ML20237K292
| ML20237K292 | |
| Person / Time | |
|---|---|
| Site: | Diablo Canyon  |
| Issue date: | 06/17/1987 |
| From: | Olshan L Office of Nuclear Reactor Regulation |
| To: | Office of Nuclear Reactor Regulation |
| References | |
| OLA-I-SC-005, OLA-I-SC-5, NUDOCS 8709040289 | |
| Download: ML20237K292 (14) | |
Text
y f
4' c. -.
f., fQ
/.
j $' "'49'o x
'?^
+
N UNITED ST'ATES l'
,, ( (',i NUCLEAR REGULATORY COMMISSION
<*";,j E -
g,,
WASHINGTON, D. C. 20555
\\'v4/
January 12, 1987 D ' N'T s mAz3-oa
. dp> 7pr
'87 /G 26 P4 :02 Docket Nos. STN 50-454 I-s& 5 c
and STN 50-455 vuu LICENSEE:
Commonwealth Edison Company FACILITIES:
Bryon Station, Units 1 and 2
SUBJECT:
MEETING StlMMARY - STRUCTURAL ANALYSIS OF SPENT FUEL POOL EXPANSION On December 22, 1986, a meeting was held in Bethesda, Maryland to discuss the structural analysis of the spent fuel pool expansion for Byron Station, Units 1 and 2.. Members of the NRC and its consultart, Brookhaven National Laboratory (BNL), and Commonwealth Edison Company and its consultants, Sargent and Lundy (S&L) and Joseph Oat Corporation, were present. Attendees are listed in.
The licensee provided the followino Joseph Oat Corporation drawings during the meeting:
1 l
E - 8119 Pool Layout ~
l D - 8123 Supports and Handle D - $122 Detail - Rack J1 D - 8243 Weld Map - Region II D - 8120 Details - Region I D - 8123 Details -' Region II provides a lisi of the NRC concerns and their resolution that were discussed at the meeting.
Sincerely, hai 'b.
d&L Leonard N. 01shan, Project Manager Project Directorate #3 Division of PWR Licensing-A
Enclosures:
As stated cc: See next page 7
K I
7 l
C 1
J
QUCLIAR QEGutAf0Rf COWISSKC Docket N 50-T'E 0L A 9;N '_I omew r,6. N
(~l b E ( 42 C y A W in the esttes of Staff ICENTIFlfD Applicant RICIIVED Intavanor RIJECTIC Lont's Ott's Coritiactor CA!E
~
'I Otteer Macss Reportet
. STYL 4
'MMMd
(
^I Mr.bennis'L.Farrar Byrnn - Station
- Commonwealth Edison Companv Units 1 and 2 CC:
. Mr. Ililliam Vortier.
Ms. Diane Chavez..
Atomic Power Distribution-528 Gregory Street Westinghouse Electric Corporation Rockford, Illinois 61108 Post Office Box 35; Pittsburgh, Pennsylvania 15230 Regional Administrator, Region III-U. S. Nuclear Regulatory Commission Michael Miller 799 Roosevelt Road Isham, Lincoln & Beale Glen Ellyn, Illinois-60137 One First National Plaza 42nd Floor Joseph Gallo, Esq.
I Chicago, Illinois 60603 Isham, Lincoln.& Beale d
Suite 1100 Mrs. Phillip B. Johnson 1150 Connecticut Avenue, N.W.-
1907 Stratford Lane Washington, D. C. '20036 Rockford, Illinois 61107 Dr. Bruce von Zellen
'Douglass Cassel, Esq.
109 Ni Dearborn Street-Department of Bioingical Sciences Suite 1300 Northern Illinois University Chicago,-Illinois _ 60602 DeKalb, Illinois 61307 Ms Pat Morrison Mr. Edward R. Crass 5568 Thunderidge Drive Nuclear Safeguards & Licensing Rockford, Illinois 61107 Saroent & Lundy Engineers 55 East Monroe Street Ms. Lorraine Creek Chicago, Illinois 60603 Rt. 1, Box 182 Manteno, Illinois 60950 Mr. Julian Hinds U. S. Nuclear Regulatory Commission Byron /Pesident Inspectors Offices 4448 Geman Church Road Byron, Illinois 61010 Mr. Michael C. Parker, Chief Division of Engineering Illinois Department of Nuclear Safety 1035 Outer Park Drive Springfield, Illinois 62704 i
L.;
__.m..
J__.___
m_
ENCLOSURE 2 MEET 1HG ATTENDEES STRUCTURAL ANALYSIS OF SPENT FUEL POO December 22, 1986 NRC
~~
COMMONWEALTH EDISON H. Ashar R. Ballard*
K. Ainger J. Stevens*
S. Gubin i
RNL g
G. DeGrassi M. Phillipoupolis S. Putnam
_ JOSEPH OAT K. Singh
- Part-time Attendance a
-J
ENCLOSURE 2 NRC CONCERNS AND RESOLUTIONS 1.
.resulting'from fuel to cell wall and baseplate impac
. uel been evaluated?-
Have these effects Resolution:
The licensee.will provide a copy of the Turkey Point he
. transcript on the integrity of the' Westinghouse Fuel under seismic loading.-
2.
of the liner due to rack foot impact?Has the pool floor evalu How was this done?
perforation Resolution:
The licensee will provide more information on ' peak stresse (strains) in the liner.
3.
Since the seismic impact loads may be sensitive to gap sizes OA procedures in place to assure that the proper inter-rack an
,are$there rack gaps are maintained during initial installation and subsequen o
loading / unloading of fuel?
Resolution:
The licensee will provide the installation procedure for the racks and the method for monitoring ~ gap sires.
4.
How was the conservatism of the single rack model demonstrat d?
appears to limit the amount of sliding and tilting of the rack betw e
The model This would not account for potential pileup of racks against caps.
mall pool wall.
Has this possibility been investigated?
e Resolution:
by the licensee.The NRC will decide whether this concern n ressed 5
within the rack cells been demonstrated?How has the c a on of the fuel The model does not appear to-response of its resonant modes. account for the flexural rigidity Has this been investigated?
e Resolution:
The licensee will provide the Fermi 2 an 6.
the model (note-p 11-25 missing from BNL c e
y r
Resolution:
The licensee provided an adeounte explanation at th
s,
7.
Uplif t of fuel from rack baseplate and resulting impact loads were not considered in the model. How was this effect evaluated?
Resolution:
The licensee provided an adequate explanation at the meeting.
- 8.
How were the rack to rack and rack to wall pap spring constants determined?
How are the impact loads evaluated?
Pesolution:
The licensee will send stiffness calculations, impact load calculations and girdle bar stresses.
9.
Have any studies been performed to test the impact load sensitivity to I
cap size?
Resolution: The licensee will provide explanation of the computer inout-output parameters.
10.
How are pool wall motions included in the analysis? Do the walls exper.ience the same motion as the floor?
Resolution:
The licensee provided an adequate explanation at the meeting.
11.
Is the buoyant force of water considered in the calculation of frictional resistance of the rack feet?
Resolution: The licensee provided an adequate explanation at the meeting.
12.
How were the floor time histories used as input generated? ' Provide the corresponding response spectra.
Pesolution:
The licensee will provide the comparison of ground motion spectra arrd the synthetic time histories that were used in the rack analysis.
13.
Clarify equation for P in Table 6.1 nn Page 6.27 of the licensing report.
g Resolution:
The licensee provided an adequate explanation at the meeting.
14 Explain the criteria used for selection of fuel to rack eccentricity values (XB1,Y)consideredinthemodels.
P Resolution: The licensee provided an adequate explanation at the meeting.
15.
Define the parameters given in the DYNAHIS computer run input and output.
Resolution:
The licensee will provide the user's manual for DYNAHIS.
16.
Rack to rack and rack to wall gaps in the computer runs do not appear consistent with gaps shown on fuel pool layout drawing.
Explain.
Resolution:
The licensee provided an adeouate explanation at the meeting.
I I
.1
AFFIDAVIT OF R. CLYDE HERRICK REGARDING THE INTERVENOR'S APPLICATION FOR STAY I, R. Clyde Herrick, being duly sworn, state as follows:
1.
I am employed by the Franklin Research Center (FRC), Division of Arvin/Calspan, Philadelphia, Pennsylvania as a Principal Engineer in the Engineering Department.
I serve as a consultant from FRC to the U.S. Nuclear Regulatory Commission (Engineering Branch, Division of PWR-A Licensing, Office of NuclearP)5eaderRegulation). A copy of my resume is attached.
2.
I have participated in the technical evaluation of the application, by the Pacific Gas and Electric Company, for high density spent fuel storage in the spent fuel pocis of Diablo Canyon Units 1 and 2.
My participation was with respect to the structural design aspects of the spent fuel racks and the spent fuel pools.
I am the principal author of Techrical Evaluation Report TER-C5506-625, dated April 30, 1986, which is included as Attachment A in the NRC Safety Evaluation Peport dated May 30, 1986, supporting the issuance of License Amendments No. 8 and No. 6 for Diablo Canyon Units I and 2, respectively, regarding the expansion of the spent fuel pools.
3.
I have reviewed the Application for Stay by San Luis Obispo Mothers for Peace and the Sierra Club, Santa Lucia Chapter, dated June 16, 1986 and the attached affidavit by Dr. Richard B. Ferguson, dated June 16, 1986.
I have addressed seven issues identified in the Ferguson Affidavit. These seven issues, discussions of which follow, form the basis of Dr. Ferguson's assertion that "the proposed reracking would significantly reduce the margin of safety for the spent fuel storage system and pose a risk to the public health and safety and protection of the environment". These issues are:
A.
Rack to Wall Interactions B.
Reck to Rack Interactions C.
Multi-Rack Interactions D.
Cushioning Effect of Water E.
Design Differences of Rack "H" F.
Fuel Pool Seismic Displacements G.
Free Standing versus Anchored Racks 4.
Rack to Wall Interactions:
The Issue:
"The Commission has not evaluated the potential for collisions of the racks with the walls of the pool."
References:
Paragraphs 12, 14, 15, 16, 20, and Appendix A
Response
In support of the licensing amendment, the analysis model used for dynamic response analysis of the rack modules was valid for the consideration of j
adjacent rack modules and adjacent pool walls. The development of the j
analysis model is discussed in Section 3.1.4 of Appendix A to the NRC Safety Evaluation Report, beginning on page 16, and is supplemented by the description provided in the response to Issue B in Paragraph 5 of this Affidavit.
The only item not specifically expressed in the analysis as applied to a pool wall was the local acceleration of the wall during contact with the rack module. Wall motions of concern are those associated with the high acceleration I
components of the acceleration time-histories. These are higher frequency components in the acceleration time-history that are associated with wall displacements acting in a time period shorter than the response time of the rack. These wall displacements could, therefore, effect the rack as local displacements in addition to the calculated rack displacement at impact.
However, my consideration of the wall displacements associated with these high acceleration components indicate that their magnitude is a fraction of the maximum computed rack displacements.
Combining these wall displacements with l'
the computed maximum rack impact displacements did not produce forces on the l
rack modules in excess of the allowable values.
Dr. Ferguson, in Paragraphs 14, 15, and 20 of his Affidavit indicates that, based on his analysis, the rack to rack impact forces and rack to wall impact I
I forces are many times the allowable loads on the rack modules.
l l
Reference to Dr. Ferguson's analysis method, as shown in Appendix A of his I
Affidavit, indicates that Dr. Ferguson has derived the well known phenomenon that a constant force (or constant acceleration) applied suddenly, and then sustained, on a simple mass-elastic system will result in twice the force (or acceleration) being realized at the mass of the simple system during the course of its dynamic response.
However ttue this might be Dr. Ferguson's fortnula l
can not describe the acceleration and dynamic response of a rack module bearing I
egainst a pool wall that is accelerating according to the acceleration time-histories developed for the analysis.
His mathematical model is inadequate.
While simple analysis can often serve well to check more complex phenomena, the major inadequacy of Dr. Ferguson's analysis was that he incorrectly assumed the wall to be accelerating at a constant rate, and did not consider the distributed mass and flexibility of the rack modules.
In using a constant wall acceleration in his analysis, he failed to consider the short time l
duration of the high wall accelerations in relation to the generally slower response times of the rack modules and fuel. Thus, Dr. Ferguson predicted excessive forces on the racks using analysis that is badly out of context for this application, i
conversely, the analysis reviewed and evaluated in Appendix A of the NRC Safety l
Evaluation Report correctly addresses the technical requirements, their proper i
application and context.
It is an adequate analysis.
1 1
0
-2 l
d 5.
Rack to Rack Interactions:
The Issue:
"The Commission has also failed to assess properly the potential for collisions of one rack with another."
- 1 Affidavit
References:
Paragraphs 12, 14, 17, 18 and 20
Response
Dr. Ferguson indicated, in Paragraph 17 of his Affidavit, that the analyses were l
based upon fallacious assumptions, but he offered no technical support of his l
statements other than an erroneous analysis of rack-wall impact (see response to paragraph 4).
On the contrary, the Comission approved analyses that carefully simulated the -
nonlinear dynamic behavior of the spent fuel rack modules as reviewed and evaluated in pages 16 to 40 of Appendix A to the NRC Safety Evaluation Report..
The analyses were based upon well established engineering principles and included the following modeling considerations:
o elastic flexibility and material strength of the rack module
.o impacts of the spent fuel assemblies oscillating in clearance -
sp6ce within the storage cells of the rack o off-center partial fuel loadings as well as full fuel load l
c a documented range of friction coefficients between the mounting peds and the pool liner o the hydrodynamic effects of water between the racks and the pool wall or an adjacent rack o the effects of impacts with an adjacent rack or the pool wall o the simultaneous consideration of three orthogonal seismic acceleration time-histories (3-dimensional analysis)
The maximum dynamic response was analyzed conservatively by choosing for
. 4 analysis the rack with the geometrical characteristics that produce high' dynamic response. The maximum forces of collisions with an adjacent rack were antlyzed by assuming the adjacent rack to be'a mirror image of the high-reponse rack unoer dynamic motion simulation ' analysis, so that the response of the adjacent rack was equally high and opposite in direction. This. assumption l
conservatively maximized the computed impact forces between two adjacent rack 1
modules, as opposed to the more realistic case where adjacent dissimilar racks would have lower dynamic response amplitudes and a different response frequency spectrum.
.i Contrary to Dr. Ferguson's statements, the analysis provided in support of the l
licensing amendment included a realistic consideration of each physical phenomenon influencing the motion of a fuel rack, and rack-to-rack impact forces were based on the high-response racks. Multi-rack interactions are addressed in Paragraph 6 below.
I 6.
Multi-Rack Interactions:
The Issue:
" Third, the Commission has ignored the potential for multi-rack collisions --."
References:
Paragraphs 12, 15, 18, and 20.
Response
Multi-rack collisions were considered and are assumed to occur. The analysis' approved by the Comission recognized that the dynamic response motion of the racks is more complex than lateral sliding of each rack unit, and that the maximum inter-rack impact occurs at the top of the rack where the sliding motion of the rack is augmented by angular motion of the rack associated with the mounting feet bouncing off the floor, i
Whereas the racks will impact randomly with adjacent racks in a complex pattern of inter-rack impact, the maximum forces on each impacting pair of adjacent racks was analyzed conservatively by considering the maximum forces resulting from the assumption of two adjacent identical, high-response, rack modules undergoing equal and opposite motions as described in the response of paragraph 5.
1 With the dissimilarity of racks in the pool-and the complex response of each rack, the possibility that two or more racks could be positioned tightly together and move in unison to damage another is extremely remote. The bouncing nature of each rack under the higher earthquake accelerations produces lateral movements at the. top of racks that will prevent tightly packed group behaviour of the racks under the acceleration levels of concern.
In Paragraph 15 of his Affidavit, Dr. Ferguson states that, " Basic physical principles predict that the forces generated in a collision with two racks sliding would be twice as large as for single racks, three times for three racks, etc..."
Again, Dr. Ferguson seems to be mislead by an overly simplistic approach as discussed in Paragraph 4 above, and appears to be treating the rack modules as rigid bodies under constant, or long-term, acceleration. Actually, the racks are flexible, mass-elastic bodies subjected to random excitation transmitted mostly by rapidly-changing friction forces between the rack mounting feet and the pool floor. As such, it would be extremely improbable that multiple racks could be bound tightly together to impact with another such group or wall.
Even if this could happen, the forces would not be as predicted by Dr. Ferguson (see paragraph 4 and 5).
~
In sumary, the analyses reviewed, evaluated and approved in Appendix A of the -
Safety Evaluation Report adequately address the considerations of multi-rack impacts.
1
- j
7.
Cushioning Effect of Water:
The Issue:
" Fourth, the Commission has over-emphasized the cushioning effect of the Water in collisions involving fuel racks."
References:
Paragraphs 12 and 19.
Response
Dr. Ferguson stated in Paragraphs 12 and 19 of his Affidavit that the cushioning effect'of the water was overemphasized and unreasonably large.
~
However, he offered no technical basis for'his conclusions, and did not provide any reasons why he believed this te be true.
It is important to consider the effects of water surrounding and between the rack modules. The virtual mass of the water adds to the mass of the rack module in determining the dynamic response of the rack module to earthquake excitation.
In addition, cushioning is derived from accelerating the mass of water from l
between adjacent racks that are moving toward each other, or from between a rack and a wall.
It should be noted that these effects constitute hydrodynamic coupling as discussed on page 23 of Appendix A to the NRC Safety Evaluation Report. These effects involve the mass of a frictionless (inviscid) fluid.
This is not fluid damping.
Fluid damping was not included in the analysis.
The hydrodynamic coupling employed in the analysis constitutes state-of-the-art analysis based upon established fluid dynamics principles. The model was reviewed specifically to determine that the coupling was realistic and conservative.
B.
Design Differences of Rack "H":
The Issue:
"Fifth, one rack in particular (fuel rack "H") has a i
different configuration from the othat racks."
References:
Paragraphs 12 and 21.
Response
In Paragraphs 12 and 21 of his Affidavit, Dr. Ferguson expressed concern about the mounting supports of rack "H" being shorter than those for the adjacent rack modules, thus providing the possibility that the protruding baseplates of adjacent racks could damage rack "H" above the baseplate.
PG&E records show that in recognition of this possibility, girdle bars were added to each "H" rack at the level of the baseplates of adjacent racks prior to issuance of the NRC Safety Evaluation Report. Therefore, with the added pirdle bars in place, any impacting of rack "H" by the baseplates of adjacent fuel racks will occur on the girdle bars provided for that purpose and not on the storage cell walls.
. I
\\
j
l
\\
9.
Fuel Pool Displacements:
The Issue:
"During the PHE, the spent fuel pools (and indeed the entire power plant) are expected to undergo' displacements of up to three feet in the north-south direction and eight feet in the east-west direction."
Reference:
Paragraph 13.
Response
In Paragraph 13 of his Affidavit Dr. Ferguson incorrectly assumes.that the spent fuel pools, are expected to undergo displacements of up to three feet in the north-south direction and eight feet in the east-west direction". He appears to have based his assumption on plots of displacement that he prepared from double integration of the time-history data to yield displacement without realizing the possible errors that may result.
Two sources of error are explained in the following paragraphs.
The first and major source of error is expected to result from a slight offset, or bias, acceleration value in the acceleration time-history data.
The acceleration tfme history was generated from the specified earthquake spectrum by a modified version of the computer program SIMQKE, a well accepted and widely used meth S Procedures in the use of the program require that a baseline correction be made to adjust the mean square of velocity, and to scale -
the peak acceleration predicted.
In the course of performing these procedures, it is possible that a very small offset acceleration was introduced to the time-history data which, upon double integration to yield displacement, led to the very large earthquake displacement cited by Dr. Ferguson.
A second, and possibly smaller, source of error may have resulted from Dr.
Ferguson's integration methods. Because the acceleration time-histories were defined as sets of acceleration values at incremental points in time.
Dr. Ferguson would probably have used a double sumation. procedure instead of a true mathematical integration of the acceleration data set. The amount of error that would accumulate from a summation procedure depends upon the smoothing, or interpolation procedures that are used. The greatest error would be realized from direct double sumation of'the time-history data set without the use of a smoothing, or interpolation procedure.
However, these errors leading to the large displacement plotted by Dr. Ferguson did not adversely affect the licensing analysis. The purpose of the acceleration time-history was to input the earthquake excitation to the dynamic analysis, which used excitation in the form of acceleration to produce forces on the rack module.
The analysis was not significantly affected by slight differences in offset acceleration.
In confirmation of the above, an inspection of the displacement curves produced by Dr. Ferguson indicated that an extremely small constant offset acceleration would account for the large displacement and the basic trend of the displacement curve.
p o
10.
Free-Standing versus Anchored Racks:
The Issues:
"The new fuel racks, unlike the origintl racks, are free-standing and unfastened to the floor]md walls of the pools".
Reference:
Paragraphs 10 and 11.
Response
Dr. Ferguson is concerned by the change to free-standing racks and states that, "The posibility of damaging collisions results directly from the replacement of racks which are anchored with new racks which are not anchored."
Free-standing spent fuel rack modules are not nev to the nuclear power industry and continue to replace anchored fuel racks on an increased frequency concurrent with the industry's need to provide additional storage capacity for spent fuel at the plants.
The advantages of free-standing spent fuel rack modules over anchored racks include the following:
o improved integrity of the spent fuel liner, in that a multitude of rack anchor members penetrating the liner to be secured in the concrete pool structure are no longer necessary, o the liner can be made smooth to better facilitate pool cleaning.
o replacement of a rack module, should it te necessary following a fuel handling accident, for example, is simp 1ified to the point that, after removal of the spent fuel assemblies, the'rodule can easily be lifted out of the spent fuel pool and a replacement module installed with less hazard and exposure to personnel.
Plants other than Diablo Canyon, have been converting to free-standing spent fuel racks for many years.
The following plants, for which I bzve participated in the review and evaluation of fuel racks for high density spent fuel storage, have received NRC approval to install free-standing fuel rack modules:
Dyster Creek Summer McGuire 1 and 2 St. Lucie 2 Turkey Point 3 and 4 Ginna Grand Gulf Millstone 2 Peach Bottom 2 and 3 4
l l
I q
11.
Conclusion:
Following careful consideration of these seven issues associated with the structural analysis of the spent fuel racks and spent fuel pool that were raised by Dr._ Ferguson, it is my professional opinion that the structural analysis of tne racks and pool provided in support of the installation of high density fnel rack modules in the pools of Diablo Canyon Units 1 and 2 are adequate, and that the racks and the pool meet the structural criteria provided by the NRC. Therefore, it is my opinion that the structural adequacy of the free-standing spent fuel racks being introduced to Diablo Canyon will not teduce the margin of safety for the spent fuel storage system and will not increase the risk to the public health and safety or to the environment.
j I
I hereby certify that the aboUt Statements are true and correct to the best of i
my knowledge.
b dw& I
/.c R. Clyde Herrick Subscribed and sworn to before me this." # day of June,.1M}6.
g.
s
~.
- j. '
.N
< <, &. ".. N
,,(
<n tvotary Public
".)
My Commission Expires: fg.
'/ff/
/
/
q l
e 8-
_ _ _ _ _