ML20035A228
| ML20035A228 | |
| Person / Time | |
|---|---|
| Site: | Prairie Island  |
| Issue date: | 03/22/1993 |
| From: | Office of Nuclear Reactor Regulation |
| To: | |
| Shared Package | |
| ML20035A227 | List: |
| References | |
| NUDOCS 9303250013 | |
| Download: ML20035A228 (3) | |
Text
.
i
[%g
(.
4 UNITED STATES
[
-)
NUCLEAR REGULATORY COMMISSION 2
WASHINGTON, D.C. 205560001
%...../
S_AFETY EVALUATION BY THE OFFICE OF NUCLEAR REACTOR REGULATION NORTHERN STATES POWER COMPANY PRAIRIE ISLAND NUCLEAR GENERATING PLANT UNIT 2 DOCKET NO. 50-306
1.0 BACKGROUND
On November 27, 1990 the licensee submitted for NRC staff review the design report of a new seismic Category I reinforced concrete building which houses two new diesel generators, D5 and D6, and other equipment.
It is stated in the report that the new D5/D6 building is structurally independent from the auxiliary building to its east side and to the turbine building on the north side, and the foundation of the building is also independent.
It is also stated that a 1/2-inch clear distance between the D5/D6 building and the turbine building, and a 2-inch clear distance between the 05/D6 building and the auxiliary building are designed to permit maximum out-of-phase displacements of the buildings due to seismic events. The staff has concluded that the design report was acceptable.
On December 20, 1990, during concrete placement of the "G" (North) wall of the D5/D6 building, an expanding stay form failed during the pouring of approximately 26 cubic yards of concrete and caused a significant distortion in the siding and girt framing of the turbine building wall immediately north of the G-wall between the 705 foot and 725 foot elevations. The turbine building has a concrete flood wall which extends up to the 705 foot elevation.
Concrete from the stay form failure created a slumped concrete on the turbine building wall, and the concrete was later removed. After the slumped concrete was removed from the top of the turbine building flood wall, it was noted that there was no compressible board installed to maintain the 1/2-inch seismic gap in the area.
The licensee drilled 4 concrete cores through the D5/D6 and turbine buildings and found no compressible board. Review of construction drawings revealed that the compressible board was not shown, nor did quality control (QC) inspection reports indicate that the board had been installed.
Concrete core drilling indicated that concrete leaking through the expanded metal stay forms had filled the 1/2-inch gap' space between the buildings in some areas resulting in contact between the concrete walls and loss of structural independence in those areas.
In a letter dated December 21, 1992, from the licensee to the NRC, it is stated that the licensee has reviewed documents which included contract information requests from Borson Corperation, the civil contractor, to Northern States Power Company regarding materials to be used to maintain isolation gaps, engineering change requests issued to clarify the gap requirements, photographs taken of in-place styrofoam used to maintain 9303250013 930322 DR ADOCK 0500 6
1 I
i clearances, and quality control sign offs for application of fiberboard to maintain gaps. Based on the review results, the licensee concluded that (a) drawings, specifications, procedures, and formal correspondence between Borson Corporation and Northern States Power company ensured that personnel involved were made aware of gap requirements between the new D5/06 building and existing buildings, (b) vertical and horizontal clearances were maintained at the foundation level between the turbine building and the 05/06 building and at the walls between the auxiliary building and the 05/D6 building, and (c) the only area where concrete to concrete contact does exist is between the turbine building flood wall and the D5/D6 building stairwell wall between elevations 695 and 705 feet. As a result of this concrete contact, the licensee has made two analyses and the results indicate that the as-built situation is acceptable.
2.0 EVALUATION The D5/D6 building and its foundation were designed to be physically separated from the existing plant structures. The gap filled with compressible material between the 05/D6 building and the adjacent structures was intended to avoid any contact of the structures during earthquakes. The D5/06 building design with the gap provisions was reviewed and accepted by the NRC staff. However, the gap was partially closed between the turbine building flood wall and the D5/06 building stairwell wall due to a construction error. The licensee evaluated the seismic loads and their effect on the turbine and D5/D6 buildings due to the lack of gap between the flood wall and the stairwell wall. The licensee made two analyses. The first analysis assumed that the flood wall of the turbine building was rigidly connected to the stairwell wall of the D5/D6 building and thus they would move together during earthquakes.
The analysis results indicate that the displacements, accelerations, and forces in the two buildings, except in the local regions of the flood wall and the stairwell wall, and their foundations remain practically unchanged from the results that the two buildings were assumed totally separated.
In the local regions of the walls that were assumed to be rigidly connected together, the seismic lateral force during an operating-basis earthquake (OBE) was increased from 3.78k to 17.7k for the flood wall and 13.lk to 22.4k for the stairwell wall where the smaller values were obtained assuming the two buildings were totally separated and the larger values were obtained assuming the two buildings were rigidly connected together. The second analysis assumed that the two buildings moved out of phase and the two walls would bang at each other. The maximum lateral forces during the bang are 34.8k on the flood wall and 44.1k on the stairwell wall during OBE. The licensee stated that these forces were less than the DBE strength of the walls. The licensee also doubled the forces generated by OBE and checked against the safe-shutdown earthquake (SSE) strength of the walls and found them acceptable.
The staff has reviewed the mathematical models used by the licensee and concurred with the assumptions used in the models. The staff also believes that the results of the analyses are reasonable. This is because the flood wall is a cantilevered wall which is flexible and the stairwell wall is also flexible and, therefore, the contact of the two walls would not affect the responses of the two buildings except in the local regions of the two walls that have contact. The two analyses performed by the licensee constitute the
i i
~
bounding cases of thc responses of the two buildings during earthquakes.
Since the OBE and SSE strengths of the two walls have exceeded the demands i
from the corresponding forces which could be applied to the walls during earthquakes, the walls are adequate even if they are connected together locally.
i
3.0 CONCLUSION
l The licensee has demonstrated that the responses of the 05/D6 building and the f
turbine building are not significantly affected by the lack of the 1/2-inch l
specified gap between the flood wall of the turbine building and the stairwell l
wall of the 05/06 building except in the regions of the walls that have contact. The licensee has also demonstrated that the walls are capable of resisting the increase of the forces due to the contact of the walls during earthquakes. The demonstration was performed through two analyses. One analysis assumed that the walls were connected together and move in phase during earthquakes; the other analysis assumed that the walls were in contact but moved out of phase during earthquakes. The two analyses have constituted the boundaries of the responses of the two buildings during earthquakes and their results have indicated that the buildings are safe and acceptable. The i
staff has reviewed the analyses and found that both the assumptions and results of the analyses are acceptable.
l
.