ML20137K162
| ML20137K162 | |
| Person / Time | |
|---|---|
| Site: | Byron, Braidwood  |
| Issue date: | 03/31/1997 |
| From: | Hosmer J COMMONWEALTH EDISON CO. |
| To: | NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM) |
| References | |
| NUDOCS 9704040256 | |
| Download: ML20137K162 (9) | |
Text
, Com.monwcahh Edison Campany 140) Opus Place Downers Grove, IL G61'>-5*01 March 31,1997 U. S. Nuclear Regulatory Commission Washington, D.C. 20555 Attention:
Document Control Desk
Subject:
Braidwood Nuclear Power Station Unit i NRC Docket Numbers: 50-454 Byron Nuclear Power Station Unit 1 NRC Docket Numbers: 50-456 Additional Information Regarding 3 Volt Interim Plugging Criteria Technical Specification Amendment Renewal Request
References:
- 1. Teleconference dated March 25,1997, between the Commonwealth Edison Company and the Nuclear Regulatory Commission Regarding the 3 Volt Interim Plugging Criteria Technical Specification Amendment Renewal Request
- 2. J. Ilosmer letter to the Nuclear Regulatory Commission dated March 13,1997, transmitting Response to Request for Additional Information During the Referenced teleconference, the Commonwealth Edison Company (Comed) and the Nuclear Regulatory Commission (NRC) discussed Comed's response to the Staff s r.xluest for Additional Information (RAI) that was transmitted in Reference 2. Spccihily discussed was Comed's response to RAI 12 which dealt with the potential effect of the multidimensional flow pattern of the position dependent pressure drop across the tube support plates in the Byron Unit I and Braidwood Unit I steam generators.
The attached response supercedes Comed's previous response to RAI 12 contained in Reference 2 and provides clarification for areas discussed during the subject teleconference.
If you have any questions concerning this correspondence, please contact this office.
Sincerely jff //H/ /
John B. Ilosmer Engineering Vice President Attachment
'I ce; D. Lynch, Senior Project Manager-NRR j
G. Dick, Byron /Braidwood Project Manager-NRR L
S. Burgess, Senior Resident inspector-Byron C. Phillips, Senior Resident Inspector-Braidwood A. B. Beach, Regional Adtninistrator-Rill Office of Nuclear Sqf,c{yjDNS g
9704040256 970331 PDR ADOCK 05000454 P
PDR Lnk byrbwd.stmgenn.3nrai3 doc A linkum Company a
12.
During the SG blowdown following a postulated main steam line break (MSLB),
the pressure drop across a TSP will be determined by the position dependent flow distribution across a TSP. Because of the differences in resistances between fluid flows parallel and perpendicular to the SG tube bundle above the topmost TSP, a multidimensional flow pattern exists. This effect will be most pronounced at the j
upper TSP. Accordingly, assess the effect of the multidimensional flow pattern on the position dependent pressure drop across the TSPs.
Response
The original analysis which was used to determine tube suppon plate loads using RELAP5M3, assumed the area around the U-Bend tubes above the top TSP to be one volume. The maximum pressure differential across the top TSP was calculated to be 2.44 psi and uniformly applied. The loss coeflicient across the U-Bend region (due to crossflow i
resistance) was added to the loss coefficient at the separator inlet. If a vanf ng loss coefficient due to the U-Bend region is considered, a multi-Dimensional flow pattern is developed during MSLB. This multi-Dimensional flow pattern will cause a non-uniform pressure distribution across the TSP.
Comed has developed a method to evaluate the effects of this multi-Dimensional pressure distribution. The method involves developing a 2-Dimensional pressure distribution across the TSP as a function of TSP radius. This was accomplished by developing four 1-dimensional flow streams modeled from the N support plate to the separator, see figure 1.
Prior to the MSLB, the fluid in the tube bundle area adjacent to the TSP is single phase.
After the MSLB begins, the liquid is subjected to decompression and acceleration forces.
The flow rate and pressure drop across the TSP can be determined by drawing a control volume around the fluid regions and solving the Bernoulli integral equation which accounts for inertial and viscous effects. The volume around the TSP was divided into four control volume flow streams. These parallel flow streams were located at widening radii from the center of the bundle. Each of these control volume streams represent a venical flow path from the N support plate to the separator. The only differences in these parallel flow streams are the bundle loss coeflicients Hydraulic losses associated with wall friction have not been included in any of the flow stream calculations. The control volume flow stream at the center of the TSP had the largest total loss coefficient due to the proximity of the U-Bend region, while the control volume stream at the outside of the tube bundle had the smallest loss coeflicient. By calculating the pressure differential across the TSP for each of these flow streams a 2-Dimensional representation in the form of a curve fit of differential pressure multiplier vs radius was created. The resulting equation was DPMult=A+Br, where A=0.87 and B=0.0041/ inch up to the point where the tube bundle edge is reached. A constant value is utilized from the bundle edge, in the non-tubed regions.
Applying this equation to concentric regions emanating from the center was performed for the following radii; 13,26, 39,52, and 61.375 inches, for use in the structural model, see table 1. Figure 2 illustrates both the hydraulically analyzed points as well as the values
utilized in the structural evaluation. Based on this table it can be see.i that the originally assumed TSP pressure differential of 2.44 psi actually decreases to 2.1 psi at the center of the bundle and increases to 2.64 at the outer edge of the bundle. This represents a 0.53 psi change from the center of the bundle to the outside edge of the bundle due to multi-dimensional influences of the tube bundle.
The method described above for determining multi-dimensional pressure is judged to be conservative based on the following:
No cross flow was assumed between the parallel flow streams. This has the effect of maximizing the pressure differences between the flow steams. Cross flow between the flow streams would cause a more uniform pressure distribution across the TSP No pressure drops or fluid inenial terms other than the separator entrance and the N and P TSPs were included. In reality, other significant pressure losses and inertia exist and serve to limit the transient velocity across the TSP more than taken credit for above.
Using the new loads listed in table 1, an analysis was performed by Westinghouse to determine their efTect during MSLB. This new analysis used the same support assumptions as the original analysis, including the extremely conservative assumption that the wedges do not provide any vertical support to the plates. Edge support is provided only by the vertical bars welded to the wrapper and / or partition plate. The new radial pressure distribution is applied only to the top plate, Plate P. The change in maximum displacement for the lower plates (less than 1%), is the result ofinte.raction between the various plates.
Table 2 lists the maximum displacement values per plate along with the original analysis displacements which assumed uniform loading. Table 2 indicates that there is a slight increase of 4.8 mils at the top "P" TSP, this represents a maximum of a 5% increase in displacement. The location of maximum displacement for the revised pressure loads is the same as the original analysis, Figure 3 and Figure 4 provide a graphical representation of the TSP displacements with the uniform pressure loading and radially distributed pressure loading respectively. These figures show that TSP displacement geometry for the two cases is essentially the same.
The design criteria of maintaining TSP displacement below 0.100 inch is maintained for all plates. Note that Plate A is the flow distribution bafile and that no IPC criteria is applied at this location.
Conclusion When the efTect of multi-dimensional flow is considered on TSP displacement, usmg j
conservative assumptions, the total TSP displacement is less than 5% and remains within the 0.100 inch design criteria.
i
l l
r l
FIGURE 1 CONTROL VOLUME DIAGRAM i
i f
i f
Separatorinlet
?
A=22.01 K=13.7 i
~
M55.45 I
l 1
~ ControlVolume/ Path I
i I
P TSP A=17 K=1.08 l
DP applicc; i
l i
1 j
A=55.45 I
1 N TSP A=17 K=1.08 I
l
_. ~
FIGURE 2 Differential Pressure vs Radiai Displacement 1.1 r-0 1.05
.5 E
=
e g
i 1
m E
L Ee I
5o 0.95 I
.!!1 E
i
+W DP Multiplier Used 0.9
- Hyd Load Calc Points i
I 0.85 O
10 20 30 40 50 60 70 I
Distance from Plate Center -inch
+
a
-,A-6.
e FIGURE 3 UNIFORM PRESSURE DISTRIBUTION MP 95>
DISPLACED GEOMETRY TIME = 0.3110 (TIME OF MAXIMUM DISPLACEMENT)
FIGURE 4 RADIAL PRESSURE DISTRIBUTION a
t e
95>
9P DISPLACED GEOMETRY TBIE = 0.3110 (TIME OF MAXIMUM DISPLACEMENT)
TABLE 1 l
l
)
i J
\\
1 1
i f
Pressure Distribution as Based on Radius for Plate P Hydraulic Analysis Structural Analysis 1
i Distance Distance from From Plate Hyd Load DP Multiplier 4
Plate Center 1
Center Calc Points Used (Inches) 4 l
(Inches) i 0
0.87 0
0.87 l
13 0.923 16.8 0.94 26 0.977 33.96 1.01 39 1.03 51 1.083 j
52 1.083 i
61.375 1.083 1
4 i
i i
i j
i g
-. _... - -. - = -..
t t
9*
. ~
Table 2 Summary of Mmwipmm Renta-tm ModelD4 Steam Ge==r=i=r l
RELAP Loads I
LoadFantar= L3 l
1 MM Y 88'8 8 8 f
RadialPmanewDistreadasfbrPisdeP i
1 j
MPasen
. pu n..u _
pm y1 A
0.1308 0.1315 C
0.0588 0.0571 F
0.0580 e.uose J
0.0745 emes _
f L*
OM75 0.0777 at o.os4s o.osso i
N 0.0848 0.0058 aname P
D.C""'
4 i
9 O
1 i
4 I
4 0
I e
4
..._.-.m..
-