Expanded

' Tube.R48C555 in KRSK0 Steam Generator No. 2

-a,L,c,c h

1-s 3-r i;

1

=-

e i

l-a L

i l.

jL l.-

F l

e L

io g -//

0771c/0122c/060883:5 60 I

w e

..-,e'ww...,

=v-..,,

4.--r--

,-,-..-9,m e m

3..,,v,

.<.e..,,y*-,.,..-.m.,,,,my,,,,--,,-,y+-,.ymy,

-,+wm,-,'ww.vy--y,,wwg,g-+,rw-,

,-gy,--wwer-t<-

m%.,.

A L

.=A.m,_w&_,

W

'l l

t I

IA$ll 0 4 l

I 15100 H2 RMS Ol5PL ACERNi$ FOR RR$KO AND 16" MODEL AT COMPARABLE MAIN FEEDWATER FLOW RATES BETWEEN 70 AND 112 PERCENT POWER l

l l

/Z

/

N-

O 4

a,44e 9

4 1

him TAM E 6-5 AVERAGE PEAK-TO-PEAK ACCELERAil0NS FOR kRSK0 AND 16' MODEL AT EOMPARABL E MAIN FEEDWATER FLOW RATES NTWEEN 70 ANO 112 PERCENT POWER

&~(h

. - ~

( /

i i

=

I l

TAOLE 6-6 i

Ga'$ FOR NR$KO AND 16" MODEL DATA AT COMPARA8tE MAIN FEEDWATER FLOW RATES BEIWEEN 70 AND !!2 PEREENT POWERIll l

l l

C-l4

449e i

k l

l T

t eIII I

Figure 6-1 Typical Accelertmeter Assembly

(

.. _.,... _ _. -, ~,., -

i g

b c

annu c

=

==

MN m

emu a

=

anmtt i

I I,

, N N "E" Level anmga

= :::: ::: lllll: L l

namza I

I I

i s

= =

1 "B-0" Level MA I

\\

)

n 4

4 1

Figure 6-2 Accelerometer Locations

a,4 46 Figure 6-3 Acceleration Spectra for KRSK0 Tube R48C55 Before and After Expansion

1 t

,,., ;. ? '

\\%

5 f **?e ****

- - - 7 U.l,'7.

.,}

., a,gp

-c n*

Q Vj.

,/

,g. u, 2.

\\

f

-e n*' a s

e~

f

. }

y

}f

\\

< u na (p\\y

&'W s

9 11 u

~~

b g,

D n**

j

.an-s r n**

W

- ge Q -.c n**

o*5

\\

+

r s.. '..,5 *-

s 4

c n**

_s o go @D ORD M ts Figure 6-4 Full Scale Counterflow Preheat Steam Generator Test W. del 160

i l

Q, bA L I

i F

f i

5 i

i l

l I

l l

Figure 6-5 XRSK0 and 16' Scale Model 0 - 200 Hz llMS Spectra Comparison Tube R49C56 5

-v v e v

a,

--e,-n wep--+,e.re---

+

c

,,e---

e--e--wv-

,e,-

v-e %-greve-em-r-w 'evw=w*-'*m-tw-e e ww = v-r w w e-e --e-e e r-r -w wv e r -, - 1

-g-+-

T

bCb So i i t '

l l

l t

I r

l Figure 6-6 KRSK0 and 16' Scale Model 0 - 200 Hz ItMS Spectra Comparison Tube M6C56 i

L

t Gy bo C f--

i i

I t

Figure 6-7 :

KRSK0 and 16' Scale Model 0 - 200 Hz RMS Spectra Comparison Tube R48C55,

____.,, _,. _.. - -... - _, _ _ -....... _ _. _..,, _ - _ _.. _... ~. _, _,

$i/bdO 1

l i

i i

I Figure 6-8: KRSK0 and 16' Scale hdel 0 - 200 Hz RMS Spectra Comparison Tube A48C55

8 o

l /

(

~

1 4

l Figure 6-9:

KRSK0 and 16' Scale 2dal 0 - 200 Hz RMS Spectra Comparison Tube R45C56.

,-,--,w,-.

3.,, -,,, -,,, __,

y-y v.

.,,w,,

-y-,n.

w

,.,.,y,

, -.,--,-,y+

w,y-

O s

S b 0, C.

i i f

l l

l r.

I i

l l

i i

l.

(

l l

l c

an I

Figure 6-10 KRSK0 and 16' Scale Model 0 - 200 Hz RMS Spectra Comparison Tube R45C56.

i

(

O k$6 r

e 1

l l

l I.

l Figure 6-11 : 16' % del SR3 Base and 11R5 Base 0-200 Hz

.,___,..m.

..,c, r

os l

t l

l

(

l-i

j l

O b> 0f r

4 16' Model SR3 Base and 11R5 Base 0-200 Hz Figure 6-12: RMS Accel. Spectra Comparison Tube 49C56 (E)

4 0.,bi&6 I

ll' t

Figure 6-13 16' Model SR3 Base and 13 TE-2 0-200 Hz RMS Accel. Spectra Comparison Tube 49C56 (B-D) 9

. - - ~

v+,.

,e.~,

-,,--,,-c-,,n,.,

-e,,-..-

o A b & '"

r i t

i l

l l

t l

l Figure 6-14 16' Model SR3 Base and 13 TE-2 0-200 Hz RMS Accel. Spectra Comparison Tube 49C56 (E)

F,.

0-f

$[

~

i 5

i f

i t

Figure 6-15 0

16 Model SR3 Base and 13TE-2 0-200 Hz RMS Accel. Spectra Comparison Tube 49C53 (B-0)

o 3

i o

' ::Y.

p y

,A

)

/,

t f f (

~

m

.1-r t-:

1 e

s 7

' ' ' f t'

i s

I

)

.g I

i 1

n

't 1

Figure 6-16 i

160 Model SR3 Base and 13TE-2 0-200 Hz RMS Accel. Spectra Comparison Tube 49C53 (E) l 4

p i

o-8,,

i:

z f

t 0

4 / / /

e it h'.i x

..,s i

e s

i t

.. [ 4 fj

,, y

. r', <

t i

e t :. r.

~

Figure 6-17

,1 l

16' % del SR3 Base and 137E-2 0-200 HzR

~

. _ ~

l t

)

n 4

......... - -. ~ :..

o s

4 t:L, b d, c i

~

il,

. -1 a-s I

i A

e

.l l

s f

l l

t 4

Figure 6-18 16' Model SR3 Base and 13TE-2 0-200 Hz RMS Accel. Spectra Comparison Tube 45C56 (E) i

' i i

i

. t

=

l O b0 f

l i

f Figure 6-19 Ga.Versus Flow Rate for Tubes R49C56, k aC56, and R49C53 ye e

e, -,.. -

,e..,

,y,,.,

y

,y...,,-..__r,....,--.9,

..-m.mw,,,w-,+-ym e,,w--

n_._..

E.o; L

7.0 REFERENCES

1 1.

" Indian Point, Unit 3 Steam Generator Sleeving Report," WCAP-10146, Westinghouse Nuclear Energy' Systems, Pittsburgh, Pa, October 1982.

,2.

Bandy, R. an'd D van R'ooyen, "St'ress-- Corrosion Cracking of' Inconel Alloy 600 in Hign. Temperature Water-An Update," Brookhaven National Laboratory

.(preprint of paper prepareo.for NACE).

3.

ASME Boiler and Pressure Vessel Code,Section III Nuclear Power Plant

~

Components,1980 Edition-through Sumer 1982 addenda.

4.

" Bases for Plugging Degraded PWR Steam Generator Tubes," Nuclear Regulatory Commission Guide 1.121, April 1,1977.

5..

Hall, J. M., Thurman, A.' t. And Truitt, J. B., 'WECEVAL, WECAN Evaluation Users' Manual", NTD-SMD-ASM, September 1981.

6.

"Counterflow Preheat Steam Ge'nerator-Tube Vibration Summary Report" SGPR-8302, Westinghouse Electric Corporation, June 1983.

r i

~

0771c/0122c/060883:5 61 7./

l i'a:n

-

ML20081A476

Contents

Text

1.0 INTRODUCTION

SUMMARY

1.0 INTRODUCTION

SUMMARY

= =

7.0 REFERENCES

Navigation menu

ML20081A476

Text

1.0 INTRODUCTION

SUMMARY

1.0 INTRODUCTION

SUMMARY

= =

7.0 REFERENCES

Navigation menu

Search