ML20198Q188
| ML20198Q188 | |
| Person / Time | |
|---|---|
| Site: | Monticello, Cooper  |
| Issue date: | 01/13/1998 |
| From: | Dangelo T NRC (Affiliation Not Assigned) |
| To: | Berlinger C NRC (Affiliation Not Assigned) |
| References | |
| NUDOCS 9801220363 | |
| Download: ML20198Q188 (14) | |
Text
_ _ _ _
. January 13, 1998 MEMORANDUM 10: Carl H. Berlinger. Chief r Containment Systems and Severe Accident Branch. OSSA FROM: Tony D'Angelo /s/
Containment Systems and Severe Accident Branch. DSSA
SUBJECT:
ISSUANCE OF THE MONTICELLO AND COOPER TRIP REPORT WITH PROPRIETARY INFORMATION DELETED This memorandum transmits the attached non proprietary version of the Monticello and Cooper trip report to the public document room. All proprietary information has been deleted but all relevant technical information has been retained. This information should be useful to the licensees involved and the public in that it provides in 'vritten form, insight to the staff's thinking, at the time of the trip, on how hydrodynamic loads should be addressed with respect to stacked disk strainers.
Docket Nos. 50-298 and 50-263
Attachment:
As stated DISTRIBUTION:
Docket Files PDR SCSB r/f (2)
DLynch TKim RHall l RElliott DOCUMENT NAME:PDRMEMO Te receive a copu of this document, indicate in the bour *C' = Copy without attachmentlencloeure *E' = Copy with attachmentrenclosure *N* = No copy 0FFICE SCSB:DSSA:NRR E! SCSB:DSSA[\\ 6 ~
NAME TD'Angelo:bw [7/fQKudrick 6 @
DATE I/#3/98 \$'98 Bl) / /98 / /98 / /98 I ~ OFFICIAL RECORD COPY NRC FiE CBiffHi gepy pa'*nni! WJ$2 appmm
. 1
- January 13. 1998 MEMORAND'JM TO: Carl H. Berlinger, Chief Containment Systems and Severe Accident Branch, DSSA l
FROM: .,oy D'Angelo /s/ j Containment Systems ano Severe Accident Branch. DSSA l l
SUBJECT:
ISSUANCE OF THE MONTICELLO NiD COOPER TRIP REPORT WITH !
PROPRIETARY INFORMATION DELETED l This memorandum transmits the attached non proprietary version of the Menticello and Cooper trip report to the public document room. All proprietary information has been deleted but all relevant technical information has been retained. This information should be useful to the licensees involved and the public in that it provides in written form, insight to the staff's thinking, at the time of the trip. on how hydrodynamic lo6ds should be addressed with respect to stacked disk strainers.
Docket Nos. 50-298 and 50-263
Attachment:
As stated DISTRIBUTION:
Docket files PDR SCSB r/f (2)
DLynch TKim RHall RElliot' DOCUMENI NAME:PDRMEMO t... .. . c. ,,oei m .- u, - i. w e. . w is. 6..: c . conv .,,%ui .eiusm.ne.e.u.. r - con, .,in .iiusm.oo...u . v . % con, Of f ICE SCSB:DSSA NRR lE SCSB.DSSAf\\ 6 l NAME TD'Angelo:bw [fM 9Kudrick O @
DATE //i3/98 \i\T98 5 ~
/ /98 / /98 / /98 I OFF 1CI AL RECOR[f COPY
. i t!
l E
MEMORANDUM T0: Carl H. Berlinger, Chief Containment Systems and Severe Accident Branch, DSSA ,
.FROM: John A. Kudrick Containment Systems and Severe Accident Rcanch, DSSA -
Anthony J. D'Angelo '
Containment Systems and Severe Accident Branch, DSSA
SUBJECT:
TRIP Rf. PORT FOR VISIT TO MONTICELLO AND COOPER ALL PR0iRIETARY INFORMATION HAS BEEN DELETED The resolution of the debris / sump strainer issue has resulted in the installation of large strainers within the suppression pool. With respect to the pressure drop across the strainer, performance data show that even when the strainer is fully loaded with debris, the pressure drop is very low.
However, introducing such large hardware into the suppression pool. involves ensuring that the design can accommodate (1) all associated loads that could be imposed onto the structure and (2) the.1 cads imparted on the torus itself. ,
Of particular concern is the method of calculation used to conipute the submerged structure loads.
At the conclusion of the staff's work on Generic Issue A 7. " Hark I Pool Dynamic Program " an acceptable methodology was documented in our Safety 4
. Evaluation Report-(NUREG 0661) for computing submerged structure loads along with all other loads within the containment of a BWR design.
In the 17 years since the methodology was published, only limited deviations have been requested. Even though the requests have been minimal, each request was reviewed by the staff before the licensee has implemented the change.
However, with the advent of the debris screen program, licensees have begun to
-install very large pump suction screens within the suppression pool under the :
umbrella of the 50.59 program. l Since the 50.59 program has no documentation requirements, until possibly after implementation, the staff is unaware of exactly what the licensees are doing relative to the supporting analyses and planned hardware. However, the staff believed until very recently that each licensee was using the methodologies and coefficients as described within NUREG 0661 to compute the j
- l
- l 2
various loads. Recently. licensees have indicated that although the l methodologies were being used, such specific input values as acceleration drag r values are being developed uniquely for each plant. The resulting loads are ,
very sensitive to the value selected. On the basis of the staff's limited understanding, we believe that the resulting forces are being reduced by values ranging from 20 to 80 percent from the values contained in NUREG 0661. :
These recent findings have led the staff to conclude that we should visit a few selected sites to obtain plant specific information. The staff would use this information to determine whether or not the unique changes to the methodologies and the input values such as the acceleration drag coefficients are appropriately under .the umbrella of 50.59 or whether an unresolved safety question is involved in the modifications. ,
Monticello and Cooper were selected for site visits. Monticello was selected because it was shut down at the time and offered an excellent opportunity to cee the strainers being installed in the drained suppression pool and to obtain on understanding of the supporting analyses. Cooper was selected because it appears that the licensee is taking the largest load reduction of any plant that the staff is aware of. ,
The intent of each site visit was to review not only results but the !
justification / bases supporting the analytical methods used to calculate the ,
loads. !
DISCUSSION The staff's first objective was to gain an understanding of how the licensee concluded that the revised methodology / input assumptions could be done under 50.59 and by what logic the licensee determined that the new strainers could be installed under that same regulation. The staff needed to understand why the very specific nature of the accepted methodology would not prohibit any deviations, in view of the fact that any plant unique deviations were -
originally. required to be addressed on a plant specific basis per the guidance of the A 7 program.
The second objective was to understand the revised drag force methodology used i at the plant. The intent was to determine whether or not portions of the ,
submerged structures continue to use the staff accepted methodology-as well as '
what structures are being evaluated using the revised methods.
[
r
3 Monticello On July 7 and 8, 1997, T. J. Kim. A. D'Angelo, and J. Kudrick went to Monticello to gather certain technical information pertaining to the project to replace the pump suction strainer. The main focus was to review and understand the engineering methodologies used in determining the drag forces associated with tne new strainer design.
To achieve these objectives, we first toured the drained suppression pool to view the various stages of installation of the new strainers. The strainer consists of a series of stacked perforated plate disks assembled onto a XX XX for structural support. Eact, strainer is about XX feet long and weighs about XX pounds. Each is constructed of stainless steel and has a flange welded to one ed. The flange is bolted to one end of a aamshead. An identical i strainer is bolted to the other end of the ramshead. The entire assembly is structurally supoorted between two ring girders acting as a simply supported beam. Extra stiffeners nave been welded to the ring girders to accept the additional drag forces associated with the larger strainers. This support l
design has a slip joint at the torus penetration to eliminate all internal l loads on the penetration. The previous design with the much smaller strainers was solely supported by the penetration: as a result, all strainer loads were transmitted to the penetration. The net result of the design change is to accually reduce the penetration loads.
The new installation appeared to be quite robust due to the degree of welding and added stiffeners to the ring girders. However, the installation approach had a major disadvantage: it required the complete draining of the suppression pool a major task.
After completing our tour of the torus, we met with licensee staff at Duke Engineering offices. Duke is the company that designed and analyzed the replacement installation. The strainer was designed and fabricated by Performance Contracting, Inc. (PCI).
The licensee explained its approach for computing submerged structure drag loads and the supporting test data. The opproach was based on the original work by Dr. Morison and has been subsequently supported by other noted authors in the field. It was previously presented to the staff in GE topical report NED0 21888 and has been used at all Mark I plants. Dr. Morison considered the hydrodynamic for:e to consist of a drag component and an inertial component, with the inertial component being the more significant by far of the two. In
t t
b 4 i o
the original methodology, coefficients were provided for various shapes that could be found in the suppression pool. - Although it was envisioned that the provided values would be used in the plant specific analyses, plant unique !
drag coefficients were not prohibited. l Monticello chose to develop plant unique drag coefficients for the new strainer configuration. Because of the. complex-geometry and the use of- ;
perforated plates on the strainer surface, it is difficult (if not impossible) :
j to establish the drag and inertial coefficients without the supporting test ,
4 data, i
Duke undertook tests in October 1996 to investigate the response of typical stacked disk strainers that were subjected to accelerated and consti.at flow :
fields. The tested strainer was similar to the one planned for Monticello, -
but it varied somewhat. For example, the test specimen used XX inch thick disks with a XX inch spacing between disks. The Monticello strainers consist ;
of XX inch thick disk with a XX inch spacing between disks. Although no ,
correlations exist to specifically calculate the differences between the t hydrodynamic masses, the staff believed that the error associated with these >
differences could be accommodated by :.onservatively selecting the values used l in the design calculations. The tests were successful in establishing the ;
constant flow or standard drag coefficients, However, these tests were ;
unsuccessful in measuring acceleration or inertial drag coefficients. The '
tests were conducted in a large tank with the strainer being dragged through >
the water. This type of testing is commonly referred to as " tow" tests. -
Because of the large strainer mass, instabilities became apparent when the test specimen was accelerated in the water. The testing was terminated because of the fear of damaging the facility hardware.
With respect to the standard drag tests, the test data were obtained to show a measured value of 1.1: the-NUREG specified a value of 1.2. During limited discussions about the actual test data, we noted that the data seem to meet
- NRC. standards.. The discussions ended when we found out that the licensee intended to continue to use the higher value (1.2).
- A totally different approach was selected to; establish the acceleration drag '
coefficients. The new tests -known as " pluck" tests, are based on measuring the resonance frequencies of the strainer and relating that measured frequency :
to a hydrodynamic mass of the strainer body. This approach is based on work- -
done by Drs. Keulegan and Carpenter of NBS (now NIST). They have shown
. -through. theoretical:and experimental: work that the hydrodynamic mass of the
=. , . - - . - ..
i 1
. 1 5
i subject body directly relates to the acceler ation drag forces that the body would be subjected to if it was placed in an oscillating fluid. The hydrodynamic mass of the strainer can. in turn, be used to calculate an ecceleration drag coefficient.
The staff accepted this approach as part of the Mark I Containment Program and documented it in NUREG 0661. In 49EG 0661, the staff indicated that acceptability of the above calculational procedures and methodology is heavily based on experimental confirmations as documented 1.1 topical reports.
On this basis, the stuff agreed with the licensee that the revised approach using " pluck" tests and the methode, logy described above to determine the acceleration drag coefficients is acceptable. The plant licensing bases as documented in the Monticello PUAR used a hydrodynamic mass coefficient of 2.0 for all submerged piping and strainers. With the current design of stacked disk strainer, the licensee, using this accepted methodology and pluck tests, has computed an average hydrodynamic less coefficient of XX. The data did vary significantly above the average value. In fact, the highest tested value was XX. In light of these large variations, the licensee. for design load calculations, elected to use a hydrodynarric mass cmfficient of XX for the new stacked disk strainer, a value that bounds all of .ne test data.
We also discussed the method of suppo' ting of the strainer. The original Monticello design strainers were 6ttac. led directly to the ECCS suction pipe entering the suppression pool sith all resultant loads transmitted to the containment penetration. The new design of the stacked disk strainers, as described above, is now completely supported from the ring girders and no torus internal piping load is transmitted to the penetration.
Strainers are bolted to each end of a ramshead piping tee end form a suction strair.er asserrbly for each ECCS. The sicciner assembly then supported at the end of each strainar b/ a trunnion, which is then welded to the ring girder flange. Hydrodynamic forces normal to the strainer centerline produce a strainer load of approximately XX kips at each trunnion and, subsequently, are transmitted to the ring girder. The licensee referred to the trunnion as the strainer connection support. Because of the magnitude of the loads, the ring girder flange, web, and web to torus shell required additional stiffening to stay within their t',ress criteria. The additional stiffening consisted of a plate welded to the girder surfaces to increase the cross section of material.
Instead of performing a complete stress analysis for the torus shell and ring
-girder, the licensee relied on the PUAR analysis for the base stress level for
_ _ - _ - ~ _- _ ._ _.___ _._ ____ _ _ _ _ _ . . _ . . _ - - .
l I
6 the ring girder during postulated events and a simplified calculation that computed the new stress within the ring girder resulting from the additional :
load of the new strainer assembly. This calculation indicated the need for j additional stiffening of the ring girder to maintain the_ girder within ASME's
, allowable limits. Results of that calculation show that the ring girder web, i flange, web to flange weld, strainer connection support, and torus shell in .
the area of the ring gi, der stress levels are at the ASME Code allowable -
4 limits. ,
I Summary of Staff Findings on Monticello
]
- The licensee chose to develop plant unique drag coefficients for the new strainer configuration. The selected approach was to perform tests on a striiner that was similar to the one that would be installed in Monticello.
The supporting analysis that determined the method of testing, as well as the ;
methods used to calculate the critical parameters, were all well recognized and had previously been found acceptable by the staff. Therefore, the staff had no difficultly in accepting the overall approach.
However, the staff had.some difficultly in readily accepting the arguments presented to show applicability of the tested strainer to the installed !
strainer. As a result, the licensee agreed to consider testing the exact strainer to resolve this issue. Since the site visit, the licensee has sent a ;
letter to the staff on July 29, 1997, committing to perform these additional tests. A schedule for this testing will be furthcoming.
With respect to the structural considerations, the licensee performed an oversimplified bounding analysis of the ring girder. The results showed that !
the stress at the ring girder was not exceeding code allowable limits, and that most of the values were at code limits. After long discussions concerning both the results and the limited analysis approach, it was agreed i that our structural experts (ECGB, Goutam Bagchi) should participate in the !
detailed discussions. Although the staff members would express their views on the subject, the official discussions would be held with ECGB.
~
i Shortly after' returning' from the site, _we spoke with the licensee by telephone about various structural matters related to the strainer installation. The staff concluded that the limited analysis of the ring girder showed that the revised ~ structure remained in compliance with the code of record (ASME).
-However, this analysis did not' address local effects of the new strainers on the_ ring girder flange. - As a result of these discussions, it was concluded that to sompletely answer the staff's concerns. it~ would be most efficient to completely redo the entire PUAR structural analysis of the ring girder only.
e
.v. 4- - , . 4-,,.s .. .-r-w,y... . ,.,. .g,, -n.. , ,c-..- .- ,,,,n-,L,m... -
n-w , -- ,. - , - - - y.
4 d
/ ;
j This analysis would use conservatively established drag coefficients _obtained j from the prototypical Monticello strainer tests. The licensee committed to complete this analysis preceding the startup from the next fuel cycle.
The staff concluded that the licensee's modification to install the stacked !
disk strainers did not involve an unreviewed safety question. This conclusion ;
was based on the fact that the licensee had devolved generic drag coefficients )
that were based on recognized analytical method that had been previously The lice.1see will ;
approved by the staff prior to making the systems operable. l confirm the plant. unique drag coefficients by test for the same size, In addition to the above, a !
configuration, and weight of strainers installed.
l complete structural analysis of the torus ring girder using the plant unique drag coefficients will be completed.
Cooper .
On July 9,1997, the Cooper staff mt with R. Hall T. D'Angelo, and j
J. Kudrick, The three NRC staff wanted to gather information concerning the analysis and installation of new stacked disk strainers at the Cooper power plant. The strainers were installed during the May 1997 refueling outage under the umbrella of 50.59. As a result, the staff had not been involved in i l
the process, nor had there been any significant documentation of the process.
Because Cooper was the first plant to install the GE stacked disk strainer, ;
the staff was very interested in the entire process. particularly in the '
methods that were developed in the calculation of the submerged structure drag -
. loads, as well as the justification / verification of these methods.
Jim Drasler of Cooper (NPPD) began the discussion with an overview of the project, as follows. The replacement project consiste:' of four RHR strainers and;two CS strainers. The RHR strainer is about XX ii.cnes in diameter, about XX inches long, and weighs about XX pounds. Its XX disks are stacked together forming the strainer assembly. At one end of the strainer is a flange with the mating flange located on the pipe from the torus penetration. When completely assembled, the strainer becomes cantilevered from the penetration.
The origina_1 RHR suction strainer consisted of two strainers per penetration.
Aramsheaddevicewas.weldedtotheendoftheprojectingpipeor-PUP.A
-strainer was attached to each end of the ramshead. The original strainers were much smaller than the newly installed stacked disk strainer. The replacement process began by cutting off the ramshead device at the weld to
- the PUP. A flange was then welded to the 20 inch diameter PUP. which is the mating flange to the flange attached.to the end of the strainer. GE stated ,
that the benefit of the GE strainer design is that the drag coefficients are l
i
- - - - , . + . - ~ + - . , ~ - - -n. -- -.--.-,- -- -
s 8
low and the resulting loads transmitted to the penetration are sufficiently low so that no additional structural stiffening of the penetration is needed.
In addition, installation of the new strainers did not require the emptying of the suppression pool and all installations that were performed under the suppression pool were possible since no underwater welding was necessary.
Final assembly simply required bolting down the two flanges. The strainer then protrudes into the suppression pool at right angles to the torus wall, but it a slightly downward angle.
A similar construction is associated with the slightly smaller CS suction strainer. It is about XX inches in diameter and XX inches long and it weighs about XX pounds. Its XX disks are stacked together forming the strainer assembly. At one end of the strainer is a flange with the mating flange locateo on the 16 inch diameter pipe from the torus penetration. When completely assembled, the strainer becomes cantilevered from the penetration.
The original CS suction strainer is very similar to the old RHR strainer. The projecting PUP also had a ramshead device welded to it and a strainer attached to each end of the ramshead. The original strainer was about 13 inches in diameter and only a little more than 14 inches long. The original RHR strainer was only slightly larger than the CS strainer; it was 17 inches in diameter and 18 inches long.
Comparing the original strainer to the new strainer is very enlightening. The new strainer is much larger and much heavier than the strainers it replaces.
As a result, one would suspect that the loads on the penetrations would also be significantly higher. However, a counteraffect that needed to be investigated was the significant open area of the strainer. To address this potential of increasing penetration loads NPPD established an important design criterion. This criterion (or goal) was to try to keep the final loads on the penetration to about the same level so that the stresses at the penetration remained about the same as in the original strainers. This important guideline meant that the penetrations would only be modified or stiffened if all other approaches became unacceptable because penetration stresses exceeded code allowable limits.
It is important to understand this program direction to gain a better under-standing of the analytical effort. To explain the program approach Nick Celia (consultant to NPPD) made the following presentation to the staff.
- The approach was to compute the strainer loads using the methodology previously approved during the Mark I Containment Program and to adjust these to account for changes in the size and position of the new strainers. The adjustments were referred to as " scale factors." Three
M
-9 ,
factors were identified: one relates to the relative position of the .
strainer with. respect to the energy source (i.e., the downcomer exit): the l second concerns the proximity to the torus shell; the third factor is !
accelerMion drag volume caused by the large hole area of the new strainar The e st significant by far was the change in the acceleration :
drag volta andprojectedarea. ,
o Detailed calculations established the acceleration drag volume of the new strainer. These calculations accounted for both the spaces between the ;
disks and the effect of the perforations in the plates covering the' disks. ;
The equation used to calculate the acceleration drag volume is identical
- to the one used in the Mark I approved methodology. However, the value of the drag _ coefficient was uniquely developed. NPPD believed that none of !
the listed values applied to a stacked disk strainer. Therefore, a1 analytical approach was developed specifically for the Cooper geometry, c Proximity factors were calculated using th approved methods; however they were not applied. The basis was that the spaces between the disks and the perforated plates make the strainer much more transparent. Therefore, NPPD concluded that the redirected flow would be so much less that this factor could be neglected. The staff pointed out that no credit was l
allowed in the LDR and proximity factors were expected to be calculated.
- In addition, the strainer is much larger than the pipe that was envisioned when proximity factors were considered. As a result, the staff view's this '
as a deviation from the approved methodology.
The discussion then moved toward the areas in which some additional conservatism was removed. The first item was the neglect of the standaro drag coefficient in the calculation of the. total drag force. The approved methodology clearly indicates that the total force is the cummation of both the standard and acceleration forces. The discussion revealed that since the acceleration force is about 95 percent of the total, the standard component could be ignored. The staff indicated that this was not a conservatism that could or should be neglected. Therefore, we could not find a basis for accepting only the acceleration force.
Another reduction in load was based on the method of summing the stresses at each . mode for_ the' piping analysis. The approved methodology clearly indicates s
that the components of load modes should be summed by the absolute sum method.
This is identified in the'LDR. The analysis for the new strainers uses the
-modified summing technique of SRSS for all modes except for the four peak k.
>- r +g. -
10 frequencies that are summed as an absolute sum. Although the LDR did not describe this method, a latter topical report that proposed this method was subsequently accepted by NRC. Therefore, since this approach was previously found acceptable, we also find it acceptable for the Cooper application. !
The drag forces associated with the PUP or extended piping from the penetration was neglected for all load calculations. Again, there was no .
justification for this simplification other than that the effect would be small. This may be true: however, it puts into question the adequacy of the calculations without an acceptable need to do so. Therefore, the staff also concluded that neglecting this component was a deviation without adequate justification.
The discussion moved into the area of other deviations from the original analyses. NPPD indicated that some relief was also given relative--to the stresses transferred to the penetration from the external piping. This is piping that is outside the torus but that is welded to the torus penetration.
- The original thermal analysis (stress analysis) was based on using the
" design" temperature. For the new strainers, the temperature was changed to ,
the " process" temperature.
The staff acknowledged that the people with the most expertise were not present at the meeting to make the most meaningful' comments on the change.
However, this was clearly a change in the licensing basis of the plant. Also, the significance or amount of stress reduction was unclear. The staff indicated that it would talk to NRR experts back at NRR.
NPPO installed the new strainers into the Cooper plant with the ueviations detailed above and without any supporting test data. The reduced drag forces were established using analytical techniques that were sound but that had not been verified by testing. This is not typical of the general approach used in establishing the approved methodology. _ All analytical approaches were supported by a sound testing base, After Cooper had returned to power, GE began testing various shapes and configurations'of strainers to verify the analytical approach selected to s coq)ute plant unique drag coefficients. Unfortunately, at the time of our meeting, most-of the test results were very preliminary. Therefore, we could only get a " feel" for the data. The results seemed to support the values used in:the analysis. However, cautions were issued'about the data since. in some cases, the information was less than a day old and the data had not been reviewed by anyone other than the originator.
- We indicated the importance of. the. test data. It is our belief that the only .
E
.%w % - -
- i 11 way to completely verify the analytical approach is via test data. This is because no analysis is totally first principled. The meeting ended with a .
summary of the staff's reaction to the material presented.
Sunnary of Staff Findings on Cooper t
The licensee chose to use the GE strainer as well as the GE methodology presented in the GE licensing topical report (LTR) " Application Methodology for GE Star.ked Disk ECCS Suction Strainer", NEDC 32721P, dated March 1997. In addition, the licensee appears to have established a general design criterion ,
to allow for the new strainer installation without any hardware modifications to the torus and to any attached piping.
To meet this general goal, the licensee changed the temperature at which the external piping is analyzed from the design temperature to the process temperature. This is a change in the plant's original licensing basis. This is a reduction in the temperature and, therefore, a stress reduction to the penetration, We said that we would take the information to our experts back at NRR since they had the needed expertise to knowledgeably comment on the issue.
Within the torus, several forces were neglected in the calculation of the overall penetration loads: the standard drag force and the loads on the PUP and attached flange.
As discussed above, deviations to the methodology must be resolved in order for the staff to find the approach acceptable. We noted that for a finding of overall acceptability, we need to rely on supporting test data. The staff would not find any method acceptable without some amount of test data. GE acknowledged the importance of test data and provided preliminary results.
GE also indicated that it would send the test report to the NRC very soon.
How the report would be submitted was unclear; however, the best guess was as an attachment to the LTR. On the basis of the deviations in the licensee's approach, the lack of test verification, and the lack of specific discussion within the 50.59 regulations, the staff believes that the licensee should readdress the 50.59 evaluation. This reevaluation should consider the specific staff comments stated.above.
Docket Nos. 50 263 and 50 298
. . j q
- i i
12 i i
On the basis of the deviations in the licensee's approach. the lack of test j verification, and the lack of specific discussion within the 50.59 ;
. regulations, the staff believes that the licensee should readdress the 50.59 !
evaluatf'1. This reevaluation should consider the specific staff coments l stated above.
Docket Nos. 50 263 !
-and 50 298 j t
ORIGINAL HEH0 DATED SEPTEMBER 9.1997 XX INDICATES PROPRIETARY INFORMATION WAS DELETED .
h T'1 N -t? # 2, -m.e w e p- ec . --wFrz, Awe e ?qr rr R7' ?NF"+ tr ^
'T ffYt"N' *"L