ML20097D459
| ML20097D459 | |
| Person / Time | |
|---|---|
| Site: | Comanche Peak  |
| Issue date: | 06/05/1992 |
| From: | William Cahill TEXAS UTILITIES ELECTRIC CO. (TU ELECTRIC) |
| To: | NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM) |
| References | |
| TXX-92252, NUDOCS 9206100129 | |
| Download: ML20097D459 (5) | |
Text
i%
Log # TXX-92252 OC File # 903.6 j
Ref. # NRC 68 08
_=
=
7UELECTRIC June 5. 1992 Wuaam J. Cahin, Jr.
c, -,y w rrnure U. S. Nuclear Regulatory Commission l
Attn: Document Control Desk Washington, DC 20555
SUBJECT:
COMANCHE PEAK STEAM ELECTRIC STATION (CPSES)- UNIT 2 DOCKET N05. 50-446 LEAK-BEFORE BREAK OVAliFICATION FOR THE ACCUMULATOR INJECTION LINES REF:
NRC letter dated May 5, 1992 to William J. Cahill, Jr.,
' Request for Additional Information Comanche Peal Unit 2, Accumulator Line Analysis (TAC No M1986)*
Gentlemen:
In response to your request for additional information referenced above, we are providing responses to your request in Attachment 1.
To assure clarity, each request is repeated and then responded to in the attachment.
af there are any questions, please call Mr. Chris E. Jensen at (214) 812-8826.
Sincerely.
William J. Cahill, Jr.
By:
D. R. Woodlan Docket licensing Manager CEJ/ds Enclosure c - Mr. R. D. Martin, Region IV w/o enclosure Resident inspectors, CPSES (2) w/o enclosure Mr. B. E. Holian, NRR w/ enclosure
,. p,, -f
\\ ov11-
- """D*"
P D
00 44/3 F
PDR l{g
i
. Enclosure to TXX-922S2 Page 1 of 4 RE0 VEST FOR ADDITIONAL INFORMATION COMANCHC PEAK VNIT 2, ACCUMULATOR LINE ANALYSIS (TAC NO. M81986)
S AFE SHUTDOWN E ARTHOUAKE (SSE) LOAD H0DELING 1.
Describe in sufficient detail the SSE methodology used for generating fable 4-1.
RESPONSE
Earthquake load Generation for the RCL and Class 1 Piping.
A.
Earthquake Inertia Effects Seismic i"put to the piping system is in the form of amplified response spectra (ARS) at the building floors as provided in specification CPES-S 1032G.
The ARS are applied in each appropriate global coordinate direction and are felt by the system through its supporting media.
When piping models span two structures or different elevations within the same structure, an enveloped ARS curve is developed and applied uniformly to the entire model.
This results in a final set of ARS. one in each global coordinate direction, to be used for analysis.
Analysis is performed using damping values in accordance with Code Case N - iil l.
Code Case N 411 damping was used for all H piping analysis performed for CPSES.
When using the response spectrum method, the effects of three-directional (two horizontal and one vertical) components of earthquake motion are considered.
The coupling effect of closely spaced modes is calculated by SRSS of the modal responses.
Closely spaced modes (within a 10% band of frequency) are accounted for by the modified grouping method.
A particular mode can only be included in one closely spaced grouping.
Groups are formed starting from the lowest frequencies,~(FSAR section 3.7N.2.7).
N4 N R* = b Rj + 2 E
RgR eg3 2
.f.1 j=1 k=My l=k+1
'-1 I
h h
\\2 y~
y egy c 11 +
W Y1 W
\\
k k
1 1 (1- ( px) 2 ) 1/2 wy=wy R
- Total unidirectional response t
l
- _. _ -... _ _,.. ~,,
1 i
Enclosure to TXX 92252 Pag'e 2 of 4
)
2 0k
- Ok *. y t, k
R, Absolute Value of response mode !
N.
Total number of modes hk Damping Ratio b
W[
Frequency t,
30 seconds S
Number of-groups of closely spaced modes M,
Lowest modal number associated with group j of closely spaced modes N,
Highest modal number associated with group j of closely spaced modes 8
Coupling factor k1 i
These formulas differ from Regulatory Guide 1,92 as discussed in Section IA(N) of the FSAR.
High frequency modes (above-33-Hz) are accounted for by the Residual Load Method:
The higher frequency contribution is calculated statically using the residual acceleration vectorl and then combined with the lower frequencies response by SPSS.
The combination of the three uni-('rectional responses is performed by SRSS to arrive at the three directional system response.
- For. all Class 1 piping attached to the Reactor Coolant loop (RCL) or flexible-equipment, the RCL or o'1uipment dynamic model is included in
-the analysis, and the RCL or c' pment support elevation is included in-developing.the ARS.
The seismic RCL model used for combination with the Class 1 auxiliary lines = represents-a simplified linearized model of the RCL piping, equipment and equipment supports which simulates the frequency content
'of the detailed RCL analysis.
Incl. vd in the simplified RCL model is the. effects of. the mainstream and feedwater lines which are represented as stiffness matrices in the model. " Computation for' Rigid Body Effects and Hermccic Excitation with KWUR0HR,' Leinbach, K.
R.,
Lauren. H. and Sterkel, H.
P., Proceedings of Sixth International SHIRT Conferen:e, K9/6, Paris 1981.
- M=
7 w-4s DM eN'ev Y
- -.t gr pyy-y tmi-y-..-w,r gpw T's 19
,my y-wmT'
-~"T'T-U'
"'F*-
9"' "f'T t-
"-*'-"T"5"'T*"re'W'-
-P+'w'"'"
W*"
t
' Enclosure to TXX 92252 Page 3 of 4 B.
Earthquake Displacement Effects There are two categories of displacements, or seismic anchor motions (SAMs), for each direction of earthquake. One category represents the rigid body displacements of the structure that are common to all points on the structure.
The second category represents the defort lon of the i
structure.
Relative displacements within the containment building occur between the containment shell and the interior concrete.
This relative displacement is considered out of phase and is accounted for by static differential-displacement analysis.
For stress evaluation. the three components of seismic anchor movement are analyzed separately and combined absolutely with the inertial load case.
The Design Specification requires that OBE SAM be included in pipe stress evaluation.
For the purpose of conservatism, SSE SAM pipe stress is included in the analysis.
This conservatism can be deleted if margin is required to qualify the system.
Loads-resulting from OBE and SSE anchor movement effects are provided for all containment penetrations and pipe support reactions.
LEAK RATE CALCULATION 1.
Supply L/D _ and f values _ used in the f rictional pressure drop g
-calculation-for node point 2041, which leads to the resulting crack length for 10 gpm leakage reported in the first row of Table 5-1; also provide the corresponding values for pressure losses /iP, and[iPg pertinent to this node point.
RESPONSE
i In reviewing the comp _ter output for the leak rate calculation, it is observed the+ certain intermediate calculational results are not printed.
-However._from the results that are printed.-it_is possible to bound the requested parameters and thus provide a-reasonable numerical-estimate.
For
-node point 2041. the ratio of channel length, L, to the hydraulic diameter, i
O, (L/D ) is reported for crack lengths of 1 - 4 in, and 1 - 5 in.
The g
g following values are reported in the computer output.
For 1 in., L/D
-_100, and leak rate - 7.7 GPM g
For 1 5_ in.. L/D
-__69..and leak _ rate - 18.7_GPM-g
.The crack length cor esponding to a leak rate of 10.0 GPM is obtained by graphic interpolation, and is determined to-be 1 - 4.3 inches.
For t'is crack length. L/Da is not reported directly: however. L/D will be between g
61 and 100 for a leak rate of 10.0 GPM.
The f. R(ion factor, f, used in the l
frictional pressure drop calculation for node 2041 was f - 0.065, and was b i
.'o ta ned from_the_Hoody Chart in the rough region, i ndependent of the Reynolds number.
L.
-- ~~.-
--.--....-,,...m.,...A.-~,.
,_o_
..e...
m..--e...
-.,--mm#.,.-m.--e.wr.r--m r-
-- r
<-r-n rn w-w-
--w,v"--
i
== -
Enclosure to TXX-92252 Page-4 of 4 4
Thefrictionloss,[iP and.thepressurelosscausedbychoking,[iP are r
g.
also not directly available from the computer output.
However, tha pressures at the choking plane for crack lengths 1 - 4 in, and 1 - 5 in, are reported, and are given below:
l For 1 - 4 in., P - 841.1 psi e
for:1
_5 in.. P,692.0 psi
. c Thus, at node 2041 for a leak rate of '.0.0 GPM (which corresponds to a crack length of 4.3 in.), the pressure at the choking plane is between 841.1 psi and 892.0 psi.
2.
Justify L/D - 40 as used in_the submittal.
Studies sponsored by HRC (documented in NUREG/CR 3475 and CR 5128) indicate that flasning occurs at L/D - 12.
ELSPONSE-
-The ratio of L/D above which both choking and frictional ef fects must be y
considered is reported as approximately 40 in paragraph 5.2.1 of WCAP-13167.
This value is consistent with previously submitted ana approved leak before break reports, including Comanche Peak Steam Electric Station Unit 1.
'Although other studies indicate that flashing could occur at L/D ratios as y
low as 12, it should be noted that the calculated L/D ratio at node 2041 is i
y between-69 and 100 (see previnus response).
Thus, since the minimum L/D y ratio of.69 is well above either value (12 or 40), both choking and
~ frictional effects-are properly considered in the leak rate calculations.
3.
Jusuf fy the number of 45' and 90' turns assumed for the flow path through
'the crack.
RESPONSE
The Westinghouse analysis for PWRs assumed that there were-no turns at_all for the flow path through the crack. The 45' and 90' turns alluded to above are appropriate for the winding type intergranular stress corrosion _ cracks found in Boiling Water Reactors (BWR's). therefore not applicable-for CPSES.
~
l E
C
-