ML19354E052
| ML19354E052 | |
| Person / Time | |
|---|---|
| Site: | Beaver Valley |
| Issue date: | 01/18/1990 |
| From: | Office of Nuclear Reactor Regulation |
| To: | |
| Shared Package | |
| ML19354E049 | List: |
| References | |
| IEB-88-011, IEB-88-11, IEIN-88-080, IEIN-88-80, NUDOCS 9001250298 | |
| Download: ML19354E052 (9) | |
Text
-.. _ _ _ _ _. - - _ - - _. _ _ _ _ _ - -.. _ _ _ _ _ - -
[e**4 S
' ~\\
UNITED STATES i
8' NUCLEAR REGULATORY COMMISSION 1
i f
wAsmwoTow, o. c. mss i
s......e SAFETY EVALUATION BY TE OFFICE OF NUCLEAR REACTOR REGULATION I
i i
THERMAL STRATIFICATION FOR BEAVER VALLEY UNIT 2 PRESSURIZER SURGE LINE WCAP-12093 DATED DECEMBER 1988 I
INTRODUCTION i
i The pressuriser surge line (PSL) in the Beaver Valley Plant Unit 2 (BV-2) is a 14" schedule 160 stainless steel pipe, connecting the bottom of the pressuriser vessel to the hot 4
leg of one of the coolant loops.
The out flow of the pres-j suriser water is generally warmer'than the hot leg flow.
j Such temperature dif ferential (AT) varies with plant opera-tion activities and can be as high as 3200F during the ini-tial BV-2 plant heat up.
Thermal stratification is the separation of cold flow stream in the horizontal portion of the PSL, resulting in temperature difference at the top and
]
bottom of the pipe.
Since thermal stratification is the direct result of the differences in densities between the pressurizer water and the generally cooler, heavier hot leg
]
water;- the potential for stratification is increased as AT increases and as the insurge or outsurge flow decreases.
Stratification in PSL was found recently and confirmed by data measured from several PWR plants.
l BV-2 original design analysis did not. include any stratified flow loading conditions.
Instead it assumed complete sweep t
of the fluid along the line during insurges or outsurges resulting in uniform thermal loading at any particular piping location.
Such analysis did not reflect PSL actual thermal condition.
It potentially may overlook undesirable line deflection and its actual stresses may exceed design limits.
In addition, the striping phenomenon, which is the oscilla-tion of the hot and cold stratified boundary, may induce high
+
cycle fatigue to the inner pipe wall and needs also to be analyzed.
Thus assessment of stratification effects on PSL is necessary to ensure piping integrity and code conformance.
5 1
5 9001250298 900118 KDR ADocK0500ggg2 P
- + -,, +,
-wr
-wwe----e
+-,c r,--.,,,---
--,e,---
~
l
\\
q Page 2 of 8
[
l STAFF EVALUATION i
i Since stratification in PSL is a generic concern to all PWRs k
an NRC Information Notice No 88-80 was issued on october 7, 1988, and then an NRC Sulletin 88-11 for the same concern was also issued on December 20, 1988.
The Licensee has submitted l
i-an evaluation report WCAP-12093 (Reference 1) which consist of presentation material and explanatory text.
Additional 1
4 information was also provided in a meeting,_between the NRC staff and the Licensee, on August 3-4, 1989, at the BV-2 site, to discuss specific portions of the report, which were included in supplement 2 of the report.
i The surge line stratification program for BV-2 consists of three major parts:
l
- a. Plant monitoring and update of design transients; l
- b. ASME III Code compliance for stress and fatigue;
}
- c. Participation in the WOG activities.
i The following is the staff's evaluation of the Licensee's efforts and information provided in the report:
A)
Plant monitorina and undate of desian transients i
The licensee had instrumented PSL and collected data for I
verifying the stratification condition in BV-2.
Such data were utilized in conjunction with data collected from PSLs of three other Westinghouse designed PWRs for creating design thermal transients and for developing flow stratification profiles.
Enveloping temperatures profiles and plantuspecif-ic profile techniques were both investigated.
)
Based on the monitoring data obtained, the design transients for heatup/cooldown, normal and upset operating conditions
'l were redefined to consider the stratification effects.
To j
determine the maximum top-to-bottom temperature differences, the system AT (maximum pressurizar and minimum hot leg temperature) was considered, i
The monitoring data obtained, indicates that during critical s
j operating transients, and heatup/cooldown operations, the i
pipe movements are consistent with analysis results.
Heat transfer analysis and test data available to date show that a stratified flow with a sharp fluid temperature gradient.at the stratification interface, will provide conservative conditions at the inside face of the pipe.
e g
Page 3 of 8 1
The normal and upset thermal transients were developed based ons 1
a.
The maximum Pressurizer to Hot Leg AT to occur in a given transient, although the maximum pressurizer tem-l parature and the minimum Hot Leg temperature for defin-ing AT may not occur at the same time.
I I
b.
Full stratification cycles for all transients except for i
steady state fluctuations, unit loading and unloading.
and reduced temperature return to power.
c.
The maximum top-to-bottom pipe temperature to be AT=
0 320 F, although none of the measured top-to-bottom pipe temperatures, in any plant, exceeded AT = 2700F even when system AT reached close to 3200F.
The envelope from measured transients in all plants was applied to define the analysis transients.
The staff con-cluded that the use of enveloped transients, based on data from BV-2 and other 3 similar PWR's, are conservative and acceptable.
B)
&SME code comoliance for stress and fatinue i
a)
Code comeliance in stress The Licensee performed a reanalysis on PSL piping and sup-ports to account for thermal stratification effects.
The analysis consisted of three parts:
(1)- global effects on stresses, moments, displacements, and support reaction: loads, based on both axial and radial variations in the pipe metal temperature, (2) local stresses due to thermal gradient, and (3) local stresses and effects on fatigue due to thermal striping.
The total thermal stratification stress.in the pipe is the result of global (structural) bending stress l
effects and the local stress effects due to thermal gradient, which means stress in items (1) and (2) above were super-imposed.
The staff agrees with the licensee's efforts and-methodology for monitoring, updating and assessing PSL for the stratification condition.
l Structural analysis of the surge line using stratification
(
induced loads determined the pipe displacements, support reaction loads, moments and forces in the piping system using the ANSYS computer code.
Although a 3200F step temperature change was assumed for stratification through out the surge line, the changes were linearized in ANSYS using a conven-i tional pipe element model.
G
.n.
.i I
s 1
I Page 4 of 8
)
Finite Element models were used to calculate local stresses due to top-to-bottom non-linear thermal gradients in the PSL.
Five (5) hot-to-cold interface locations were analyzed using eleven (11) cases of thermal stratification, to calculate piping' response under all required loading conditions, reflecting temperatures differences up to 3200F.
Other cases j
were obtained by interpolation.
Nozzle loads at locations connecting the surge line to the pressuriser and the hot leg-l to the main coolant loop were also evaluated.
Loads were found to be acceptable for'all design conditions.
The li-l consee reported that their best estimate analytical results compared favorably with measured displacement data observed i
during BV-2 monitoring.
Furthermore, no discernable distress l
was observed in the entire PSL including supports, during a recent visual examination.
{
The licensee also reported (Reference 2) that a trapeze i
assembly consisting of two snubbers was removed to accommo-date the unanticipated thermal movements observed during>the 2
hot functional testing (HFT),-and the power ascention testing (PAT), conducted in late 1986.
In addition, plant operating i
restrictions were implemented to alleviate worst case situa-tions.
The efforts for the snubber removal and operating.
restrictions were approved by NRC in a meeting held between i
Duquesne Power & Light Personnel and NRC's Office of Nuclear Reactor Regulation (NRR), in February 1989.
The calculated downward displacement value of 2.5" at at approx. mid-span location was found acceptable when compared to the measured value of 2.4" in the vertical direction for a stratification case of AT= 1800F.
]
i Local axial stresses were developed due to restraint of axial i
expansion or contraction.
The worst case being a step change t
in fluid temperature from hot to cold and occurs at the i
stratified interface.
This causes higher metal temperature gradients at the inside surface rather than the outside surface of the pipe.
i A 2-D finite element model was developed and the analysis i
considered unit loads of pressure and bending.
The resulting
)
stresses were scaled using actual pressure transients and moments from the global structural analysis.
Results of j
L global and local stresses at any point in the pipe cross section were combined using superposition methods.
t J
a l
Page 5 of 8 The primary plus secondary stress intensity range of equation 12 of ASME section III, NB-3600 was calculated to be 54.1 ksi and occurs at the hot leg nossle.
This is less than the code allowable value of 3.0 sa or 55.5 kai.
Stresses were inten-sified by "C" factors, to account for the worst case concen-tration for all piping elements in the PsL,- except for the RCL nozzle which used the results.of the finite element analysis in lieu of the code stress indices.
The staff agrees with the approaches used by the licensee for perform-ing PSL reanalysis.
b)
Code connliance in fatinue The total fatigue to PSL is the result of stratification induced global and local cyclic stress effects and striping induced local cyclic stress effects.
To account for the thermal striping effects to PSL, flow model test results performed for the Liquid Metal Fast Breeder Reactor primary loop and for the Mitsubishi Heavy Industries Feedwater Line cracking, were reviewed to establish the boundary condition.
These. test results were used to define striping oscillation data, amplitude and frequencies, for evaluating high cycle fatigue.
Portions of PSL which experience stratification and striping were defined based on measured results.
A frequency of.30 HZ with an attenuating amplitude of 100% AT were selected for the analysis.
Thermal striping stresses are the result of differences between the pipe inside surface wall and the average through wall temperatures which occur with time due to the oscilla-tion of the hot and cold stratified boundary.
This stress is defined as a thermal discontinuity peak stress and is re-quired by the ASME Code for the calculation of fatigue usage factor.
The peak stress range and stress intensity values were calculated from a 2-D finite element analysis stresses and were intensified by X to account for the worst stress concentrationforallpipfngelements.
The maximum fatigue usage factor obtained is 0.68 and occurs at the RCL nozzle safe end.
The Licensee reported that considering AT attenuation with time, and a frequency of.30 HZ, a usage factor of less than
.20 was determined as the worst casa due to striping alone, even when the stresses were intensified by "K" factors.
L l
l I
i Page 6 of 8 t
Furthermore, the total calculated usage factor for striping I
alone was increased by 50% to account for any uncertainty in l
the selection of frequency and other variations.
A surface film coefficient of 500 BTU /hr-sq. ft CF was used and it was based on a flow rate of 90 gpa which was assumed to be con-stant throughout all striping analysis.
Although the data i
used in the assessment were obtained from 1/5 scale test model which showed that the frequency can range from.10.50 l
HZ., the frequency corresponding to the flow rates which closely approximate the Richardson's number for the pressur-iter surge line were used.
The staff agreed that the stresses will be higher with the lower frequency and the.30 HZ. average frequency is justi-fled.
otential due to a film coefficient of 500 BTU /hr-sq.However, ths thermal striping gF and attenuation ft.
is questionable, but at this time no other better number exists and therefore this represents the best judgement.
If other information will be available, based on the ongoing s
efforts by EPRI or possible future NRC research work, it will be utilized and further assessments will be made for assess-ing the striping effects of PSL.
t With the thermal transients updated to reflect stratification effect, new fatigue usage factors were calculated.
To deter-aine the new fatigue usage factors by stratification alone,
~
the more detail techniques of ASME III, NB-3200 were en-ployed.
Due to the non-axisymmetric nature of stratification loading, stresses due to all loadings were obtained from Finite Element analysis and then combined on a stress com-ponent basis.
Five (5) levels of thermal stratification at five worst case points were calculated using the WECEVAL program.
Except in nozzle locations, for most components in the surge line (girth butt welds, elbows), no gross structural'discon-tinuity is present and therefore the contribution of this code defined "Q" secondary stress is zero.
As a result, the equation 10 stress value for these components is due to pressure and moment only.
For the RCL hot leg' nozzle, the results of the 3-D finite element WECAN analysis was used to determine the secondary gross structural discontinuity con-tribution for transients with stratification in the nozzle.
When the appropriate stress indices were used, the following maximum values were obtained:
e b
r a
Page 7 of 8 Location Stream Value fK.E.I.)
Allowable fK.S.I.)
RCL Nozzle og 12 = 54.1 3 Sa = 55.5 RCL Nostle og 13 = 33.6 3 Sa = 48.6 Usage factors at the inside and outside pipe wall were evalu-ated at 17 sections through the pipe wall.
For the cases which eq. 10 exceeded the 3Sm. limit, a simplified elastic-plastic analysis was performed as per NB-3653.6.
Values of Eq 12, 13, Re and thermal stress rachet were checked.
Eq 13 is not affected by thermal stratification in the pipe where no gross structural discontinuities exist, but was checked at' the nottles.
The licensee reported that the fatigue usage factors, without including the striping effects of.20 at the elbow to the vertical pressurizer drop and of.48 at the RCL nozzle safe end were calculated.
Based on this, a worst case cumulative usage factor (CUF) at the RCL nozzle of.68 is established when the striping effects are included.
The staff agrees with the approaches used by the licensee for calculating the usage factor.
CONCLUSIONS Based on our review, we conclude that the information pro-vided by the licensee in References 1, 2, and 3 are compre-hensive and acceptable.
The licensee of Beaver Valley Unit 2 had made acceptable efforts to meet all requested actions as delineated in the NRC Bulletin 88-11.
Their efforts have demonstrated that based on plant specific temperatures pro-files defined by available stratification data on Beaver Valley Unit 2, the PSL meets the ASME Code Section III accep-tance criteria.
PRINCIPAL CONTRIBLRORS Shou Hou, NRC reviewer F. Vasiliadis, NRC consultant DATED January 1990
1 i
e Page 8 of 8 l
i REFERENCES 1.
Westinghouse Report WCAP-12093 (Proprietary),
" Evaluation of Thermal Stratification for the Beaver Valley Unit 2 Pressuriser surge Line" December 1988.
j Supplement 2, ' Additional information in support of the evaluation of Thermal Stratification for the Beaver i
Valley Unit 2 Pressuriser Surge Line", dated August 1989.
j 2.
Letter from Duquesne Light Company to H.R.C.
i "Special Report" Dated May 16, 1989.
ND3MNOt1884 i
3.
Letter from Duquesne Light Company to N.R.C.
j
" Response to NRC Bulletin 88-11" Dated May 17, 1989.
l i
I I
l
)
1
?
w
,e Enclosure'2
$aMg 0rqDa./v/
ObM Bian TekAh
~ kc '
Wz-m RP/
\\
M
//6ims L ddC
/.1/
'r's2-6 Y2-no o x 26 o konaId A. h erj bLC si 393 - O '70
(( b 44
/Kd WA-4 E7-I M M-fer S T4m A/4, C.
30/-49'h-/3*7 e
$ ' $dn'rjd WE/$'dik:.
(2/f)f/2-7274~
9 6m S.
[ALus w/
Web h C41t)2svbes5
)
~f~i / fn WN 76/ -4f2 -f pf){
V 1
6.
(M NA<--
3 s / - 4 4 -a 7
- 4-
{- llposiMiJ AltcGwsafc4r 401-62a-f W I
\\
35e A..z%7 Wa-s7f-c.zo7 2r= we U newsc.
%- m-yw 9
[N. lta.-
(Ales Sms - (kE 4-I2.- 3C4 $4 KC.Chy W*styA* se -
triz - ns -6s*f E. L. GMkNfcKo WesrtHs. House Alz. - 234 - GA Sc6 6.R. M UTy6L A
\\N ES n N 4 Hou.f e 4-l 2.
- 2. 7 4 - 6 3 oJ' J$WCLl'kSTA' D( C f/2 - 393 -av.2.-
i Ss.we K.
Muwworps DLc 412 - 391 - 773 7 l
c,efLE bWeDd
}LC.
4 l 2. -34 3 - 77I b i
h
-