ML19329A544
| ML19329A544 | |
| Person / Time | |
|---|---|
| Site: | Oconee  |
| Issue date: | 11/03/1972 |
| From: | Peltier I US ATOMIC ENERGY COMMISSION (AEC) |
| To: | Deyoung R US ATOMIC ENERGY COMMISSION (AEC) |
| References | |
| NUDOCS 8001060059 | |
| Download: ML19329A544 (10) | |
Text
l-
.A
... ~ -
NOV 3 7372 Docket No. 50-269 R. C. DeYoung, Assistant Director for Pressurized Water Reactors, Licensing A. Schwencer, Chief, Precsurized Water Reactors Branch NdirfAsaMgg TilRUt A. Schwencer liEETING WITH DUKE POWER COMPAhT AND BABCOCK AND WILCOX COMPAhT - OCTOBER 25, 1972 B&W TOPICAL REPORTS CN THE SUBJECT OF REACTOR VESSEL INTERNALS - OCONEE UNIT 1 Enclosed is a sunnary of the meeting held on October 25, 1972 with Duke Power Company and the Babcock & Wilcox Company. An attendance list is also enclosed.
I El
.I. A. Peltier, Project Manager Prassurized Water Reactors Branch No. 4 Directorate of Licensing
Enclosures:
1.
Meeting Su==ary 2.
Attendance List cet R. S. Boyd D. Skovholt D. Knuth DISTRIBUTION R. Maccary Docket H. Denton PWR-4 Reading PWR Branch Chiefs RP Reading R. W. Klecker IAPeltier M. Rosen R0 (3)
D. Lange C. Hou H. Schierling R. Bernero B. Buckley i
H. Faulkner J. H. Sniezek K. Wichuan K. Economos
. ?:t:7:::
c omer >
g/;,,q nEC PDR L:PWR-4 L::WR SURNAMr>
Wen er 11LDl.12._ J1/..3.1.7 2....
oart >
Form AEC-Sin (Rev.9 53) AECM 0240
- 13. s. covEnutzt engxTnes orrrct asto o. 4os.34s 8001060 O Q g
ENCLOSURE NO. 1 MEFTING WITH DUKE POWER COMPANY AND BABCOCK & WILCOX COMPANY MEETING
SUMMARY
October 25, 1972 i
Summary The questions raised by MEB on B&W topical reports BAW-10037,10050, and 10051 and the answers presented at the meeting by B&W are enclosed with the original and related copies of this report (Docket File, Project Manager, MEB and PDR copies).
Although the topicals along with the supplemental information provided by B&W will permit MEB to complete its review of the Oconee Unit 1 internals redesign, approval of the topical reports depends on the success of Oconee Unit 1 tests. B&W and the applicant will receive letters stating the con-ditional acceptability of the reports.
Per Safety Guide 20, in order for Oconee Unit 1 to be a valid prototype B&W must establish its ability to predict the forcing functions which can be correlated with the response of the sytem. This effort requires a blending of theoretical and experimental work and because of the scaling problems the 1/6 scale flow model tests are not useful in establishing dynamic fluid characteristics for the large system.
In this regard B&W stated that it is currently considering tests in a full scale vessel capable of testing full scale production guide tubes and nozzles. B&W is i
considering.unning both the old and the new design guide tubes and nozzles in this facility.
B&W also stated its intention to perform in air vibration tests on the new internals using the SMUD hardware at Barberton just as the old internals were tested during the failure investigation.
B&W suggested that thermal shield forcing functions may be developed from the original Oconee Unit 1 hot functional tests. These tests in conjunction with the full scale internals tests could in B&W's judgement provide the
' basis for establishing Oconee Unit 1 as a valid prototype per Safety Guide No. 20. Another possibility discussed is the instrumentation of the thermal shield in Oconee Unit 2.
MEB reserved judgement on these approaches until it has had an opportunity to review the entire package of efforts being conducted by B&W. B&W said it would attempt to provide this package to the AEC by the end of November 1972.
i
ENCLOSURE No.-2 o.
MEETING WITH DUKE POWER COMPANY AND BABCOCK AND WILCOX COMPANY - 10/25/72 ATTENDANCE LIST DUKE POWER COMPANY K. S. Canady BABCOCK & WILCOX B. A. Karrasch E. O. Hooker G. E. Kulynych A. H. Lazar R. R. Steinke R. N. Edwards R. S. Baker H. J. Fortune G. D. Lindstrom AEC
- 1. A. Peltier J. . Sniezek C. hou D. Lange A. Schwencer K. Wichman K. Economos V. Potapovs
TOPICAL REPORT BAW-10037 REVISION 1 VESSEL MODEL FLOW TESTS in equation 1-1.
Verify the possible omission of the flow area term 1.
QUESTION:
rrect.
Equation 1-1 in B&W Report BAW-10037, Revision 1, is co idual meter The meter flow coefficient, k, includes the ind h Flow
RESPONSE
Factor a dimensionless variable distribution The flow frequency content and the related energythe one-sixth i
was not determined by the measurements dur ngIdentify the co 2.
-QUESTION:
functions for response scale model testing.
testing to the postulation of forcingProvide the bases for the use of th AW-10051 to compute prediction analysis. simple equation set forth on page 3-4 of B the shedding frequency, since this metho a simple flow condition.
ith our con-A study of available literature and discussions wd energy distri-sultants indicated that the flow frequency an
RESPONSE
be butions determined from the 1/6 scale model could notFurthe correlated to actual vessel conditions. st l components of model were not simulated in the model, measurements were not attempted.
structural response the velocity Data from the 1/6 scale model was used to predictsel and inter and static pressure distributions within the ves i functions This data was then used as the basis for the forc ng described in'BAW-10051.
h dding The use of the simple equation to determine vortex s e internals is only a part of the structural analysis of theThe components.
l'
TOP 1 CAL REPORT BAW-10050 EVALUATION OF OCONEE REACTOR COMPONENT FAILURE 1.-
-QUESTION:
As stated in page 4-12, the first mode frequency of the
~
instrument guide tube _is 250 Hz while the vortex shedding frequency.is approximately 385 Hz, therefore, the first mode response may be excluded as a failure mode. However, higher modes may be in the range of the vortex shedding frequency or other forcing frequencies.
(a)
Provide a' comparison of the higher mode guide tube frequencies with the shedding frequency.
RESPONSE
The second mode of the instrument guide tube would be 21000 Hz.
This is substantially above the vortex frequency of 385 Hz.
This would also be an unsymetric mode and would not be excited as easily.
On page 4-11 of BAW-10050 in paragraph 4.4.3 it was stated the velocity could be as high as 60 fps.
1 A velocity of 39 fps would give vortex shedding frequency of 250 Hz.
With less than 4 pump operation the velocity would be less than 60 fps and could therefore excite the instrument guide tube at it fundamental frequency of 250 Hz.
B&W still believes that vortex shedding was a contributing factor leading to the failure of the tubes.
(b)
Provide the criteria that was used for the redesign of the instrument guide tubes.
RESPONSE
Instrument guide tubes will not be installed in the Oconee Reactors, therefore, no criteria for redesign are presented.
(c)
Provide a discussion of other possible causes of failure, such as the mentioned random excitation of turbulence and the reactor coolant pump excitation.
Include the effect of the pump shaf t frequency of 20 Hz (Page 4-9).
RESPONSE
The pump shaf t frequency of 20 Hz is not believed to be a major contributer to the excitation of the instrument guide tubes. The second blade passing frequency of the pump of 280 Hz also mentioned on Page 4-9 is sufficiently close to the naturally frequency of the tube (250 Hz) to have excited the tube at its natural frequency.
There was also a 190 Hz frequency identified in the hot functional test data, which could have also caused excitation.
E
BAW-10050 Page 2 Provide a discussion on the following possible failure mode on 2.
QUESTION:
the core instrument nozzels: The core structure vibratory motion and the cross flow loading may produce a rotational vibration mode in the guide tubes and associated lateral defromation of the lower tips.
The lateral motion may produce vibratory contact with the inserted tip of the incore instru-ment nozzle and result in cyclic bending stresses at the bottom of the nozzle to failure.
An investigation was conducted to detensine whether mechanical
RESPONSE
coupling of the incore instrument nozzles with the guide tube assemblies could have caused or contributed to the failures.
The interior of the incore instrument guide tube extensions were examined to determine if failure of an incore nozzle had occured without signs of contact between the nozzle and the
[
guide tube.
Evidence of contact occured in most cases.
However,'in four cases little or no contact was indicated.
This substantiated the conclusion that the incore instrument nozzles could f ail without excitation by the reactor internals.
4 TOPICAL REPORT BAW-10051 DESIGN OF REACTOR INTERNALS AND INCORE INSTRUMENT N0ZZLES FOR FLOW INDUCED YIBRATION 1.
QUESTION:
Describe the loading combinations and the analytical cethods used to confirm the structural integrity of the instrumentation guide tubes. Provide the basis for the criteria that redesign is not necessary if two guide tubes fail during hot functional testing.
RESPONSE
Instrument guide tubes will only be installed on two B&W Reactors.
For these reactors, a redesign of the instrument guide tubes will be presented in the respective FSAR's.
2.
QUESTION:
As shown in table 3-3 (page 3-261 the cantilever part of the guide tube and the flow distributor assembly (vertical) have approximately the same first mode frequencies. The configuration shown in figure 3-3 indicates that the vertical motion of the flow distributor may produce rotation and therefore lateral motion of the lower tip of the guide tube.
Provide a summary of the dynamic analyses used to account for possible dynamic coupling of the Guide tube and the flow distributor assembly.
Include the effects of cross flow on the cantilever portion of the guide tube. The associated cyclic bending stresses at the incore instrument nozzle should also be provided.
RESPONSE
It is true that rotations as a result of vertical motion of the flow distributor will occur at the attachment points of the' cantilevered portions of the guide tubes and the flow distributor. This was recognized as a potential source of lateral excitation on the guide tubes and was investigated as a part of the analysis.
that these rotations w6uld be quite small ( %10Theresultsindicateg, rad
~
- however, the resulting lateral guide tube loadings would be considerably less than the conservatively assumed cross-flow loadings.
In light of the above, this source of excitation was assumed to be included in the cross-flow loadings.
The results of any dynamic interaction between guide tubes and flow distributor will be measured during the h9t functional testing. These results, contrelated with the results of the in-air testing of an identical set of internals, will identify the amount of dynamic coupling.
1
BAW-10051 Page 2 3.
QUESTION:
The shedding frequency used for computing the 6 value of the drag force acting on the incore instrument nozzle was actually based upon a 2 inch diameter (page 3c5 ) of the lower portion.
Since the upper portion is a 1-1/8 inch diameter (8=1),
provide a summary of the analysis to show that excessive response amplittads of the instrument nozzle will not occur.
RESPONSE
This possible effect was fully evaluated during the original design effort. The question correctly points out that reso-nance will apparently occur due to vortex shedding from the reduced diameter of the upper portion of the nozzle, with the assumed flow conditiens. This is indicated by_the_ frequency ratio B = e /W = 1 for the fluctuating component of the drag s n force. This is simply an anomaly arising from the accund.ation of worst case assumptions and is not representative of the actual conditions. These assumptions included:
a.
A cross-flow past the nozzle is assumed to be uniform over the entire length. Actually, the reduced diameter portion is inside the lower end of the guide tube and the cross-flow past it is considerably reduced.
b.
The velocity past the entire nozzle was assumed to be 40 ft/ rec. Actually, this value represents the conservative upper bound of the peak velocity at any point in the lower head of the vesael.
Also, this velocity was assumed to be croso-flow (at right angles to the nozzle) rather than acting at a skewed angle which is the real case.
c.
The Strouhal number "S" was assumed to be 0.45 which represents the upper bound of possible values.
This parameter is actually a strong function of Rey-nolds number and hence, is dependent on the velocity and the diameter of the tube.
d.
The natural frequency of the nozzle used in the cal-culations of 8 was 425 Hz.
A more detailed analysis indicates that the actual ffEquency~of thw nozzle will probably be 550-600 Hz.
All these factors are combined to eliminate the possibility of resonance between the forcing functions and the structure.
BAW-10051 Page 3 4.
' QUESTION:
Provide the basis for assuming that.the lowest mode deflection of the thermal shield is 0.06 inches.
5.
QUESTION:
Provide the basis for assuming that the amplitude of other predominate modes of the thermal shield are a function of the ratio of the frequencies squared to the first mode (page 3-14).
6.
QUESTION:
Provide tha basis for neglecting the combined modal contribu-tion effects in predicting the maximum radial deflection of the thermal shield under the hot functional testing and normal operational loadings (Table 3-5).
RESPONSE
Questions 4, 5 and 6 all deal with the thermal shield and the assumptions used in predicting its response, and will therefore be answered by a single p. jfp /p Evaluation of the measured response data from the first Oconee Hot Functional Test indicated that the maximum amplitude of vibratory motion of the Thermal Shield was about.006" (before wear of the supports led to subsequent damage). To account for the ' possibility that we may not have measured amplitude at the peak location or time, a safety factor of 10 was placed on this measured value to establish a criteria for redesign. The resulting.060" amplitude was assumed to occur in the most probable (lowest ) mode of operation of the thermal To evaluate ts effecE"of response in other modes, the
- h shield.
stresses in the supports were determined for comparable ampli-tudes, using the ratior.of frequency squared. This corresponds to an assumption of constant input force or acceleration. We expee the total amplitude to be considerably less than 0060",
so t'.e results were not combined for the various modes. This escamption of amplitude is conservative in comparison to that ceasured for other operating reactors.
The revisions for the support design for the thermal shield increased its stiffness and greatly increased the strength of the support.
In addition, removal of the vertical legs of the flow baffles reduced the peak annulus velocity and the forcing functions acting on the shield. Both of these changes are in the direction of further reduction of thermal shield motion.
,