ML20034A970
| ML20034A970 | |
| Person / Time | |
|---|---|
| Issue date: | 03/20/1990 |
| From: | Wegner M NRC OFFICE FOR ANALYSIS & EVALUATION OF OPERATIONAL DATA (AEOD) |
| To: | |
| Shared Package | |
| ML20034A967 | List: |
| References | |
| TASK-AE, TASK-E90-04, TASK-E90-4 AEOD-E90-04, AEOD-E90-4, NUDOCS 9004250155 | |
| Download: ML20034A970 (5) | |
Text
. _. _ _. _. _ _ _ _ _. _.. _ _.
j
- i AE00.[NGINEERING EVALUATION
{
UNIT:
Multi le EE REPORT N0.: AE00/E90-04 DOCKET NO:
Multi le DATE: March 20, 1990 LICENSEE:
Multi le CONTACT: Mary $. Wegner
{
4 f
h
SUBJECT:
$ WELLING AND CRACKING IN HAFNIUM CONTROL RODS 0
1 SUMARY Hafnium was introduced as a neutron absorber for use in control rods in 1980.
It was meant to be used in addition to carbon tetraboride (84C) as a replace-mentforsilver-indium-cadmium (Ag-In-Cd). Hafnium control rods were intro-ducedasoriginalequipmentinseverallaterpressurizedwaterreactors(PWRs).
3 i
They have experienced swelling and cracking as determined by several-licensees i
usingeddycurrenttesting(ECT). Of the 14 plants licensed to use hafnium four rods, six have replaced or will replace them, two never installed them, hem.
have recently or may soon install them, and two are continuing to use t I
J Analyses by Westinghouse (2) of the effects of swelling of control rods predict
.I that neither physical interference with insertion nor mechanical resistance i
l' which would increase rod drop time will be of a large enough magnitude to invalidate FSAR safety analysis conclusions. Eddy current test results to date j
have upheld these analyses.
Given the limited examinations to date, it would be prudent to require all l
plants licensed to use hafnium control rods to perform NDE of these rods to i
and wear over at least determine the extent if any of cracking, swellinglife of the hafnium rods.
i h
several fuel cycles corresponding to the useable DISCUSSION Events Wolf Creek November,1988(1) l Eddy-current testing (ECT) of Wolf Creek's hafnium absorber control rods were performed by Cosbustion Engineering, the non-destructive examination (NDE) l contractor, using encircling and profiling coils in the differential mode. The acceptance criteria stated that a minimum of 50 percent cladding cross-sectional area at any axial location must exists any rodlet showing wear-i through of cladding would either be replaced or positioned axially at an eleva-I tion other than at a guide card (a section of the guide tube that guides each i
rodlet) when the reactor control cluster assembly sRCCA) was fully withdrawn; and the free rotation of the RCCA wculd be demonstrated during startup testing and periodically during operation.
1 The ECT results identified ' numerous anomalous indications on essentially i
every rodlet of each RCCA.' These anomalies consisted of cladding I.D. indi-
}
cations and rodlet swelling. The indications included average cross-sectional clad area losses of up to 40 percent with some areas approaching through-wall.
Rodlet swelling was at least 15 mils in some cases.
l l
9004250105 90043g POR ORG NEXD PDC
{
I i
.t.
Rodlets in four of 53 RCCAs had cladding loss which exceeded the 50 percent minimum cross-sectional area requirement, with the worst condition being an a> proximately 80 percent loss. Later the NDE contractor infomed the 1< censee tiat the 15 mil swelling was closer to 30 mils with one rodlet in four exhibi-ting this condition. [Kach rodlet is 0.381 inch, 0.D., and the guide card i
holes through which the rodlets pass are 0.420 inch, I.D.
Swelling of 0.039 inches or more would interfere with rod motion.]
Foreign September,1988 (3)
The unit was at 75 percent. power when an electrical fault caused a turbine trip and reactor scram. During the scram, an RCCA failed to fully insert.
The rodlets were found to have swelling, cracking, and wear-through. An end i
plug of one of the rodlets was found in an adjacent fuel bundle. The other RCCAs showed cracking and swelling.
Shearon Harris November, 1989(4,5,6)
The stainless steel tip of one rodlet in a D-bank RCCA was missing during inspection of RCCAs for swelling and the assembly exhibited an additional 150 pounds of withdrawal tension. Two other assemblies had an additional 200 pounds increased withdrawal tension over the normal 470 pounds.
Background
i Hafnium was introduced as a neutron absorber for use in control rods in 1980.
It was meant to be used in addition to B4C as a replacement for Ag-In-Cd.
In-reactor experience with hafnium as an absorber was limited to Yankee-Rowe (operating), Indian Point 1 (shutdown), and Shippingport (shutdown).
The hafnium rods were designed to replace.the Ag.In-Cd rods on a one-to-one l
basis without major design changes.
It was claimed'that' hafnfum would provide improved drop time margin, had excellent material properties, and had less reactivity worth depletion. Hafnium control rods were introduced as original equipment in several later PWRs, in particular, Wolf Creek, Callaway, and Shearon Harris.
Because of the limited experience with the use of control rods with hafnium absorber in commercial nuclear power plants,l inspection of a sample of the the NRC staff decided that a surveillance program was necessary. A visua control rods at Wolf Creek and Callaway, the first two plants to use the hafniumrods,wasrequired.(7)
The visual inspection of the rods was carried out at Wolf Creek, but the results were not meaningful. The subsequent ECT, however, proved very useful. The calibration of the standards used in the Wolf Creek ECT were for 0.396 inches which makes the reports of swelling beyond that diameter questionable.
Following a re-review of the data, the NDE contractor decided that no swelling beyond 0.410 inches had existed and that, for similar reasons, the average wall cross-sectional area loss was as high as 40 percent.
3 i
l i
Westinghouse attributed the swelling to hydriding of the hafnium.
Hydriding occurs when hydrogen comes into contact with hafnium. Hydrogen from the RCS diffuses through the stainless steel cladding of the rodlet and reacts with the hafnium to form hafnium hydride. Hafnium hydride has a volume increase of 1
j nearly 15 percent over that of hafnium metal.
If the entire volume change of 1
hafnium hydride occurred so as to increase the diameter of the rod, the theo.
j retic maximum swelling would be 0.026 inches. This would still be less than i
thepassagewayintheguidecardsbyatleast0.010 inch.(2) i Other Results i
Following the receist of the Wolf Creek inspection results, the licensee for Callaway reviewed tie previous inspection results for the hafnium control rods installed at that plant and determined that, although the equipment used was not calibrated for ID indications and swelling, there was reason to believe that swelling had occurred. The coil used in the inspection had an I.D. of 0.400 inches. Therefore the detected swelling could not have exceed 0.400 inches.(8)
Millstone 3 reports similar findings following their June 1989 ins section of hafnium RCCAs to that of Wolf Creek. Millstone plans to replace tie hafnium rods when they can no longer be used with Ag-In-Cd rods. Other information on the use of hafnium rods in PWRs may be found in the Appendix. (10) l FINDINGS 1.
Hafnium control rods in use in certain pressurized water reactors have experienced swelling and cracking as detemined by several licensees using eddy current testing (ECT).
2.
Of the 14 plants licensed to use hafnium rods, six have replaced or will replace them, two never installed them, four have recently or may soon install them, and two are continuing to use them.
3.
Analyses of the effects of swelling of control rods postulate that neither physical interference with insertion or mechanical resistance which would increase rod drop time would be of a large enough magnitude to invalidate FSAR safety analysis conclusions. ECT findings to date have upheld these analyses.
CONCLUSIONS Given the limited examinations to date, it would be prudent to continue to i
require all plants licensed to use hafnium control rods to perform NDE of these rods to determine the extent if any of cracking, swelling, and wear over at least several fuel cycles corresponding to the useable life of the hafnium rods.
9 4
REFERENCES 1.
NRC inspection Report 50/482-88/035 2.
WCAP-12204 FProprietary), " Hafnium RCCA Issues," March,1989 WCAP-1220$ (Non-Propretary), ditto 3.
[ Proprietary Report) 4.
Region !! Mornin9 Report, $1/15/89 5.
NRC Inspection Report 50/400-89/018 6.
NRC Inspection Report 50/400-89/028 7.
Minutes of NRC Meeting May 13,1982 with Union Electric, Kansas Gas and Electric, et al.
8.
Letter from W. J. Johnston, Westinghouse to C.E. Rossi, NRC dated 12/8/88 9.
Memorandum from S. S. Wu, NRR to V. W. Hodges, NRR dated 2/17/89
- 10. TelecopyofInternalReport(Draft)fromNortheastUtilities,3/13/90 l
l
l APPENDIX i
4 Table 1.
Status of Plants Which Have Hafnium Control Rods Installed Plant Status of Control Rods Braidwood 142 Licensa.1 for Hf 7/89 q
Byron !&2 Licensed for Hf 7/8g l
Callaway Continue to use Hf rods, inspect i
Comanche Peak 182 Will not use Hf rods i
Millstone 3 Will replace remaining hafnium rods w/Ag-In-Cd rods when the hafnium rods crack or swell beyond acceptable limits Shearon Harris Swelling, cracking, broken tip To replace Hf rods South Texas 1 Replaced Hf rods 2
Hf rods to be replaced Vogtle 1&2 Hafnium rods to be replaced Wolf Creek Continue to use Hf rods, inspect ll l
I i
l l
F
.