ML17219A261
| ML17219A261 | |
| Person / Time | |
|---|---|
| Site: | Saint Lucie  |
| Issue date: | 12/12/1986 |
| From: | FLORIDA POWER & LIGHT CO. |
| To: | |
| Shared Package | |
| ML17219A259 | List: |
| References | |
| NUDOCS 8612150319 | |
| Download: ML17219A261 (35) | |
Text
ATTACHMENTA Marked-up Technical, Specification;,Pages:
3/4 4-8 B
3/4 4-3 961Z150319, 5000~35 gg 1212 pDR ADOCK'
@DR P
FJW2/024/3 REmiLATORII BUCKET Fii,E l;OPy
t fr i.,w,..lCi 1
~ s)(.(ga).)l ~,i~'>>>
~.
~f. c<">> M~ s (., -, i;~~.
Q \\
)
REACTOR COOLANT SYSTEM SURVEILLANCE RE UI REAGENTS Continued 5.
Defect means an fmperfectfon of such severity that ft exceeds
%earp uggfng limit.
P tube containing a defect fs defective.
Any Cube which does not permft the passage of the eddy-curl cnt inspection probe shall be deemed a defectfve tube.
6.
7.
P1u in Limit means the imperfection depth at or beyond w
cn C e tu e shall he removed from service because ft may become unserviceable p {or to the next fnspectfon and fs annal to ~40%* of tha rorslnal tubs wall th$ cknsss~
Unserviceable descrfbes the condition of a tube ff ft leaks df h9 ghi ff Alt t
integrity fn the event of an Operating Basfs Earthquake, a
loss-of-coolant accident, or a steam lfne or feedwater lfne break as specified in 4.4.5.3.c, above.
The stean generator shall he determined OPERABLE after completfng the corresponding actfons (plug all tubes exceeding the plugging lfmft and all tubes contafr.fng throu h-wall cracks) required by Table 4.4-2, b.
esca pr ruhr ram'currcrrtrw arm'rr rs eqoa4 7p 5D/as pyrrR r.OCULOTerri/O'7 OR rIt8OVE'h'H r~r pARrrqz suppoRr pzrirc rrtr rude Ro~~
rr!7 r o rect usrvF 4.4.5.5
~Rs orts 8.
Tube Ins ection means an fnspectfon of the steam generator u
e rom the point of entry (hot leg sfde) completely around the U-bend to the top support of the cold leg.
a.
Followfng each fnservice fnspect on o
steam generator tu es, e
number of tubes plugged fn each steam generator shall be reported to the Commission wfthfn 15 days.
The complete results of the steam generator tube fnservfce inspec-tion shall be fncluded fn the Annual Operatfng Report for the period fn whfch this fnspectfon was completed.
This report shall fnclude:
Number and extent of tubes fnspected, R.
Location and percent of wall-thickness penetration for each fndfcation of an fmperf'ectfon.
3.
Identification of tubes plugged.
'This 40 p
t fs not applicable during the cycle 1 opera to Joe 30, 1986.
If at any is perfod t ers any Modes other than Modes 1
and 2, or Mode 3 4 hours4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br />, the unit shall be placed fn cold n
the tubes wfth fndfca than 40% throu h-ratfon shall be removed from servfce prior to ex ST.
LUCIE UNIT 1
3/4 4-8 Amendment No.
13
KACTQR COOLANT SYSTEN 3 4.4.5 ST AN G NERATORS ontfnued Thc planC fs expected Co be operated fn a manner such that the secondary coolant wfll be mafntafned wfthfn those parameter lfmfts found to result fn.neglfgfble corrosfon of the steam generator tubes.
If the secondary coolant chemfstry fs not mafntafned wfthfn these parameter lfmfts, localfzed corrosfon may lfkely result fn stress corrosfon cr'ackfng.
The extent of crackfng 4urfng planC operatfon would be lfmfted by Che lfmftatfon of steam generator tube leakage between the prfnary Coolant system and the secondary coolant syster (prfmary-Co-secondary leakage
~
1 gallon per mfnute, total),
Cracks havfng a pr fmary-Co-secondary. leakage less than Chfs lfmft durfng Operatfon efll have an adequate margfn of safety to wfthstand the loads fmposed durfng normal operatfon an4 by postulated accfdents.
Operatfng plants have demonstrated that prfmary-to-secondary leakage of 1 gallon per mfnute can readfly be detected by radfatfan monftors of sCeam generator hlowdown.
Leakage fn excess of thfs lfmft wfll requfre plant shutdown and an unscheduled fnspectfoni durfng whfch the leakfng tubes wfll be located and plugged.
wastage-type defects are unlfkely wfth the all volatfle treatment (AVT) of secondary coolant.
- However, even ff a defect of sfmflar type should develop fn servfce, tC wfll be found durfng scheduled fnservfce steam generator Cube examfnstfons.
Plgggfng wfll be r ufred of a11 Cubes wfth fmperfectfons exceedfng the lu (n lfeft fc,
y e
n on of Specfffcatfon 4.4,5.4.a s
o
~
be nomfnal wa)l thfcknes Steam generator Cube fnspec on o
operatfng p an s
ve emonstrated the capabflfty Co relfably detect degradatfon that has penetrated 2R of the orfgfnal tube wall thfckness.
ST.
1.VCIE - VNIT 1 8 3/4 4-3 Amendment No.
69
ATTACHMENTB SAFETY EVALUATION This is a request to revise Section 4.4.5.4 (steam generator ) Acceptance Criteria of the Technical Specifications for St. Lucie Unit 1.
0ESCR IPT ION Technical Specification 3/4.4.5 requires steam generator operability and specifies Surveillance Requirements to verify steam generator integrity.
The current steam generator tube plugging limit as defined in 4.4.5-4 is 40% of the nominal tube wall thickness.
The proposed change will.speci-fy a tube plugging limit of 54% for all regions except that a limit of 50'ill apply to locations at or above the top support plate for tube rows 117 through 120, inclusive.
The proposed change specifies a more accurate tube plugging limit based on analysis to the criteria of Regulatory Guide 1.121,"Bases for Plug-ging Oegraded PMR Steam Generator Tubes".
In addition, the proposed change prevents unnecessary (1) plugging of tubes, (2) associated high.
personnel radiation exposure and (3) decreases in the steam generator heat transfer surface,area.
The proposed steam generator tube plugging acceptance criteria have been established by analysis in CENC-1747 (Attachment 0) in accordance with the provisions of Regulatory Guide 1.121.
It was determined that the limiting event as regards to tube loading was a combination'f Loss of Coolant Accident (LOCA) and Safe Shutdown Earthquake (SSE).
The limit-ing event was found to control the allowable tube degradation in the region above the top support in tube rows 117 through 120.
The analysis shows that tube degradation of up to 59% is acceptable for this region.
I The allowable degradation in the remainder of the tube bundle, including the regions of tube rows 117 through 120 below the top support, is gov-erned by the normal operating differential pressure criteria of Regula-tory Guide 1.121.
The criteria states that during normal operating condition no tube will a) be stressed beyond the elastic range of the tubematerial and b) display a factor of safety of less than 3 against tube rupture.
The analysis shows that tube degradation of up to 63%
meets the nornmal operating differential pressure criteria.
'he allowable tube wall degradation was'lso determined for the effect's of Hain Steam Line Break (MSLB) combined with SSE.
It was determined, for this event, that tube degradation of up to 66% is acceptable; there-fore MSLB combined with SSE is. not controlling.-
In establishing the Technical Specification limits for St. Lucie Unit 1, the above allowable degradation was reduced by 10% to cover Eddy Current Testing (ECT) measurement error and continued degradation through the next'uel cycle.
Justification for the 10% allowance is provided in Attache.nt C.
ATTACHMENT C JUSTIFICATION FOR 10'X UNCERTAINTY ALLOWANCE FOR ECT ERROR AND CONTINUED OPERATION FOR THE DURATION OF NEXT CYCLE
ATTACHMENT C, PART I ALLOWANCE FOR ECT ERROR Accuracy in measuring the depth of tube defects using the Eddy Current Testing (ECT) method is a variable, depending on the size (voltage) and depth of the tube defect as well as the nature of tube material attack.
Large volume defects produced by mechanisms such as chemical "wastage" or mechanical wear are usually overestimated by ECT and the defect sizing requir es no additional allowance.
This position has been supported by CE-sponsored testing on behalf of Consumers Power and is referenced in Regulatory Guide 1.121, Footnote 4 (Reference Cl-3).
Small volume deep pitting defects are sometimes underestimated by ECT.
CE demonstrated in I
Reference cz-5 however, that tubes with clusters of nearly through-wall I
pitts, still possess more than adequate margins of safety.
Therefore, only crack-like defects (both IGSCC and IGA) have the potential for uncertainty, regarding the conservatism of the ECT estimate.
In Reference C1-4, CE and FPSL demonstrated that IGSCC defects present no safety concern due to their characteristic "leak before break" behavior, when occurring in CE's very ductile (35 ksi < yield stress
< 55 ksi) Inconel 600 tubes.
The purpose of this document is to present significant evidence obtained through prototypically degraded tube testing by both CE and Battelle-Pacific Northwest Laboratories (within an NRC funded project, Ref-erencee Cl-l), that ECT testing of tubes with chemically induced k(f 1
i
~f defect depths.
This means that while ECT may underestimate somewhat the deepest isolated penetration of IGA defects, the tube's burst pressure performance will be indicative of the ECT estimate, usually with some addi-tional margin beyond that required from Reg.
Guide 1.121 (Reference C1-3).
This phenomenon can be explained by examining the photo-micrograph shown in Figure C2-1.
The depth of general attack (represented by the dashed line) will dictate the burst pressure.
This general attack is closely predicted
by ECT examination.
The deep isolated penetration (shown within the cir-cle) is typically what is sought and reported as the defect depth obtained from metallography.
This deep isolated penetration will have virtually no effect on the burst pressure performance of the defected tube, due to the reinforcement provided by adjacent unattacKed material.
It is for the E
E fd
~f which to predict tube burst pressures and use as a basis to calculate safe-ty margins.
Tables Cl-1 and Cl-2 contain data that was specifically generated to evalu-ate IGA typical of that found, through removed tube metallography, in the St.
Lucie 1 steam generators.
Table Cl-3 contains IGA data previously generated for IGA patterns found in the Palisades steam generator as well as burst data from nondegraded tubes for purpose of control and comparison.
Data in all three tables was generated at high temperature (temp.
600 F) using chemically induced IGA degraded tubes.
The burst pressure versus defect depth data, estimated by ECT, matched extremely well with theoreti-cal predictions, as can be observed in Figure Cl-l. It is significant to note from Figure Cl-1 that the region of tube differential pressure opera-tion (with respect to tube plugging limits of either the present 40% or the proposed 54% was fully outside of the defected tube burst pressure data band.
This separation includes the safety factor of three (3) on burst pressure required by Reg.
Guide 1.121, Reference Cl-3.
Several additional points should be made regarding Figure Cl-1.
The refer-ence burst pressure curve for local tube wall degradation is determined from the following relationship:
PB = 2 Vu t:(b-a)/(b+a)j Where PB
= Burst Pressure 0
V' = Tube Ultimate Strength
= 80 ksi 8 600 F
b = Tube Outside Radius (as degraded) a = Tube Inside Radius
If a nondegraded tube or a uniformly thinned tube is being considered, the relationship for predicting burst pressure must be corrected to account for 51 d
yielding before burst.
The following correction is made:
is subsituted for 7 in the above equation where:
u
. 7
= 2V(3) ~
= 0.85V" for n
=
1 -~/~
y u
and V
= Tube Yield Strength
= 35 ksi 9 600 F
It is for these considerations that the nondegraded test samples and Pali-sades circumferential IGA test samples (360 by 3.0 inches long) burst data fell between the two reference curves.
St. Lucie IGA samples, which were local in nature, displayed burst data above the upper curve, typically.
-5,2.7d-22~fi 11d iidd5 f nominal tube wall or less.
Figures Cl-2 through Cl-6, illustrate similar results obtained by Battelle-Pacific Northwest Laboratories (see Reference ct-f). Figures C1-2, Cl-3 and Cl-4 illustrate the uncertainty with regard to the ECT estimate of the EDM slot depth for the PNL test samples.
- However, as was the case for 2
2 <<1, 2
1 *1
~f conservative for purposes of predicting the burst pressure.
Figure Cl-5 illustrates the conservatism of the data (a portion of which is EDM crack simulation) when compared to tube plugging limits similar to that sought by Florida Power 8 Light for the St.
Lucie 1 steam generators.
Figure Cl-6 compares the St. Lucie 1 operating tube differential pressure (including a safety factor of three on burst pressure) and tube geometry with the Battelle-Pacific Northwest Laboratory results for EDM cracks.
The PNL data supports an allowable tube degradation of up to 63'X for St. Lucie 1 parame-ters.
This comparison corroborates the testing and analysis results devel-oped by CE as previously discussed.
Based on the foregoing discussion, ECT estimates of IGA degradation of tubes typical of those used in the FPSL St.
Luct'e 1 steam generator, have b ~fi 11 f
f ff 1
b b
pressure.
In the majority of cases, the data was conservative over and above the factor of safety of three, required by Regulatory Guide 1.121.
The maximum nonconservative prediction by ECT, with regar d to burst pres-
- sure, was 5% of tube wall.
The results of CE's evaluation was confirmed by data obtained by PNL as part of an NRC funded research project dealing with degraded tube burst pressure and ECT examination.
REFERENCES Cl-1.
NUREG/CR.-0718, PNL-2937, "Steam Generator Tube Integrity Program, Phase I Report,"
by J.M. Alzheimer, et al.,
September 1979.
Cl-2.
NUREG/CR-3561, PNL-4695, "Eddy Current Round Robin Test on Laborato-ry Produced Intergranular Stress Corrosion Cracked Inconel Steam Generator Tubes,"
by R. L. Bickford, et al., January 1984.
Cl-3.
Regulatory Guide 1.121, "Bases for Plugging Oegraded PWR Steam Gen-erator Tubes," August 1976.
Cl-4.
CENC-1740, "Tube Burst and Leakage Testing of IGA and IGSCC Oefects Representative of Those Found in the St. Lucie, Unit One Steam Gen-erators,"
by W. J. Heilker, et al., August 1986.
Cl-5.
CENC-1497, Revision 3, "Millstone II Steam Generator Analysis for Allowable Tube Oegradation due to Pitting Attack Between Tubesheet and First Support,"
W. J. Heilker and P.
L. Anderson, January 1982.
"Steam Generator Tube Burst and Collapse Predictions,"
by Milton
- Vagins, Presented at the Sixth Water Reactor Safety Research Infor-mation Meeting Held at the National Bureau of Standards, Gaithersburg,
- Maryland, November 6-9, 1978.
IGA TUBE DEFECTS"CORRELATION WITH BURST PRESSURE FP&L ST.
LUCIE 1 S.G.
TUBE BURST TEST DATA - SUPP.
2 TABLE C1-1 Tube Sample I.D.
Defect Dimensions Axial Circum.
Length Extent
( In. )
(Deg. )
Tested Standar d ECT Basi s Burst Defect Predi ct.
PTTest Pressure Depth PB t
P dt Burst Prdt.
(psi)
(X Wall)
(psi)
Metallography Basis (Max. Depth)
Defect Predict.
PTTest Depth PB t
P dt.
(%%d Wa11)
(psi)
Testing Conducted at 600 2-6 2"8 2-13 2-17 2-23 2-1 1.0 7.5 2-4, 1.0 7.5 2-5 1.0 7.5 1.0 7.5 2-7 1.0 7.5 1.0 7.5 2-12 1.0 7.5 1 '
7.5 2-14 1.0 7.5 1.0 7.5 2"21 1.0 7.5 1.0 7.5
>9150 6400 7750
>4200
>8500
>4600 4800 6600 7250 4100 7800 4700 65 48 56 55 28 56 61 45 50 58 40 53 4007 NA 33 5882 1.09 59 5005 1.55 41 5115 NA 46 8031 NA 35 5005 NA
'54 4452 1.08 68 6208 1.06 54 5664 1.28 63 4785 0.86 74 6749 1.16 65 5335 0.88 66 7499 NA 4674 1.37 6641 1.17 6100 NA 7285
. NA 5225 NA 3672 1.31 5225 1.26 4230 1.71 2996 1.37 4007 1.95 3895 1.21 The above twelve samples comprise all burst test data.
missing numbers represent unsuccessful ZGA flaw implementation.
Maximum safe test vessel pressure was reached.
The data are plotted since they represent lower bound burst pressure.
These samples developed small through wall leaks and could not be pressurized to burst.
While the data are not
- plotted, they are meaningful in that they demonstrate "leak before break" behavior.
IGA TUBE DEFECTS-CORRELATION WITH BURST PRESSURE FPSL ST.
LUCIE 1 S.G.
TUBE BURST TEST DATA - SUPP.
1 TABLE C1-2 Tube Sample I.D.
Defect Dimensions Axial Circum.
Length Extent
( In. )
(Deg. )
Tested Standard ECT Basis Burst Defect Predict.
PTTest Pressure Depth PB t
Pp Burst Prdt.
(psi)
('X Mall)
(psi)
Metallography Basis (Max. Depth)
Defect Predict.
PTTest Depth PB tst pp dt
(% Wall)
(psi)
Testing Conducted at Approx.
565 F
1-2 1-3 1-5 1-6 1.0 90 1.0 90 1.0 90 1.0 90 1.0 '0 1.0 90
>9600
>8500 6600
>8000 3500 4650 15 35 45 50 65 85 9396 NA 18 7285 NA 20 6208 1.06 69 5664 NA 40 4007 0'7 74 1742 2.67 78 9083 NA 8873 NA 3559 1.85 6749 NA 2996 1.17 2542 1.83 1-7 1-8 1-9 1-10 1.0 180 1.0 180 1.0 180 1.0 180
>8500
>8000 6650 3450 20 40 50 90 8873 NA 13 6749 NA 42 5664 1.17 63 1166 2.96 75 9603 NA 6533 NA 4230 1.57 2883 1.20 1-12 1-13 1-14 1.0 360 1.0 360 1.0 360 1.0 360
>8500
>8500 5900 5550 25 40 50 70 8348 6749 5664 3447 NA 1.04 1.61 32 17 58 63 7606 NA 9187 NA 4785 1.23 4230 1.31
IGA TUBE DEFECTS"CORRELATION WITH BURST PRESSURE CPC PALISADES S.G.
TUBE BURST TEST DATA TABLE C1"3 Tube Defect Dimensions Axial Circum.
Sample Length Extent I.O.
( In. )
(Oeg. )
Tested Standard ECT Basis Burst Defect Predict.
PTTest Pressure DePth PD S
Pp de Burst Prdt.
(psi)
(% Wall)
(psi)
Metallography Basis (Max. Depth)
Defect Predict.
PTTest DePth PBurst Prdt.
(X Mall)
(psi)
P"1 3 0" P-12 3 0" P-13 3.0" P-14 3.0" 2 Samples 3.0" Uniform 360 360 360 360 360 Circumferential 4625 53 4125 56 5950 43 6000 45 450 IGA Tested at 600 F (1974) 4535*
1.02 58 4255*
0.97 58 5461" 1.09 46 5277*
1.14 43 100 4067" 1.14 4067" 1.01 5185" 1.15 5461" 1.10 0
N/A Local 20%
P-39 0 5" 75 P"40 0.5" 75 P"42 0 5" '5 3 Samples 0.5" 75 Wall Thinning + IGA Tested at 600 F (1974) 7500 49 5773 1.29 54 6500 55 5115 1.27 77 5300 50 5664 0.94 70 2400-100 3300 5225 1.44 2656 2.45 3447 1.54 0
NA-A-1 A-2 A-3 A-4 A-5 8"1 B-2 NA NA NA NA NA NA NA Nondegraded Tubes NA 9550 NA 9600 NA 9850 NA 10, 150 NA 10,150 NA 10,200 NA 10,350 Tested at 600 F (1975) 9300*
1.03 9300" 1.03 9300*
1.06 9300*
1.09 9300" 1.09 9300*
1.10 9300*
1.11 NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA Note* - Based, on General Surface Attack (Corrected for Tube Smelling)
I
I
~
IGA TUBE BURST TEST DATA -
LABORATORY DEFECTS DEFECT KEY: 0 ST.
LUCIE IGA ST.
LUCIE IGA (UNBURSTED)
PALISADES WASTAGE + IGA P
PALISADES CIRCUM.
IGA CE NONDEGRADED 0.048 IN. WALL GEOMETRY:
TUBE.O.D. = 3/4 INCH WALL t
= 0.048 INCH MATERIAL:
!NCONEL 600 ANNEALED CE SPEC.:
55 KSI v a> > 35 KSI MILL TEST'u 90 KSI 9600 F:
80 KSI o
I
-'1 I
- jl.::;.::;
~
~
12 Ije:
!Il ii!iIi!!j}ii'i" I:jl I;~
.Ie'.
'ri ~
- 'h!!
~I(:.Ill (I l::!
a I ~ ~
~ I<<<<
- II
~.".
1st:I all j i(l}
~ ~ a i"
':I:~ ~ I
- I ~
t'ai
~ ~
~ ~
REFERENCE BURST PRESSURE FOR LOCAL TUBE WALL DEGRADATION
~ ~
I:I' (1<<
<< ~
- I( 1
- Il e
~ ~
~ ~
~ ~ ~
~a<<
~ ~
~ 1
~ ie
~
~
~ ! ~
10 I
I:(s
~ 1<<i "I:
~
~ I~
le::
ie
- III ts ~
I ~
~
I
~
- l. I
~
~
~ I
~(;
~ ~ ~
~
~
\\ ~
~ ~
'. I"a
~ ~ le
~'i
- !Il el)
I lt(1
~
I l t'i i"i!!
~
.!I,l f1<<
~
~
~
~ ~I;:
~ Ii
~I
~ ~
~
~
ae
~ oll e
~ ~ ~
<<:I I
I ~
Its II:
ll~ I Ijl I
I (
~
~
~
~ ~
ifII
<<I I
~I I I I
~
11
'l1
~ I",
~ ~ ~
~I
~i(
1 ~ I
~
1 6
~ ~ ~ ~
I'
~
, I I
~ ~ ~
~
~
~
- I!
ll".
ll"e I.::
~ ~
(
~
I ~<<
~
~
i-!i:
REF BURST PRESSURE ii'OR UNIFORM TUBE i.:!:
WALL THINNING
~ ~
a(I
~ ~ ~
~ ~
~ ~
\\
'!ll 1st.!
et:I
!I!i
~
~
( ~:I 24 oP
,'I I It(
~ !-i.! Ili I
~
~
~ ~
I ~
~ I af:I
~ ~
~ ~ ~ ~ (I ~ ~
~ ~
~ ~
I ~
~ ~ I
~ a
~(;
~
~
~
- ~ '
(
~ ~
('<<
it I ~
o I il:.I 1 I
~ ~
~ti Iel: (
<<P
~ ~ I
~
I
~!
( II
~ ~
- tl a'it:
REGION OF OPERATION FOR 40K TUBE
- !.::! (.
PLUGGING LIMIT elj:
REGION FOR 54%
'IMIT
~ ~
~
~
I((i
~ I
}el I ~
f:js (e(f IIII t(II ai
~ ~
I '
I".
~ I
~
~
1 ~
1 ~ (1 11
. :II(
~
~ l I'I o ~I..
~
Je.l
<< ~
'l}il!;i Igi
+sos<<
1(,...
!'!I"I
,I I
~ ~ I
- It
~
1!ei: (1 It(
I(t I!Is 1 ~
I
!I<<'Ij
~I
~
i~
~
0 0
10 20 30 40 50 60 70 80 90 100 g TUBE WALL DEGRADATION BY ECT
1.0 0.8 CT "
0.8 C3ill TL UNCONSERVATIVE
~
~
~ 0 ~
~
~
~
N 0
~ y
~
~
CT. TTO 0,4 CI ~
~ 0 0
~
~
~ +
0.2
~
~
CONSERVATIVE O.f 0.0 0.2 O.4 0.8 tACTUAl) hP 0.8 1.0 FIGURE Cl-2 Comparison of Predicted and Actual Burst Pressure Parameters for All EDM Slot Specimens f0 5
8.
<<u M
U 30 fOM SLOT~
TIIBf 551E i O.BT5 a0.050IHTAT B,f,fl
~ 0. 750 a 0. IH3 IHTAT C I
~ 0.350 c0.04$ IHTAT Ol
~ O.Q5 r0.03I IHTATGl IaIA5 IIHOTTTCTTO
~ ~
~
~
C1 ~
~l o+
~
~
~
~
~
~
~
~ ~
TO 5 ~
30 20 30 AO 50 60 l0 NT 00 TIBT ACTUALOffKCT OfPTH IS <<Alll FIGURE CI-3 Eddy-Current Indicated Defect Depth Versus Actual Depth for. EDM Slot Specimens
10 UNDETECTED DEFECT DEPTH
% WALL TS-X ss.eD TD SS DEFECT LENGTH ECT MULSURED ACTUAL 0
0 0
IA I
0'.25 0.50 0.75 1.00 L25 L50 DEFECT lENGTH Eisschesl FIGURE Cl-4 Burst Pressure Versus Defect Length for EDM Slot Heat B (0.875 x 0.050)
Tubing
'90 I
BD 70 50 40 Ci 30 20 10 i ~
o g
~
~
~
o ~o
~
"~ "REGION.",',
'-".'OF"'1978,""
PLUGGING"
'".P. ACTI'CE
~ 0
~
~
~ o
~t
~I
~ ~
~ 0
~
0 1 ~
~ ('5,',
~
~
~
~
1~a
~
~ ~
~
~
~
~ ++
~
~
~
~
~ ~
~
0
~
~ g
~
~
~f
~
~~ W ~ 04 1
2 3,
4 5
D 7
B 9
10 11 12 BUAST PAESSUAE Tksll FIGURE Cl-5 ECT Indicated Flaw Depth Versus Burst, Pressure for Various Flaw Types, Ref. C1-6.
1.0 0.8 h/t = 0.25 0.6 hP DPo 0.4 ST.
LU 1
S.G.
TUBE h/t 1.00 h/t = 0.50 h/t = 0.63 h/t = 0.76 0.0 0
4 6
8 10 12 14 16 I./~t L
{PREDICTED) = 1 -h+~e o
t FIGURE Cl-6. Burst Pressure Parameter Prediction Curves for EDM Slot Specimens - Length Parameter Variation
ATTACHMENT C, PART II IGA GROMTH RATES FOR THE ST.
LUCIE STEN GENERATORS Intergranular attack (IGA) has occurred at locations within the sludge piles and at tube support locations in the steam generators at St.
Lucie-1.
The postulated environment producing this IGA is acidic sulfates which apparently entered the steam generators with make-up water.
Acidic sulfates have been identified in several studies (g
, g
, g) as being capable of producing IGA in Alloy 600 steam generator tubes.
Although the ability of acid sulfates to produce IGA has been demonstrated.,the.,rate, at,.whiqh sulfate induced IGA defects propagate has not been established'-.=.~-'~,
A4jk IGA is the most pervasive form of corrosion affecting the secondary side of steam generator tubes.
Although seawater sites are not immune to this type of corrosion, IGA is more common at freshwater sites.
Essentially all fresh water site PMRs with significant operating times have experienced IGA.
At these locations, the fresh water used for condenser cooling tends to become alkaline when concentrated and,,aa ae lt '~yys~of.Qg~djga~g,~have used alkaline environments.
TabTi,:
nacteMVnomu"-.':fkjAl~~~opag<<i on rates for Alloy 600 at 650 F in various alkaline environments.
These data indicate IGA propagation rates of from 2 to approximately 8 mils per year.
C-E conducted a literature survey to determine if propagation rate data were available for Alloy 600 in acid sulfate environments at typical operating temperatures.
Limited data were available.
Reference (3) presented data on IGA and other forms of corrosion in Alloy 600 heat transfer tubes in a, 19 tube model boiler test which operated for 358 days with a
secondary environment severely faulted with simulated acidified (H2S04) cooling tower water.
The nominal sulfate level for this test was 40 ppm.
Post-test destructive examination indicated IGA was present
-jn Alloy 600 in various.IQallurgi
-conditions.
At the locations
- examined, the depth of IGA was
@7< 2-.0. aod
- mil's which is equivalent to propagation rates of 0.7 mpy, 1.0 mpy and 2.2 mpy.
Reference (4) described a
pot boiler test designed to assess the effects of sul furic aci d on the corrosion of steam generator material s under high temperature heat transfer conditions.
This test operated for 122 days with an average sulfate level of 270 ppb.
IGA was noted in two locations in the post-test examination'.
The estimated depths of attack were I and 2 mils which corresponded to propagation rates of 3 and 6 mpy, respectively.
In a third test (5),
a pot boiler operated at an average sulfate level of 85 ppm for 214 days.
IGA, with a
maximum depth of 4.6 mils, was present at two locations examined during the destructive examination.
This corresponds to a propagation rate of 7.8 mpy.
However, at these locations, the average depth of IGA was 2-4 grains
( 1.5 mils) which is equivalent to an average growth rate of 2.5 mpy.
Table 2 tabulates the IGA growth rate data discussed above.
The data show some variation, as expected.
Comparison with Table I also suggests somewhat lower growth rates for Alloy 600 in sulfates as compared to caustic solutions.
This reflects a
difference in the corrosivity of the environments and lower temperature for the sulfates test which will result in reduced growth rates.
Based on a
review of the
- data, an assumed IGA growth rate of 2.5 mils/year is appropriate for establishing a
cycle-to-cycle operation allowance for the St.
Lucie-I steam generator tubes if IGA is presumed to be progressing and if the conditions promoting IGA still exist in the steam generators.
This number is judged conservative because changes to the make-up water system at St.
Lucie-1 has resulted in a
reduction in the quantity of sulfates entering the steam generators.
Recent re-analysis of ECT data from several St.
Lucie-I steam generator inspections indicates that there has been no substantial defect growth for a least two cycles (6).
C-E judges that this is the result of improved water chemistry with respect to reduction of sulfates entering the steam generator via the make-up water systems.
This action may have eliminated the conditions required for the development and propagation of IGA.
This data supports a
zero growth per cycle operational allowance.
- However, for conservatism, a
5 percent per cycle allowance is appropriate.
REFERENCES l.
EPRI NP-3046,
. valuation of Condensate Poli shers EPRI Research Pro 'ect 623-3 Final Re or
, June 1983.
EPRI NP-3138, PWR Model Steam Generator Corrosion
- Studies, EPRI Research Pro 'ect 623-Final Re ort, June 1983.
3.
EPRI NP-3044, Corrosion Performance of Alternative Steam Generator Materials and Desi ns Volume 3, July 1983.
4.
CE-NSPD-188, The ffect of Sulfuric Acid on the Corrosion of Steam Generator Materials Under Hi h Tem erature Heat Transfer Conditions.
5.
Unpublished C-E data.
6.
Private communication, Frank Carr of FPKL.
Table 1
IGA GROWTH RATES DATA FOR ALLOY 600 IN CAUSTIC SOLUTIONS Environment Duration Da s
Average Corrosion Rate mils ear Caustic 84-91'.5 Caustic
+
12'1o NaZS04 12'i02 12% Na2HP04 12% Na2C03 12A NaF 21 - 180 4.5 - 8.3 Caustic
+ 12K'aCl 12'l 12/o MgS04 125 CaSO4 63 - 180 2-3
~
~
Table 2
IGA GROWTH RATE DATA FOR ALLOY 600 IN ACID SULFATE SOLUTIONS Test ur ation Da orrosion Rate mils ear 358 0.7 1.0 2.2 122 3.0 6.0 214 2.5*
2.5
- Average value, maximum value was 7.8 mpy
DEPTH OF GENERAL ATTACK
- i
~
DEEP ISOLATED m%)~(, P.":.'- '.:':l--'. '....
..." ~ '.
FIGURE C2-1 IGA ATTACK - TUBE SAMPLE 78110, 50X MAX PENETRATION - 56K, 1 to 3
VOLTS DEPTH OF GENERAL ATTACK.
"~,
P
~
1
. ~:
~
FIGURE C2-2 IGA ATTACK - TUBE SAMPLE 78109, 50X AVG PENETRATION - 25K, 0.1 to 0.4 VOLTS
ATTACHMENT D
CENC-1747, St. Lucie 1 Steam Generator'Allowable Tube gall Degradation, October 1986.