ML25112A275

Nbsr Facility - Licnse Amendment Request

ML25112A275
Person / Time
Site:	National Bureau of Standards Reactor (TR-5)
Issue date:	04/22/2025
From:	Newton T US Dept of Commerce, National Institute of Standards & Technology (NIST)
To:	Office of Nuclear Reactor Regulation, Document Control Desk
References
Download: ML25112A275 (1)

Text

April 22, 2025 U.S. Nuclear Regulatory Commission ATTN: Document Control Desk Washington, DC 20555-0001

Subject:

License Amendment Request Ref: NBSR Facility License TR-5, Docket 50-184 Sir/Madam:

UNITED STATES DEPARTMENT OF COMMERCE National Institute of Standards and Technology Gaithersburg, Maryland 20899-6100 On December 19, 2024, NCNR submitted a license amendment request for a change in the Technical Specifications to revise operability requirements for primary coolant flow channels (NRC Accession No. ML25014A397). Attached please find supplemental information of a memo analyzing the consequences of loss of flow.

Please contact me if you have any questions.

Deputy Director and Chief, Reactor Operations and Engineering NIST Center for Neutron Research cc: Justin Hudson, NRR/DANU/UNPL NISI

Document Type: Technical Memo Status: Approved Document

Title:

Consequences of a Failed Plenum Flow Meter during Loss of Flow Accident Document Number: NCNR-RE-TM-000651 Date: 12/12/2024 NIST Test Reactor (NBSR) Technical Memo Consequences of a Failed Plenum Flow Meter during Loss of Flow Accident Approved An analysis on the feasibility of operating with DWV-2 closing without the FRC-4 flow meter.

The analysis also shows the consequences of operating without FRC-3 while DWV-1 is closing Name Signature Author:

Elsalamouny, Noura R.

(IntlAssoc)Weiss, Abdullah G. (Fed)

Reviewer:

Mattes, Daniel A. (Fed)Seiter, Jacob A. (Fed)

Approver:

Sahin, Dagistan (Fed)

Document Type: Technical Memo Status: Approved Document

Title:

Consequences of a Failed Plenum Flow Meter during Loss of Flow Accident Document Number: NCNR-RE-TM-000651 Date: 12/12/2024 Page 1 of 27 Executive Summary The NIST Test Reactor (NBSR) Technical Memo investigates the possible consequences of a hypothetical accident scenario where the DWV-2 valve, which controls coolant flow to the inner plenum of the reactor core, closes unexpectedly and the flow rate indicator FRC-4 is bypassed. The memo utilizes the RELAP5/MOD3.3 modeling approach to simulate coolant flow dynamics within the NBSR primary cooling system and analyze the impact of DWV-2 closure on coolant distribution and thermal hydraulics in the reactor core.

The Loss of Flow Accident LOFA refers to a sudden or unexpected reduction or complete loss of the flow within the system [1]. The LOFA accident would occur due to pump failure, valve malfunction, or a break in the pipeline. The LOFA accident is different from the Flow blockage incident [2] where the flow blockage could occur due to an obstruction within the system, which prevents the fluid from flowing past a certain point. Loss of Flow Accidents (LOFAs) have been studied in various research reactors (MTRs) using thermal-hydraulic codes such as RELAP5/MOD3.2, RELAP5/MOD3.3, and RELAP5/MOD3.4.

Simulation results indicate that the consequences can be severe for MTR reactors operating at 20 MW.

Specifically, when the flow is reduced to less than 20% of nominal rates, the cladding temperature can exceed blistering temperatures and approach melting temperatures [1][3]. Other simulations were performed on different MTRs with 5 MW and 10 MW operating powers. The consequences were not as severe as in 20 MW reactors. However, partial damage and overheating in the fuel were observed when the flow was reduced to less than 20% of nominal rates.

In the examined scenario, the closure of DWV-2 leads to a severe loss of flow in the inner plenum, which can potentially cause fuel damage before an automatic low-flow scram is executed. Similarly, the closure DWV-1 results in a severe loss of flow in the outer plenum, posing the risk of fuel damage prior to scram activation. In both cases, the model indicates that the temperature of the fuel plates significantly increases, exceeding the blister temperature and nearing or above the melting point of the aluminum cladding, thereby suggesting that severe damage to the fuel plates is expected during both scenarios.

As amended on March 15, 2023 (Amendment No 15), the current technical specifications require either one of the FRC-3 or FRC-4 to be operational. However, based on the analysis results, the reactor cannot be safely operated when either the FRC-3 or FRC-4 flow meter is bypassed. Both FRC-4 and FRC-3 must be required for all operations. The engineering recommendation is to complete a LAR in a timely manner and update the technical specifications to require both inlet and outlet plena low flow scrams to be operable as a limiting condition for operation. Necessary updates and modifications of operating instructions and increasing FIA-1 low flow alarm limit to initiate a scram due to closure of the inner plenum isolation are also recommended. For the immediate term, at the discretion of the CRO, a detailed SSI could be written to limit operations that require both flow meters.

Document Type: Technical Memo Status: Approved Document

Title:

Consequences of a Failed Plenum Flow Meter during Loss of Flow Accident Document Number: NCNR-RE-TM-000651 Date: 12/12/2024 Page 2 of 27 Introduction The primary cooling system in the NBSR is essential for maintaining safe and efficient operations. It features four main pumps responsible for circulating coolant to both the inner and outer plena of the reactor core. Coolant flow into the core is divided through two valves: DWV-1 and DWV-2. Flow through DWV-1 provides coolant to 24 outer plenum elements, while flow through DWV-2 cools the 6 elements in the inner plenum. Flowrate is monitored directly for the inner plenum via FRC-4, the outer plenum via FRC-3, and the total flow is measured at the reactor outlet via FIA-40. Per Technical Speciations TR-5 (as Amended 15, March 2023) 3.2.2 Reactor Safety System Channels, the reactor shall not be operated unless the scram channels for the reactor outlet low flow and at least one of either the inner or outer plenum low flow are operable.

Recently, a query arose regarding the possibility of inadvertent closure of the DWV-2 valve during normal operation with FRC-4 is bypassed, which is currently a permissible configuration. The concern raised was whether fuel damage could occur before an automatic low flow scram was executed. This raises important questions about the subsequent flow dynamics through DWV-1 and its impact on the system. Understanding the indications of flow through DWV-1 post-closure of DWV-2 is crucial for assessing the system's behavior and ensuring optimal reactor cooling and operational safety. The reactor initiates SCRAM due to low flow when the flow is less than 1200 US gpm in the inner plenum or less than 4700 US gpm in the outer plenum or the total flow is less than 5900 US gpm. This work discusses an analysis of this hypothetical accident. The potential causes of unintentional valve closure are not analyzed in this report and remain otherwise undefined.

Methodology RELAP Model The analysis was conducted using RELAP5/MOD3.3, incorporating the current model as incorporated in the latest FSAR as part of the license Amendment 15, ML23055A300 (March 2023). Note that the RELAP model includes critical components such as the main pumps designated as Pump component 20, DWV-1 represented as Valve component 51, which regulates the flow between the Pump and the Outer Plenum, and DWV-2 represented as Valve component 50 which connects the component 40 to Inner plenum 51, controlling flow between the Pump and the Inner Plenum (see Figure 1). A minor modification was done to the model to complete the analysis in this report. A Valve component (50) is added to mimic DWV-2. This change does not affect the model. Nevertheless, model verification was completed to ensure accuracy of results.

This integrated modeling approach allows for detailed simulation and analysis of coolant flow dynamics within the NBSR primary cooling system. This approach ensures consistency and accuracy in predicting system behavior and response to operational changes or scenarios, such as valve closures and their impact on coolant distribution and thermal management within the reactor core.

Document Type: Technical Memo Status: Approved Document

Title:

Consequences of a Failed Plenum Flow Meter during Loss of Flow Accident Document Number: NCNR-RE-TM-000651 Date: 12/12/2024 Page 3 of 27 Figure 1. NBSR nodalization scheme (from [4]) with valve component 50 added.

There is a controlled system in the model that was developed to control the reactor scram logic [4]. The models system scrams the reactor when any one of these TRIPS are true (see Table 1 and Figure 2). Table 1 shows the Trips' number, function and conditions, Figure 2 present the control of NBSR scram system.

Each TRIP has a condition to be true or false triggered or not triggered. When a TRIP condition is true, it means that the NBSR or its control mechanism has been triggered or activated and the NBSR would scram at certain time for the safety. When a TRIP condition is false, it means that the system or control mechanism has not been triggered. The system works normally without any need to shut down or scram.

Figure 2 presents the relevant control systems logic for the NBSR reactor scram[4], where each TRIP has a condition to be triggered in order to activate the TRIP (see Table 1). For example, the low inner plenum flow TRIP is 521 which should be activated when the flow rate is less than 1200 US gpm and controlled by a control variable 908 as seen in Figure 2.

During steady state operation, the TRIPs are defined to be false or not activated and the reactor operates normally. In the Reference calculation, only TRIP-401 is activated true as it controls the DWV-2 closure time. In the accident scenario, the TRIPs 401 and 522 are activated which they control the DWV-2 closure, and low inner plenum flow in addition to disable the scram logic system in order to simulate the accident with the maximum power of the NBSR (20 MW).

50 0

Document Type: Technical Memo Status: Approved Document

Title:

Consequences of a Failed Plenum Flow Meter during Loss of Flow Accident Document Number: NCNR-RE-TM-000651 Date: 12/12/2024 Page 4 of 27 Table 1. Reactor Trip system [4].

TRIP Function Status in steady state conditions Reference Accident 401 Accident of throttling of coolant flow to the inner plenum is simulated by closing the valve at the inlet of the inner plenum (using TRIP-401) at t =

0.0 s.

False True True 402 Accident of throttling of coolant flow to the outer plenum is simulated by closing the valve at the inlet of the outer plenum (using TRIP-402) at t =

0.0 s.

False False False 403 This accident is simulated by stopping the primary pumps (using TRIP-404), closing the at the discharge lines of the primary pumps valves (using TRIP-405), and stopping heat transfer between the primary and secondary systems (Using TRIP-403) at t = 0.0 s.

False False False 404 405 406 Accident of seizure of one primary coolant pump is simulated by reducing the primary pump speed by 2/3 (using TRIP-406) at t = 0.0 s.

False False False 410 Accident of maximum reactivity insertion is simulated by inserting reactivity (using TRIP-410) at t = 0.0 s.

False False False 501 Controls the reactor power False False False 502 Controls TRIP 501 (Reactor Scram due to high reactor power)

False False False 521 Controls the flow to inner plenum False False False 522 Controls TRIP 521 (Reactor Scram due to low flow of the Inner Plenum)

False False True 531 Controls the flow to outer plenum False False False 532 Controls TRIP 531 (Reactor Scram due to low flow to Outer Plenum)

False False False 541 Controls the total flow to the primary system False False False 542 Controls TRIP 541 (Reactor Scram due to low total primary flow)

False False False 601 Reactor Scram when high reactor power scram or low inner plenum flow scram occurs.

False False Disabled (These are the SCRAM logic system, and are interdependent) 602 Reactor Scram when high reactor power scram, low inner plenum flow cram, or low outer plenum flow scram occurs.

False False 603 Reactor Scram when high reactor power scram, low inner plenum flow scram, low outer plenum flow scram, or low total primary flow scram occurs.

False False

Document Type: Technical Memo Status: Approved Document

Title:

Consequences of a Failed Plenum Flow Meter during Loss of Flow Accident Document Number: NCNR-RE-TM-000651 Date: 12/12/2024 Page 5 of 27 Figure 2. Control system of NBSR reactor Scram [4].

Document Type: Technical Memo Status: Approved Document

Title:

Consequences of a Failed Plenum Flow Meter during Loss of Flow Accident Document Number: NCNR-RE-TM-000651 Date: 12/12/2024 Page 6 of 27 Analysis Approach The RELAP model used in this work was introduced in the NBSR licensing as part of the license Amendment 15, ML23055A300 (March 2023). A minor adjustment was made as described earlier (i.e.,

adding valve component 50) for simulation of DWV-2. The modified model is verified by simulating steady-state operations, ensuring its fidelity in representing the reactor's behavior during normal operations.

This model was then leveraged for transient analysis, enabling investigation into the scenario of DWV-2 closure which aimed at evaluating the consequences of severe events such as heat-up incidents and Loss of Flow Accidents. Similarly, the same model was leveraged for transient analysis to investigate the consequences of DWV-1 closure scenario. The transient analysis and evaluated consequences of DWV-1 closure is presented in Appendix A.

The steady state calculation The model results were compared with the updated FSAR data in the Steady Sate. The calculations were performed at the highest NBSR power ~20 MW. The most important parameters to be investigated for this run are the flow rate of the Inner and Outer Plenum, Reactor outlet temperature, and total flow rate (see Table 2) and the model captures the steady-state values of the reactor parameters. As demonstrated in Table 2, the differences are acceptable, which yields sufficient confidence to proceed with this models results while considering a 3.3% validation error in the reactor outlet temperature.

Table 2. Validation of the model using steady state conditions (data taken from Section 2.1 and section 4.2.3.7 in the updated FSAR Appendix A) [5].

Parameter FSAR Steady state Calculations (this works model)

% Difference Steady state operation power 20 MW 20 MW 0.00%

Total flow rate 8700 US gpm 8704 US gpm 0.05%

Inner Plenum flow rate 2300 US gpm 2297.8 US gpm 0.1%

Outer Plenum flow rate 6400 US gpm 6407.3 US gpm 0.11%

Outlet reactor temperature 45.6 44 3.30%

Transient calculation The NBSR model was leveraged to perform a transient calculation and investigate the consequences of DWV-2s closure. Two different calculations were performed for this analysis:

The 1st calculation is a replica of the analysis in the FSAR for throttling of primary coolant flow to the inner plenum, and it is considered a reference case. In the reference case, we simulate the closure of DWV-2, which reduces the D2O flow rate to the inner plenum. This closure was performed by activating TRIP 401 to be true as its condition is triggered Figure 3. Figure 3 shows the reactor control trips and their status from the RELAP5 output. When the number is 0.0 or greater than 0.0 means that the trip condition is triggered, and the TRIP is activated at this certain time. If the value is -1.0 means that the TRIP condition is not triggered, and the trip is not activated yet. In this analysis the trip which controls the NBSR scram were activated at 15.5s. Note that in this work, the calculation is extended beyond the time range analyzed for the FSAR. This is done to ensure that there are no subsequent failures (i.e., overheating) beyond what the analysis in the FSAR considers.

Document Type: Technical Memo Status: Approved Document

Title:

Consequences of a Failed Plenum Flow Meter during Loss of Flow Accident Document Number: NCNR-RE-TM-000651 Date: 12/12/2024 Page 7 of 27 Figure 3. NBSR Scram TRIP system when DWV-2 is fully closed Reference at the end of calculation.

The 2nd calculation is considered as the Accident calculation, it simulated the closure of DWV-2 which is controlled by making the condition of TRIP 401true to initiate the scenario. Also, the TRIP 522 which controls the Inner Plenum flow was activated too and its condition was triggered. The TRIP 522 is true when the flow rate of the inner plenum is less than 1200 US gpm, but in the model the limit is changed to be 0.0 US gpm. The trip status can be checked from the RELAP5 output (see Figure 4), the TRIP is active when the TRIP condition equal to 0.0 or a number greater than 0.0 and the TRIP is not active when TRIP condition = -1.0, In the accident scenario, the TRIPs (601, 602 and 603) which control the reactor scram are disabled. And only two TRIPs are activated as mentioned before (401 and 522).

Figure 4. NBSR Scram TRIP system when DWV-2 is fully closed Accident at the end of calculation.

Results Closure of DWV-2 As previously mentioned in the Transient calculation subsection, two calculations were performed to investigate the closure of DWV-2.

Reference:

the DWV-2 is closed and the NBSR scrams when the flow of the inner plenum is less than 1200 US gpm.

Accident: the DWV-2 is closed, no flow indictors are active, and no scram occurs throughout the accident.

The transient calculation Reference of DWV-2 was mentioned in SAR; the DWV-2 takes approximately 30 seconds to fully close after receiving the closure signal. This gradual closing reduces the mass flow rate that feeds the inner plenum. When the mass flow rate is less than 1200 US gpm, the Reactor Scram occurs.

As the flow level in the inner plenum decreases, the TRIP system activates at ~ 15.5 s, causing the NBSR to scram. After the scram, NBSR power drops from 20 MW to 0.9 MW at 20s. The power takes ~ 5 s to drop significantly.

The calculation Reference is maintained successfully without any severe/unrecoverable issues. Note that because of the inner plenums flow rate reduction, noticeable increases in the outer plenum flow rate can be observed.

Document Type: Technical Memo Status: Approved Document

Title:

Consequences of a Failed Plenum Flow Meter during Loss of Flow Accident Document Number: NCNR-RE-TM-000651 Date: 12/12/2024 Page 8 of 27 The Accident scenario occurs when DWV-2 is closed and the flow indicator of the inner plenum is not functional, which allows a scenario that disables the ability of the reactor to scram due to low flow in the inner plenum. With FRC-4 bypassed, the only potential scram source remaining is FIA-40, which does observe reduced flow due to the increased system head-loss, at 6700 US gpm. The gradual closing of DWV-2 reduces the flow rate in the inner plenum while the power is constant at its maximum value 20 MW. With the reduction of inner plenum flow rate, the fuel plates temperature increases from 320 to 900 K, which exceeds the blister temperature ~723 K and approaches the melting point of Aluminum ~ 933 K. Therefore, core melting would likely begin at ~ 20-25 s. The RELAP5 code is not able to calculate after 24 s due to the significant voiding, wherein the cladding gets to temperatures beyond the range accommodated by the thermal properties used by the code. As such, the transient fails to compute beyond 24 s because of the extremely high cladding temperatures reached in the accident.

Table 3 presents a comprehensive comparison of evaluation results between the "Reference" and "Accident" scenarios for various parameters, including the pressure and mass flow rate of the inner and outer plena, total mass flow rate and pump pressure. As shown by the model, the minimum flowrate measured at FIA-40 is 8267 US gpm, still well above the scram setpoint of FIA-40. It is therefore concluded that no automatic action will occur with FRC-4 bypassed in this scenario. This comparison provides insights into how these critical parameters vary under the inner plenum throttling accident in the FSAR (Reference) versus during a similar accident scenario without flow indication in the inner plenum (Accident).

Table 3. Comparison between Reference and Accident calculation results.

Parameter Initial conditions Reference Accident calculation Steady state operation power 20 MW SCRAM occurs at 15.5 s.

Power Reduced to 0.9 MW at 20 s.

Power reduced to 0.75 MW at 35 s No SCRAM occurs Constant power 20.0 MW Status of DWV-2 Fully opened DWV-2 fully closed at 30 s DWV-2 not fully closed Calculation failed at 24 s At the end of calculations Inlet flow rate 8700 US gpm 8167.0 US gpm 8276.0 US gpm Inner Plenum flow rate 2288 US gpm 0.0 US gpm 415.0 US gpm Outer Plenum flow rate 6400 US gpm 8167.0 US gpm 7861.1 US gpm Pump pressure 55.5 psi 43.5 psi 43.5 psi Inner Plenum pressure 41 psi 20.6 psi 22.14 psi Outer Plenum pressure 44.4 psi 37.9 psi 52.97 psi Outlet pressure 21.5 psi 21.5 psi 21.5 psi Outlet flow rate 8700 US gpm 8167.0 US gpm 8276.0 US gpm

Document Type: Technical Memo Status: Approved Document

Title:

Consequences of a Failed Plenum Flow Meter during Loss of Flow Accident Document Number: NCNR-RE-TM-000651 Date: 12/12/2024 Page 9 of 27 Figure 5 provides a detailed comparison of the NBSR's power and reactivity between the "Reference" and "Accident" scenarios. In the Reference calculation, the NBSR undergoes a scram successfully at ~15.5 seconds, causing a sudden drop in power. Reactor power decreases to 0.9 MW at ~20 seconds, within 5 seconds of the scram. Subsequently, the power further decreases to approximately 0.75 MW at 35 seconds in the simulation, or 20 sec after the scram. Concurrently, the reactivity, reflecting the deviation from nominal criticality, initially drops below 0 dollars (indicative of a negative reactivity insertion) immediately after the scram. It then stabilizes at a constant value of -18.00 dollars by 20 seconds.

In contrast, the Accident scenario depicts the maximum operational power of the NBSR at 20 MW with a reactivity value of 0 dollars (indicating nominal criticality). This scenario represents the peak power output under normal operating conditions without any reactivity deviations. In other words, this demonstrates that without the flow indication in the inner plenum, the reactor will not automatically scram if DWV-2 inadvertently closes.

Figure 5. Power & reactivity evaluation.

The pressure of the inner plenum, outer plenum and pump for the Reference and Accident calculation is presented in Figure 6. The inner and outer plena pressure of Reference and Accident calculation have the same behavior. In the case of the References inner plenum pressure, as the DWV-2 is closing the pressure decreases until the DWV-2 is fully closed at ~30 s then the References inner plenum pressure reaches a constant value of ~20.6 psi. In the case of Accident calculation, the inner plenum pressure decreases in an identical fashion to the Reference calculation, until the calculation fails at ~24s due to the excessive heating SCRAM

Document Type: Technical Memo Status: Approved Document

Title:

Consequences of a Failed Plenum Flow Meter during Loss of Flow Accident Document Number: NCNR-RE-TM-000651 Date: 12/12/2024 Page 10 of 27 and fuel (likely) melting. At 24 s, the pressure reaches 22.12 psi, which is consistent with the Reference calculation. The References outer plenum pressure is increases until it reaches ~55.1 psi (after which it settles), while the Accidents outer plenum pressure reaches ~52.9 psi before the calculation fails.

Figure 6. The pressure evaluation.

Figure 7 provides an assessment of the D2O mass flow rates in both the inner plenum (DWV-2) and outer plenum (DWV-1) during the Reference and Accident scenarios. The closure of DWV-2, which regulates flow to the inner plenum, significantly impacts these flow rates. In the Reference scenario, the DWV-2 flow rate gradually decreases as the valve closes. By about 30 seconds, the DWV-2 flow rate reaches 0.0 US gpm, indicating complete closure. Concurrently, the flow rate in DWV-1 steadily increases and reaches 8100 US gpm as DWV-2 closes.

In contrast, during the Accident scenario, the simulation fails prematurely at 24 seconds, before the full closure of DWV-2. Despite this failure, the flow rate in DWV-2 remains at approximately 415 US gpm.

Meanwhile, the flow rate in DWV-1 increases to approximately 7861 US gpm before the simulation fails.

This indicates that the fuel may reach melting at low flow in the inner plenum, prior to the complete loss of flow in the inner plenum. More specifically, it appears that the inner plenum flow are reaches almost 420 US gpm, which is not sufficient to prevent fuel plate damage. Note that 420 US gpm is 35% of the current low flow setpoint for the inner plenum, and ~18% of the nominal flow rate. In other words, dropping below 20% of the nominal flow rate in the inner plenum could cause a fuel damage. This is consistent with the TRACE simulations ran by BNL in an attempt to simulate the February 3, 2021, conditions, where it was found that ~80% flow blockage is sufficient to cause significant voiding in a fuel element. In this Accidents case however, the flow blockage would be to six elements, not just one; therefore, this may presumably

Document Type: Technical Memo Status: Approved Document

Title:

Consequences of a Failed Plenum Flow Meter during Loss of Flow Accident Document Number: NCNR-RE-TM-000651 Date: 12/12/2024 Page 11 of 27 lead to a similar partial fuel plate melting accident as February 3s. The only difference is that in this Accident scenario, all six elements may experience a plate melting.

Figure 7. Mass flow rate evaluation.

An essential parameter for investigating the Accident calculation is the temperature of the fuel plate cladding (heat structures), which is depicted in Figure 8. Initially, the Reference fuel plate cladding temperature rose to 360 K and then dropped following the scram at approximately 15.5 seconds. The maximum reference temperature for the fuel plates is 356 K. In the event of an accident, the temperature of the fuel plates begins to rise gradually in the first 20 seconds, reaching approximately 400 K. Afterward, a rapid increase in temperature is observed, with the fuel plates reaching as high as 890 K before the simulation fails. When considering the temperature results, it is important to recall that the NBSR U3O8 fuel plates blistering temperature is ~ 723 K and the Aluminum claddings melting temperature is ~ 933 K. During the Accident scenario, the maximum cladding temperature (~890 K) is very close to the melting temperature and significantly exceeds the blistering temperature. This indicates that severe damage of the fuel plates and U3O8-Al relocation in the inner plenum is expected during the Accident scenario.

Document Type: Technical Memo Status: Approved Document

Title:

Consequences of a Failed Plenum Flow Meter during Loss of Flow Accident Document Number: NCNR-RE-TM-000651 Date: 12/12/2024 Page 12 of 27 Figure 8. Inner core temperatures (all six fuel elements around the center of the core).

The temperatures of the hot channels within the inner plenum have been analyzed both in the Reference scenario (Figure 9) and the Accident scenario (Figure 10). In the Reference case, the hot channel temperatures show a distinct pattern following the reactor scram. Initially, temperatures decrease from 340 K to 320 K. Subsequently, beginning around 35 seconds after the scram, as the reactor operates at approximately 0.75 MW power, there is a noticeable rise in temperatures. This increase continues until temperatures reach the boiling temperature of 374 K within the time frame of 50 to 60 seconds. After this peak, temperatures gradually decrease, stabilizing between 360 K, and 340 K after 60 seconds.

Document Type: Technical Memo Status: Approved Document

Title:

Consequences of a Failed Plenum Flow Meter during Loss of Flow Accident Document Number: NCNR-RE-TM-000651 Date: 12/12/2024 Page 13 of 27 Figure 9. Hot channel temperatures in the Reference scenario.

In Figure 10, the temperature of the hot channels during the Accident calculation demonstrates a notable trend. Initially, the temperatures rise steadily over time, reaching their peak at approximately 20 seconds.

However, at ~ 20-second, a critical event occurs: the inner plenum experiences a failure. This failure is directly linked to the temperatures of the fuel plates, which have exceeded the blister temperature and are nearing the cladding melting point. This elevated temperature causes significant changes in the material properties, including relocation and even melting of the fuel materials. The root cause of this critical failure is identified as a severe loss of flow within the inner plenum.

Document Type: Technical Memo Status: Approved Document

Title:

Consequences of a Failed Plenum Flow Meter during Loss of Flow Accident Document Number: NCNR-RE-TM-000651 Date: 12/12/2024 Page 14 of 27 Figure 10. Hot channel temperatures in the Accident scenario.

Figure 11 and Figure 12 showed the Reference and accident void fraction in the inner plenum. The extensive voiding occurred in the zero-flow depicting the axial void distribution in the lower and upper portions of the inner plenum. In the reference case the void fraction started at 41 seconds (11s after the closure of DWV-2). In the Reference case the voiding fraction is between 0.0 to 0.006, which has small values and could be neglected. So, the inner plenum Reference appears to be in no voided state. In the accident scenario, the voiding is extensive due to the zero flow in the upper and lower cores of the inner plenum.

The transient nature of the void response is evident in the periodic fluctuations of the axial void profile.

The inner plenum appears to be mostly in voided state. The voiding in the Upper part of the inner plenum started 2s earlier than the voiding in the lower part of the inner plenum.

Document Type: Technical Memo Status: Approved Document

Title:

Consequences of a Failed Plenum Flow Meter during Loss of Flow Accident Document Number: NCNR-RE-TM-000651 Date: 12/12/2024 Page 15 of 27 Figure 11. Reference case void fraction in the inner plenum.

Document Type: Technical Memo Status: Approved Document

Title:

Consequences of a Failed Plenum Flow Meter during Loss of Flow Accident Document Number: NCNR-RE-TM-000651 Date: 12/12/2024 Page 16 of 27 Figure 12. Accident scenario void fraction in the Inner Plenum.

Discussion and Conclusions The accident scenario following the closure of DWV-2 with no flow indicators has severe consequences on the fuel elements cladding integrity. As discussed, a complete loss of flow accident is initiated due to the closure of DWV-2 in the inner plenum. The severe events are concluded in the significant increase of fuel plates temperatures reaching the melting point. There is a high probability that the Inner Plenum failure and loss Inner core integrity would occurr. As the power remains constant and considering the quick evolution of the accident, it would be nearly impossible for operators to observe and respond to the initiative events of the accident. It might be better to monitor not only the flow reduction in the inner plenum but to also the flow rate increases in the outer plenum. Once the flow rate of the inner plenum reaches ~2000 US gpm, the outer plenum reaches ~6600 US gpm. As such, it would be wise to consider a warning for high outer plenum flow. Action would still be needed to be taken very quickly (within less than 10 seconds). At this moment the reactors must scrammed immediately to eliminate the severe consequences.

In short: Within the evaluated hypothetical accident scenario, the consequences of accidents previously analyzed in the updated FSAR is substantially different. Therefore, continuous flow must be provided to both the inner and outer plenum fuel elements while the reactor is operating. With the current system configuration, the reactor cannot be safely operated when FRC-4 is bypassed. Both FRC-4 and FRC-3 have to be operational.

Document Type: Technical Memo Status: Approved Document

Title:

Consequences of a Failed Plenum Flow Meter during Loss of Flow Accident Document Number: NCNR-RE-TM-000651 Date: 12/12/2024 Page 17 of 27 References

[1] N. Akhal, etal. Loss Of Flow Accident (LOFA) with protection in NUR nuclear research reactor; three-dimensional analysis of a Fast LOFA and a Slow LOFA,Progress in Nuclear

Energy, Volume

162, 2023,

104779, ISSN 0149-1970, https://doi.org/10.1016/j.pnucene.2023.104779.

[2] A. Lyria, etal. Flow blockage accident analysis in a multi purposes research reactor using Relap5 system code, Progress in Nuclear Energy, Volume 168, 2024, 105019, ISSN 0149-1970, https://doi.org/10.1016/j.pnucene.2023.105019.

[3] Y. Guo, etal. Thermal hydraulic analysis of loss of flow accident in the JRR-3M research reactor under the flow blockage transient, Annals of Nuclear Energy, Volume 118, 2018, Pages 147-153, ISSN 0306-4549, https://doi.org/10.1016/j.anucene.2018.04.014.

[4] National Institute of Standards and Technology (2015) NBSR Calculation Notebook RELAP5 Models for Non-LOCA Events. Nuclear Science & Technology Department Brookhaven National Laboratory Upton, NY 11973.

[5] Safety Analysis Report (SAR) for License Renewal for the National Institute of Standards and Technology Reactor -NBSR 14 Appendix A, Sections 2.1 and 4.2.3.7.

[6] Safety Analysis Report for the National Institute of Standards and Technology Reactor -

NBSR, NBSR 14.

[7] Instrumentation & Controls Calibration Procedure ICP 3.1.3 Reactor Outlet Flow FIA-40.

Document Type: Technical Memo Status: Approved Document

Title:

Consequences of a Failed Plenum Flow Meter during Loss of Flow Accident Document Number: NCNR-RE-TM-000651 Date: 12/12/2024 Page 18 of 27 Appendix A. Analysis for DWV-1s Inadvertent Closure A.1.

Transient Analysis of DWV-1 Closure The NBSR model was leveraged to perform a transient calculation and investigate the consequences of DWV-1s closure. Two different calculations were performed for this analysis:

The 1st calculation is a replica of the analysis in the FSAR for throttling of primary coolant flow to the outer plenum, and it is considered a reference case. In the reference case, we simulate the closure of DWV-1, which reduces the D2O flow rate to the outer plenum. This closure was performed by activating TRIP 402 to be true as its condition is triggered. Figure A 1 shows the reactor control trips and their status from the RELAP5 output. When the number is 0.0 or greater than 0.0 means that the trip condition is triggered, and the TRIP is activated at this certain time. If the value is -1.0 means that the TRIP condition is not triggered, and the trip is not activated yet. In this analysis the trip which controls the NBSR scram were activated at 21.9 s. Note that in this work, the calculation is extended beyond the time range analyzed for the FSAR. This is done to ensure that there are no subsequent failures (i.e., overheating) beyond what the analysis in the FSAR considers.

Figure A 1. NBSR Scram TRIP system when DWV-1 is fully closed Reference at the end of calculation.

The 2nd calculation is considered as the Accident calculation, it simulated the closure of DWV-1 which is controlled by making the condition of TRIP 402 true to initiate the scenario. Also, TRIP 532 which controls the outer Plenum flow was activated too and its condition was triggered. The TRIP 532 is true when the flow rate of the outer plenum is less than 4700 US gpm, but in the model the limit is changed to be 0.0 US gpm. The trip status can be checked from the RELAP5 output (Figure A 2), the TRIP is active when the TRIP condition equal to 0.0 or a number greater than 0.0 and the TRIP is not active when TRIP condition = -1.0, In the accident scenario, the TRIPs (601, 602 and 603) which control the reactor scram are disabled. And only two TRIPs are activated as mentioned before (402 and 532).

Figure A 2. NBSR Scram TRIP system when DWV-1 is fully closed Accident at the end of calculation.

Document Type: Technical Memo Status: Approved Document

Title:

Consequences of a Failed Plenum Flow Meter during Loss of Flow Accident Document Number: NCNR-RE-TM-000651 Date: 12/12/2024 Page 19 of 27 A.2.

Results for DWV-1 Closure As previously mentioned in the Transient calculation subsection, two calculations were performed to investigate the closure of DWV-1.

Reference:

the DWV-1 is closed and the NBSR scrams when the flow of the outer plenum is less than 4700 US gpm.

Accident: the DWV-1 is closed, the FRC-3 flow meter is bypassed, and no scram occurs throughout the accident.

The transient calculation Reference of DWV-1 was mentioned in SAR; the DWV-1 takes approximately 60 seconds to fully close after receiving the closure signal. This gradual closing reduces the mass flow rate that feeds the outer plenum. When the mass flow rate is less than 4700 US gpm, the Reactor Scram occurs.

As the flow level in the outer plenum decreases, the TRIP system activates at ~ 21.9 s, causing the NBSR to scram. After the scram, NBSR power drops from 20 MW to 1.9 MW at 23 s. The power takes ~ 3 s to drop significantly.

The calculation Reference is maintained successfully without any severe/unrecoverable issues. Note that because of the outer plenums flow rate reduction, noticeable increases in the inner plenum flow rate can be observed. It was noticed that the trip of the outlet flow is activated at ~ 36 s when the outlet flow is less than 5900 US gpm.

The Accident scenario occurs when DWV-1 is closed and the flow indicator FRC-3 of the outer plenum is not functional, which allows a scenario that disables the ability of the reactor to scram due to low flow in the outer plenum. The gradual closing of DWV-1 reduces the flow rate in the outer plenum while the power is constant at its maximum value 20 MW. With the reduction of outer plenum flow rate, the fuel plates temperature increases from 401 to 1400 K, which exceeds the blister temperature ~723 K and exceeds the melting point of Aluminum ~ 933 K. Therefore, core melting would likely begin at ~ 30-35 s. The RELAP5 code is not able to calculate after 34 s due to the significant voiding, wherein the cladding gets to temperatures beyond the range accommodated by the thermal properties used by the code. As such, the transient fails to compute beyond 34 s because of the extremely high cladding temperatures reached in the accident.

Table A 1 presents a comprehensive comparison of evaluation results between the "Reference" and "Accident" scenarios for various parameters, including the pressure and mass flow rate of the inner and outer plena, total mass flow rate and inlet pressure. As shown by the model, the minimum flowrate measured at FIA-40 is 8267 US gpm, still well above the scram setpoint of FIA-40. It is therefore concluded that no automatic action will occur with FRC-3 bypassed in this scenario. This comparison provides insights into how these critical parameters vary under the outer plenum throttling accident in the FSAR (Reference) versus during a similar accident scenario without flow indication in the outer plenum (Accident).

Table A 1 Comparison between Reference and Accident calculation results.

Parameter Initial conditions Reference Accident calculation Steady state operation power 20 MW SCRAM occurs at 21.9 s.

Power Reduced to 1.9 MW at 23 s.

Power reduced to 0.9 MW at 50 s No SCRAM occurs Constant power 20.0 MW

Document Type: Technical Memo Status: Approved Document

Title:

Consequences of a Failed Plenum Flow Meter during Loss of Flow Accident Document Number: NCNR-RE-TM-000651 Date: 12/12/2024 Page 20 of 27 Status of DWV-2 Fully opened DWV-1 fully closed at 60 s DWV-1 not fully closed Calculation failed at 34 s At the end of calculations Inlet flow rate 8700 US gpm 4000.0 US gpm 6108.0 US gpm Inner Plenum flow rate 2288 US gpm 4000.0 US gpm 3626.0 US gpm Outer Plenum flow rate 6400 US gpm 0.0 US gpm 2481.1 US gpm Inlet pressure Pump 55.5 psi 96.3 psi 85.5 psi Inner Plenum pressure 41 psi 79.6 psi 68.7 psi Outer Plenum pressure 44.4 psi 21.9 psi 22.9 psi Outlet pressure 21.5 psi 22.5 psi 22.1 psi Outlet flow rate 8700 US gpm 4000.0 US gpm 6108.0 US gpm Figure A 3 provides a detailed comparison of the NBSR's power and reactivity between the "Reference" and "Accident" scenarios. In the Reference calculation, the NBSR undergoes a scram successfully at ~21.9 seconds, causing a sudden drop in power. Reactor power decreases to 0.9 MW at ~50 seconds, within 3 seconds of the scram. Subsequently, the power further decreases to approximately 0.78 MW at the end of simulation (80 s). Concurrently, the reactivity, reflecting the deviation from nominal criticality, initially drops below 0 dollars (indicative of a negative reactivity insertion) immediately after the scram. It then stabilizes at a constant value of -18.00 dollars by 21.9 seconds.

In contrast, the Accident scenario depicts the maximum operational power of the NBSR at 20 MW with a reactivity value of 0 dollars (indicating nominal criticality). This scenario represents the peak power output under normal operating conditions without any reactivity deviations. In other words, this demonstrates that without the flow indication in the outer plenum, the reactor will not automatically scram if DWV-1 invertedly closes.

Document Type: Technical Memo Status: Approved Document

Title:

Consequences of a Failed Plenum Flow Meter during Loss of Flow Accident Document Number: NCNR-RE-TM-000651 Date: 12/12/2024 Page 21 of 27 Figure A 3. Power evolution.

Figure A 4 presents the assessment that evaluates the D2O mass flow rates in both the outer plenum (DWV-

1) and inner plenum (DWV-2) under Reference and Accident scenarios. The closure of DWV-1, which controls the flow to the outer plenum, has a significant impact on these flow rates. During the Reference scenario, the flow rate through DWV-1 decreases gradually as the valve closes, reaching 0.0 US gpm by approximately 60 seconds, indicating that the valve is fully closed. Meanwhile, the flow rate through DWV-2 increases steadily, reaching 4000 US gpm as DWV-1 closes.

In contrast, during the Accident scenario, the simulation fails prematurely at 34 seconds, about 25 seconds before DWV-1 fully closes. Despite the premature failure, the flow rate in DWV-1 remains at approximately 2480 US gpm, while DWV-2s flow rate increases to about 3626 US gpm before the simulation stops. This suggests that the fuel reached a melting point at a relatively low flow rate in the outer plenum, prior to the complete loss of flow. Specifically, with DWV-1s flow rate nearly at 2480 US gpm39% of the current low flow setpoint and about 30% of the nominal flow rateit is evident that falling below 30% of the nominal flow rate in the outer plenum can lead to fuel damage.

Document Type: Technical Memo Status: Approved Document

Title:

Consequences of a Failed Plenum Flow Meter during Loss of Flow Accident Document Number: NCNR-RE-TM-000651 Date: 12/12/2024 Page 22 of 27 Figure A 4. Evolution of mass flow rate.

Figure A 5 shows the pressure of the inner plenum, outer plenum and pump for the Reference and Accident calculation. The inner and outer plena pressure of Reference and Accident calculation have the same behavior. In the case of the References outer plenum pressure, as the DWV-1 is closing the pressure decreases until the DWV-1 is fully closed at ~60 s then the References outer plenum pressure reaches a constant value of ~21.6 psi. In the case of Accident calculation, the outer plenum pressure decreases in an identical fashion to the Reference calculation, until the calculation fails at ~34 s due to the excessive heating and fuel melting. At 34 s, the pressure reaches 22.9 psi, which is consistent with the Reference calculation.

The References inner plenum pressure increases until it reaches ~79.1 psi (after which it settles), while the Accidents inner plenum pressure reaches ~68.7 psi before the calculation fails.

Document Type: Technical Memo Status: Approved Document

Title:

Consequences of a Failed Plenum Flow Meter during Loss of Flow Accident Document Number: NCNR-RE-TM-000651 Date: 12/12/2024 Page 23 of 27 Figure A 5. Evolution of the pressure.

An essential parameter for assessing the Accident scenario is the temperature of the fuel plate cladding (heat structures), as shown in Figure A 6. Initially, the temperature of the Reference fuel plate cladding increases to 390 K before dropping following the scram at approximately 21.9 seconds. The maximum reference temperature for the fuel plates is 401 K. In the Accident scenario, the fuel plate temperature gradually rises over the first 30 seconds, reaching about 425 K. Subsequently, the temperature increases rapidly, peaking at approximately 1400 K before the simulation fails. For context, the blistering temperature of NBSR U3O8 fuel plates is around 723 K, and the melting temperature of aluminum cladding is approximately 933 K.

During the Accident scenario, the cladding temperature significantly exceeds both the blistering and melting temperatures. This indicates that severe damage to the fuel plates occurred, with U3O8 and aluminum likely undergoing relocation in the outer plenum due to the extreme temperatures.

Document Type: Technical Memo Status: Approved Document

Title:

Consequences of a Failed Plenum Flow Meter during Loss of Flow Accident Document Number: NCNR-RE-TM-000651 Date: 12/12/2024 Page 24 of 27 Figure A 6. Cladding temperature evolution.

The temperatures of the hot channels within the outer plenum have been analyzed both in the Reference scenario and the Accident scenario see Figure A 7. In the Reference case, the hot channel temperatures show a distinct pattern following the reactor scram. Initially, temperatures decreased from 342 K to 320 K.

Subsequently, beginning around 35 seconds after the scram, the reactor operates at approximately 0.9 MW power, there is a noticeable rise in temperatures. This increase continues until temperatures reach the boiling temperature of 352 K within the time frame of 55 to 75 seconds. After this peak, temperatures gradually decrease, stabilizing between 340 K, and 335 K after 80 seconds. In contrast in the accident scenario, as the reactor power is stabilized at 20 MW, the temperature started to rise from 342 K until reached 391 K (exceeds the boiling temperature), then it drops as the numerical calculations failed at 34 S.

Document Type: Technical Memo Status: Approved Document

Title:

Consequences of a Failed Plenum Flow Meter during Loss of Flow Accident Document Number: NCNR-RE-TM-000651 Date: 12/12/2024 Page 25 of 27 Figure A 7. Hot channels temperatures.

Figure A 8 illustrates the void fraction in the outer plenum during the accident scenario. Extensive voiding is observed at zero flow, reflecting the axial void distribution in both the lower and upper sections of the outer plenum. The voiding is particularly pronounced due to the zero-flow condition in the upper and lower cores of the outer plenum. The transient nature of the voiding response is highlighted by periodic fluctuations in the axial void profile. The outer plenum appears to be entirely voided. Notably, voiding in the upper section of the outer plenum begins 2 seconds before voiding occurs in the lower section. Table A 2 provides a detailed explanation of each part of the fuel rod and its representation in the nodalization scheme.

Document Type: Technical Memo Status: Approved Document

Title:

Consequences of a Failed Plenum Flow Meter during Loss of Flow Accident Document Number: NCNR-RE-TM-000651 Date: 12/12/2024 Page 26 of 27 Table A 2. outer plenum fuel rod nodalization.

Channel Represent Fuel rod from NBSR nodalization scheme 403 Lower part of the fuel rod 405 Middle part 407 Upper part of the fuel rod 409 Upper end of the fuel plate

Document Type: Technical Memo Status: Approved Document

Title:

Consequences of a Failed Plenum Flow Meter during Loss of Flow Accident Document Number: NCNR-RE-TM-000651 Date: 12/12/2024 Page 27 of 27 Figure A 8. Void fraction of the outer plenum Accident.