NOC-AE-13003007, RETRAN-02 MOD005.3 Code Version
| ML13165A028 | |
| Person / Time | |
|---|---|
| Site: | South Texas  |
| Issue date: | 06/04/2013 |
| From: | Dunn R South Texas |
| To: | Christopher Jackson Office of Nuclear Reactor Regulation |
| References | |
| NOC-AE-13003007, STI: 33702432 | |
| Download: ML13165A028 (11) | |
Text
Nuclear Operating Company South Teas Pro/ect Electric Generating Statlon PO. Bar 289 Wdsworth, Texas 77483 -A- -
June 4, 2013 NOC-AE-13003007 STI: 33702432 Attention: Mr. Christopher P. Jackson Reactor Systems Branch, NRR U.S. Nuclear Regulatory Commission 11555 Rockville Pike Rockville, MD 20852 South Texas Project Units 1 and 2 RETRAN-02 MOD005.3 Code Version.
The attachment to this letter provides notification to the NRC regarding the release of RETRAN-02 MOD005.3 Code Version.
There are no commitments in this letter.
If you should have any questions on this matter, please contact Mr. Charles Albury at (361) 972-8901.
Roland Dunn Manager, Nuclear Fuel &
Analysis Engineering MK/
Attachment:
RETRAN-02 MOD005.3 Code Version Notification of Code Release
01 NOC-AE-1 3003007 Page 2 of 2 cc:
(paper copy) (electronic copy)
Regional Administrator, Region IV A. H. Gutterman, Esquire U. S. Nuclear Regulatory Commission Morgan, Lewis & Bockius LLP 1600 East Lamar Boulevard Arlington, TX 76011-4511 Balwant K. Singal U. S. Nuclear Regulatory Commission Balwant K. Singal John Ragan Senior Project Manager Chris O'Hara U.S. Nuclear Regulatory Commission Jim von Suskil One White Flint North (MS 8 B1) NRG South Texas LP 11555 Rockville Pike Rockville, MD 20852 Kevin Polio Senior Resident Inspector Richard Pena U. S. Nuclear Regulatory Commission City Public Service P. 0. Box 289, Mail Code: MN1 16 Wadsworth, TX 77483 C. M. Canady Peter Nemeth City of Austin Crain Caton & James, P.C.
Electric Utility Department 721 Barton Springs Road C. Mele Austin, TX 78704 City of Austin Document Control Desk Richard A. Ratliff U. S. Nuclear Regulatory Commission Texas Department of State Health Washington, DC 20555-0001 Services Robert Free Texas Department of State Health Services
NOC-AE-1 3003007 Attachment Page 1 of 9 May 30, 2013 Mr. Christopher P. Jackson Reactor Systems Branch, NRR U.S. Nuclear Regulatory Commission 11555 Rockville Pike Rockville, MD 20852
Subject:
RETRAN-02 MOD005.3 Code Version Notification of Code Release
Dear Mr. Jackson:
The purpose of this letter is to inform the NRC that the RETRAN-02 MOD005.3 code version has been released by the RETRAN User Group for use by the industry. RETRAN-02 is a non-LOCA system thermal-hydraulic transient analysis code that is widely used by licensees in the U.S. and also internationally. The Electric Power Research Institute originally developed RETRAN, but has authorized the RETRAN User Group to provide ongoing maintenance and support.
Based on previous discussion with the NRR described in a November 24, 1997 letter to Mr.
Timothy E. Collins from Mr. Gregg Swindlehurst of Duke Power Company who was the Chairman of the RETRAN Maintenance Group, NRC review and approval of the RETRAN-02 MOD005.2 code version was not required since no new models were included. Changes included in MOD005.2 contained only error corrections that were developed under an Appendix B QA program. Similarly, changes made. in preparation of the MOD005.3 code version contain only error corrections (12) and no new models, and were also done under an Appendix B QA program. Therefore, the NRC is hereby notified that RETRAN-02 MOD005.3 has been released for general use by the industry, consistent with the limitations of the NRC's prior safety evaluation report issued for RETRAN-02 MOD005.0 and subsequently extended to MOD005.1 and MOD005.2.
For your information, the attachment summarizes each error corrected, the associated code modification, and the affect each modification had on the results for a. set of standard sample problems. The testing procedure and standard sample problems that were used for the MOD005.2 code version were also used to test the MOD005.3 code version. They are discussed in the attachment to the November 24, 1997 letter noted previously.
Should you or your staff require any additional information, please do not hesitate to contact me at (361) 972-8901. Thanks for your attention to this matter.
Charles Albury Chairman - RETRAN User Group STP Nuclear Operating Co.
P. 0. Box 289, M/S N3010 Wadsworth, TX 77483
NOC-AE-1 3003007 Attachment Page 2 of 9 Attachment RETRAN-02 MOD005.3 Error Correction and Testing Summary RETRAN-02 MOD005.3 contains 12 modifications that were made to RETRAN-02 MOD005.2.1 to correct identified errors. Table I summarizes error reports and modifications made to resolve the errors. It should be noted that RETRAN-02 MOD005.2.1 is the MOD005.2 code version with code modifications required to correct syntax errors trapped by newer compilers and modifications required to transport the code to newer operating systems. No model error corrections or enhancements were included in RETRAN-02 MOD005.2.1. It is functionally equivalent to MOD005.2.
The modifications summarized in Table I were added sequentially to MOD005.2.1. Testing was performed after each modification was added and consisted of running the 10 standard sample problems and then comparing the results with those from RETRAN-02 MOD005.2.1. The comparisons were made by use of an automated program COMPAR, which compares auxiliary output files containing time histories of key output variables for each problem. Any differences between the revised code and the MOD005.2 results that exceeded 0.1% were reported. The number of differences reported for each sample problem are shown in Table I1.They provide a macro-level view of how the modifications affect the standard sample problems, which use-typical RETRAN-02 modeling and options.
Table III gives more detail about the magnitude of the observed differences between the MOD005.2.1 and MOD005.3 results. The modification dependent differences ere consistent with the error correction and were expected. None of the corrections resulted in significant differences in results for the standard sample problems.
NOC-AE-1 3003007 Attachment Page 3 of 9 Table I RETRAN-02 MOD005.2 Trouble Report and Associated Modification Summary Trouble Report Modification Number Trouble Report Description Number Modification Description 431 Failure in junction properties occurs for a negative fill 406 The slip convection term for the negative fill was not junction. The flow is zero and the valve that is closed, being set to zero after the valve at the fill junction was closed and the flow stopped. The modification sets the slip convection term to zero when the valve is closed.
436 In subroutine QDOT, the Prandtl number is discontinuous for 405 A correction has been developed for subroutines htrc and wall temperatures exceeding 1500 degrees F. In the qdot based on similar changes to RETRAN-3D. In the Bennett problem, this resulted in conduction solution original c6ding, the Prandtl number was set to a constant converge warning messages to be printed during steady- if the wall temperature was greater than 1500 F since state initialization. 1500 F represents the upper limit for the water properties: however, the value was not consistent with the computed value at 1500 F, introducing a discontinuous transition in the Prandtl number. The limiting Prandtl number value is now computed based on properties evaluated at a temperature of 1500 F.
438 A failure in restart is encountered when using the 407 The modification resets the values for the IPTN flags for temperature transport delay option. restart cases by checking the PHASE flag for the volume as well as the restart flag LDMP on the problem dimension card and if these conditions are both true, the IPTN flag is changed to the negative of its absolute value. This turns temperature transport off for the two-phase volume.
439 If the actinide breeding ratio is not supplied on the 140000 408 Subroutine INRKEN was modified to set the value of data card, the code defaults to a value of 0.0. This is not UDUF to 1.0 if it is the default case (UDUF=0.0) as what is discussed in the RETRAN input manual. stated in the RETRAN-02 User's Manual, EPRI NP-1850-CCM-A, Volume 3 Revision 6.
440 When a system turbine trip calculation is exercised, error 414 The coding for calculating the mixture velocity and messages relating to junction property evaluation are resulting kinetic energy was revised to insure that the observed. Review of the information indicates a problem in junction flow rate and area are at consistent time levels.
the calculation of the kinetic energy at junctions that have This prevents unreasonable kinetic energy values from time-dependent flow areas. resulting in out or range enthalpies.
443 While reviewing the RETRAN-3D pressurizer model and its 411 The sign of two derivative terms in term DAMP were performance relative to RETRAN-02, an error was corrected in subroutine PRZR.
NOC-AE-1 3003007 Attachment Page 4 of 9 Trouble Report Modification Number Trouble Report Description Number Modification Description discovered in the formulation of the liquid region work term.
This error also exists in RETRAN-02. In equations VIII.5-29 and VIII5.-30 (EPRI-1850-CCM, Vol. 1 Rev. 6) the partial derivatives for the liquid region volume with respect to vapor and liquid energy are transposed incorrectly. The coding in the subroutine PRZR is also incorrect.
445 The behavior of the general transport system in some large 409 The changes in concentration reflect the explicit nature of analysis models was inconsistent with expected behavior, the generalized transport model. A more implicit The sample deck illustrates that the impurity (e.g., boron) formulation was achieved by moving the generalized concentration of an accumulator varies significantly during transport calculation (subroutine GENTRN) after the blowdown. The expected behavior would be to have a material properties have been updated to new time-step constant boron concentration in the accumulator during values (as in RETRAN-3D) in subroutine TRAN. Also, for blowdown. Changes in boron concentration were noted reported case with the extreme conditions, the transport when flow rate or time-step size changes significantly, and medium mass ratio (Mm previous/ Mm current) may be the potential exists when the implicit convergence criteria are far from 1.0 By including the transport mass ratio term, not satisfied. the transport model can better handle these extreme condition cases; however, it is still possible to have such extreme conditions that the explicit formulation for the generalized transport model will exhibit numerical errors.
Code wa's added to GENTRN to include the transport medium mass ratio. FTB File ID71 was modified to save the previous time-step transport medium mass for each volume. Logic was added to SAVIMP, RESET and EDATA4 to insure these new values are reset properly for the iterative numerical solution and saved to allow RESTART.
450 The vector momentum option contains an error for models 410 Subroutine injun was modified to write an error message using junction angles that are not 0, 90, 180, or 270 degrees. and terminated code execution if angles other than 0, 90, Use of any of the above angles in any combination will not 180, or 270 are provided by the user.
encounter the error. This error applies to all versions of RETRAN-02 prior to and including MOD005.2 and RETRAN-3D MOD003 and prior versions.
451 At 900 seconds, the vessel head (Volume 10) is all vapor yet 413 The test on low flow conditions to allow condensation in the heat transfer mode for Conductor 23, the metal mass in subroutine htrc was removed and replaced with a test on Volume 10, incorrectly indicates forced convection to liquid. ihtmap=0, which indicates the forced convection only
NOC-AE-1 3003007 Attachment Page 5 of 9 Trouble Report Modification Number Trouble Report Description Number Modification Description map is used. If the combination heat transfer map is selected (ihtmap#0) and if twall < tsat then condensation heat transfer is used.
453 While evaluating the compiler upgrade on a Sun platform, 415 A modification was made in WAT9 to test that the relative this case showed a floating point exception (divide by zero.) volume is greater than zero as well as smaller than the The transient finishes without a code failure, and the results minimum value. A modification was also made so that if appear reasonable, but the code reports that a single divide the assumed state in region 1 is two-phase, the by zero has occurred. calculated enthalpy is forced toward a two-phase value.
This eliminates the inconsistency as the region transitions from barely two-phase to single phase liquid.
456 A bad junction enthalpy causes the code to fail in the critical 414 The coding for. calculating the mixture velocity and flow model. The error is the result of small valve areas that resulting kinetic energy was revised to insure that the result in very large kinetic energy terms that lead to bad junction flow rate and area are at consistent time levels.
junction enthalpies. This prevents unreasonable kinetic energy values from resulting in out or range enthalpies.
457 The junction enthalpy exiting hot leg volumes 101, 102, 103 417 The convective term in subroutine TRNSPT (temperature demonstrate a peculiar (but small) change in the junction transport delay time model) used variables HSPIN and enthalpies when the temperature transport delay model is HSPOUT. These variables include the junction enthalpy active, and an enthalpy effect for the elevation change between the center of the upstream and downstream volumes, respectively. They were replaced with the junction enthalpy, which corrects the reported problem.
458 An error in the decay heat model can result in under 419 The prompt power is now calculated as the product of the prediction of power for models using 1-D kinetics (generally sum of the yield fractions (the steady-state prompt power BWRs). fraction) and the amplitude as was implemented in RETRAN-3D mod_368.
460 The pressurizer liquid region is not re-established after the 418 Subroutine WAT8 contained logic that prevented the level goes to 0, followed by an in-surge, liquid region from re-establishing unless a minimum Iiquid mass appears during a single time step. With small time -step sizes, the amount of liquid mass that can be rained out of the vapor region is proportionately smaller, causing the minimum mass test to fall. When this occurs, a single region pressure search is performed and the liquid region does not appear. Removing the quality limits from WAT8 resolved the problem.
When the liquid region is re-established, some of the
NOC-AE-1 3003007 Attachment Page 6 of 9 Trouble Report Modification Number Trouble Report Description Number Modification Description partial derivatives used in subroutine PRZR to determine the liquid volume change (DVOL) are zero (not defined) because two regions did not previously exist. The liquid volume change is computed using the product of the liquid mass and saturated liquid specific volume. A similar calculation was added for the situation where the vapor region re-establishes (taken from RETRAN-3D).
NOC-AE-13003007 Attachment Page 7 of 9 Table II Summary of RETRAN-02 MOD005.3 Modification Differences Standard Sample Problem Differences Modification 1 2 3 4 5 6 7 8 9 10 mod spI vo18 sp5 tita ttwob urcw fl2d turb ttqxl atws 405 0 0 0 0 0 6 0 0 0 0 406 0 0 0 0 0 6 0 0 0 0 407 0 0 0 0 0 6 0 0 0 0 408 0 0 0 0 0 6 0 0 0 0 409 0 0 0 0 0 6 0 0 0 0 410 0 0 0 0 0 6 0 0 0 0 411 0 0 0 0 11 24 0 0 0 61 413 0 0 0 0 11 24 0 102 0 61 414 12 314 4 7 1138 97 0 358 144 66 415 12 314 4 7 1138 97 0 358 144 66 417 12 314 4 7 1138 96 0 358 144 66 418 12 314 4 7 1137 96 0 358 144 78 419 12 314 4 7 1137 96 0 358 169 78
NOC-AE-1 3003007 Attachment Page 8 of 9 Table III Summary of Differences Observed by Modification Modification Sample Number Problem Observed Differences 405 ucrw No effect.
The differences reported in Table II were due to a change in the compiler version used and not the modification itself.
The differences are indistinguishable when both results are plotted on a single chart.
406 all No effect.
407 all No effect.
408 all No effect.
409 all No effect 410 all No effect.
411 Differences in results were expected as a result of this code correction.
ttwob One flow value at the steam line entrance was out of tolerance prior to the peak power at -0.8 seconds.
It is 1 volume removed from the two-region nonequilibrium volume that will be affected by the modification. All other differences were after the scram with the largest error being in the void reactivity at 0.45%, which has no effect on the long term power.
ucrw -The surge line flow was the only parameter that was out of tolerance. During one time step it differed by 0.9% and the rest of the time it differed by 0.2% to 0.45%.
atws No parameters are out of tolerance for the first 88 seconds. Between 88 and 100 seconds there are minor differences in the secondary side mixture level, relief valve flows and SG heat transfer, but they are generally less than 0.9%. These differences go away between 100 and 140 seconds. After 140 seconds some minor difference in the SG mixture reappear, but the level is essentially zero.
At 90 seconds, the pressurizer pressure differs by 1.1% (expected), but returns within tolerance after 93 seconds.
412 Not included in .MODO05.3 - deemed to be a model change.
413 Differences in results were expected as a result Of this code correction.
turb All differences are in the feedwater heater where the pressure differs by a constant 0.4%. No other parameters are out of tolerance.
NOC-AE-1 3003007 Attachment Page 9 of 9 Modification Sample Number Problem Observed Differences 414 The correction will affect the enthalpy for any junction containing a valve that opens or closes.
spl Prior to 0.4 seconds there are no significant differences (marginally greater than the tolerance).
Between 0.4 seconds and 0.46 seconds when t the problem terminates, the pressure and flow at the center of the pipe segment differ by up to 0.3%.
vol8 When the break flow initiates, it differs by up to 1.5% for the first few time steps, but there after returns to tolerance. The pump exit flow differs most of the transient by up to 7% for short periods of time but returns to differences less than 1%.
- sp5 The differences are minimal on only marginally greater than the tolerance.
tlta The differences are minimal on only marginally greater than the tolerance during a very short period near the beginning of the transient.
ttwob There are many parameters out of tolerance for the duration of the transient. Overlay plots of the RETRAN-02 MOD005.2 and MOD005.3 results illustrate that they results are in very good agreement Most parameters show two lines that are seldom more than a line width apart except for the void and Doppler reactivities following the scram when their contributions are insignificant.
ucrw The pressurizer pressure differs by 0.7 psia at the end of the transient and the secondary side pressure is approximately 1 psi lower (0.1% difference). The surge line flow differences are similar to those introduced by modification 411.
turb The condenser pressure differs by up to 0.65% and the feedwater heater pressure differs by up to 0.2%.
The high pressure turbine extraction flow differs by up to 5%.
ttqxl The steam line flow differs up to 5% for one time point but is typically less than 0.45%. The power differs by 0.33% at the peak and returns within tolerance shortly thereafter.
atws Only slight changes from modification 411 results. Very similar.
415 all No effect.
416 Not included in MOD005.3. Modification already included when installing MOD005.2 on Windows with updated Intel compiler. Resulting code version is identified as MOD005.2.1.
417 all No effect.
418 It was expected that this modification could affect models with pressurizers.
atws Only slight changes from results for modifications 411 and 414.
419 As expected, this modification only affects decay heat for the 1-D kinetics model.
ttqxl The power differs by 0.33% at the peak (same as modification 414) but increases to 0.4% at the end of the transient (1 second). This slight difference at the end of the transient is due to the error correction.