ML20100G414
| ML20100G414 | |
| Person / Time | |
|---|---|
| Site: | Crystal River  |
| Issue date: | 02/29/1996 |
| From: | STRUCTURAL INTEGRITY ASSOCIATES, INC. |
| To: | |
| Shared Package | |
| ML20100G412 | List: |
| References | |
| SIR-96-016, SIR-96-016-R00, SIR-96-16, SIR-96-16-R, NUDOCS 9602230161 | |
| Download: ML20100G414 (16) | |
Text
i 1
U. S. Nuclear Regulatory Commission Attachm:nt 1 3F0296-07 Report No.: SIR-96-016 Revision No.: 0 Project No.: FPC-02Q File No.: FPC-02Q-401 ;
February 1996 Input to Items A and C of NRC's Questions on Relief Request for Inspection of Transition Piece to Bottom Head Weld at Crystal River Unit 3 Preparedfor:
Florida Power Corporation f
Preparedby:
StructuralIntegrity Associates,Inc. I San Jose, CA l
l l
9602230161 960216 4 i PDR ADocK 05000302 StructuralIntegrity Associates, Inc.
l G PDR 1
I 9
ITEM A I
- 1. EVALUATION OF POTENTIAL DEGRADATION MECHANISMS i i
The transition piece to bottom head weld is subjected to operation at an elevated temperature of approximately 575'F for long times (on the order of 200,000 hours0 days <br />0 hours <br />0 weeks <br />0 months <br />). The potential effects of thermal aging, irradiation embrittlement, and corrosion must be considered as potential degradation mechanisms at the bottom head weld at Crystal River Unit 3. The change in
^
section from the adjacent transition piece in this area may lead to increased stresses, making it susceptible to thermal fatigue. Finally, the impact of fabrication-related defects must be evaluated.
Corrosion As noted in EPRI report NP-5461 (1), potential corrosion mechanisms include: ;
- General corrosion [
t
Intergranular (IGSCC)
Transgranular(TGSCC)
Irradiation assisted (IASCC)
- Erosion-corrosion
- Crevice corrosion l
- Pitting
- Intergranular attack i
.
- Hydrogen embrittlement
- Microbiologically influenced corrosion (MIC) )
' l The entire inside surface of the reactor vessel is clad with stainless steel with nominal l composition and properties equivalent to those ofTypc 304 [2]. The purpose of the claddmg l
1 l
4
)
SIR-96-016, Rev. 0 1
{ structuralIntegrity Associates, Inc.
i
is to provide corrosion protection to the low alloy steel plate and weld metal and to assist in maintaining reactor coolant water purity l The general corrosion rate of the stainless steel in
. a PWR environment is extremely low. Thus, the contributions of corrosion products to' degradation of fuel rod heat transfer, to turbidity of the water, or to acting as a source of activated products that might contribute to dose rates are minimized. The stainless steel is ,
1
' also extremely resistant to pitting or crevice corrosion in the controlled purity PWR coolant. l 1
Whde stainless steels have been shown to be subject to SCC in Boiling Water Reactors, the j resistance ofstainless steel weld metal to SCC is much greater [3]. Further, the susceptibility j of the reactor pressure vessel base material to growth of SCC is extremely low, especially l under the non-oxidizing conditions in the PWR reactor vessel. In the extremely unlikely event ;
' that reactor water does contact the low alloy steel corrosion degradation would be limited to j general corrosion or galvanic corrosion in the low alloy steel at rates much less than 1 mpy l
[4). The presence of stainless steel cladding, a material with extremely high resistance to velocity effects, will preclude erosion-corrosion.
1 j Hydrogen embrittlement most often occurs when materials are charged with hydrogen as a i
. result ofprocessing (e.g., plating operations), exposed to a service environment with a high f 3
- p. partial pressure of hydrogen and elevated temperature (e.g., refinery vessels), or as a result :
ofhydrogen produced from corrosion. Environmentally assisted cracking may be the result f
l of SCC, the active interaction between stresses and corrosion processes, or hydrogen
, embrittlement, where hydrogen produced from the corrosion process embrittles the structure
, at the crack tip producing a brittle type fhilure. The latter mechanism may be considered to produce che same level ofdegradation as that described for SCC in the paragraphs above (i.e.,
- degradation due to SCC or hydrogen embrittlement is considered to be unlikely for this location). Since the amount of corrosion that occurs at the cladding / coolant interface is so l small, very little hydrogen will diffuse through the stainless steel to the low alloy steel.
, The presence of stainless steel cladding effectively eliminates corrosion as a degradation
- mechanism of the low alloy steel structural components.
I i
SIR-96-016, Rev. 0 2 I f StructuralIntegrityAssociates,Inc.
i e
- - <w,r:-
Neutron Embrittlemtal 1
As shown in Table 1, for the head transition piece, a fluence of 4.79E16 n/cm2has been j 1
reported [5]. For purposes of this evaluation, this fluence has also been conservatively )
assumed for the lower shell to head transition weld (WF-154). Using the methodology provided in Reg. Guide 1.99, Rev. 2, this corresponds to a fluence factor of 0.052, resulting
- in a reference shift of 6.9'F and adjusted reference temperature (ART) of 16.9F'. This ,
reduction in fluence results in a shift which is significantly less than what is reported for the beltline materials, as can be seen in Table 1.
Thermal Embrittlement
. Ferritic materials, including pressure vessel steels such as those used for the Crystal River Unit 3 reactor pressure vessel, can be subject to a loss of toughness as a result oflong term exposure to elevated temperatures. Temper embrittlement, produced by the segregation of impurities such as arsenic, antimony, and tin to grain boundaries is a strong function of i
temperature. The effect is observed in the range 750*F to 900'F. The exposure temperature for the Crystal River Unit 3 pressure vessel (as for other LWR reactor vessels) is far too low to produce thermal embrittlement.
Fatigue As noted above, the change in section at the transition piece to bottom head area makes it a potential area of susceptibility for thermal fatigue. However, the following discussion of 1 stresses demonstrate that this vessel location will not be subjected to transient events, either in large numbers or of a magnitude that will produce thermal fatigue. Fatigue crack growth will be contributed mainly by startup/ shutdown transients. Crack growth is expected to be very small for the remaining 167 startup/ shutdown cycles at Crystal River Unit 3 [6].
SIR-96-016, Rev. 0 3 h StructuralIntegrityAssociates,Inc.
l l
Fabrication Defects
]
l With respect to Crystal River Unit 3, both preservice, as well as the first 10-year ISI inspections, did not identify any reported flaws. As such, in the enlikely event that fabrication-related defects are found, they are expected to be relatively small and insignificant.
Conclusion.a_ l i
This evaluation has demonstrated that service-induced degradation of'the transition piece to l bottom head weld as a result of corrosion, fatigue, or thermal embrittlement mechanisms is extremely unlikely. The primary contributor to the presence of flaws in that weld is due to fabrication. Prior vessel inspections did not identify any flawc and as such the existence of fabrication related defects in the bottom head weld is unlikely. Relief from inspection of the !
transition piece to bottom head weld in the Crystal River Unit 3 reactor vessel appears to be justified. :
- 2. STRESSES ANDLOADS Stresses acting on the transition piece to bottom head weld consist of the following:
- Pressure
- Thermal Transients Expansion / Contraction Stresses of Cladding Loads Associated with Welded Attachments (Flow Stabilizer Lugs) !
- Stresses Resulting from Bottom Head Penetrations The significance of these loads in relation to the weld location are discussed below.
SIR-96-016, Rev. 0 4 f StructuralIntegrityAssociates,Inc.
-. ~. - - . , . . ..- --.
I Pressure An analysis' due to internal pressure of 2240 psig was performed in Reference 7. The l geometrical details of the reactor pressure vessel shown in Figure I were used to construct i the finite element model shown in Figure 2. The resulting through-wall pressure stresses in the transition piece to bottom head weld location are shown in Table 2.
t Thermal Transients !
Several transient are described in the original Reactor Vessel Stress Reports [2].' However, ;
it was determined that the most significant transients in the bottom head region are those associated with plant heatup and cooldown. Analyses were performed using the finite j element model shown in Figure 2 to determine the stresses and temperatures associated with l the heatup and cooldown transients. Maximum through-wall thermal stresses are shown in :
Table 2. These stresses are relatively small compared to the pressure stresses. Due to the adjacent thickness change of the transition piece, the thermal stress distributions are altered '
from those expected for a shell remote of a discontinuity.
Exoansion and Contragtion Stresses The expansion and contraction stresses result from differences in the coefficient of thermal expansion between the stainless steel clad and the low alloy pressure vessel steel. As required by the ASME Code, during construction, post weld heat treatment (PWHT) was performed on the vessel at Crystal River Unit 3 following welding operations, including the application of the cladding. PWHT is intended to reduce residual stresses arising from welding. This was achieved by subjecting the entire vessel to a temperature of about 1100* - 1200'F for 12 to 48 hours5.555556e-4 days <br />0.0133 hours <br />7.936508e-5 weeks <br />1.8264e-5 months <br /> and then gradually cooling it uniformly to room temperature. i I
SIR-96-016, Rev, 0 5
{ StructurallatogrityAssociates,Inc.
i
L i
It is generally believed 'that due to PWHT, the vessel is stress free at the operating temperature of approximately 575"F [8]. However during cooling to room temperature, tensile residual stresses are generated in the stainless steel cladding because stainless steel has !
a higher coefficient of thermal expansion than the low alloy pressure vessel steel. It has been established that the tensile stress in the stainless steel at room temperature is on the order of f
yield stress (30-35 ksi for stainless steel) [9]. The cladding stress is compensated by an essentially uniform coswdve force in the low alloy base metal The postulated distribution of the clad stress distribution at room temperature is shown in Figure 3 for the 5" thick bottom head. Because of this stress distribution, the associated stress intensity factor tends I to have a very sharp exponential decay through the wall of the vessel for inside surface cracks. Since the stress in the base metal is compressive, cladding induced stresses are not a factor for embedded flaws, the type of flaw most likely to be encountered during inspection.
At normal operating temperature (575'F), the effects of cladding stresses diminish due to the ,
relative coefficient of thermal expansion. It is predicted that the cladding-induced tensile stress reduces to a value of approximately 5 ksi at the normal operating temperature, with a l similar percent reduction in the base metal.
Loads Associated with Welded Attachments (Flow Stabilizers) i In the original design, there are twelve flow stabilizer plates welded to the lower head in the j vicinity of the transition piece to lower head weld with a size of approximately 37 inches by 10 inches. Before plant operation, these plates were cut-out and leaving only 1 inch of the original 10 inch width protmding from the surface of the cladding. As such, the mechanical loads on these lugs are insignificant. The plates were made from stainless steel and as such there is no differential thermal expansion stress between the plate and the stainless steel cladding. The location of these plates extend 1 inch below the centerline of the weld, i
i SIR-96-016, Rev. 0 6 h StructuralIntegrityAssociates,Inc.
l l
1 Loads from Bottom Head Penetrations
-i The loads on the bottom head due to the fifty-six (56) 3/4-inch penetrations were analyzed
, ' in the original Siress Report (10). These penetrations are at least 10 inches away from the
! transition piece to bottom head weld and, as such, the weld location is unaffected by the loads .
- from these penetrations.
ITEM C '
- l. FRACTURE TOUGHNESS AND ASME CODE, SECTION III, APPENDIX G, FLAW f
! SIZE 1 ,
}
- The critical flaw size in the bottom head region is typically larger than the critical flaw size i 1
- . m the beltline region because the bottom head, unlike the beltline, does not experience significant irradiation effects from exposure to neutron fluence. Although the thickness of the ;
! bottom head is smaller than the beltline region, the stresses due to internal pressure are [
i L qurtailed in the bottom head because the smaller thickness is compensated by the spherical shape (as opposed to the cylindrical shape of the beltline). For the bottom head weld in question, there are no penetrations or appurtenances immediately adjacent to the weld that - .
j would necessitate the use ofa stress concentration factor in determining the stresses. Thus,
, with lower stresses and a higher available toughness, the flaw tolerance capabilities of the .
bottom head region are expected to be significantly higher than the beltline region.
L 4
- . To demonstrate the increased flaw tolerance of the bottom head region, analyses were i performed in accordance with ASME Code,Section XI, Appendix G to establish curves of ,
allowable pressure versus fluid temperature for various flaw sizes. These analyses utilized the limiting reference temperature for all plate and weld material in the bottom head region l
(RTm = 16.9 F). Three flaw sizes were evaluated: 1/4T,3/8T, and 1/2T. Various fluid f
- temperature heatup and cooldown rates ranging from that identified on the current P-T Tech. !
SIR-96-016, Rev. 0 7 l
{ StructuralIntegrityAssociates,Inc. I
r Spec. curve up to 100 F/hr were evaluated with respect to both outside surface and inside surface flaws. The limiting 100 F/hr results are shown in Figure 4 for all three flaw sizes, and sre compared against the currently established Tech. Spec. P-T curve based on the limiting beltline region, for the hydrotest condition.
. The results shown in Figure 4 demonstrate that the current P-T curve for the limiting beltline region significantly bounds that of the lower head region. For flaw sizes as large as 1/2T, significant margin remains. The primary reason for the significant difference between the bottom head and beltline regions is a result of decreased beltline toughness caused by irradiation. This effect is apparent when the adjusted reference temperature (ART) of 208'F (using the fluence specified on the Tech. Spec. P-T curve) for the limiting beltline material (upper shell plate) is compared to the 16.9'F value described above for the limiting bottom head material (head transition piece). Thus, the fracture toughness capability of the bottom head region is significantly bounded by the existing Tech. Spec. P-T curve for the beltline region.
1 SIR-96-016, Rev. 0 8 f StructuralIntegrity Associates, Inc.
l l
REFERENCES
- 1. J.F. Copeland, et. al., " Component Life Estimation: LWR Structural Materials Degradation Mechanisms", EPRI report NP-5461, prepared by Stmetural Integrity Associates, Inc.,
September,1987.
- 2. Stress Analysis Report #13," Thermal / Mechanical Analysis of the Reactor Vessel Shell for j Florida Power Corporation, Crystal River Unit 3", Rev.1, January 1974, B&W Contract No.
620-0007-51.
- 3. W. S. Hazelton, W. H. Koo, " Technical Report on Material Selection and Processing i Guidelines for BWR Coolant Pressure Boundary Piping", NUREG-0313, Rev.' 2, USNRC,
- January,1988. l
- 4. G. J. Licina," Nuclear Power", Authorized Reprint from ASTM Manual 20,1995. i
- 5. B&W-2108, Rev.1, " Fluence Tracking System", B&W Nuclear Service Company, May 1992.
- 7. SI Calculation Package, FPC-01Q-301, " Reactor Pressure Vessel Finite Element Stress Analysis", Rev. O.
- 8. ASME Section XI Task Group on Reactor Vessel Integrity Requirements, " White Paper on Reactor Vessel Integrity Requirements for Level A and B Conditions" EPRI TR-100251, January 1993.
- 9. ASME Boiler and Pressure Vessel Code,1989 Edition,Section III, Appendices.
- 10. Design Report #6 " Thermal Mechanical Stress Analysis of Reactor Lower Head & Support Skirt", Vol. 2 of 2.
SIR-%-016, Rev. 0 9 f StructuralIntegrityAssociates,Inc.
i l
1 i
Table 1 l Initial and Adjusted RTm7 of Selected Plates and Welds at Crystal River Unit 3 - )
l l
Estimated Chemistry Chemistry Adjustments For 1/4 T Part Name & MK Heat Slab initial RT , Factor ARTwot Maripn ARinor i Meterial No. No. No. *F Cu wt % Ni wt % *F *F o s, 'F o *F *F >
Upper Shell A1-2071 C4344 2 20 0.20 0 54 141 80 154.5 17.0 0.0 208 5 SA-533-65 Grade B, Class 1 i (omhm Benw ,
Head Transition - 124W295VA1 - 10 0.10 0 80 67.00 35 1.7 00 16.9 Piece
?
SA-508-64, Class 2 (Code Case 1332-3)
_ Unh@
( Bottom Henry Wan Thickness (inches) Attenuation.1/47 Fluence Fluence Factor. FF Full 1/4T Fluence at 10 e 4 2*' 1/4 T foamm 3
Upper Shell 8 44 2.11 2.29E+19 nlcm 0.603 1.38E+19 nicm3 1.090 Head Transition Piece 5 00 1.25 4.79E+18 n/cm' 0741 3 55E+16 n/cm' O.052 1
B I
1 SIR-96-016, Rev. 0 10 h StructuralIntegrityAssociates,Inc.
1 1
1 i
l I
Table 2 Stresses for Bottom Head-To-Transition Piece Weld Pressure (2240 psi) End of licatup 1001/hr l End of Cooldown 100Tihr Stress. Esi Stress, ksi Temp.. "F Stress. ksi Temp T Distance, in Axial lioop Axsal floop Noic 1 Axial floop Note 2 0.00 19.670 15.080 0.152 -13.220 572.6 -2.536 13.730 301.1 0 50 19.500 15.030 0.423 -11.780 592.7 -2.227 12.530 306.3 1.00 19.290 15.000 0.573 -10.420 587.7
-1.803 11.390 311.0 l.50 19.080 14.970 0.575 -9.235 583.3 -1.277 10.390 315.1 2.00 18.860 14.930 0.405 -8.246 579.6 0.616 9.573 318.7 2.50 18.650 14.900 0.056 -7.458 576.4 0.194 8.936 321.7 3.00 18.470 14.870 -0.460 -6.862 553.9 1.146 8.472 324.1 3.50 18.330 14.850 1.132 -6.461 572.0 2.229 8.188 325.9 4.00 18.250 14.840 -1.936 -6.249 570.7 3.423 8.081 327.1 4.50 18.220 14.860 -2.857 -6.243 569.9 4.706 8.172 327.7 5.00 18.190 14.850 -3.804 -6.371 569.7 4.968 8.375 327.9
- 1. RCS Temperature at 6041
- 2. RCS Temperature at 280T SIR-96-016, Rev. 0 11
{ StructuralIntegrityAssociates,Inc.
' ,1 a
i A dL I j db JL A A A A A l' '
$ f l w l- 4 , "- ; l l ' '
, , L i
I l
' 'O L I l . '
E ' 12.125" min -> +- @
I m - Q, '
I -
.' g 84.1875" R , h e
l I d ir h !
- Qm h l
I
' 8
" n
. l k E I E '
n i ui R l N M t 4
'
- I Q
I i 8.438" min -+ 4-i e i
y 85.5" R ,
e i 4
. e ;
I l i t,
3 t
,,es, J ,
i l
v l
l 5"s S./ '
l 77mamr l LX v s l Crystal River Unit 3 - Reactor Pressure Vessel.
95091r2 Figure 1. Vessel Geometry at CR Beltline and Bottom Head Regions M i m in #r ssociates, Inc. !
SIR-96-016, Rev, 0 12
1 l
i 1
N s
.s 4
i l g s
s T
t
\
s
/
i l Figure 2. Reactor Vessel Axisymmetric Finite Element Model l
l SIR-96-016, Rev. 0 13 StructuralIntegrity Associates, Inc.
l
. Stress (ksi) a 3116" 5" 35 kal 7,
~
/ ~
A ~
1.3 W/H//////////////// -1.3 seame Figure 3. Clad Residual Stress SIR-96-016, Rev. 0 14 h StructuralIntegrity Associates, Inc.
__=_____________-
6 9
/
6
/
2 0
0 7
e v
r 0 0
T 4 8 2 T T C a 6
/ / /
1 3 1 c T
e 2
- r. r. r. p /
1 h/ /h h S
/
d F F F
- *
- h n
O 0 0 c a d 0 0 e T, i 1 1 T
- - 8
- - /
0 3
- - - 0
- - - 5 ,
- - - T 4
/ -
1 -
f o -
s e
i z
S 0 w 0 ) a 4 F* l F
( r fo e e r
v r t u n u ar i o
C e g p e T- m R P e d 0
0 T a 3 e H
m
/
t o
7' B t
o r
fo
.I 0 e 0 v r
. I
- ' llI 2 u
.I . .*f.. C l
- I*f llf / T-
- * . .f.I. lp . P
- l. I./ / *
/
4
- t.
- e
.. f. f.
- r
~- * /. 0 u
g
- i
'., 0 i
'f. / '. 1 F
]
e v
0 ru G 0 0 0 0 O C 0 0 0 0 0 T-5 0 5 0 5 P 2 2 1 1 [
S L
n, c,e3 e oWmo n. X T-P ManEi EE=,h DIOi5, i
i S ni ~y
$N6eb *.pQ.O G ll!l1 t