ML20198N025
| ML20198N025 | |
| Person / Time | |
|---|---|
| Site: | Monticello  |
| Issue date: | 01/12/1998 |
| From: | Kudrick J NRC (Affiliation Not Assigned) |
| To: | Berlinger C NRC (Affiliation Not Assigned) |
| References | |
| NUDOCS 9801200223 | |
| Download: ML20198N025 (9) | |
Text
.\\
,pnun,9
,y
- t UNITED STATELS 3
j NUCLEAR REGULATORY COMMISSION l
WA.4HINGTON, D.C_20M.54ut.
- Jawary t2, 1998 i
l MEMORANDUM TO: Carl H. Beninger, Chief Containment Syst M and Severe Accident Branch, DSSA
)
l M
i FROM:
Jack Kudrick
/[
Containment Sy fms and Severe Accident Branch, DSSA Tony D' Angel Containment Sy t and Severe Accident Branch, DSSA
SUBJECT:
NRR AUDIT OF MONTICELLO STRAINER TESTS INTRODUCTION On Monday December 15 and Tuesday December 16,1997, three staff members (Tony D'Ange:o, Rob Elliott, and Jack Kudrick) went to the EPRI NDE test facility located in Charlotte, N. Caroline to witness a series of tests for the Monticello strainers that are currently installed. Representatives of Monticello and Brunswick, along with staff from Duke Engineering were also present during our meetings. An attendees list is attached. The tests were establaned to develop a data base for the stre'ner design installed in the Monticello Nuclear Power Plant.
The purpose of this data base was to provide the necessary test data to allow calculation of the acceleration drag forces on the strainer. These tests are commonly known as " pluck" tests.
DISCUSSION On Monday we witnessed a series of 11 air tests and 'i4 submerged water tests that were conducted on the Monticello strainer. These pluck tests are designed to obtain the natural frequency of the strainer in both air and water. For the air test, the natural frequency of the strainer is established based on the mass of strainer.
For the water strainer test, the mass of water interacting with the strainer in addition to the strainer mass can be calculated. Once the water mass is known, the acceleration drag can then be calculated. The large number of duplicate tests were run to determine the repeatability of the tests as well as to provide some degree of statistical confidence in the test data, g
ff 9901200223 990112 PDR ADOCK 0S000263 P
POR
',a
C 6
method of archival. This last item shculd include such information as who will be the keeper of the source data, in what form the data will be kept (i.e., tape, paper, etc), and for how long will the information be kept.
The report should include how this data is to be used f, asign purposes. For example, a discussion of test repeatability and uncertan it of the data should be-provided. Ur certainties should include instrument error and calculational uncertainties. Using this data base, a discussion of the safety margins and their rationale for these values to go from the data base to design based values should be included in the report.
The final discussion area was centered about how'NSP will use this data. The NSP
-representative indicated that the test report will be on site for NRC review but they
- did not intend to put it on the docket. Nor did it appear that any reanalysis would be performed. Rather, the new data would be used to validate the values used in the existing analysis. The staff indicated that we would look to the commitments identified in their letter to the stuff dated 11/12/97. With this final item completed, the meeting was called to a close.
Docket No. 50-2G3 DISTRIBUTION:
Central Files SCSB r/f (2)
PDR TKim
. CCarpenter GHolahan SNewberry RElliott JStang DLynch
' 0OCUMENT NAME: :EPRI.TST
- v.,
..,.e m so.u
. inac.
w.: c. copv.w e.n.c%m.nv.ncio...
e. copy. m.u. chm.nv.ncio... v. nocorv 0FFICE.
SCSB:DSSA:NRRnIE SCSB: DSS & [6 NAME AD'Angelo:bwAg)t/ JKudrick @
01/ A/98; 01/a/98 7
/ /98
/ /98
-/ /98 DATE
/
0FFICIAL RECORD COPY
?
7 2
The instrumentation consisted of three accelerometers mounted on the strai:..a.
Data was recorded on a FM recorder and plotted on two strip recorders at the rate of 50 data points per second. On one of the recorders, the reco. led amplitude as a function of time was strip charted for each of the three accelerometers. The second and more important recorder provided one composite amplitude trace as a function of time. In effect, it represented the composite trace of the three signals.
In addition, it also provided tha primary frequency of the major modes. For the Monticello strainer, the natural frequency in air was computed to be higher than that of the water tests. These vrilues were taken directly from the strip recorder, it should be noted that the frequency was highly reproducible from run to run.
From what we witnessed, the variation appeared to be less than 0.01 HZ.
In addition, a second peak was noticed on all traces. For air, this peak appeared to beIt 3an 10 percent of the fundamentr.1 frequency. But in water, this second peak was more than 50 percent of the primary mode. While a final conclusion could not be made, it was assumed to represent a slightly rocking motion of the strainer due to the preload being slightly off the center of gravity.
At the end of the water testing, the water was drained from the tank and the strainer was wrapped with a thin sheet of aluminum and allowed to dry overnight.
The next day, contact paper was applied to both strainer ends, thereby completely enclosing the water mass within the strainer. Tests showed that the enclosure significantly increased the water muss and thereby also reduced the natural frequency in water.
After having witnessed 10 water tests with the solid strainer system, we discussed the supporting analytical approach. Using fundamental eqe : ions of motion for an oscillator (Newton's Law of Motion), one can show a w ionship between the natural frequencies of the strainer in both air and water to the hydrodynamic mass coefficient. While final results were unavailable, Monticello staff said during the meeting that the tests appear to be showing a decrease in drag of about a f actor of three from the value used in the design loads calculations.
However, it was unclear as to how Monticello plans to use this new information.
As a minimum, a report will be written ared located on site to support the design analysis. But it was unclear as to whether or not a reanalysis would be performed, i
in support of the site visit, the staff had iorwarded a series of 4 questions concerning the testing facility. These questions are contained below. During the meeting, each question was discussed in depth. A summary of their responses to each of the questions is provided below.
o
l t
3 l
- 1. What is the test objective?
The objective was to generate the drag parameters needed specifically for the Monticello design.
- 2. When will the test report become available?
The report will be provided to NSP in about 30 da es. At the present time, NSP does not plan to put the report on the docket. However, they will be more than willing to share the data with the NRC. How this will be achieved was left for future discussions. The staff noted that there probably will be an ACHS meeting some time in the summer to discuss this issue of loadin0 on the strainer. Considering the importance of this meating, the staf f indicated a need to establish finai design criteria for the hydrodynamic strainer loadings. It is from this view point that the staff l
Indicated a need to have open discussions with all the licensees including NSP concerning the need to put the data and evaluation on the docket. In addition, the staff believes that open discussions will allow for the development of appropriate criteria without resulting in excessive design margins.
- 3. -Provide a description of the test facility This information will be provided in the test report, in addition, GA related information including calibration, recording of data, and the archiving of the data will be in::luded in the report. With respect to the pnriod of archiving of data, NSP responded thet they had not given much thought on this matter, but they did not see this.as an issue. In response to equipment calibration, they indicated that the instruments were calibrated to 4.6 % at
% HZ and 2.5 % at 10 HZ before tne tests. After the tests, a single point test was done with a hand held instrument. Measured frequency response were within the calibration range of the instruments.
4.
Test Plan and Preliminary Results The staff discussed with Duke Engineering the " Pluck Tests" observed over the past two' days and had 'a frank discussion on the tests conducted. The degree of pre load placed upon the strainer was discussed to understand the sensitivity of the " Pluck Test". Duke had performed a calculation to Edetermine the amount of pro load needed for the test such that the desired 1
- amount of displacement would be achieved. The displacement needs to be sufficient so that strainer interaction with the water is sufficiently u
f kk.
m.
. 4 m....,
m
--w.-.,1, hr,
.w y
oc -.,,
,,,--y...
y e
c
\\
4 prototypical to assure an accurately calculated strainer drag. In addition, the staff discussed with Duke the cimilarities of the tested strainer with that of the Monticello installed strainer. The tested straine; differed with the Monticello strainer in that the core tube is fabricated from carbon steel instead of stainless steel. Additionally, the ramshead was simulated by a blind flange.
ANALYi.JAL BASIS The method chosen by DE&S to determine the hydrodynamic mass of the Monticello specific strainer was based on the principle that added mass to a known oscillator willlower the natural frequency of that oscillator. If the ECCS suction strainer is assumed to be the oscillator, then the water mass which interacts with
. the strainer can be determined by measuring the change in natural frequency of the strainer. That water mass which interacts with the strainer as described above is that water in the suppression pool which is accelerated by a LOCA or SRV discharge and induces a drag force upon the strainer. The notion of a combination water drag forces acting on submerged bodies in a pool of water was first discussed by Dr. J. R. Morison and others in a 1950 paper, "The Forces Exerted by Surf ace Waves on Piles," Petroleum, AIME, Vol. 189, 1950. A further discussion specific to drag forces acting on bodies submerged in a typical suppression cool was presented by Dr. F. J. Moody, " Forces on Submerged Structures in Unsteady Flow Fields," Thermal Reactor Safety Meeting,1977. Both papers form the basis used today for estimating the drag forces acting upon the strainer, however in the past, strainers were treated as solid bodies.
The task at har d presently is too estimate the drag force actireg on the strainer knowing that the strainer consists of a minimum of 30% open area. A sound defendable method was needed for determining the total drag forces a-ting on a body constructed of perforated plate with a minimum of 30% open area. The method chosen by DE&S was to calculate the hydrodynamic water mass of the strainer, then using the work by Drs. Morison, Moody and others, relate that water mass to drag forces acting upon the strainer.
DE&S referenced works by Drs. Sarpkaya and Isaecson, " Mechanics of Wave Forces on Offshore Structures," Van Nostrand Reinhold &Co.,1981 and Dr. Blevins, " Formulas for Natural Frequency and Mode Shape," Malabar, Fla.:
R. E. Krieger, 1984,1979. Those' works discussed a approach for measuring experimentally added mass on a structure, in this case the strainer, within a still fluid. If that added rness can be determined, then a coefficient of hydrodynamic mass (Cm) could be calculatedas well as the total drag of the strainer.
l
, 1 5
CONCLUSION
.The staff concluded that these tests are a significant improvement from the previous tests from two points of view. The major improvement was the close similaritics between the Monticello installed strainer and the tested strainer. The second improvement, but also very important was the amount of tecting. Eleven air tests and 15 water tests were conducted. With the addition of the air tests, a basis was established for determining the added mass of water which would interact with the strainer. Looking at the strip recorders we noted that the test results were highly reproducible and should be more than sufficient to establish a firm data base.
The analytical approach used for determining the added water mass is bcsed on established work and supported by works o' others. For the calcule. ion of total
- drag forces, the licensee should continue to use the ; avious NRC staff approved method such as those contained in the Long Term Containment improvement Program (LTCIP). Specifically, the submerged drag loads were calculated using the method described in General Electric Co, topical report, NEDO 21471,
" Analytical Model for Estimating Drag Forces on Rigid Submerged Structures Caused by LOCA and Safety Relief Valve Ramshead Air Discharges," 1977 The exception from that report would only be for determining the hydrodyramic mass and the associated acceleration drag volume of the strainer. Therefore, we concluded that the results should resolve the last remaining issue on the installed strainers.
From this prospective, the staff discussed the contents of the data report to assure that all aspects of the testing would be addressed. We agreed that the report should clearly describe the analytical procedures to go from the tested natural frequency data to the acceleration drag values. The analytical methodology should be referenced and allinput values entered in tabular form. Supporting this documentation should be the strip chart data along with the internally calculated parameters such as accoieration and frequencies. Based on these discussions, the staff concluded that the tested strainar was similar enough to establish an acceptable data base.
The elements necessary to show that this test series was a properly conducted O/A tests were discussed. Based on these discussions it was agreed that the following would be considered. The items that will be included in the report are a description of the instrument calibration with traceability to a known standard, how data was recorded and what type of internal calculations were performed to obtain th(, processed output, the selection process of the test preload, and the
O 6
method of archival. This last item should include such information as who wFl be the keeper of the source data, in what form the data will be kept (i.e., tape, paper, etc), and for hnw long will the information be kept.
1 The report should include how this data is to be used for design purposes. For example, a discuission of test repeatability and uncertainty of the data should be provided. Uncertainties should lnclude instrument error and calculational uncertainties. Using this data base, a discussion of the safety margins and their rationste for these values te go from the data base to design based values should be included in the report.
The final discussion area was centered about how NSP will use this data. The NSP
- representative indicated that the test report will be on site for NRC review but they did not intend to put it on the docket. Nor did it appear that any reanalysis would be performed. Rather, the new data would be used to validate the values used in the existirig analysis. The staff indicated that we would look to the commitments identified in their letter to the staff dated 11/12/97. With this finalitem completed, the meeting was called to a close.
Docket No. 50 263
Mr4c# sTr4,uust Tvsr 7 I'l-l l9 l91
!!iEf'El tJotr Pact LiTV NM@_
'k$ Ll ATled WC
"[5M Mel Dtrys c.it-ns-m 100 kr[ewb/1 TE46 (770)44t-sza8
((Sio) G+9 -If b a s i d.. it.\\la m esI l mc W
3d WS-225 '
Ts.ak Kude; ch asnec
'Tbnf $ '/fmye/o US/YEC.
(3o y/5'-225f
//?fMW fdtLh' D knE Poh'[A IM 3 fi l - / / C l) boow-%
ace ew os wu_
~ _ 2. _
cm
,_y,_,
re i o,a m
-t rz we c.xe me / u 7- >v < >
Drws ?cacne a Nf? -/h%r,c neu G / z - 29r-a vr 02eg //sLL
() E J 5 0 3v-775 atie a
t s
6 method of 6rchival. This last item should include such information as who will be the keeper of the source data, in what form the data will be kept (i.e., tape, paper, etc), and for how long will the information be kept.
The report should include how this data is to be used for design purposes. For example, a discussion of test repeatability and uncertainty of the data should be provided. Uncertainties should include instrument error and calculational uncertainties. Using this data base, a discussion of the safety margins and their rate >nale for these values to go from the data base to design based values should be included in the report.
The final discussion area was centered about how NSP will use this data. The NSP representative indicated that the test report will be on site for NRC review but they did not intend to put it on the docket. Nor did it appear that any roanalysis would be performed. Rather, the new data would be used to validate the values used in the existing analysis. The staff indicated that we would look to the commitments identified in their letter to the staff dated 11/12/97. With this final item completed, the meeting was called to a close.
Docket No. 50-263 DISTRIBUTION:
Central Files SCSB r/f (2)
PDR TKim CCarpenter GHolahan J Newberry -
li5lliott
-JStang DLynch DOCUMENT NAME:
r.n = *,
.io.c.w w c -c m._.m ui.n.c,o,.ov.ncio...
c. cm on
.n.c.,m.no.ncio...
v. u, OFFICE.
SCSB:DSSA:NRR E SCSB: DSS Q 6 l
NAME AD*Angelo:bwA4)t/ JKudrick rd M 01/ A/98
~
01/f2/98 7
/ /98
/ i98
/ /98 DATE
/
~
0FFICIAL RECORD COPY I
-.-.--s
..--a.
w
~
e
-