ML20206G204
| ML20206G204 | |
| Person / Time | |
|---|---|
| Site: | South Texas  |
| Issue date: | 05/04/1999 |
| From: | Alexion T NRC (Affiliation Not Assigned) |
| To: | NRC (Affiliation Not Assigned) |
| References | |
| TAC-M98914, TAC-M98915, NUDOCS 9905070124 | |
| Download: ML20206G204 (14) | |
Text
,
- May 4, 199cl MEMORANDUM TO: File i
FROM: Thomas W. Alexion, Project Manager, Section 1 ORIGINAL SIGNED BY Project Directorate IV & Decommissioning l Division of Licensing Project Management 1 l
SUBJECT:
SOUTH TEXAS PROJECT, UNITS 1 AND 2 - DISCUSSIONS WITH LICENSEE ON ITS 10 CFR 50.59 EVALUATION OF MAIN STEAM LINE BREAK (MSLB) REANALYSIS AND EFFECT ON ISOLATION VALVE i CUBICLE (IVC) AND REACTOR CONTAINMENT BUILDING (RCB) !
(TAC NOS. M98914 AND M98915)
During a December 1,1998, phone call between the NRC staff and the licensee (STP Nuclear Operating Company), the staff informed the licensee that the staff needs additional information on the licensee's 10 CFR 50.59 evaluation of the MSLB reanalysis and its effect on the IVC and RCB in the structural area. The staff also indicated which specific items of the licensee's 50.59 evaluation the staff was interested in obtaining additional information on. (A copy of this 50.59 evaluation was placed in the public document room on June 13,1997, ACN 9706160259),
in order to facilitate future discussions between the licensee and the staff, the licensee provided the draft information in the attachment. The purpose of this memorandum is to place the attachment in the public document room.
Docket Nos. 50-498 and 50-499
Attachment:
As stated r
et F e PUBLIC PDIV-1 r/f DOCUMENT NAME: 1:\SOTEXAS\MEM98914.WPD To recetve a ecpy of the document, indcate n the box C= Copy w/o attachment / enclosure E= Copy with attachment / enclosure N = No copy OFFICE PDIV-1/PMJ PDIV-D/LA E PDIV-1/SC M s NAME TAlexio h CJamersonk RGramm b DATE h/b /99 Y / 2S /9 (/ 9 /99 OFFICIAL RECORD COPY 9905070124 990504
APR-22-1999 .09154 NUCLEAR LICENSING 512 972 9298 P.02/14 f-eer Cperatine pany m
NOC-AE-000512 Pile No.: G25 v
3f J]M JU ,
10CFR50.59 U. S. Nuclear Regulatory Commission Attention: Document ControlDesk Washington, DC 20555 South Texas Project Units 1 and 2 Docket Nos. STN 50-498, STN 50-499 Resnonse to NRC Ouestions Concernine Revised Main Steam Line Break Analysis Pursuant to a telephone conference conducted on December 1,1998, the South Texas Project submits the attached responses to questions from Nuclear Regulatory Commission staff reviewers regarding Justification for Continued Operation 93-0004.
"Ibe South Texas Project understands that some of the questions are no longer current because they are in response to statements referring to the analysis as it existed previously or to engineering judgments used to justify operation during the mterval between identification of the problem and revision of the calculation. Because the stated intent of the Nuclear Regulatory Commission staff review is to evaluate the design basis as it stands today, the South Texas Project has identified those questions that are not applicable.
If there are any questions, please contact either Mr. P. L. Walker at (512) 972 8392 or me at (512) 972-7162.
S. E. Thomas Manager, Design Engineering PLW Attehment: Response to NRC Questions Concerning Revised Main Steam Line Break Analysis Attachment
APR-22-1999 09:54 NUCLEAR LICENSING 512 972 6290 P.03/14 Auht NOC AE.000512
]/ lb f }U J ,
Pile No.: 025 Page 1 of 6 I
RESPONSE TO NRC QUESTIONS CONCERNING REVISED MAIN STEAM LINE BREAK ANALYSIS l BACKGROUND: I Justification for Continued Operation 93 0004 addressed the Main Steam Line Break blowdown model used by the South Texas Project which was determined to be not conservative. During the l
interval between identification and correction of the problem, the South Texas Project used l engineering judgment and evaluation of preliminary calculation results to justify full-power operation. The analysis was revised, and review of the changes in accordance with 10CFR50.59 demonstrated that the Main Steam Line Break issue does not result in an Unreviewed Safety Question at the South Texas Project. Justification for Continued Operation 93-0004 was closed following completion of the necessary calculation revisions.
NUCLEAR REGULATORY COMMISSION OUESTIONS:
L What method and assumptions were used in obtaining the new dynamic loadfactors?
Response: .
Attachment A is an excerpt from a dynamics textbook (" Introduction to Structural Dynamics,"
by J. Biggs) ahowing the value of dynamic load factor (DLF) as a function of the ratio (t, / T),
where T is the stmetural period and t, is the rise time of the load. For this model, the load is assumed to rise linearly from zero to some value and stay constant thereafter. In order.to use this model to calculate DLF, two values are needed, namely the period of the structure and the rise time of the applied load. The goal is to select DLF conservatively. A general observation gleaned from Attachment A is that conservative values of DLF are obtained using amall values of the ratio (tr / T).
P.uind -- Conservatism in the calculation of DLF is achieved by calculating the period T as an upper bound. This is achieved by selecting the most flexible portion of the Isolation Valve Cubicle. The largest span of any concrete slab in the Isolation Valve Cubicle is 27 feet. This span is evaluated as a one-way simple-span beam to calculate the lowest natural frequency as 13.2 Hz. This corresponds to a period of T = 0.076 seconds. The period is a property of the structure that is not changed by new pressure loads.
Rima Time (tA --Conservatism in the calculation of DLF is achieved by defining the rise time t, as the earliest time at which a defined peak pressure occurs, corresponding roughly to the time at which the roof blows off. Attachment B shows an example P(t) curve [ plotted on a semilog scale) taken from Figure 3.6.A-3 of the UFSAR for region 7 of the Isolation Valve Cubicle.
Superposed on the P(t) curve in Attachment B are the selected values of rise time (t, = 0.011 see) and peak pressure (P = 27.7 psia = 13.0 psig). This means that the ratio (t, / T) is equal to (0.011/0.076) = 0.15. Entering the figure in Attachment A using this value, the dynamic load factor is determined to be approximately 1.95. Hence, the structural design pressure for this example case is 1.95 x 13 = 25.4 psi.
us bemea -- -ti..- ..ie--- aw._d, eg ,4 Awemw ar -w s*-44 ..emum--=
- PR-22-1999 09:54 NUCLEAR L1CENS1NG 512 972 8298 P.04/24 l me e2 Attachment ;
yl NOC-AE 000512 i
, ._) . File No.: O25 Page 2 of 6
- 2. Provide analysis or evaluation that supports the statement: "If a localized area is overstressed by 10% in this concrete cor6guration, the stresses will be redistnbuted, and thus the loads would be carried by other structural areas or members. "
Response:- This is not applicable. De reference to overstress is an engineering judgment limited in scope to Justification for Continued Operation 93-0004. The current design calculation shows that all stresses are within American Concrete Institute code allowables. I Bere is no overstress.
- 3. Justification for Continued Operation 93-0004 states: "ne design strength of the concrete (fe) is 4000 psi whereas actual strength of the concrete is in excess of 6000 psi. l Actual strength of Grade 60 reirtforcing steel is typically S to 10% above the minimum '
required 60 ksi. " Provide examples of material test data thatjustify [use of strengths in i excess of)4000 psi and 60 ksi. '
Response: His is not applicable. The higher material strengths mentioned in the question are only applicable to Justification for Continued 6 >eration 93 0004. De current design calculation uses 4000 psi for concrete strength and 60 ksi' r reinforcing steel yield strength.
- 4. Justificationfor Continued Operation 93-0004 states: "However, based on a review of the existing analysis, DED has concluded that the maximum stress in the building is strongly correlated to the maximum pressure in the break node, and thefact that the non-peak pressures have increased is of relatively little sigmficance. " Explain the rationale behind the conclusion with reference to existing analysis.
Response: The question refers to an engineering judgment used in Justification for Continued Operation 93-0004. It is not directly applicable to the current design basis calculation. The validity of this judgment is confirmed by the results of the detailed final analysis. (See responses to Questions 6 and 7 for discussion of the final analysis.)
S. Discuss the engineering basis for the statement: "De small increase [results of RCB short-term MSLB pressure analysis] has been evaluated and found to be within the existing margin of the structure." ,
Response: Table 6.2.1.2-13 of the UFSAR has been updated to show the new slightly higher pressures. This table also lists the higher pressures conservatively used in the structural design calculation and the design margins. It is readily apparent from the table that all calculated pressures are well below the structural design values. The minimum margin shown in the table is 116% (meaning the pressure used as input for structural design is 116% greater than the new higher calculated pressures).
- 6. Provide a copy of calculation CC 6251.
Response: During the December 1,1998, teleconference, the South Texas Project advised the NRC that the calculation includes roughly 10,000 pages (mostly computer output). The NRC
, CI"'
o w
.APR-22-1999 09:55 NUCLEAR LICENSING 512 972 8298 P.05/14 Atr*4mmt T g n NOC AB-000512 h Pile No.: 025
. J -
Page 3 of 6 agreed to withdraw the request for a copy of the calculation and accept a written summary of the calculation instead. .
Summary ofInidst ca1ruindan CC-6251 The original structural analysis of the Isolation Valve Cubicle used the Finite Element Method.
The loads, load factors, and load combinations used in the analysis are listed in Table 3.8.4-1 of the UFSAR. These loads included dead, live, seismic (Operating Basis Earthquake and Safe Shutdown Earthquake, each in three directions), wind, tornado, flood, pressure, thermal, and piping loads. Note that pressure is the only one of these loads to change as a result of Justification for Continued Operation 93 0004. The calculation lists all of the load combinations in UFSAR Table 3.8.4-1 and then explains why several of them can be deleted as being obviously bounded by other cases.
Six load combination cases were identified as being potentially governing. The Finite Element Method analysis used only those six load combinations.
The finite element model included more than 2000 phte elements to represent reinforced concrete walls and floors. In addition, nearly 200 beam elements were used to represent the reinforced concrete columns and beams. The steelphtforms located at several elevations in the building were not included in the model The soil was represented in the model with linear springs attached to the plate elements adjacent to soil.
The pressure loads P(t) were calculated in separate calculations, using methods described in detailin Section 3.6.A of the UFSAR. Several different accident scenarios were considered to establish the accident pressure to be used for structural design. Calculation CC-6251 tabuktes the peak pressures from the various cases considered (breaks of Main Steam Line, Feedwater, Auxiliary Feedwater, and Steam Generator Blowdown lines) and demonstrates that the governing accident is a Main Steam Line Break (MSLB). The pressure calculations divided the Isolation Valve Cubicle into nine regions, as shown in Attachment C. The main steam line passes through regions 6 and 7, so the highest design pressure occurs in one of these two regions.
A break in region 7 produced the higher pressures. Outside the break region, both cases were considered and the higher of the two pressures was used. The pressure load in each region was applied to all of the finite elements in that region as a uniform load of magnitude P. =
(maximum pressure for the region) x (DLF). The dynamic load factor used was 1.32, which is 1.2 conservatively increased by 10%.
The output of the Finite Element Method analysis was forces and moments in each element. A computer analysis of each element compared these loads to the American Concrete Institute code allowables calculated based on the element thickness and reinforcing steel content. The calculation demonstrated that the elements met the code.
Summarv of Revisions to Calculation CC-6251 The calculation was revised in two steps:
Initial Portion of Revision -- New pressures were calculated on a preliminary basis (unverified) to be used as input for structural design. The pressures were provided to Bechtelin
, m O
.APR-22-8999 09:55 NUCLEAR LICENSING 512 972 0390 P.06/14 ao in. w 1
,I 7.h 8 i 5 NOC-AB 000512 j 1l Liu J . File No.: 02S i
- Page 4 of 6 the form of eighteen P(t) functions showing pressure vs. time in each of the nine regions of the j Isolation Valve Cubicle, with one set of nine for an assumed MSLB in region 6 and one set for j an assumed break in region 7. Bechtel selected the higher of the two pressure peaks for each j region, and calculated the dynamic load factor as described in the answer to Question 1. The i following table summarizes the results. !
i HM NX iMisfjj M E M E2 hIf0f N i
&$l Q?C JK:Q] gQWO ;&f Q]T9G 1 7.0 1.32 9.2 11.6 1.0 11.6 25.5 2 7.0 1.32 9.2 11.1 1.0 11.1 20.1 l
-3 6.0 1.32 7.9 11.0 1.2 13.2 66.7 l 4 8.6 1.32 11.4 11.1 1.2 13.3 17.3 j 5 8.6 1.32 11.4 11.3 1.2 13.6 19.5 i 6 8.6 1.32 11.4 10.3 1.2 12.4 8.9 i 7 14.0 1.32 18.5 13.3 1.9 25.3 36.7 8- 5.3 1.32 7.0 9.3 1.2 11.2 59.5 9 5.3 1.32 7.0 9.8 1.0 9.8 40.1 (Note: pressure is given in psig; P. = (P) x (DLF), where P is maximum pressure and P. is the design basis accident pressure.) j The table compares new and old values of design pressure (peak pressure multiplied by DLF).
In the break region, the design pressure increased by 37%, due entirely to an increase in DLP from 1.32 to 1.9. In the non-break regions of the Isolation Valve Cubicle, the design pressure increased by different amounts in each region, with 67% being the maximum increase in any region.
The most highly stressed elements from the Finite Element Method analysis were reevaluated.
All elements in region 7 were considered to have their calculated element forces and moments increased by 37% and were shown to be still within the allowable limit. All elements in regions other than region 7 were considered to have their calculated element forces and moments increased by 67% and were shown to be within the allowable limit. Hence, the Isolation Valve Cubicle was shown to be acceptable with the new pressure loads. ' Itis analysis is conservative because it amplifies all of the loads (including seismic) by the same ratio as the increase in pressure.
Finni Ponion of Revision -- After Bechtel completed this analysis, the new calculated pressures were verified. During the verification process, the pressures were recalculated and revised. Rather than mpeat the new analysis that Bechtel had just done, the South Texas Project compared the final P(t) curves (shown in the UFSAR as Figure 3.6.A-3) to the ones used by Bechtel and concluded that Bechtel's analysis was bounding. 'Ibe following facts are documented in the calculation to support this conclusion:
m 7 c, L J'
I
'APR-22-1999 09:35 NUCLEAR LICENSING 512 972 8299 P.07/14
- ^
Auehmem e "P O 1 NOC-AE-000512 b}
h f- Pile No.: G25 Page 5 of 6 1
In each of the nine regions, the final (verified) peak pressure was lower than the unverified value used by Bechtel.
The rise time used to calculate dynamic load factor was very similar in both cases.
Unlike the original curves and the unverified curves, the verified P(t) curves reached their maximum values at the end of the time history (see Attachment B). This means that the peak occurs at a time when the behavior is nearly static. However, the dynamic load factors were calculated based on the earlier peak and conservatively multiplied by the late maximums to get the design pressure.
Also, as mentioned above, the Bechtel analysis was conservative in that it implicitly increased all Joads, not just pressure loads.
- 7. Justification for Continued Operation 93 0004 states: "Upon receipt of the final pressure / temperature analyses from Nuclear Fuel and Analysis, DED Civil / Structural Group will reevaluate the Isolation Valve Cubicle structurefor the efects of the pressure loading." What were the results of the reevaluation? .
Responsei The results from the reevaluation show that the original design basis is preserved.
All strvetural components of the building meet the American Concrete Institute code allowables for the loads and load combinations specified in the UFSAR for design of Category I structures.
The UFSAR specifies these requirements in Table 3.8.4-1.
- 8. What changed in the thermohydraulic calculations from the Justificadon for Continued Operation to the final analysis used to determine the subcompartnent pressure loads?
Response: The attached table is a summary of what changed in the thermal-hydraulic calculations in going from the Justification for Continued Operation to the final analysis used to determine the subcomparttnent pressure loads.
= ,seea N-N=6-
,APR-22-1999 09:56 NUCLEAR LICENSING 512 972 029 U /14 Attachmant 3p A C NOC AE 000512 File No.: 025
. [ Page 6 of 6 j TABLE 1 DIFFERENCES BETWEEN l
JUSTIFICATION FOR CONnNUED OPERATION 93-0004 AND FINAL ANALYSIS l
- JCO 93-0004 Mnal Analysis The pessure response versus thne for the Isolation Valve ne mass and energy release for the revssed gessure Cubical (IVC) and Reactor Containment Building (RCB) response analysis combined the 100% and 4% quality was calculated using separate mass and energy releases steam into one bounding mass and energy release rate.
1 assam ng 100% quality steam and 4% quality steam. De bounding mass and enury release rate is paud of the highest energy from the 100% and 4% releases for each time step considered.
1 For both the IVC and the RCB models, the nodalization For both the IVC and the RCB models, the nodahzatica parameters used (volume, junction areas, etc.) were the parameters were revised to mcre accurately simulate the same as those in the original calculation of record, ne actual subcompartments and still povide conservative parameters in the original calculation were developed results. De nadsbation parameters were revised based using plant drawings. on drawings and plant walkdowns. Plant walkdowns identified minor plant features not readily apparent on the j plant drawings used in the criginal analysis.
2 For se IVC model, flow areas of three junaions were decreased (between 12% and 37%), and one junction was j
mereased by -26%. Minor changes were made to the -
m
' ertia lengts and friction factors.
For the RCB model, flow areas of four jtmetions were decreased (between 24% and 63%). Minor changes were made to the inertia lengths and friction factors.
IVC Model- IVC Model:
De Controlled Release Roofing Panels were modeled to The Controlled Release Roofing Panels were modeled to 1 blow off at a pW~=~ ed setpost beed on hand blow off dynamically using an in-house developed code
=
I calenlations. De hand calculations took into accou't the bued on the guidelines given in AppendixE of design blow-off pressure plus extra delay time to account ANSUANS 56.101982. He code (Stect. Travel) and the for the weight of the panels and the time to clear the IVC methodology used are diu-d in UPSAR Section roof to obtain the full vent area to the atmosphere. 3.6.A.7. The code has been verified and documented 3 De Pands wen inodeled to remain on the roof for a under the station software quality assurance gogram.
period of time following the set pessure. The panel blow-off delay time was conservatively estimated using hand i calculations based on the guidelines gesented in ANSUANS 56.101982, Appendix E. The delay time was !
ses equal to the time required for the panels to rise to the elevation at whidi the vent area was equal to the full open
]
vent area. At the end of the delay time, the panels were l
modeled to blow offinstantly. '
p P
]
1
APR-23-1999 09:56 NUCLEAR LICENSING 512 972 8298 P.09/14 i
.o -
i I
F)f9 L Lie M u3 9IJ ;
a l
Attachment A {
a Calculation of Dynamic Load Factor f 1
i
[Appendir 7 of Calculation CC-6251, page 14] l
)
i i
i l
l s ![ii i/Tl u
. . - . . . . _ . - . _ _ . . _ _ . , _ _ _ . _ . . . . . ~ . .
APR-22-1999' 09:56 NUCLEAR LICENSING. 512 972 8290 P.10/14
' - e7 h J0 dua p 1 lpJ
\
Attachment A -
l 1
. l WWM% $
- e c a m -o a s.e tan s.serasa.:c a m. .y tf
"; IIIII TLF = 19F u Y -
\\
~
/!
1" *
E u
\
T
~
\ / X ~ .-
V \/ NCsu <a sa u
~/y U '
IT '*
e
- llll ..
l s e, 4
f-- l g, 4-(
. .. _ - w e
3
\
- f. 5 --
3 3
\ \ * ' -
u 4 g a 4 m
uus. ,-..r a.L.i* w u . ac .s. a-, en..r
.63.+4 t a 6 s..
z.a, >
Consider avw a synametabt triangular pulse which starts at mere and remebes a snazhaum at W the total duration (Hg.2Ac).". In this '
case /Eq. (2.15)'n' asst be applied la three stager, taking for1beJaitis!
eendits .t men stace the samt velostr sad.c-;-
.er h yie.
ading stace. n th= tudentowt.'be used ane as ten ":
Fwy: '$dw%..fx4% k $Nu = A W e4'h y
=
. 3. syp, uqs~ m , %4 t.
j t
J
,a P
.APR-22-1999 09:56 NUCLEAR LICENSING 512 972 8290 P.11/14 1, i 4,@ $. @u Attachment B 1
Example of P(t) Function (Region 7, from UFSAR Fig. 3.6.A 3]
1 I
i t
e
$ uu 4
,meo.4%s=w.,
e APR-22-1999 09:56 NUCLEAR LICENSING 512 972 8290 P.12/14 g .
p p n U nd s Ju .
4 Attachment'B STP: MSLB IN IVC, PRESSURE RESPONSES -
ABSOLUTE PRESSURE IN NODE 7 32 i . . . . .
2, _.......;................:.......4........:.......... ,\
s
/l max =27.74 I 27.7 : .; ; ;
^
ic :
........:........ ..... .. ,.........:......... (m.na).pm
- 2. _.......
W '
cc . . .
o : . : : : : i
$ 23 --------I - ---- ---------------l--------:------'
k 5 i :
w 20 - - * - - - - '
1-------t-------t-------;-------'
~ '
17
.----- .;----4.--------4.-------4.------:-----' .
y . . . r ee.t een ...? . . i . ...t . . .....! , , , ,init . .....
0.001 .01 0.1 1 10 100 1000 TIME (sec)
{r* 0.011 SOUTH TEXAS PROJECT UNITS 1 & 2 PRESSURES IN IVC DUETO MSLB NODE 7 (BREAK NODE 7: CASE 2)
. (Sheet 7of 9)
CD1006781 rigure 3,s A.3 nedston s m t2 0 0 b 0 ! I
APR-23-1999 09:57 NUCLEAR LICENSING 512 972 8298 P.13/14
.f m rp C b IE s
f) u 4
Attachment C 1 Regions of the Isolation Valve Cubide (Appendix 7 of Calculation CC-6251, page 32]
l i
l
.TRi1" s .
1
e APR-22-1999 09:57 NUCLEAR LICENSING 512 972 8290 P.14/14 Attachment C -
SOUTH TEXAS PROJECT CC b 1El - 2. , ,% o, 3'1 JOB NO.14926 i CALCULATION SHEET emoc no. Ne "b t2 susses I tf t' Pl 9 la P'D A A.C tvi v.t 4.o CnQtic msgr no, arv. onicawAtoa oave cuscusa care nev, oniciuAvon cars enscuss oare
(- 3 % T4 . . . ~feit.
W \Q)gt<
I 1
, @ s , - . a p.<
- d. v . es ' a s; w
.. ,p a
=
f'ggq.. .-+ ,.,, ,a.yc.-G5 ~ ""T...' .@
- n. e ,... -= ==. $,, .r' a P
11 I
g.ag g.
12 $g *. s e m, ,
m
~ iT.;'.W -
v .
(= 1 v
-+%.
pu, s,, ., s..
s
. a,. .
1
~
g,r-o -
mj m
m m , _
~,a,4, m
u:c
. ________________.____________ ___ _ _ _ _____ p), , Q 3 ,. ,,
,, c4 s.. .
" 3 ,
=
$4
= l M
M i
=
1l:;19p
" I
! b t 1 i
. zo ,
- m ..
ETF 1310 f7/94 m q l 1 0 TOTAL P.14
- - - - -- -. - . _ .