ML20138E347
Text
{{#Wiki_filter:. ENCLOSURE 2 lj ITS/NRC/93-10 b TECHNICAL EVALUATION: RETRAN Model Qualification Decrease in Heat Removal by the Secondary System NTH-TR-01 Florida Power and Light Comany Turkey Point and St. Lucie Stations I i Prepared for U.S. Nuclear Regulatory Commission Washington, D.C. 20555 Under NRC Contract No. NRC-03-90-027 FIN No. L1318 m \\ --,-,m w International Technical Services, Inc. 420 Lexington Avenue New York, NY 10170 j3-{2-17 S yo8023b
- munr
ITS/NRC/93-10 TECHNICAL EVALUATION: RETRAN MODEL OVALIFICATIO_M DECREASE IN HEAT REMOVAL BY THE SECONDARY SYSTEM NTH-TR-01 FOR THE FLORIDA POWER & LIGHT COMPANY TURKEY POINT AND ST LUCIE STATIONS
1.0 INTRODUCTION
NTH-TR-01, dated July 1989 (Ref. 1), documents results of a series of thermal-hydraulic transient analyses performed by Florida Power & Light Company (FPL) to obtain NRC approval for their use of RETRAN-02 M00004, an NRC reviewed and approved code (Ref. 2), in licensing actions associated with transients within one category of transient events: Decrease in Heat Removal by the Secondary System. Additional information was provided in References 3, 4, 5 and 6. FPL submitted the subject topical report in response to the Safety Evaluation Report (SER) issued by the NRC on the previous FPL submittal " Florida Power & Light Company - Topical Report on RETRAN" (Ref. 7) for its use of RETRAN-02 for St. Lucie Units I and 2 (Combustion Engineering plants) and Turkey Point Units 3 & 4 (Westinghouse 3-loop plants). The staff stated in that SER that further qualification of FPL's RETRAN model was necessary before it can be used in support of licensing action. In particular, the staff required that FPL (1) justify plant nodalization, (2) compare the computed prediction with plant operating data, (3) perform sensitivity studies to demonstrate user specified parameters are adequately conservative and (4) justify licensing assumptions. Furthermore, underlying all the stated conditions is the fundamental requirement implicit in NRC Generic Letter 83-11 (Ref. 8) that FPL demonstrate its thorough technical comprehension of the RETRAN computer code and its modeling. Therefore, our review focused upon examination of the degree of thoroughness and accuracy with which FPL explained its use of the code and the code options, as well as evaluation of the adequacy of FPL's models and submittals with respect to each of the aforesaid requirements for the subject category of transients. 2.0 REPORT
SUMMARY
FPL presented comparison of RETRAN analysis to plant operational transients in order to demonstrate the adequacy of the base plant model for both Turkey Point and St. Lucie plants. The limited justifications presented by FPL for its plant nodalization, input selection and selection of particular correlations built into the code were described but no supporting plots or analysis were presented. FPL developed one Turkey Point base model for both units while separate base 1 1
i models were developed for each of St. Lucie units, each consisting of a two-loop plant model with an optional steam generator multi-node nodalization. Descriptions of the plant nodalizations and models selected for use in the analysis were presented in Chapter 1 of the topical report. Two sets of benchmark analyses against operational transient data in the category of transients at issue were presented. For Turkey Point benchmark analysis, a loss of inverter event was analyzed, and for St. Lucie benchmark analysis a partial loss of feedwater event was analyzed. Specifics of modeling features which may impact the analysis were reviewed and evaluated. 3.0 EVALUATION Adequacy of FPL's application of the RETRAN computer code for thermal-hydraulic calculations of the transient behavior of its Turkey Point and St. Lucie nuclear plants is discussed below. In order to demonstrate the adequacy of these plant models in performing l plant transient analysis in this particular category, FPL compared code results with plant operational transient data from its Turkey Point loss of inverter event and St Lucie Unit 1 partial loss of feedwater. In support of licensing assumptions, FPL presented comparisons of the global results with those of the current FSAR analyses for the same set of eight transients in the category of interest for each of St. Lucie Unit 1, Unit 2 and Turkey Point Units 3 and 4. Since similar methodology was developed for both Turkey Point and St. Lucie plant
- analysis, the RETRAN analysis methodology was discussed in generalities, with distinctions made where applicable.
3.1 RETRAN Nodalization FPL developed two-loop base model nodalizations for Turkey Point and St. Lucie analyses. However, no thorough RETRAN base model nodalization qualifications or justifications for either Turkey Point and St. Lucie models were presented. Several features are common to both plant models. In both models, complex controllers describing the steam bypass control systems are modeled using tables of flow versus demand. The demand is computed by the respective steam bypass control systems. Allowance for pressure dependence of the discharge flow was ignored unless analysis requires the demand as a function of pressure. Description of how the demand was computed is discussed as part of each plant model description in later sections. Multi-node steam generator models were developed for both plant models as options to be used when finer details are necessary. However, nn sa b tical justification or demonstration of the accuracy or sufficiency of such models were presented although FPL reported that they compared the results of two changes in nodalization. The non-equilibrium option was used in the upper 2
i'. downtomer region. The bubble rise model was used in the separator volume. Although theses choices are not necessarily inappropriate, no justification was presented for these model choices. For both plant models, the pressurizer (PZR) was modeled as a single node, non-equilibrium volume. Heat transfer between the PZR wall and the liquid is not modeled, therefore is ignored. The PZR spray option which does not de-superheat the vapor region was chosen for the FPL models to allow the peak pressures to be predicted higher during insurge transients than the other spray option would. SG safety valve modeling does not account for accumulation and hysteresis effect. The discharge flow is computed using a fill table of flow versus pressure based upon the Moody critical flow correlation. For the Turkey Point model, one second ramps are used in the PZR safety valve modeling and half-second ramps are used for the secondary safeties. Instantaneous actuation is assumed for St. Lucie. Five percent SG tube plugging was assumed in the St. Lucie base model whereas the values of fifteen and seventeen and one half percent plugging were incorporated into the St. Lucie Unit I and 2 models, respectively. All of these values were stated to be well above the current plugging levels in the l steam generators. A feedwater controller model was developed for the Turkey Point RETRAN model; however, no similar model has yet been developed for the St. Lucie model. 3.1.1 Turkey Point RETRAN Plant Model The base two-loop Turkey Point plant model consists of two separate loops each containing one hot-leg, a steam generator (SG) and one cold-leg. The steam generator in the base model is nodalized with equal length volumes in the primary side, and the secondary side of the SG is modeled with a single j node volume. FPL also developed a multi-node model incorporating two axially parallel volumes with three vertically stacked nodes in each where each volume interacts with a half of the primary side tubes. The two paralle) volumes representing the secondary side are completely separate and do not communicate with each other. Little justification was given for this nodalization and no studies were presented to demonstrate its accuracy or its convergence although FPL reported (without presenting any supporting data or analysis) that combining these two channels into one channel did not improve comparison to plant data (Ref. 5). Although FPL stated that this model was intended to allow better mixture level tracking in the loss of AC transient, 1 no explanation or justification was given as to how this nodalization accomplished this goal. Comparison of the single node SG vs. multi-node SG models was presented in an effort to qualify SG nodalization for use with the loss of AC transient analysis. However, the results obtained with these two SG secondary models were significantly different and no demonstration was given that either was more accurate than the other or that the multi-node model was sufficiently 3
+ ,a 8 detailed to be either physically accurate or numerically converged. FPL i stated that it will use the multi-node SG model for analysis of those transients for which accurate tracking of the SG level is required since the j low SG 1evel_setpoint is used to generate the react r trip and the auxiliary feed...ter signals. However, absent justification of either model, there is no demonstration that this multi-node model is adequate. The SG mixture level 'is computed through the use of a RETRAN controi block in which an algorithm is used to relate the mixture volume in the SG with an equivalent liquid level. For the benchmark calculation performed, the ligdd mass computed by RETRAN was said to be less than that used by the vendor for the FSAR analysis when the low level - signal was predicted to occur, thus lower than that predicted in the FSAR analysis. This algorithm was neither explained nor justified. Rather FPL relied upon the global results that the timing of the trip signal generation was predicted by the FPL model to be within one second of that predicted by the vendor. 3.1.2 St. Lucie RETRAN Plant Model FPL developed two separate RETRAN plant models for St. Lucie, one for each of two units. The basic nodalization for both models is nearly the same: a two-loop model consisting of two separate loops each containing one hot leg, a steam generator and two cold legs. The major difference in the nodalization is that for St. Lucie Unit 2 the upper plenum / head flow paths are slightly different from that developed for Unit 1. The SG is nodalized with four equal length volumes in the primary side and a single node secondary side. The St. Lucie multi-node SG model is nearly the same as that developed for the.. Turkey Point. steam generators, except that' it contains two vertically stacked nodes, instead of. three, in each of the parallel volumes for the secondary side as well as the primary side in both directions of the U-tubes. The primary side nodalization is. identical to that used in the base plant model. No justification was given for this model. Therefore,.it is recommended that FPL demonstrate at the time of first licensing submittal that the use a particular SG model for these transients results in conservative prediction. The SG mixture level is said to be computed by RETRAN in the same manner it is done for the Turkey Point analysis. FPL stated that "an earlier SG low level trip signal" is generated. Whether or not this early prediction of. trip signal is conservative may be transient dependent. -Therefore, FPL should demonstrate that for those transient for which an early reactor trip .is not. conservative, the resulting analysis still remains conservative and meets acceptance criteria. 3.2 Comnarison with Goeratina Data For the purpose of qualification of the best-estimate base model, FPL selected.two. sets of operation transient data; one from Turkey Point and the other for St. Lucie. Some analysis assumptions and protection system delay times were adjusted to match the actual time necessary to actuate and to 1 generate responses. 4 i .l 1
] [ i d i 3.2.1 Turkey Point Unit 4 - Loss of Inverter i FPL analyzed the loss of inverter event which occurred at Turkey Point Unit 4 while it was operating at full power. The loss of inverter resulted in a turbine runback. It also led to de-energizing the pressurizer heaters and initiated letdown isolation. Neither of the two power operated relief valves 4 (PORV) was available. The "A" SG feedwater controller was transferred to manual from automatic, and the feedwater to SG "A" remained at the 100% power i flow during the early stage of the transient, while feedwater flows to SG's "B" and "C" remained in automatic mode for the duration of the event. Both SG feedwater pumps were tripped due to reaching the high SG level setpoint about one minute.fter the reactor trip. Operator actions were taken to mitigate the cooldown caused by the feedwater transient. I This transient was selected to demonstrate the ability of the Turkey Point base model to simulate a primary to secondary heat generation / removal mismatch sufficient to result in a primary pressurization to the reactor trip setpoint. In addition, it was .alyzed because this event was similar to the ) licensing type transient. a 4 For RETRAN simulation the first 60 seconds of the event were analyzed. The following components and events were modeled: feedwater and auxiliary j feedwater flows, pressurizer spray and heaters and the turbine runback to 70% 4 power at the design rate of 200% per minute. 3 i The RETRAN initial conditions matched most of the plant conditions as recorded. Three parameters were adjusted to better match other available key j i plant conditions, SG recirculation ratio, RCS flowrate and main feedwater flow. 1 j~ The RETRAN predictions generally followed the trends of the plant data (peaks lagging by about 4 seconds); however, some predicted key parameters were outside of the range of plant data uncertainty. As noted by the licensee, l the responses exhibited by the secondary side parameters were slower than j i those measured during the event. i Although FPL performed sensitivity studies with respect to some of the code models, options and input, results were found to be insensitive to changes in i, these variables. This is because sensitivity studies were performed 1 examining the effect of variables which were largely irrelevant, such as i i components which were unavailable during the event or variables which were known ' to have minor effect. FPL reported that the use of the temperature i transport delay option in the primary loop resulted in improved prediction of !4 plant results but provided no data or analysis. No sensitivities were presented with respect to relevant parameters governing primary-to-secondary l heat transfer such as steam generator nodalization and model (e.g. bubble rise model) selection. 3.2.2 St. Lucie Unit 1 - Partial Loss of Feedwater A loss of power to the "B" main feed regulating valve (MFRV) controller i g 1 1 l
caused the "B" HFRV to shut and a loss of feedwater to the "B" steam generator to occur, which led to a reactor trip due to the low SG level setpoint. 4 For analysis of this event FPL used the multi-node steam generator model for the stated purpose of to better tracking the changes in the SG mixture level. Although the RETRAN initial conditions matched those measured plant values, some parameters were either adjusted to yield desirable values in other parameters or matched against historical values since data were not available. In either case, the changes in adjusted variables were not expected to impact the transient significantly. The first 100 seconds of the transient was analyzed and compared against the recorded data. As with the comparison discussed in 5 3.2.1, FPL was able to capture the global trends with the prediction. However, despite the fact this event was analyzed using the multi-node SG in order to track the SG level better, the predicted SG level did not compare well with the data. Differences between the predicted and measured were attributed to the modeling of the MSSV's and to a lesser extent to the shorter delay time assumed in the analyses. Upon re-analysis performed with a modified MSSV modeling, FPL was able to obtain SG pressures and loop temperatures closer to the data but did not reproduce the SG mixture level data. No sensitivities were presented with respect to relevant parameters governing primary-to-secondary heat transfer such as steam generator nodalization and model (e.g. bubble rise model) selection. 3.3 Safety Analysis Methods FPL described the process by which they intend to identify the limiting transients and select the key input parameters. The criteria used in the described process followed the requirements of the Standard Review Plan (SRP). For the catepry of the Decrease in Heat Removal from the Secondary System, the acceptance criteria are (i) to maintain peak pressures, both primary and secondary, below 110% of design; (ii) to maintain an adequate water level in the SG for the loss of normal feedwater type events to remove decay heat; and (iii) to examine for a potential for the PZR to go water solid. In addition, 1 departure from nucleate boiling ratio (DNBR) was to be evaluated. 3.3.1 RETRAN Modelina Since performance of licensing-type analyses requires changes to the best-estimate base model to reflect the necessary degree of conservatism, FPL idendfied the variables which modified for that purpose. For the category of transients in this submittal, the models which FPL identified as requiring changes are (1) safety valves, (2) temperature transport delay, (3) multi-node steam generator model, and (4) use of the slip model instead of the bubble rise model. 6
3.3.1.1 Safety Valves The safety valve model s were modified to incorporate accumulation and hysteresis and also to limit the discharge flow to the design value. The accumulation pressure is set at opening setpoint +3%. Based upon the experimental data which showed that the operation of these valves were nearly 1 instantaneous, FPL changed the RETRAN fill table which was time-dependent in the benchmark analysis to that of pressure-dependent. However, the valves are lumped together into a single valve, in terms of setpoints and flowrates. , 3.3.1.2 Temoerature Transoort Delav Ootion Since the effect of use of this option was found to be small, this option was not selected in the licensing model (Ref.1) but in Reference 5, FPL reported that this option will be used for the loss of AC transient analysis but provided no data or analysis. 3.3.1.3 Steam Generator Nodalization All transients, except the loss of AC, in this category for Turkey Point are analyzed using the single-node SG model since accurate tracking of the SG level is not required. For the loss of AC, FPL uses the un, justified multi-node SG to better predict the SG level. As before, no demonstration was provided to support the assumption that this model tracks the level better or i accurately. 3.3.1.4 Slio Model In Reference 5, FPL reported that the slip model should be included in certain circumstances. The algebraic slip model has not been approved or qualified and may not be used without qualification (Ref. 2) 3.3.2 Limitino Transient Determination To demonstrate the adequacy of the FPL RETRAN models for licensing applications and to develop a methodology for determination of the limiting event within the category, FPL presented qualitative discussion of each of seven FSAR-type events described in the Standard Review Plan under the Decrease in Secondary Side Heat Removal for Turkey Point and St. Lucie (Units I and 2) plants: (1) Loss of External Load; (2) Turbine Trip; (3) Loss of Condenser Vacuum; (4) Main Steam Isolation Valve Closure; (5) Steam Pressure Regulator Failure; (6) Loss of Non-emergency AC Power to the Station Auxiliaries; (7) Loss of Normal Feedwater Flow; and (8) Feedwater Line Break. FPL selected and reanalyzed a limiting transient from this category for each of St. Lucie Units and one for Turkey Point Units 3 and 4 using RETRAN, FPL compared sequences of events but did not present comparative plots of the RETRAN results to the respective current FSAR analysis. 1 The RETRAN built-in ANS 1973 decay heat curve was used where decay heat modeling is necessary. 7
bp. l F. ) 3.3.2.1 St. Lucie Unit 1 FPL analyzed the loss of condenser vacuum 'as a limiting transient for this category of transients. for the following reasons:.0f the eight transient-types in this category, loss of external load, turbine. trip, main steam isolation valves closure, steam pressure regulator failure events.and the }. loss of normal.feedwater flowrate are said to be bounded by the loss of condenser. vacuum. The loss of non-emergency AC power to the station auxiliaries is stated to be bounded by the loss of forced reactor coolant flow for the first 10 seconds in the event with_ respect to DNB. The loss of normal.feedwater event would result in the heat-up only.after there is a significant loss of SG inventory from continuation of steam flow. Therefore, i' the RCS pressurization is expected to be less than that from the loss of condenser vacuum event. Long-term cooling capability for both. of these i events relies on the ability of the AFW to maintain the necessary heat' sink. i Review of this issue is beyond the scope of this review. The feedwater line break for St. Lucie is a cool-down event and therefore bounded by the main 4 steam line break. This approach-is reasonable. 4 3.3.2.1.1 Loss of Condenser Vacuum (LCV) Loss of external load, turbine trip and loss of condenser vacuum are similar transients in which power / steam flow mismatch results in rapid pressurization 3 of the primary and secondary sides. The primary difference between the loss d od load and turbine trip events is the timing of the turbine trip. In both cases, the respective bounding events are postulated if credit is not taken for the. reactor trip on turbine trip or action. of the steam dump-bypass system (SDBS). The difference between these two transients and the LCV event is that in the latter event, LCV disables the steam dump bypass' system and results in a rampdown of main feedwater. Therefore, loss of condenser vacuum bounds the other two transients. Similarly,.the main _ steam isolation valve (MSIV) closure and steam pressure j regulator failure are bounded by LCV since the time it takes for the MSIV to close or. the regulator to. control turbine valves is much greater than the rapid action of the turbine-stop valves assumed in LCV. Assumptions designed to result in conservative predictions were used in the analysis. Comparison of the sequences of events between the RETRAN predicted and the I current FSAR values-indicated that the RETRAN predicted peak RCS pressure is about 100 psi less than the current FSAR value and occurred about 3.5 seconds later, while the predicted SG pressure was 20 psi higher than the FSAR value and occurred 1.7 seconds later. This indicates greater primary to secondary heat transfer in the RETRAN model. Differences were attributed to the difference in the moderator temperature coefficient used in the analysis and the pressurizer safety valve opening setpoint tolerance used in the FSAR analysis. FPL stated that both of these values are consistent with the current Technical Specifications for this plant. 8 i 1
-.~. 2 3.3.2.2 St Lucie Unit 2 y Similar to St. Lucie Unit 1. LCV bounds the loss of external load, turbine trip, main steam isolation valve closure, steam pressure regulator failure and loss of normal feedwater flow. FPL presented analysis of feedwater line break to demonstrate that it is a cool-down event and that it is bounded by the main steam line break. Since 4 the RETRAN model developed and discussed in this topical is intended only for the analysis of heat-up transients, review of this particular event analysis 4 is beyond the scope of this review and therefore was not reviewed. 3.3.2.2.1 Loss of Condenser Vacuum for loss of condenser vacuum, conservative assumptions, designed to maximize primary side pressure were selected, l Comparison of the sequences of events between the RETRAN predicted and the current FSAR values indicates that the RETRAN predicted peak RCS pressure is about 75 psi less than the FSAR value and occurred about I second later, and i the predicted SG pressure was 60 psi less than the FSAR value and occurred 1 second earlier. ] Differences were attributed to the difference in the computer codes (CESEC vs. RETRAN) used for the analysis. FPL stated that those initial conditions whose. values. are different from those used in the FSAR ure selected in order i to be more conservative by following the Technical Specification limits. i 3.3.2.3 Turkey Point Units 3 and 4 Similar to the St. Lucie cases, of the eight transients in this category of events, loss of external load, turbine trip, main steam isolation valves closure, steam - pressure regulator failure events and the loss of normal feedwater flowrate are determined to be bound by the LCV event whereas the feedwater line break is bounded by the steam line break since it is a cooldown event for Turkey Point plants. Therefore, FPL reanalyzed two event types: loss of condenser vacuum and the loss of offsite power with concurrent turbine trip. 3.3.2.3.1 Loss of Condenser Vacuum Loss of condenser vacuum generates a turbine trip signal, which causes the turbine stop valves to close. Concurrent to a turbine trip above 10% of full l power, a reactor trip signal is also generated. The loss of condenser vacuum j will result in a gradual rampdown of main feedwater. 1 FPL selected initial conditions were stated to be based upon conservative assumptions and according to the Technical Specification limits. A single node SG model was used in the analysis. Resulting sequences of events, when compared with the FSAR values, indicated 9 1
4 i j that the predicted peak RCS pressure was about. 50 psi higher than the FSAR i value and occurred 1.7 seconds earlier and the peak SG pressure was also 4 i computed to be about -20 psi higher than the FSAR value and occurred 4.7 i seconds later. 3.3.2.3.2 Loss of Non-Emeroency AC Power to the Station Auxiliaries The loss of non-emergency AC. (LOAC) power to the station auxiliaries was } defined as a complete. loss - of offsite power with concurrent turbine trip. Therefore, the event would result in a simultaneous-loss of load, a loss' of FW, a loss of forced RC flow ~and reactor scram. After reactor scram, decay i heat is removed by the operation of the turbine driven AFW pumps and the SG { safety valves to prevent the PZR from becoming water solid. i For this transient analysis FPL used both their single and multi-node SG models. The results were significantly different in all relevant key [ parameters computed. FPL attributed the major difference after approximately { 200 seconds to the main steam safety valve (MSSV) modeling, however,.no L analysis or parametric studies were presented to justify this statement. Furthermore, FPL presented no information to indicate or demonstrate that further changes in SG modeling would not cause further changes. Therefore no conclusion can be reached regarding the accuracy or adequacy of either j. model. i When compared with the FSAR sequence of events, FPL's. predicted peak pressurizer water volume is 280 ft3 less than computed in FSAR analysis and .the timing of such occurrence was predicted at 720 seconds into the transient i by RETRAN while it was 3720 seconds in the FSAR prediction. These differences were attributed by FPL to code differences but no effort was made i to assure accuracy or convergence of the FPL RETRAN models.
4.0 CONCLUSION
S The subject submittals were reviewed in the context of whether or not the conditions set forth in the SER issued by the NRC in respect of Reference 6 +. had been satisfied and to determine the adequacy of FPL's RETRAN plant models for licensing applications for the transient analysis with a decrease in heat removal by the secondary side. i We find that FPL has not satisfied the following conditions: I 1. FPL has not justified its SG nodalizations for either Turkey Point or St. Lucie. .l However, regarding the FPL single node SG nodalization, we are of the view, based upon our own engineering judgement and experience, that the particular nodalization and model selection are appropriate for best-estimate and licensing modeling of those transients in the category of Decrease in Heat Removal by the Secondary Side for which (1) a depletion of the secondary side mass inventory is not expected, (2) heat transfer degradation is not expected due to inventory loss and (3) accurate tracking of 10
l i 4,. 3 J secondary side mixtura level is not' required for actuation of other q safety functions. 1
- j FPL's multi-node SG models have not been qualified for any use and j
therefore may not be used in either best-estimate or licensing i analyses without qualification and demonstration that the selected ii SG modeling generates conservative and numerically converged results. 2. Although analytical methodology development approaches can be shared in St. Lucie and Turkey Point RETRAN - model development, l q' actual model qualification and justification should be performed on [ individual plant-design basis due to difference in plant designs and operations. including control systems, SG and vessel designs. i Wrefore, separate qualification should be performed for each of a 'u following: (1) SG nodalization, (2) SG mixture level computation algorithm and (3) use of the slip model.
5.0 REFERENCES
q 1.- "RETRAN Model Qualification Decrease of Heat Removal by the Secondary System," NTH-TR-01, July 1989. 2. Letter from A.C.. Thadani (USNRC) to R. Furia (GPU), " Acceptance for Referencing Topical Report EPRI-NP-1850 CCN;A, Revisions 2 and 3 Regarding RETRAN02/M00003 and M00004," October 19, 1988. 3. Letter from W.H. Bohlke (FPL) to USNRC, " Response to Request for Additional Information Related to Topical Report NTH-TR-01, RETRAN Model Qualification - St. Lucie Plant Unit Nos. I and 2 and Turkey Point Plant l Units Nos. 3 and 4 (TAC Nos. 75082, 75083, 75084 and 75085)," May 2 j 1991. { 4. Letter from W.H. Bohlke (FPL) to USNRC, " Response to Request for Additional Information Related to Topical Report NTH-TR-01, RETRAN Model Qualification - St. Lucie Plant Unit Nos. I and 2 and Turkey Point Plant j Urtits Nos. 3 and 4 (TAC Nos. 75082, 75083, 75084 and 75085)," May 19, 1992. 5. Letter from W.H. Bohlke (FPL) to USNRC, " Response to Request for [ Additional Information Related to Topical Report NTH-TR-01, RETRAN Model Qualification - St. Lucie Plant Unit Nos. I and 2 and Turkey Point Plant Units Nos. 3 and 4 (TAC Nos. 75082, 75083, 75084 and 75085)," February i 16, 1993. 6. Letter from W.H. Bohlke (FPL) to USNRC, " Response to Request for i Additional Information Related to Topical Report NTH-TR-01, RETRAN Model Qualification ~ - St. Lucie Plant Unit Nos. I and 2 (Docket Nos. 50-335 and 50-389) and Turkey Point Plant Units Nos. 3 and 4 (Docket Nos. 50-4 250 and 50-25?i sTAC Nos. 75082, 75083, 75084 and 75085)," September 28, 1993. l i 11 i e J
7. Letter from G.C. Lainas (USNRC) to W.F. Conway (FPL), " Florida Power and Light Company - Topical Report on RETRAN (TAC ?!o. 60550) and Topical Report on PWR Physics Methodology (TAC No. 60549," April 19, 1988. f 8. Licensee Qualification for Performing Safety Analjses in Support' of. licensing Actions (Generic Letter llo. 83-11), USNRC, February 8,1983. I 1 l l l.. l \\ i l r i 12 i 4}}