ML20198J499
| ML20198J499 | |
| Person / Time | |
|---|---|
| Site: | Oyster Creek |
| Issue date: | 12/22/1998 |
| From: | Roche M GENERAL PUBLIC UTILITIES CORP. |
| To: | NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM) |
| Shared Package | |
| ML20198J502 | List: |
| References | |
| 1940-98-20695, GL-96-06, GL-96-6, NUDOCS 9812300200 | |
| Download: ML20198J499 (11) | |
Text
.-
[n *
.s
{
GPU Nuclear. inc.
(
U.S. Route #9 South orked ve NJ 731 0388 Tel 609-9714000 December 22,1998 1940-98-20695 U.S. Nuclear Regulatory Ccmmission Attention: Document Control Desk Washington,DC 20555 Gentlemen:
Subject:
Oyster Creek Nuclear Generating Station (OCNGS)
Docket No. 50-219 Facility Operating License No. DPR-16 Reouest For Additional Information Concerning Generic Letter 96-06 Pursuant to your letter of September 17,1998 and our discussion with Mr. Ron Eaton on September 28,1998 conceming a revised submittal date, please find attached the requested
/
- information.
If there are any questions or additional information is required, please contact Mr. Joseph D. Lachenmayer of our staff at 973-316-7971.
Very truly yours, bh0 Michael B. Roche Vice President and Director
{j1 Oyster Creek cc:
Administrator, Region I
/
NRC Senior Resident Inspector AOD '/
NRC Project Manager
% 12300200 981222 T
PDR ADOCK 05000219
.B P
PDR Q
D r o d u e 4 K 'c t fu d i t' "n"h"'I
'C;f("\\
v
)
4
.o REQUEST FOR ADDITIONAL INFORMATION FOR RESOLUTION OF GENERIC LETTER (GL) 96-06 ISSUES AT OYSTER CREEK NUCLEAR GENERATING STATION Backaround i
By letter dated June 3,1997, GPU Nuclear stated that it believed the Shutdown Cooling System Supply (inlet) Line Penetration, the Reactor Water Cleanup Supply Line Penetration, and the Recirculation Loop Sampling Line Penetration did not require corrective actions. Our preliminary calculations indicated that these penetrations were not susceptible to overpressure i
failures in that the calculated stresses were within 120% of the allowable stress at the postulated temperatures and pressures. During the course of performing design verification of the calculations it was found that calculated stresses for the Recirculation Loop Sampling Line Penetration do exceed 120% of the allowable stress. As a result, GPU Nuclear will modify the Recirculation Loop Sampling Line Penetration to install overpressure protection. The modification is addressed in the Integrated Schedule pursuant to license condition 2.C.(6) of the Full Term Operating License and is scheduled to be completed in 18R. The pre-existing operability determination that concluded the Recirculation Loop Sampling Line is operable is still applicable.
With respect to the Shutdown Cooling System Supply (inlet) Line Penetration, and the Reactor Water Cleanup Supply Line Penetration, GPU Nuclear has completed design verification of the calculations and has verified that these penetrations were not susceptible to overpressure failures.
The information requested concerning these penetrations is provided below.
Issue i The applicable design criteriafor the piping and isolation valves, and include the required load conditions.
Response
l The design criteria for the piping is based on the original piping design code which is B31.1,1955. Section 123(b)(2) of B31.1,1955 states that either pressure or temperature may exceed nominal design values if the ccmputed stress in the pipe wall does not exceed 20% above the allowable stress values for 1% of the operating period.
Based on these code requirements, the piping acceptance criteria is that the calculated pipe stresses based on postulated accident internal pressures and stresses due to occasional loads are less than 120% of code allowable stresses at the postulated temperatures.
The original valve specification specifies that the valves are to be in accordance with B31.1,1955 Section 124(c). Section 124(c) specifies that all valves shall be to l
Manufactures Standards or equal, for the respective pressure and temperature. The i
original valve specification also specifies that valves are to be designed for internal pressures to MSS SP-66-1964 which documents applicable methodology for determining design pressure. MSS SP-66-1964 indicates that the applicable stress allowable for valve pressure retaining materials shall be to ASME Section 1.
I r
.s Attachment 1940-98-20695 Page 2 of 4 Given the Section 124(c) and MSS SP-66-1964 requirements, the calculated stresses within the valve walls, based on postulated accident intemal pressures, are to be less than 120% of code allowable. Although the criteria ofless than 120% of stress allowable was applied; results for these valves showed that in all cases the calculated stresses were less than the code allowable stress.
Issue 2 A drawing ofthepiping run between the isolation valves. The drawings shouldinclude the length ofthe runs and thickness ofthe piping segments, and the thickness ofthe insulation.
Response
Enclosed are isometric drawings 3E-214-A2-1000, Sheets 1 and 3 concerning the Shutdown Cooling System and 3E-215-A2-1000 Sheet I concerning the Reactor Water Cleanup System.
Please note that the drawings do not indicate insulation thickness. The Reactor Water Cleanup Supply Line is insulated with 2" of Nukon fiberglass blanket insulation inside the drywell. The Shutdown Cooling System Supply Line is insulated with 21/2" inches of Nukon fiberglass blanket insulation inside the drywell.
Issue 3 The maximum calculated temperature andpressurefor thepipe run. Describe, in detail, the method used to calculate these pressure and temperature values. The description should include a discussion ofthe heat transfer model used in the analysis and the basisfor the heat transfer coeficients usedin the analysis.
Response
A description of the analysis and the maximum calculated temperatures and pressures is provided below.
Penetration System-Line Maximum Maximum Temperature Pressure X-10 Reactor Water Cleanup 196 *F 2287 psi Supply Line X-8 Shutdown Cooling System 176 F 1262 rsi Supply (inlet) Line Maximum temperatures and pressures were calculated in two phases. First, the maximum calculated temperature was determined using the Gothic 4.1.c computer code. The model consisted of a volume representing the drywell volume with the subject piping volumes associated with the systems of concern modeled as thermal conductors. The conductors were allowed to cool down and heat up with the volume representing the containment atmosphere.
The conductors were modeled using the actual piping materials, thermal conductivity, line sizes and lengths. The model was then run using containment temperature and pressure profiles based upon the existing mainsteam line break (MSLB) accident. The portions of the system piping inside the drywell were allowed to heat up, while the portions of the' system piping outside the
s s
Attachment 1940-98-20695 Page 3 of 4 drywell were not. No credit was taken for piping insulation because the survivability of the piping insulation during a MSLB cannot be assured. This is a conservative assumption since it maximizes heat transfer from the environment to the fluid in the pipe.
l l
Initial temperature of the fluid inside the piping _ was chosen based on normal operating temperatures. The initial temperatures for the systems, which are only susceptible to the phenomena j
stated in GL 96-06 when they are out of service, were calculated using a GOTHIC model assuming l
a conservative Drywell temperature of 125 'F. The thermal conductors, described above, are allowed to cooldown and exchange heat with the containment atmosphere. The initial temperature l
for the RWCU System was conservatively determined to be 150 F. This was based on a out of
(
service time period of 4 days which is considered to be the maximum time the system may be out L
of service without adverse chemistry trends. Since the Shutdown Cooling System line is typically I
not in service during power operation, its initial temperature was assumed to reflect the Drywell temperature of 125*F.
During the design basis MSLB each piping section of concem is being symmetrically heated. As a L
result, the entire surface area of the pipe is at the same temperature and the heat transfer is in the radial direction. The mode of heat transfer inside the pipe was assumed to be pure conduction. To facilitate this, each piping section was modeled as a solid cylinder (rod) with the divided layers l
designated as metal (stainless steel or carbon steel) or water.
Correlations for free convection inside a horizontal cylinder with uniform heating or surface temperature are difficult to obtain because the process is inherently unsteady. Handbook of Heat Transfer Fundamentals, by Rohsenhow, Hartnett, and Ganic, Second Edition 1985, provides a correlation for quasi-steady convection in an enclosure following a step change in wall temperature.
The quasi-steady regime becomes established after an initial transient in which the heat flow is l
dominated by conduction. The time period required to establish the natural convection made of l
heat transfer is not specified in this reference. To vedfy the peak temperatures calculated ush the l
conduction method described above, an altemate method was applied using guidance from NAl, the l
author of the GOTHIC code, and a correlation from the previous reference, for natural convection inside a cylinder. The Shutdown Cooling System was modeled as a control volume with a constant low pressure (100 psia) boundary condition. This boundary condition is needed because the GOTHIC code cannot be used to calculate the intemal pressure of the volume. The boundary l
condition prevents the pressure from getting too large. This approach will slightly under predict the temperature rise because the water specific heat decreases slightly at high pressure and low l-temperature. It is estimated that the error in temperature rise due to this neglected effect is less than 1.0%. Calculation results showed that results of the two analysis methods were consistent. The conduction method was proven conservative and appropriate. In one particular case, the peak l
temperature using the altemate method was 153.5 F while the peak temperature using the conduction method was 155.2 'F.
The heat transfer model for the exterior surface of the thermal conductors (piping sections) was the
- direct model for convection heat transfer contained in the GOTHIC code. In accordance with the j
Gothic User's Manual, the direct model provides good results when a condensation model is 1
)
.=
Attachment 1940-98-20695 Page 4 of 4 required. The Uchida heat transfer coefficient was used for the condensation heat transfer correlation.
The second phase of the analysis calculated maximum pressures using models developed in the Mathcad Software. Based on the maximum temperatures identified by the Gothic modeling, the Mathcad Software was used to solve equations for fluid response and the piping stress-strain relationships, it was determined that the penetrations are not susceptible to overpressure failure when fluid pressure, based on the thermodynamic properties of the fluid, equaled the strain required to satisfy the thermodynamic pressure. Fluid propedies were determined using software libraries of ASME steam table information. The final values obtained were verified by comparison with the ASME steam tables. Stress vs. Strain relationships were determined based on the material propedies of the existing piping and using thin walled cylinder equations, assuming constant stress across the pipe thickness.
9 e
O
1 1
\\
OVERSIZE
?
DOCUMENT'^
l PAGE(S) PULLED i
l l
SEE APERTURE CARD-FILES '
i i
l i
APERTURE CARD / PAPER COPY AVAILABLE THROUGli NRC FILE C
............................................................. ENTER i
.....................................................(S)
.h NUMBER OF OVERSIZE PAGES FILMED ON APERTURE CARD i
ACCESSION NUMBERS OF OVERSIZE PAGES:
1 0N b0
$9W620//>f_
990/0 70/(o (o ga DWed
_-