ML19256G314
| ML19256G314 | |
| Person / Time | |
|---|---|
| Issue date: | 12/12/1979 |
| From: | Office of Nuclear Reactor Regulation |
| To: | |
| Shared Package | |
| ML19256G309 | List: |
| References | |
| NUDOCS 7912310024 | |
| Download: ML19256G314 (5) | |
Text
.
UNITED STATES y
'1 NUCLEAR REGULATORY COMMISSION s.
WASHINGTON, D. C. 20555
~
A
/
SAFETY EVALUATION ATYPICAL WELD METAL ENGINEERING BRANCH DIVISION OF OPERATING REACTORS Introduction During 1978, B&W initiated work contracted with the B&W Owners Group on a program for evaluating the material properties of "early vintage" 177-fuel assembly reactor vessel welds. One of the work phases in this program had the objective of characterizing the chemistry of reactor vessel (RV) beltline welds. Extensive chemical analyses of the archive sources of RV welds have:been performed as part of this work. Two samples of test weldments made for the Crystal River 3 reactor vessel surveillance program were part of the weld metal archives subjected to chemical analysis. The resul+.s of these analyses, performed by the Mt. Vernon Works Qaulity Assurance Laboratorf, indicated that one of these samples had atypical concentrations of nickel.and silicone, while the concentrations of the other elements were in the normal range for MnMcNi:Linde 80 submerged-arc RV weldments. The other sample had the nominal chemistry.
The atypical weld was made with weld wire designated by heat number 72105. This heat of weld wire was used in the fabrication of 12 reactor vessels. These vessels and the location of possible atypical welds are listed in Table 1.
Charpy V-notch tests on the atypical weld metal resulted in a higher than normal value of RTNDT, partially because of unusually high scatter. Therefore, we re-quested that the licensees of the above plants administratively apply revised pressura-temperature operating limits that reflected the possible presence of atypical weld metal.
In calculating these limits the atypical we.d was assumed to have an unirradiated RTNDT of 1200F and radiation damage is predicted by the upper limit line in Regulatory Guide 1.99.
Currently all the affected plants are operating under such revised limits.
Discussion 10 CFR Part 50, Appendix G " Fracture Toughness Requirements", requires that pressure-temperature limits be established for reactor coolant system heatup and cooldown operations, inservice leak and hydrostatic tests, and reactor core operation. These' limits are required to ensure that the stresses in the reactor vessel remain within acceptable limits. They are intended to provide adequate margins of safety during any condition of normal operation, including anticipated operational occurrences.
The pressure-temperature limits depend upon the metallurgical properties of the reactor vessel materials. The properties of materials in the vessel beltline region vary over the lifetime of the vessel because of the effects of neutron irradiation. One principle effect of the neutron irradiation is that it causes the vessel material nil-ductility temperature (RTNOT) to increase with time.
The pressure-temperature operating limits must be modified periodically to account for this radiation induced increase in RTNDT by increasing the temper-ature required for a given pressure. The operating limits for a particular operating; period are based on the material properties at the end of the operating Byperiodicallyrevisingthepressure-temperaturelimitsYgcountfor" #
period.
1660 23
=
radiation damage, the stresses and stress intensities in the reactor vessel are maintained within acceptable limits.
The nagnitude of the shift in RTNDT is proportional to the neutron fluence that the materials are subjected to. The shift in RTNDT can be predicted from Regulatory Guide 1.99.
To check the validity of the predicted shift Sur-in R"NDT, a reactor vessel material surveillance program is required.
veillance specimens are periodically removed from the vessel and tested. The results of these tests are compared to the predicted shifts in RTHDT, and the pressure-temperature operating limits are revised accordingly.
Since the unirradiated RTNDT of the atypical weld metal was determined to be high, and it was assumed to be sensitive to radiation damage, the atypical weld metal would generally be the limiting vessel material. Therefore, all licensees with vessels that might have been fabricated with atypical weld metal were required to revise their pressure-temperature operating limits to reflect the possibility that atypical material was used in their construction.
Evaluation To resolve the atypical weld issue, B&W has conducted an extensive investigation of records, metallographic examinations, chemical analyses, and fracture mechanics tests on both unirradiated and irradiated atypical weld material. The results of this program are presented in BAW-1556.
Since 1966, 42 heats of submerged-arc weld wire have been purchased for RV and surveillance specimen fabrication at Mt. Vernon, and, except for the discovery of the partial-thickness off-chemistry conditions in the second CR-3 surveil-lance block, there is no evidence that atypical weld wire reached the shop floor.
The results of more than 2000 chemical analyses have been reviewed relative to the 42 wire heats. All, except for the one batch of Crystal River material, have been within the normal ranges. These' include through-thickness tests from seven RVs fabricated at Mt. Vernon and tests of wire currently in inventory.
Detailed metallographic examinations were performed on seven fractured Charpy specimens. Both macro-and micro-examination techniques were employed, as well as a fractographic examination with a scanning electron microscope.
Relatively little porosity was noted in any of the weld material examined.
Examinations revealed columnar grains outlinad by proeutectoid ferrite. The orientation of the grains and the unusually hi h amount of proeutectoid ferrite are believed J
to be the cause of the high scatter in the Charpy data.
Numerous chemical analyses were performed on the atypical weldment. The bulk of :hese analyses were obtained usirg a Jarrel-Ash emission spectrometer. The con:entrations of 10 elements were rxaasured by this technique. X-ray floures-cense analysis was used to measure the concentrations of nickel, molybdenum, and copper in irradiated Charpy specimans. Results show that the copper content was high, averaging between 0.4 to 0.5%.
The chemistry cf atypical material is :ompared to typical material in Table 2.
Charpy V-notch tests were performed on both unirradiated and irradiated material.
The irradiated specimens were irradiated in the Crystal River 3 reactor vessel.
Dynamic and static fracture toughness tests were conducted on one inch thick 1660 232 f
compact tension specimens at room temperature. Although the dropweight NDT' 0
is -20 F, the results of the Charpy tests show that 50 ft-lbs of energy is absorbed at 1500F, therefore the unirradiated value of RTNDT is 90 F.
Using RTNDT equal to 900F, the toughness properties obtained from the fracture mechanics tests, KIc (static) and kid (dynamic), are conservative (lie above) the KIc curve in ASME Code,Section XI and the KIR curve in ASME Code, Section 0
III respectively. Using an RTNDT of -20 F (the dropweight NDT), the fracture mechanics data fall within the scatter of data on normal material used to obtain the KIc and KIR curves. This indicates that the RT DT value of 900F N
is conservative.
The effect of irradiation on the mechanical properties of atypical material have been avaluated, using the test results on Crystal River 3 sury illancg speci-mens. These specimens were subjected to a fluence of 1.1 X 10 o n/cm'. This fluen:e produced an increase in RT DT of 350F.
N Frcm aur review we conclude that the probability that atypical weld metal was used in fabricating the subject vessels is very low.
However, we conclude that in calculating pressure-temperature operating limits for these vessels, the properties of atypical material should be considered. As discussed above, we have c eternined that an initial value of RTNDT of 900F is a very conservative value TheincreaseinRTDTduetoiradiatjonshouldbebasedonthemeasured N
value of 35 F at a fluence of 1.1 X 10 8 n/cm and the damage prediction _ slopes in Resulatory Guide 1.99.
We also conclude that the administratively applied pressure-temperature operating limits may be removed from those plants whose Technical Speci-fications were not modified to include limits based on the atypical weld material.
The operating limits in the Technical Specifications for Browns Ferry 1 are presently being reviewed and will include limits based on the atypi:al weld metal.
The Technical Specifications for Rancho Seco have operating limits based on the atypical material that are more restrictive than they would be if based on the criteria developed from this review.
Midland 1 is being reviewed for an Operating License and its Technical Specifications have not been finalized to date.
The pressure-temperature limits for Three Mile Island 2 should be revised to reflect the possible I
use of atypical material in this vessel.
This poses no irmiediate problem since this plant ;is not curre. '.ly operational.
The Technical Specifications of the other eight subject pla 's contain pressure-temperature operating linit ; that.are in accordance with Appendix G,10 CFR Part 50 based on both typicil and atypical weld metal properties.
At these plants we have found that atypic,11 material is not currently limiting due to the low amount of irradiation damage.
The staff will continue to monitor the effects of radiation on the prok ies of tte atypical weld material. Six capsules containing the atypical weld metal are in the Crystal River 3 surveillance orogram. One of these capsules has already been removed and tested. Also, there is enough atypical material in st3 rage at B&W to fabricate fracture toughness specimens up to 1.0T compact fracture toughness specimen size.
Detaileo information collected from our review, including our calculations, are retained in our Branch files.
1660 233
. TABLE 1.
LOCATION OF POSSIBLE ATYPICAL WELDS t
PLANT LOCATION OF WELD B&W Oconee 3 Center Cire. Beltline TMI 1 Upper Cire. Beltline Lower Circ. Beltline TMI 2 Dutchman to Lowerhead ANO 1 Head to Flange and Nozzle to Shell Midland 1 Center Cire. Beltline CR-3 Center Cire. Beltline Rancho Seco Vertical Seam Beltline WESTINGHOUSE Zion 1 Inter to Lower Circ. Beltline Zion 2 Vertical Seam Beltlire 0
(o and 180 )
Turkey Pt. 4 Nozzle Shell to Interm. Circ.
a gE, Br. Ferry 1 Shell to Flange and Longitudinal Weld in Beltline Quad Cities 2 Closure Head to Flange
{ () 60 9
\\
l TABLE 2.
ATYPICAL WELD CHEMISTRY C
Mn P
S Si Cr Ni Mo CR-3 Weld
.08 1.f> 5
.021
.013 1.0
.0/
.10
.45 Mn-Mo-Ni
.08 1.6
.018
.015
.5
.07
.60
.40 (Typical) 1660 235 1
,