Text

University of Texas - Austin - Response to the Request for Additional Information Regarding License Renewal - Redacted
	ML15211A638
Person / Time
Site:	University of Texas at Austin
Issue date:	07/15/2015
From:	Whaley P; University of Texas at Austin
To:	Michael Balazik; Document Control Desk, Office of Nuclear Reactor Regulation
	Hardesty D, NRR/DPR, 415-3724
References
	TAC ME7694
	Download: ML15211A638 (93)
	v • d • e

UNIVERSITY OF TEXAS AT AUSTIN RESEARCH REACTOR LICENSE NO. R-129 DOCKET NO. 50-602 Response to Request for Additional Information Regarding License Renewal Request.

JULY 15, 2015 REDACTED VERSION*

)

Deparnnent of Mechanical Engineering THE UNIVERSITY OF TEXAS AT AUSTIN Nuclear Engineering7iaching Laboratory Austin, Texas 78758

512-232-5370 -FAX512-47.1-zi589-http://ýwuwme.iitexas.edu/-netll SAUS1 July 15, 2015 ATTN: Document Control Desk, U.S. Nuclear Regulatory Commission, Washington, DC 20555-0001 M. Balazik Project Manager Research and Test Reactors Licensing Branch

SUBJECT:

Docket No. 50-602, Request for Renewal of Facility Operating License R-129 REF: UNIVERSITY OF TEXAS AT AUSTIN - REQUEST FOR ADDITIONAL INFORMATION REGARDING THE LICENSE RENEWAL REQUEST FOR THE NUCLEAR ENGINEERING TEACHING LABORATORY TRIGA MARK II NUCLEAR RESEARCH REACTOR (TAC NO. ME7694)

Sir:

Attached are Analysis of Neutronic and Thermal Hydraulicperformanceof the University of Texas TRIGA Mark II Nuclear Research Reactor, and response to referenced Request for Additional Information (RAI) related to items 7, 8.3, 10, 11, 13, 14, 15, and 35. We respectfully request an additional 90 days to support completion of the remaining RAIs.

Please contact me by phone at 512-232-5373 or email whaley@mail.utexas.edu if you require additional information or there is a problem with this submittal.

Thank P. M. Whaley /

Associate Director Nuclear Engineering Teaching Laboratory The University of Texas at Austin I declare under penalty of perjury that the foregoing is true and correct.

Executed on July 15, 2015 R .iegalski Ac* 9(

NETL Director ATT: 1. Request for Additional Information Items 7, 8.3, 10, 11, 13, 14, 15, and 35 with Responses

2. Analysis of the Neutronic and Thermal HydraulicPerformanceof the University of Texas TRIGA Mark II Nuclear Research Reactor, July 2015

Attachment I: Request for Additional Information Items 7, 8.3, 9, 10, 11, 13, 14, 15 and 35 with Responses

RAI 7

The guidance in NUREG-1537 Section 4.5, "Nuclear Design," requests that the licensee provide a detailed description of analytical methods used in the nuclear design, including computer codes used to characterize technical parameters pertaining to the reactor. UT SAR Section 4.5 states that the "characteristics and operating parameters of this reactor have been calculated and extrapolated using experience and data obtained from existing TRIGA reactors as bench marks in evaluating the calculated data." Please provide comprehensive analysis of UT TRIGA behavior. Please describe the methods used for steady state neutronic (steady-state and kinetics) and thermal-hydraulic analysis and include comparisons with UT TRIGA measurements that demonstrate that those methods are appropriate to analyze the limits imposed by the UT TRIGA TS.

Response

The methods used for steady state neutronic (steady-state and kinetics) and thermal-hydraulic analysis of the UT TRIGA reactor are documented in Analysis of the Neutronic and Thermal Hydraullc Performanceof the University of Texas TRIGA Mark II Nuclear Research Reactor,July 2015. The models used in analysis are validated by (1) comparison with experimental data, (2) comparison to comparable analyses, and/or (3) comparison to results using alternate codes.

RAI 8.3

[The guidance in NUREG-1537 Section 4.5.1, "Noirmal Operating Conditions," requests that the licensee define the limiting core configuration (LCC) which defines the highest power densities and temperatures achievable.] Please provide the technical parameters including analysis of "reactor kinetic behavior, basis reactor criticality, control rod worth, definition of the limiting core configuration (LCC), [etc.]" (NUREG-1537, Section 4.5.1). State whether the comparison of calculated and measured values demonstrates acceptable model development.

Response

Reactor kinetic behavior, reactor criticality, control rod worth, and the limiting core configuration are addressed in Analysis of the Neutronic and Thermal HydraulicPerformanceof the University of Texas TRIGA Mark II Nuclear Research Reactor,July 2015. Validation by comparison of calculated and measured values assures acceptable model development.

RAI 9

UT SAR Section 4.5.3 refers to calculations made by General Atomics (GA) and the calculations are said to be applicable to UT TRIGA core parameters because of their similarity. The GA-4361 unit cell parameters are displayed and compared with UT TRIGA core parameters. Please provide the technical parameters that are applicable to UT TRIGA.

Response

Reference to the GA-4361 unit cell parameters is removed, superseded by the Analysis of the Neutronic and Thermal HydraulicPerformanceof the University of Texas TRIGA Mark II Nuclear Research Reactor,July 2015.

RAI 10

UT SAR Table 4.21, "Limiting Core reactivity," displays Reference and Current control rod worths. In Table 4.14, please explain the origin of the values listed under the "Reference" column. Given the difference between the "Reference" and "Current" values of excess reactivity and shutdown margin, which values are being used in the UT TRIGA TS. The reference refers to the initial control rod worths as indicated in the 1992 SAR, as indicated in the description of Table 4.13.

Response

The paragraph preceding the table says "Based on the control rod worth values noted in Table 4.13 [4.14, Control Rod Worth (for the current core configuration)] and calibration data from June 29, 2011, the ability of the control rods to meet the specified limits is demonstrated in Table 4.21. When significant changes to the core configuration are made, verification that the core meets current requirements is accomplished including evaluation that the control rod calibration is valid or reestablishing the cone control rod worth calibration." Therefore the purpose of this table is document that control rod worth under the current core configuration meets requirements, noting the requirement to revalidate if the core configuration changes.

Nevertheless, this information is superseded by Analysis of the Neutronic and Thermal Hydraulic Performance of the University of Texas TRIGA Mark II Nuclear Research Reactor,July 2015.

RAI 11

The guidance in NUREG-1537 Section 4.5.3, "Operating Limits," requests that the licensee define the operating limits for its facility. However, the UT SAR does not provide sufficient information in this regard. Therefore:

(1) Please describe any limits or conditions on the evaluation of excess reactivity contributors, such as those due to temperature variations and poisons (e.g., xenon and samarium).

Please provide calculations of full power reactivity defects for power, xenon, and samarium.

(2) Please provide calculations for excess reactivity and control rod worths, and evaluate whether they are in agreement with the analytical model and with UT TRIGA performance.

Provide a discussion that describes the evaluation of these calculations to demonstrate acceptable reactor shutdown and shutdown margin. Include consideration of experiment reactivity.

(3) Please provide "a transient analysis assuming that an instrumentation malfunction drives the most reactive control rod out in a continuous ramp mode," (NUREG-1537, Section 4.5.3) of the reactor using a rate of withdrawal consistent with proposed UT TRIGA TS values of the maximum control rod withdrawal speed, reactivity rate, and the control rod scram time including uncertainties.

(4) Please provide all other applicable technical parameters, "excess reactivity, control rod worth, temperature coefficients, [etc.]" (NUREG-1537, Section 4.5.3).

Response

(1) There are no limits or conditions required related to excess reactivity contributors.

(2) Calculations and comparison of analytic models to observed parameters are provided in by Analysis of the Neutronic and Thermal Hydraulic Performanceof the University of Texas TRIGA Mark II Nuclear Research Reactor,July 2015. Experiment limits are established to maintain potential transients within the limits of a pulse at the maximum reactivity addition.

The addition of experiments with negative reactivity is not credited in shutdown margin calculations as contributing to shutdown margin. The removal of experiments with positive reactivity associated with experiments is not credited in shutdown margin calculations as contributing to shutdown margin.

(3) There is no reactivity transient possible with the control rod drives that would exceed the reactivity transient from a reactor pulse.

Nevertheless, methodology based on MATLAB SIMULINK in an RAI response by the DOW TRIGA reactor was implemented using NI LabVIEW Design and Simulation Suite (figure following) with minor modifications:

Initial neutron generation time based on reactivity, (2)

" Specific heat capacity taken from GeneralAtomics E-117-833 The U-ZrHx Alloy. Its Propertiesand Use in TRIGA Fuel- Simnad, 1980,

LabVIEW implementation of Runge-Kutta as a variable time solver,

Termination of rod withdrawal at scram, and

" Reactivity insertion as a ramp function.

Since the maximum historical control worth at the UTTRIGA was slightly more than $4, integral rod worth was used as $5. Scram time was set to occur at 60 seconds to avoid mitigating interaction. Control rod speeds were varied from 0.1% $Ak/k-s to 1.0% Ak/k-s in 0.1% increments.

The results are documented in Analysis of the Neutronic and Thermal Hydraulic Performanceof the University of Texas TRIGA Mark II Nuclear Research Reactor,July 2015.

As expected, peak power and peak temperature were bounded by maximum reactivity addition in pulsing analysis. Maximum fuel temperature for the fastest addition rate was 435 0 C at the maximum reactivity addition rate, below the scram setpoint.

(4) The specified technical parameters are documented in Analysis of the Neutronic and Thermal HydraulicPerformanceof the University of Texas TRIGA Mark II Nuclear Research Reactor, July 2015.

Imp~~

E3-..

F4 "

~'

40ig~

.9--

--- cc~

-. =L1111 41?-

~S'9'

. ......

rL* *...

U ~UL... i

RAI 13

UT SAR Section 4.5.4, Subsection C describes the use of the Bernath correlation which is typical for TRIGA analysis, but later UT SAR Section 4.6 states that the Biasi correlation is actually used for the evaluation of the DNBR. Please confirm that the Bernath correlation is used to characterize DNBR for UT TRIGA, or demonstrate the applicability of Biasi correlation to UT TRIGA analysis.

Response

Thermal hydraulic analysis was updated using the Bernath correlation and TRACE, see Analysis of the Neutronic and Thermal Hydraulic Performanceof the University of Texas TRIGA Mark II Nuclear Research Reactor,July 2015.

RAI 14

The guidance in NUREG-1537 Section 4.5.2, "Reactor Core Physics Parameters," states that the supplied analysis should show that reactivity coefficients are sufficiently negative to prevent or mitigate any reactor transients. UT SAR Section 4.6 states, "Limits on reactivity are based not on the peak pulse power level, but on the final equilibrium power level associated with the reactivity." It also provides a series of statements on page 4-46 regarding pulse reactivity acceptability.

(1) The analysis provided in UT SAR, Section 4.6 is based upon Figure 4.21 which is said on page 4-34 (UT SAR) to display the fuel element reactivity coefficient. This figure does not display the fuel element reactivity coefficient. A version of this coefficient, believed to be from GA-7882, is displayed in Figure 4.2B. However, the applicability of this figure to UT TRIGA has not been established (e.g., there is no consideration of power level, burnup, etc.). Please provide clarification as to the relationship of the reactivity coefficient with the analysis provided in UT SAR Section 4.6.

(2) The basis for TS 3.2 "Pulsed Mode Operation," states that the reactivity limits are established so as to meet fuel temperature limits. However, this is inconsistent with the statements in UT SAR Section 4.6 as described above. Please revise the discussion in UT SAR Section 4.6 to support the UT TRIGA TS.

(3) UT SAR, Section 4.6 (p. 4-46) provides a series of statements regarding pulse reactivities and responses that are not supported by analyses. Please provide analysis supporting these statements in sufficient detail so that a confirmatory analysis can be performed.

Response

(1) The Ul-TRIGA fuel temperature coefficient was calculated in Analysis of the Neutronic and Thermal Hydraulic Performanceof the University of Texas TRIGA Mark II Nuclear Research Reactor,July 2015, and used in analyses.

(2) The statement in Section 4.6 is incorrect, and will be revised to agree with the Analysis of the Neutronic and Thermal HydraulicPerformanceof the University of Texas TRIGA Mark II Nuclear Research Reactor,July 2015.

(3) Section 4.6 will be revised to be consistent with See Analysis of the Neutronic and Thermal HydraulicPerformance of the University of Texas TRIGA Mark II Nuclear Research Reactor, July 2015.

RAI 15

The guidance in NUREG-1537 Section 4.6, "Thermal-Hydraulic Design," requests that the licensee provide information and analyses of thermal-hydraulic conditions in its reactor demonstrating that sufficient heat removal capacity exists for steady-state or pulsing operation at the maximum licensed power level. The fuel element and coolant conditions should be conservatively calculated, the DNBR correlation should be properly defined, and the resulting steady state DNBR at licensed power should be greater than 2.0.

(1) UT SAR Section 4.5.4 states that the TRACE analysis of the UT TRIGA DNBR used the Bernath correlation and determined that the DNBR is 5.9 for a 45 kW fuel element occurring at an elevation of 87 percent of the active fuel element height. However, typical analysis of TRIGA DNBR indicate that the power corresponding to DNBR of 1 is approximately 50 kW.

Please review and revise these calculations, or provide the detailed results that confirm the calculated DNBR of UT TRIGA.

Please describe the analytical methods used to determine the DNBR, including the core inlet and exit conditions assumed and other assumptions and correlations employed. This analysis should describe the parameters determined from the LCC such as peaking factors and limiting coolant inlet temperature and that the inlet temperature used for DNBR is a limiting value by showing how it corresponds to the primary pool water temperature measuring channel value.

(2) NUREG-1537, Section 4.6 describes that the licensee provide detailed "analyses for a pulsing reactor containing descriptions of the core configurations; the bases of the feedback coefficients; the calculational model and assumptions; the thermal-hydraulic evolution during a pulse; core, transient rod, and fuel characteristics that determine the shape and magnitude of a pulse; and the safety considerations that establish limits to pulse sizes."

The UT SAR Appendix 4.1 provides a discussion of phenomenology; however, it does not demonstrate the acceptability of pulsing in UT TRIGA using parameters of importance that are demonstrated to be fully applicable to UT TRIGA.

Please provide a comprehensive description of the calculational methods and the results that demonstrate the acceptability of design assumptions and TS for pulsing at UT TRIGA (e.g., the LCC, the approved power level, the pulse of reactivity inserted by the transient rod as allowed by TS, the value of the fuel temperature coefficient, the effective delayed neutron fraction, the prompt neutron lifetime, etc.). The analysis should demonstrate reactor overall behavior (e.g., power vs. time and fuel element temperature vs. time, etc.)

and compliance of the leading fuel element in the LCC to safety limits (SL) as stated in the TS.

Response

(1) and (2) Sections 4.5 and 4.5 will be revised to be consistent with See Analysis of the Neutronic and Thermal Hydraulic Performanceof the University of Texas TRIGA Mark II Nuclear Research Reactor,July 2015.

RAI 35

The guidance in ANSI/ANS-15.1-2007 Section 3, "Limiting conditions for operations," provides guidance and recommendations for the specifications pertaining to the limiting conditions for operation (LCO). This guidance is supplemented by NUREG-1537 Appendix 14.1. Some deficiencies and differences with the proposed UT TRIGA TS are described below. Please discuss these deficiencies and differences and revise accordingly.

(1) Section 3.1 of the guidance describes having specifications for fuel burnup, core configurations, and reactivity coefficients (if such coefficients establish required conditions).

Such specifications are not present in the proposed UT TRIGA TS.

(2) Section 3.1 of the guidance describes that limits be placed on core excess reactivity and the corresponding regulatory interpretation includes provisions that account for experiment worth. Proposed UT TRIGA TS 3.1 "Core Reactivity," Specification A excludes consideration of experiments having positive reactivity.

(3) Section 3.1 of the guidance describes that limits be placed on the shutdown margin and states that this value "should be large enough to be readily determined experimentally, for example, ->0.5%Ak/k or ->0.50dollar." Please provide an analysis and evaluation that demonstrates the ability to repeatedly measure core reactivity with sufficient accuracy to justify this small value of the shutdown margin.

(4) Section 3.2 of the guidance describes that a limit be established for the maximum control rod reactivity insertion rate for non-pulsed operation. The proposed UT TRIGA TS do not provide such a specification. This rate, and the control rod scram times, are typically justified through the analysis of an uncontrolled, control rod withdrawal transient.

(5) Section 3.2 of the guidance describes a specification for permitted bypassing of channels for checks, calibrations, maintenance, or measurements. Proposed UT TRIGA TS 3.3, "Measuring Channels," does not specify when it is permitted to bypass channels for checks, calibrations, maintenance or measurements.

(6) Proposed UT TRIGA TS 4.3 "Measuring Channels," contain Surveillance Requirements for the Fuel Temperature Channel and the Upper Level Radiation Monitor. However, there are no associated LCO specifications.

(7) Section 3.3 of the guidance describes specifications for leak or loss of coolant detection and a secondary coolant activity limit. No such specifications are found in the proposed UT TRIGA TS.

(8) Section 3.8.2 of the regulatory interpretations states that containers for experiments containing known explosive materials shall be designed such that the design pressure of the container is twice the pressure the experiment can potentially produce. Proposed UT TRIGA TS 3.6 "Limitations on Experiments," does not include such a specification.

Response

(1) No specified coefficients establish required conditions.

(2) In absolute terms, experiments with negative reactivity increase actual shutdown margin, but experiments with positive reactivity reduce actual shutdown margin. Experiments at UT are not credited in such a way as to reduce shutdown margin. This is conservative.

(3) Shutdown margin calculations are based on control rod reactivity worth. Control rod reactivity worth is based on a fit to a curve using data points from individual, independent measurements of time-dependent response to discrete control rod movements. Typical errors for individual data points in the curve fit are less than 1%, with a nominal maximum of 1-2% error.

Experience has shown reactivity balances to be repeatable a few cents of reactivity when considering comparable experiment configurations and burnup. Reactivity balance data for the first quarter of 2105 was collated and the data compared, including reactivity balance data (1) on sequential days of operation with shutdown periods long enough to allow significant xenon decay following power operations, and (2) sequential startups at low power where temperature was not a factor.

REACTIVITY BALANCE DATA, 1 ST QUARTER 2015 DATE $ A$

03/16/15 -0.50052 0.01297 03/13/15 -0.48755 03/06/15 -0.19882 03/04/15 -0.18428 02/26/15 -0.53977 02/24/15 -0.55564 02/19/15 -0.5396 02/17/15 -0.50962 02/12/15 -0.61461 0.02238 02/10/15 -0.59223 02/03/15 -0.19229 0.00684 02/02/15 -0.18545 -0.02301 01/30/15 -0.20846 01/26/15 -0.17187 0.01457 01/26/15 -0.1573 01/16/15 -0.14302 01/15/15 -0.11108 Therefore the $0.29 limit, as currently approved for the UT TRIGA reactor, is appropriate.

(4) The "maximum control rod reactivity insertion rate for non-pulsed operation" and "control rod scram times" are not credited in analysis.

(5) Channels are not bypassed during operation for checks, calibration, maintenance or measurements. Automatic actions occur during pulse mode, including: (1) the gain of the NPP 1000 changes to capture the larger signal from the pulse, (2) the NP 1000 is bypassed, and (3) the NM 1000 is shifted a calibration mode (Mode 7). When exiting pulse mode these changes are automatically reversed.

(6) Fuel temperature has a limiting safety system setting which requires a measuring channel.

The measuring channel has an operability requirement. Area radiation monitors are required to support personnel protection.

(7) The design of the secondary (chill water) coolant system is engineered to prevent pool to chill water leaks. Technical Specifications requires differential pressure that prevents such leakage. If the requirement is not met, Technical Specifications requires correction or shutdown and isolation of the chill water.

(8) The design and review occurs prior to operation. The design of experiments is incorporated in "Design Features," Section 5.4, "Experiments." Experiments that do not meet design requirements are not installed in the reactor, and are therefore not subject to limits on operations. The design features of experiments do not have surveillance requirements except as established though experiment approval (with the exception of reactivity limits) which is an administrative requirement. The Technical Specifications, Section 6.4 requires an experiment review, and the design criteria for encapsulation of known explosive material will be added to Section 5.4.

ATTACHMENT II ANALYSIS OF NEUTRONIC (PART I) AND THERMAL HYDRAULIC (PART II) PERFORMANCE OF THE UNIVERSITY OF TEXAS TRIGA MARK II NUCLEAR RESEARCH REACTOR LICENSE R-19 DOCKET 50-602 SUBMITTED TO THE NRC IN SUPPORT OF THE UT TRIGA LICENSE RENEWAL JULY 15, 2015

NEUTRONIC ANALYSIS OF THE UNIVERSITY OF TEXAS (UT) TRIGA REACTOR TABLE OF CONTENTS 1.0 Purpose and Discussion 1 2.0 SCALE MODEL 1 2.1 Materials 3 2.1.1 General 3 2.1.2 UZrH Densities 4 2.1.3 Fission Product Density 6 2.1.4 Control Rod Poison Density 7 2.2 GEOMETRY 7 2.2.1 General 7 2.2.2 Reflector and Core 7 2.2.3 Fuel Elements and Control Rods 8 3.0 Calculations 10 3.1 Initial Core 11 3.1.1 Materials 11 3.1.2 Initial Startup 13 3.1.3 Operational Loading 13 3.2 Configuration Changes Over Core Life 15 3.3 Reactivity Calculations 16 3.3.1 SCALE Calculations 17 3.3.2 Reactivity Measurements 18 3.3.3 Reactivity Calculations and Reactivity Surveillance Data 19 3.4 Fuel Temperature 21 3.4.2 SCALE and TRACE Calculations for Power Distribution 21 3.4.3 Core Fuel Temperature Based on Measurement 24 3.4.4 Comparison 24 3.6 Fuel Temperature Reactivity 25 3.7 Pulsing 26 4.0 RESULTS 30 4.1 Nuclear Data 30 4.1 Peaking Factors 31 4.2 Fuel Temperature 32 4.3 Control Rod Worths 34 4.4 Pulsed Reactivity Response 34 4.5 Control Rod Speed 36 5.0 Conclusion 36 Appendix 1: Fuel Element Core Locations

NEUTRONIC ANALYSIS OF THE UNIVERSITY OF TEXAS (UT) TRIGA REACTOR LIST OF TABLES Table 1: Material Specifications: Standard Composition Library Values 3 Table 2: Standard Added Nuclides 6 Table 3: Unit Cell Structure: Geometry 8 Table 4: Component Dimensions 8 Table 5: Standard Fuel Element Unit Model 9 Table 6: Fuel Follower Control Rod Unit Model 10 Table 7: Transient Rod Unit Model 10 Table 8: Summary: Initial Core Fuel Element Groups 12 Table 9: Core Configurations and Burnup 16 Table 10: T-6 Burn Parameters 16 Table 11: Calculated Reactivity Values 17 Table 12: Reactivity Surveillance Data 18 Table 13: Comparison of Reactivity Calculations to Surveillance Data 19 Table 14: Observed Fuel Temperature Data 21 Table 15: Average and Monitored Fuel Temperatures (°K) 25 Table 16: Comparison Measuring Channel to Calculated Temperatures 25 Table 17: Pulse Data 26 Table 18: LCC Reactivity Values 34 Table 19: LCC Response to Pulsing from Power 35 Table 20: Simulation of Continuous Control $5 Rod Withdrawal 36

NEUTRONIC ANALYSIS OF THE UNIVERSITY OF TEXAS (UT) TRIGA REACTOR LIST OF FIGURES Figure la: Top View 7 Figure 1b: Isometric 7 Figure 1c: Core and Air Spaces 7 Figure 2: Standard Fuel Element 9 Figure 3: Fuel Follower Control Rod 9 Figure 4: Transient Rod 10 Figure 5: Core Positions 11 Figure 6: Initial Criticality 14 Figure 7: Operational 1992 Core 14 Figure 8: 114 Core Peaking Factors 22 Figure 9: Radial Peaking Factor 22 Figure 10: Axial Peaking Factor 23 Figure 11: Monitored and Average Fuel Temperatures for Channel Power 23 Figure 12: Calculated Core Average and FT2/FT2 Temperatures: 114 Element Core 24 Figure 13: Reactivity and Average Fuel Temperature 26 Figure 14: Correlation of Current 114 Element Core Average Fuel Temperature to Measuring Channel Temperature 28 Figure 15: Pulsed Power Levels 29 Figure 16: Pulsed Fuel Temperatures 30 Figure 17: LCC Excess Reactivity and Fuel Temperature 31 Figure 18: LCC Peaking Factors 80 Element Core 31 Figure 19: LCC Axial Peaking Factors 32 Figure 20: LCC Radial Peaking Factors 32 Figure 21: LCC Correlations of Fuel Temperatures and Element Power 33 Figure 22: LCC Fuel Temperatures at Core Power for the 80 Element Core at Limiting Pool Level and Temperature 33 Figure 23: LCC 80 Element Core Pulsed from Power 36

THERMAL HYDRAULIC ANALYSIS OF THE UTTRIGA REACTOR TABLE OF CONTENTS 1.0 Introduction 2 2.0 General Description of Heat transfer at the UTTRIGA 2 3.0 Power Distribution 3 3.1 General 3 3.2 Criticality Calculations 4 3.3 Power Distribution within Fuel Elements 6 4.0 Thermal Hydraulic Modeling: Unit Cell Geometry and Thermal Hydraulic Characteristics 8 4.1.1 Unit Cell Geometric Parameters 9 4.1.2 Unit Cell Thermodynamic Loss Factors 11 4.2 Physical UTTRIGA Thermal Hydraulic Model 13 4.2.1 Fluid System Component Modeling 14 4.2.2 UTTRIGA Application 15 5.0 Model Validation 20 5.1 Temperature Calculations 20 5.1.2 Operating Data 20 5.1.4 Comparing FT1 and FT2 Measurements to UTTRIGA Model Calculations 21 5.1.3 FT1 and FT2 Comparisons 23 5.1.5 Summary 24 5.2 Comparison to Reference ThermalHydraulic Analysis 24 5.2.1 Coolant Flow Rates 25 5.2.2 Critical Heat Flux Ratio Calculations 25 5.2.3 Summary 26 6.0 Results 27 6.1 Fuel Element Power at Maximum Core Power 27 6.1.1 Core Configurations and keff 27 6.1.2 Maximum Fuel Element (Hot Channel) Power 28 6.2 Fuel Temperature 29 6.2.2 Fuel Temperature Measuring Channel Protective Action 30 6.2 Critical Heat Flux 32 7.0 Limiting Core Configuration 33

THERMAL HYDRAULIC ANALYSIS OF THE UTTRIGA REACTOR LIST OF TABLES Table 1: Geometry for Fuel Segments 6 Table 2: Fuel Element Axial Peaking Factors 8 Table 3: Summary of Principle Thermal Hydraulic Values 10 Table 4: Channel End Geometry 12 Table 5: K Classical factors 12 Table 6: K Factors 13 Table 7: Pressure Boundary Condition 16 Table 8: Down-comer Pipe 16 Table 9: Connecting Pipe 16 Table 10: Specifications for Unit Cell Flow Channel 17 Table 11: TRIGA Fuel and Zirconium Material Properties .19 Table 12: Operating Data 21 Table 13: Fuel Temperature Data 22 Table 14: Limiting Temperatures 31 Table 15: LCC B Ring Peaking Factors 35

THERMAL HYDRAULIC ANALYSIS OF THE UTTRIGA REACTOR LIST OF FIGURES Figure 1: Excess Reactivity and Average Rod Power 5 Figure 2: Maximum Channel Power at 1210 kW 5 Figure 3: Peak to Average Power for 114 Element Core at 600 'K 7 Figure 4: 114 Elements 600 *K: B04 Radial Power Distribution 7 Figure 5: 114 Elements 600 °K: B04 Axial Power Distribution 8 Figure 6: Flow Channel for UT TRIGA 9 Figure 7: Moody Diagram 13 Figure 8: TRACE Model 15 Figure 9: Cold Leg to Flow Channel Connector 16 Figure 10: Fuel Element Temperature Radial Profile 20 Figure 11: Fuel Element Temperature Axial Profile 21 Figure 12: FT2 Temperature Response to Core Power Level 22 Figure 13: Comparison of Calculated Flow Rates for UTTRIGA and Reference Calculation 25 Figure 14: Comparison CHFR for Reference and UTTRIGA Model 27 Figure 15: Criticality Considerations 28 Figure 16: Maximum Element Power for 1210 kW Core (MCNP & SCALE at 300 & 600°K) 28 Figure 17: Fuel Temperatures as a Function of Element Power Level 29 Figure 18: Fuel Element Temperature Profiles for Selected Element Power Levels 30 Figure 19: Maximum & Monitored Fuel Temperatures at Maximum Element Power 31 Figure 20: B03/FT1 Effects 32 Figure 21: TRACE Critical Heat Flux: Limiting Pool Conditions 32 Figure 22: Highest Calculated Hot Channel Power 33 Figure 23: Limiting Core Configuration Hot Channel Critical Heat Flux Ratio 34 Figure 24: LCC Hot Channel Radial Temperature Profiles (Upper: Lower & Mid Heated Length) 34 Figure 25: Limiting Core Configuration Hot Channel Axial Temperature Profiles (Heated Length) 35

. PART I ANALYSIS OF NEUTRONIC PERFORMANCE OF THE UNIVERSITY OF TEXAS TRIGA MARK II NUCLEAR RESEARCH REACTOR

PART I NEUTRONIC ANALYSIS OF THE UNIVERSITY OF TEXAS (UT) TRIGA REACTOR 1.0 Purpose and Discussion Neutronic analysis evaluates operational characteristics of the reactor over the license term. Analysis is validated by benchmarking the model against historical performance from the initial core load to current values.

Analysis of the UT TRIGA reactor nuclear characteristics was performed using the SCALE code package 1 ,

originally developed for the U.S. Nuclear Regulatory Commission, with some SCALE calculations supported by MCNP calculations. SCALE is an integration of codes in a plug and play framework that provides problem-specific calculations. SCALE includes deterministic and Monte Carlo radiation transport solvers, nuclear data libraries, and various computational and data-processing modules that process and integrate data shared between code routines and options.

Depletion and criticality calculations are performed for the UT TRIGA reactor. These calculations use the KENO Monte Carlo code for neutron transport, with the results of the KENO calculations used in ORIGEN for depletion/buildup calculations.

Fission and neutron absorption during reactor operation results in fissionable isotope depletion and buildup of transuranic isotopes and fission products. Depletion calculations (SCALE TRITON sequence) characterize material composition of the fuel from initial values to the composition at end of fuel life. Benchmarking is based on composition at specific core burnup values. The SCALE TRITON sequence requires multigroup cross-section libraries, which requires additional processing to correct for self-shielding.

Criticality calculations (CSAS6 sequence) are used to determine keff values for specific material compositions (with modifications by depletion calculations). The UTTRIGA SCALE model is configured to allow different control rod configurations; calculated keff values are used to determine excess reactivity and the integral (full-in to full-out) reactivity worth of individual control rods. The SCALE criticality sequence calculation can use either multigroup or continuous energy libraries, with continuous energy library calculations preferred.

Calculations of integrated neutroinc and thermal hydraulic performance required the use of data from TRACE, addressed separately.

2.0 SCALE MODEL SCALE model development requires standardized input specifications for materials and physical geometry. The SCALE Standard Composition Library (SCL) defines material constants and isotopic composition for elements, specific mixtures, and compounds. The SCALE Generalized Geometry Package (SGGP) provides a standard method for developing a geometric model used in the sequences.

1 Scale: A Comprehensive Modeling and Simulation Suite for Nuclear Safety Analysis and Design, ORNL/TM-2005/39, Version 6.1, June 2011 (ORNL Radiation Safety Information Computational Center, code CCC-785) 2

PART I 2.1 Materials 2.1.1 General Material density in SCALE is based on standard material composition libraries, in units of either atomic density (atoms/cm3 normalized to 1 barn), or mass density (g/cm 3 ). A user specified "density multiplier" adjusts from the standard library density to a non-standard specification. Standard composition library and density multiplier values for the UT TRIGA calculations are provided in Table 1.

SELF-SHIELDING CALCULATIONS Neutron interaction is a function of material density and cross-section. The SCALE TRITON module is restricted to multigroup (energy-average) cross-sections, while the CSAS6 module is capable of using either multigroup or continuous energy cross-sections. Resonance self-shielding is not well characterized in multigroup calculations, and processing routines account for self-shielding. Flux varies spatially, and it is necessary that each material in each physical structure be uniquely identified for self-shielding correction routines. For example, although the cladding is essentially identical for all fuel elements, individual material specifications are required for each element to support self-shielding corrections. A single material specification is adequate for a continuousenergy calculation.

Table 1: Material Specifications, Standard Composition Library Values Material ID Density SCL SCL Density Reference Multiplier Density Zirconium 1011-1127 at121 6.49 M8.2.2, Isotopes and Their natural alt. 12 Abundances h2o 3 2011-2127 1 0.9982 M8.2.4, Compounds alt. 2' ss304S4 311-427 1 7.94 M8.2.5, Alloys and Mixtures alt. 31 graphite 4 1 2.3 M8.2.4, Compounds Aluminum 5 1 2.702 M8.2.2, Op. Cit.

dry-air alt. 61 1 1.2E-3 M8.2.5, Alloys and Mixtures h (h 2 ) 611-727 2.02E-3 1 M8.2.2, Op. Cit.

b4c 8 1 2.52 M8.2.4, Compounds mo 9 1 10.2 M8.2.2, Op. Cit.

h-zrh2 6 11-127 0.104-0.078 1 M8.2.3 Elements and Special nuclide Symbols zr-zrh2 11-127 5.904-4.431 1 M8.2.3, Op. Cit.

2 Continuous energy calculations; group energy libraries requires material specifications for resonance treatment 3 Includes S(ct,3) scattering kernels 4 C (0.08%), Si (1.0%), P31 (0.045%), Cr (19.0%), Mn55 (2.0%), Fe (68.375%), Ni (9.5%)

5 C (0*1026%), N14 (76.5081%), 016 (23.4793%)

6 Hydrogen in zirconium hydride with S(a,p) thermal kernel 3

PART I CORE MATERIAL LABELING Unit cells are defined in all core positions where fuel might be used, consisting of the fuel element and contiguous spaces or structures. As noted above, self-shieldingcalculations require unique identification for every material in a geometric unit where self-shielding is important. Although zirconium, air, stainless steel, and water in and surrounding a fuel element can be represented by single material specifications in continuous energy calculations (material indices 1, 2, 3, and 6 respectively),

multigroup calculations require unique specification for each material in each fuel element. For convenience, labels for materials in multigroup calculations are correlated with the fuel material identification in the element. For example, the element with fuel labeled as material 11 has zirconium fill rod material number 1011, water material number 2011, stainless steel material number 311, and hydrogen material number 611. Labeling is summarized in Table 1.

2.1.2 UZrH Densities The initial of mass 231U and 23S U-enrichment for each TRIGA fuel element received at the UT facility is documented in special nuclear material records to have a range of grams of 235 U with enrichments ranging from 19.58% to slightly over 20%. Nominal value for TRIGA fuel uranium loading is 8.5%, but the mass in each fuel element but there is no specific assay value for (1) zirconium hydride mass, (2) actual fuel loading, or (3) total ZrH-U. There is no record of dimensional values for individual fuel elements, machined to fit cladding. Cladding is a commercially manufactured product with standardized engineering tolerances. Fuel expands slightly during reactor operation until it is well bonded to cladding, distributing mass across the available volume. Therefore nominal cladding dimensions are used to determine fuel volume, and nominal uranium loading to determine fuel mass.

The cladding of the fuel in fuel follower control rods is smaller than standard fuel elements, with correspondingly different volumes. The instrumented fuel element (IFE) configuration is machined to accommodate thermal couples in the fuel meat. IFE fuel volume is on the order of a few per-cent lower than standard fuel elements.

The UZrH fuel is manufactured as three in ( cm) cylinders with a central in ( cm) hole.

After processing to increase hydride content, the central hole is filled with a zirconium rod. Outer diameter of a standard fuel element is in ( cm), with the fuel follower diameter in

( cm). The volume of the three fuel cylinders isgiven by equation 1:

V = Lz. (. - r2 ) 1 Where L is the total length ( cm), ro is the outer radius, and ri the inner radius. Standard fuel elements therefore have a volume of cm 3, and FFCR fuel has a volume of cm 3 . However, the edges of the fuel are chamfered at a radius of in ( cm), reducing the as-built volume slightly. Since (noted above) fuel expands during power operation the volume of the fuel is characterized as if the chamfers were not present.

Because of variation in composition (indicated above) and variation in fuel element power histories prior to use in the UT TRIGA located at the NETL (discussion to follow), fuel element materials were modeled 23 238 for each element. Since assay values are available for the mass of 1U and enrichment, 23'U and U density is calculated directly as mass distributed over fuel volume. The SCALE standard library density for isotopes is I g/cm 3, the density multiplier therefore is the calculated density. The density of the 4

PART I uranium 235 component in the fuel matrix is the mass (from assay value) distributed over the volume of the fuel, where the volume of the fuel is based on nominal dimensions (equation 2):

m232 P235,Fuel -- 23 VFuel 235 The density of the uranium 238 component in the fuel matrix is determined by the U mass and the enrichment (equation 3):

__M3

_ m235 P238,1;icl E25-VFuel Simnad 7 notes density of uranium of 19.07 g/cm 3, with the density of pure ZrH for H:Zr ratios 1.6 (ratio represented by A) or greater (equation 4):

PZn - O. 1706 + 0.0042. X ForX=1.6, PzrH=5. 6 40 g/cm 3; for X=1.7, PzrH=5. 6 2 6 g/cm 3. Although the total mass of ZrH in each fuel element is not known explicitly, the ZrH density and volume in a fuel element can be used to determine the mass (equation 5).

71 l ZrH,Fui = PZrHFuel ' VFuel With the volume of the uranium distributed across the fuel element given by (equation 6):

V = M6 Pu The mass of the ZrH in the fuel is calculated as (equation 7):

nlZrHFuel = PZrH J'§7uel ) PýrH 7 V

KFuel

( u Pu)

Finally, the density of the ZrH distributed in the fuel element is given as (equation 8):

VFuel Vit,,ie =Az "1 P, *Vr,,,, )

Pu VFV The mass and enrichment of 235U are known from assay, leading to density calculation (equation 9):

7 GA-117- 833 The U-ZrH Alloy: Its Propertiesand Use in TRIGA Fuel, Simnad (1980) 5

PART I PZrH Fuel = P7,H r iIE3 PU

Vue, )

9 Using ZrH ratio and assay values, the density of ZrH in the fuel matrix is calculated (equation 10a):

"OZrH.uel = 0.1706 + 0.0042" X pu *Vuei ' ia The mass ratios of Zr and H in ZrH are used to separate the density of the ZrH in the fuel matrix into components of Zr and H (equation lob), where Xrepresents the ratio of Hto Zr:

91.224 and X. 1.00794 91.224 + X. 1.00794 91.224 + X. 1.00794 lob Mass densities for 235U, 238U, ZrH, Zr, and H in fuel were calculated for individual fuel elements.

However, groups of elements in the initial 1992 core were sufficiently similar in mass, enrichment, and burnup that the total number of material data sets was reduced from 87 to 40, discussion to follow.

2.1.3 Fission Product Density The SCALE TRITON sequences automatically include specified fission product nuclides (in the ADDNUX subroutine) that will be tracked in coupled depletion (ORIGEN) and transport (KENO) calculations 8. A large fraction of moderation in TRIGA reactors is attributed to hydrogen in the fuel. Scattering properties for hydrogen in zirconium hydride are substantially different than the elemental hydrogen and zirconium cross-sections. Therefore, hydrogen cross-sections in TRIGA fuel are specified as hydrogen in zirconium hydride (and zirconium specified as zirconium in zirconium hydride) with appropriate cross-section modification to simulate the UTTRIGA neutron physics. Some stable zirconium isotopes in TRIGA fuel are also high-yield fission products, which may not be well represented by the Zr in ZrH data. Isotope cross-sections can only be used once in ORIGEN so that using standard composition libraries for zirconium-hydride in conjunction with the zirconium implicitly included with ADDNUX will result in a fatal conflict in ORIGEN. It is therefore necessary to disable the automatic nuclide addition in SCALE depletion calculations with TRIGA fuel, and manually specify fission product nuclides that exclude stable zirconium fission products. Fission product contribution to total zirconium inventory is not large, and this approach is not expected to significantly affect calculations.

Table 2: Standard Added Nuclides U-235 U-238 U-234 U-236 B-10 B-11 N-14 0-16 1-135 1-127 1-129 Np-237 Pu-238 Pu-239 Pu-240 Pu-241 Am-241 Cm-242 Pu-242 Am-243 Cm-243 Nb-93 Tc-99 Sn-126 Xe-131 Cs-133 Cs-134 Cs-135 Xe-135 Cs-137 Nd-143 Pr-143 Ce-144 Nd-145 Nd-146 Nd-147 Pm-147 Sm-147 Pm-148 Pm-149 Sm-149 Sm-150 Eu-151 Gd-152 Eu-153 Eu-154 Gd-154 Eu-155 Gd-155 Gd-156 Gd-157 Gd-158 Gd-160 Cm-244 Kr-83 Mo-95 Nd-148 Sm-151 Sm-152 Nb-95 Zr-95 Mo-97 Mo-98 Mo-99 Xe-133 La-139 Ba-140 Ce-141 Pr-141 Ce-142 Ce-143 Nd-144 Sm-153 Eu-156 Zr-93 8 SCALE Manual T1.3.7.5 6

PART I The standard added nuclide sets in ADDNUX were developed for SCALE (Table 2) to represent the fission products with significant interaction probabilities. There are additional sets of fission product nuclides available to incrementally improve calculations, but calculations performed with the additional nuclides for both fresh core material composition and a heavily burned core showed insignificant absorption rates for the additional nuclides.

2.1.4 Control Rod Poison Density The B4 C poison schematic (TOS2508226.C) specifies a minimum of 2.48 g/cm 3. Standard composition library density for B4C is 2.52 g/cm 3. User specified density is therefore 98.4% of SCALE standard density.

2.2 GEOMETRY 2.2.1 General SCALE geometry is based on nesting standard geometric shapes that contain standard materials, with complex geometries specified by excluding intersecting shapes. A collection of assembled geometric shapes is labeled as a "unit." SCALE standard geometries used in this model include cylinders (specified by radius, upper elevation, and lower elevation), hex-prisms (specified by the inner radius, upper elevation, and lower elevation), and truncated cones (specified by upper radius, upper elevation, lower radius, and lower elevation). Lattice structures are represented as specific geometric shapes positioned in an array format. The UTTRIGA model is displayed in Fig. la-ic, with graphite (reflector), water, and aluminum removed in Fig. 1c.

Figure la: Top View Fig 1b: Isometric Fig 1c: Core and Air Spaces 2.2.2 Reflector and Core The core volume is fabricated from an array of hexagonal-prism units (referred to as array or core positions), bounded by a hexagonal-prism extending from the top of the core to the bottom (representing the core shroud). Generally, each array unit includes a prism simulating the upper and lower aluminum grid plate, and a prism for the space between. Units that can be assigned to array positions include peripheral position units (simulating water and solid grid plates), fueled units, graphite rod units, water void units (simulating water and grid plates with penetrations), and the central thimble unit.

Sections of grid plate at the periphery of the core have no penetrations other than alignment holes and a set of small on-axis and radial holes that allow insertion of flux probes; these grid plate sections are 7

PART I modeled as solid aluminum plates. Grid plate representations in peripheral positon are therefore solid structures, while other units have grid plate penetrations to accommodate fuel, graphite rods, water voids, or the central thimble. Grid plate unit geometry is provided in Table 3. Units with graphite dummy rods, control rods, or fuel elements are specified as the overall hexagonal structure in Table 3, modified to add graphite rods, control rods, or fuel to the central thimble.

Table 3: Unit Cell Structure, Geometry Geometry ID Radius Upper Lower Hexprism 30 2.177 32.309 -36.424 Hexprism 40 2.177 32.309 30.734 Cylinder 41 1.911 32.309 -36.424 Hexprism 50 2.177 -33.249 -36.424 2.2.3 Fuel Elements and Control Rods Representation of fuel elements and control rods based principally on GA schematics T135210D210 (Fuel Element Assembly), T135210C212 (Reflector - Fuel Element), TOS210B229 (Disc), TOS210C232 (Cladding - Fueled Element), TOS210D213 (Fuel), TOS210B217 (Rod- Fuel Element), 21717-002 (Element Standard, , Standard Fuel Element assembly ), TOS250D225 (Control Rod - Fuel Follower),

and TOS250B226 (Poison). A complete listing of fuel and control rod dimensions is provided in Table 4.

Fuel elements are modeled as cylinders with conical shapes simulating top and bottom end fittings.

Component dimensions from Table 3 were developed into unit specifications for fuel elements as indicated in Table 5.

Table 4: Component Dimensions Dia. Dia. Radius Length Length Vol. 9 3 TYPE COMPONENT REFERENCE (in.) (cm) (cm) (in.) (cm) (cm )

ZrH Fill Rod FFCR ZrH-U UT SAR 5/91 4-65 FUEL SFE ZrH-U UT SAR 5/914-62 Lower graphite T13210C212-1 Upper graphite T1321oC212-1 Mo Disk Mo Disk TR AIR FLWR. UT SAR 5/914-65 SFE (SS) T135210D210 Inner Outer less thickness FFCR (SS) UT SAR 5/91 4-65 Inner Outer less thickness TR (AI) UT SAR 5/914-65 Inner Outer less thickness POISON FFCR B4 C T052508226 B4 C TR B4C TOS2508226 9 Volume column values for fuel only 10Volume excludes zirconium "filler" rod 11U 8

PART I FUEL ELEMENTS GA schematic T135210C212 indicates the axial reflectors are in ( cm) in diameter with length of in ( cm) at the lower position and in ( cm) in the upper position.

-

- I Figure Figure 2: Standard Fuel Element Table 5: Standard Fuel Element Unit Model COMPONENT Unit Radius UpperZ Radius12 Lower Z Excluded Media (cm) (cm) (cm) (cm) Units End fitting 60 na 3 Cladding 61 -62 3 Air Gap 62 64 6 67 Graphite 63 na 4 Fuel 64 -65 FUEL Zr Fill Rod 65 na 1 Mo Disk 66 na 9 Graphite 67 na 6 End Fitting 68 na 3 FUEL FOLLOWER CONTROL RODS The B4 C poison in TRIGA fuel follower control rods is manufactured to a diameter of in. and length 13 of in. (drawing TOS2508226.C). Schematic values are used in analysis . Fuel Follower Control rods are modeled as a set of cylinders (Table 6).

12 For truncated cones, with surfaces specifying an upper and a lower radius 1

9

PART I Table 6: Fuel Follower Control Rod Unit Model UpperZ Lower Z Length Length media COMPONENT Unit Radius (cm) (cm) (cm) (in) (cm)

CLADDING (SS) 60 3 (SS) 60, 61-65, 67 (UPPER FIT) -- --

UPPER AIR 61 6 (Air) 61 (WELD) -- --

POS. AIR GAP 62 6 (Air) 62 POSION 63 7 (B4C) 63 WELD -- --

FUEL AIR GAP 64 6 (Air) 64 FUEL 65 FUEL 65 -66 ZIRC FILL ROD 66 1 (Zr) 66 (WELD) -- --

AIR FLLOWER 67 6 (Air) 67 (BOTTOM FIT)

TRANSIENT (PULSE) ROD The transient control rod is modeled as a set of cylinders (Table 7). The transient rod is protected by a guide tube. The guide tube surface has Y2 in penetrations above the upper grid plate and below the lower grid plate to reduce the effects of fluid compression during operation (drawing T13S210D150).

These penetrations were not modeled.

': 7 Air

,

I'e Figure 4: Transient Rod Table 7: Transient Rod Unit Model Radius UpperZ Lower Z Length Length media COMPONENT Unit (cm) (cm) (cm) (in.) (cm)

CLADDING (AI) 60 5 (AI) 60-61 -63 (UPPER PLUG) -- --

Air at B4C 61 6 (Air) 61-62 POSION 62 7 (B4C) 62 (weld) -- --

AIR FOLL. 63 6 (Air) 63 LOWER PLUG -- --

3.0 Calculations Material composition of individual fuel elements vary in manufacturing (previously noted) and also from fission and transmutation during operation of the reactor. The neutron flux is not uniformly distributed across the core, so that co-irradiated individual fuel elements experience different burnups. Therefore 10

PART I individual element composition was tracked according to the initial composition and specific core locations (Appendix 1) over core life to support model validation.

3.1 Initial Core The UT TRIGA reactor became operational in 1992. Most of the initial core load was transferred from a previous UT TRIGA reactor located on the main campus.

3.1.1 Materials For convenience, core positions (Fig. 5) are referred to as rings, labeled A (central thimble, CT) through G. The B ring is shaded green, C ring light blue-green, D ring rose, E ring light brown, F ring light blue, and the G ring yellow.

G26 G27 G28 G29 G30 G24 F21 FU P23 F24 F25 F26 SRC G23 E17 E8 E19 E20 E21 F27 G33 E20 G22 F19 E16 D14 F22F28 PNT 621 F18 Els F29 635 G20F7 F24 F30 G36 F16 E1 C7 3 E03 603 G18 15E02 G02 G17 F12 E (0 0 3 E 3 G16 2 Ilo D06 P04 F4 604 615 F12 E09 E08 E07 P06 E05 FOS G05 G14 F11 FlO P09 FOB F07 P06 G06 G12 Gll G10 G09 G08 Figure 5: Core Positions With the exception of elements in fuel followers (i.e., fuel follower control rods), the fuel elements in the initial core had prior burnup in other facilities. Most of the previously burned elements were used in the 250 kW TRIGA previously located on the main UT campus. A smaller set of fuel elements were manufactured more recently and irradiated in facilities that have higher rated power levels.

Burnup records for these previously irradiated elements include only total burnup (grams), with no information about the specific core configuration or location. Burnup records are based on a ratio of grams burned for each MWD, therefore the total burn (MWD) was determined as grams 235U depleted divided by 1.05 g/MWD. Elements with operating history at the previous UTTRIGA were simulated as a burn at 250 kW for a time interval calculated as total burn (MWD) divided by 0.25 MW. A small quantity of lightly burned fuel was transferred from a GA TRIGA reactor and from a TRIGA reactor in Illinois. The 11.TRIGA had a rated power of 1 MW. Some of the Ill. fuel had an initial H:Zr ratio of 1.7. All lightly-burned fuel elements decayed at least 1 year prior to use at the current UT TRIGA, and a 1-year decay intervals were included in burnup calculations.

In characterizing lightly burned fuel element composition for the initial core, similar elements were grouped (Table 8) based on 235U loading, enrichment, and total burnup (MWD). Depletion calculations were performed on a single fuel element representative of each group to develop material data for 23 5 subsequent calculations. Material data for these calculations for U, 238 U, Zr, and H were formatted as 11

PART I mass density while fission products (without stable zirconium isotopes) were introduced at an atom concentration of 1E-20 atom density (normalized to a barn), as recommended by the ORNL SCALE group (private communication). Mass densities (from the OPUS report module) for each representative element were used as materials for each fuel element in the "initial operational core".

Table 8: Summary, Initial Core Fuel Element Groups POS. g W% g 235 GROUP ELE. POS. g 235 W% g 235 GROUP ELE.

235 depl. depl.

A 6143 B-04 19.79 0.15 2983 F-26 20 0.69 B 5916 C-02 19.79 0.22 2979 F-25 20 0.69 C 5917 C-03 19.69 0.2 (M) 2983 F-26 20 0.69 D 5920 D-15 19.79 0.24 2906 E-06 20 0.69 5921 B-01 19.79 0.24 2928 E-16 20 0.69 5922 B-02 19.79 0.24 2932 E-19 20 0.69 6886 B-05 19.69 0.24 N 2962 F-14 20 0.69 6924 C-04 19.69 0.24 5283 B-06 20 0.69 6926 C-05 19.69 0.24 2977 F-24 20 0.69 6927 C-06 19.69 0.24 2974 F-21 20 0.69 6928 C-08 19.69 0.24 2929 E-17 20 0.70 6929 C-09 19.79 0.24 2950 F-05 20 0.70 6930 C-10 19.69 0.24 2902 E-02 20 0.71 6889 C-11 19.69 0.24 2905 E-05 20 0.72 6932 C-12 19.69 0.24 2911 E-09 20 0.73 5912 D-08 19.79 0.24 2912 E-10 20 0.71 5913 D-09 19.79 0.24 2915 E-12 20 0.71 5914 D-10 19.79 0.24 2918 E-13 20 0.71 5915 D-11 19.79 0.24 2946 F-02 20 0.71 5918 D-12 19.69 0.24 2969 F-18 20 0.71 5919 D-13 19.79 0.24 2947 F-03 20 0.72 6923 D-17 19.69 0.24 2975 F-22 20 0.72 6925 D-18 19.79 0.24 2955 F-09 20 0.72 5981 B-03 19.90 0.29 2940 E-23 20 0.72 6142 D-16 19.79 0.29 2971 F-20 20 0.72 5904 D-07 19.79 0.3 2913 E-11 20 0.73 5902 D-04 19.79 0.31 2927 E-15 20 0.73 5903 D-05 19.90 0.31 2930 E-18 20 0.73 5844 D-01 19.79 0.32 2935 E-20 20 0.73 5845 D-02 19.79 0.32 2938 E-21 20 0.73 5846 D-03 19.79 0.32 2954 F-08 20 0.73 2984 F-27 20 0.66 2957 F-10 20 0.73 2985 F-28 20 0.67 2968 F-17 20 0.73 2899 E-01 20 0.68 2970 F-19 20 0.73 12

PART I Table 8: Summary, Initial Core Fuel Element Groups POS. g235 W% g 235 GROUP ELE. POS. g235 W% g235 GROUP ELE. depl. depl.O 2965 F-16 20 0.68 2908 E-07 20 0.74 2910 E-08 20 0.68 W 2948 F-04 20 0.74 2939 E-22 20 0.68 2960 F-13 20 0.74 L 2959 F-12 20 0.68 2976 F-23 20 0.74 2964 F-15 20 0.68 X 2903 E-03 20 0.75 2925 E-14 20 0.69 Y 2952 F-07 20 0.77 2941 E-24 20 0.69 Z 2904 E-04 20 0.78 M

2944 F-01 20 0.69 AA 3513 F-30 19.6 0.99 2979 F-25 20 0.69 AB 5198 F-29 19.6 4.93 3.1.2 Initial Startup Initial criticality for the UT TRIGA reactor at the Nuclear Engineering Teaching Laboratory (NETL) was accomplished on 02/13/1992. Criticality was attained with 3 fuel follower control rods (control element fully withdrawn, fuel fully inserted) and 56 standard fuel elements (including two instrumented fuel elements). Total mass of 235 U was in the standard fuel elements and in the three fuel followers. Critical fuel loading is displayed (Fig. 6) with labels:

0 CT for the central thimble (water void) 0 SFE for standard fuel rods 0 TC for instrumented fuel elements a GR for graphite rods 0 S for the neutron source

WV for water voids

52, S1 and RR for the fuel follower control rods (Shim 1 and 2, Regulating Rod) 0 TR for the transient rod A direct comparison between critical masses of the initial 1992 UT TRIGA core and the historical GA TRIGA cores is complicated by (1) differences in reflection, i.e. graphite rod and water void configurations, (2) prior power history for the UT TRIGA fuel elements, and (3) fundamental difference in 235 core geometry. Nevertheless, the U in the UT TRIGA compares well with the (approximately water moderated) of fresh elements required in the prototypical GA TRIGA reactor.

3.1.3 Operational Loading Fuel was loaded in the UT TRIGA reactor to support operation at 1.1 MW on 03/16/1992. The core contained 84 standard and 3 fuel follower elements (Fig. 7, labeling consistent with Fig. 6) with kg 23 5 U. As previously noted, all the standard fuel elements in this configuration had prior power history, with 7.91 MWD generated at the UT Taylor Hall facility in 46 elements, and 41.07 MWD generated at General Atomics facilities in 56 elements; 15 fuel elements had operating history in the reactors at both The University of Texas at Austin TRIGA and General Atomics reactors.

13

PART I WV WV WV WV WV wvv v w wvwwwv s WV WV WV WVS imSWV WVW WV WV SEE S2E iE E IE WV WV WV WV SEE' 'SFESFE TC;S SFE WV WV WV WV SfE SOSFE SijE SFE WV WV WV: stV SF SE 4 SEWVWV s*E WV wV SSEE S1WS*E SPE S WV WV wVV WV SF SFE SFE SF SF*E VV WV WV WVN SWV WWVWVW WV W WV WV WV Wv WV Figure 6: Initial Criticality WV GR GR GR WV WV SE SFE SEE SEE SEE SFE S WV SFE SFE SEE SEE SFE SFE SFE WV wV SEE SFEE S2 SiEE SEE WV WV sF sr F ET V F WV WV SFE W'I FU S $4E E WV GR FE SFE SKE SEE SE SFE GR GR SF SF*E S1 SFE SF GR GR SF SFE SF SF SFF SE SFE GR GR GR SF SF SF SF G GR GR GR GR GR GR Figure 7: Operational 1992 Core The initial operational core configuration existed from 01/28/92-03/16/92, with few configuration changes (principally changes in IFE configuration). Major changes to core configurations over the NETL TRIGA operating history (since 1992) include:

" Instrumented Fuel Element (IFE) relocation and replacement

" Insertion and removal of a three-element in-core experiment facility (D17, E22 and E23)

Fuel insertions and movements to compensate for burnup

Reflector changes (potential reactivity effects, although not a core change as defined) o Beam port and reflector flooding o Reflector replacement

Comprehensive core reconstitution following reflector replacement

Insertion and removal of a second 3-element in-core experiment (Ell, F13, F14), and

" Insertion and removal of a seven- element in-core experiment facility (E03, E04, F03, F04, FOS, G04, G05).

14

PART I 3.2 Configuration Changes Over Core Life Instrumented fuel elements were moved or replaced 4 times between 1992 and 1998. The original three-element facility was installed for various operating periods between 1992 and 2001. Three-element facility installation required fuel movement from the core position to alternate locations with additional fuel added to compensate for the reactivity loss from displaced fuel. (Operating periods for experiment facilities range from days to several months, based on experiment needs.)

Table 9 provides a summary of information applicable to each configuration along with the date the configuration was established, the range of dates for which the configuration was active, the number of fuel elements in the configuration, experiment facilities in use, and the total burnup for the configuration. Note that prior to reflector replacement in 2004 there were only two configurations that did not use the intermittently installed three-element experiment facility, and dates for these configurations consequently overlap. After reflector replacement, the experiment facility positions were configured and left in place even when an experiment was not being performed, and experiment configurations after reflector replacement were not intermittent over short periods.

Reflector deformation was observed in 2002 from an apparent increase in internal pressure caused by water intrusion, which lead to beam port flooding. The reflector was intentionally flooded, then replaced in 2004. An attempt was made to model flooding as water in the reflector and beam port air filled spaces. The start date for simulating flooded spaces is a best estimate, not precise. As part of the reflector replacement, new core grid plates were designed incorporating two additional facilities, a three-element facility (in a lower reactivity-worth location) and a seven-element facility (at the core periphery). The new three-element facility has been used through several operating periods since 2004, and the seven element facility during one extended operational period.

Changes in core configuration were generally conducted in conjunction with fuel inspection or periodic maintenance outages. Fuel inspection and core reconfigurations occurred in 2002, 2004 (associated with the reflector replacement), 2005, 2006 (multiple times) and 2007.

This complex operating history was modeled as core configurations operated over discrete intervals.

Fuel movement logs at the beginning of each configuration were used to identify which elements were located in core positions at the start of the configuration. Burnup for all operations was simplified as a single interval for which configuration was constant. When alternating configurations existed for periods during which an experiment facility was intermittently used, the burnup that occurred in each configuration was evaluated. Some configuration dates overlap as alternating positions were used to support experiment facility utilization.

During each burnup interval, operations occurred at a wide variety of power levels up to a maximum nominal power level of 950 kW. A T-6 SCALE depletion calculation was performed with a fuel temperature of 600°C for each configuration to develop a material data set (23 .U, 238U, and fission products) representative of fuel element composition at the end of each burnup interval. Lightly burned fuel elements were depleted in separate calculations (to develop a material data set based on initial assay and burnup data) prior to UT TRIGA burnup calculations.

15

PART I Table 9: Core Configurations and Burnup Configuration Dates No. of Installed Core Burnup Acc. Burnup Date Active Elements Facilities (MWD) (MWD) 03/16/1992 1992-1999 87 Fully Fueled 31.07 31.07 03/04/1999 1999 90 3EL(A) CD 0.24 31.31 04/27/2000 2000-2001 89 Fully Fueled; 5.51 36.82 RefI. flood 06/29/2000 2000-2001 92 3EL(A) 8.90 45.72 07/22/2002 2001-2002 95 3EL(A) 35.47 81.19 11/22/2002 2002-2004 103 3EL(A) 24.93 106.12 07/15/2004 2004-2005 102 3EL(B) )-

NeL 15.70 121.82 New Refl.

07/13/2005 2005-2007 104 3EL(B 64.71 186.53 07/24/2007 2007-2008 108 3EL(B) 18.35 204.88 06/11/2008 2008-2009 110 7EL(B) 21.29 226.17 06/23/2010 2010-2014 114 3EL(B) 52.88 279.05 Depletion sequence calculations assumed a constant power level over the time interval for the configuration (Table 10). Since the initial 3-element facility was used intermittently, time intervals prior to 2004 are a best estimate based on records of daily power history and core data. The last step in each depletion calculation is a 7-day decay time to assure minimal xenon poisoning in criticality calculations.

Table 10: T-6 Burn Parameters CORE START END MWD DAYS AVE PWR 87 03/16/92 10/26/99 31.07 2780 0.0112 90 03/04/99 10/26/99 0.24 236 0.2500 89 04/27/00 01/31/01 5.51 279 0.0197 92 06/29/00 07/18/01 8.90 384 0.0232 95 07/22/02 11/07/02 35.47 108 0.3284 103 11/22/02 03/17/04 24.93 481 0.0518 102 07/15/04 06/30/05 15.70 350 0.0449 104 07/13/05 07/02/07 64.71 719 0.0900 108 07/24/07 06/04/08 18.35 316 0.0581 110 06/11/08 06/14/10 21.29 733 0.0290 114 06/23/10 07/25/11 11.90 397 0.0300 114 06/23/10 08/02/12 24.54 771 0.0318 114 06/23/10 10/14/14 52.88 1574 0.0336 3.3 Reactivity Calculations The UTTRIGA was modeled in SCALE as previously described. Calculations with SCALE were compared to integral control rod worth data.

16

PART I 3.3.1 SCALE Calculations UT TRIGA criticality calculations using the CSAS6 sequence are based on the same model and material composition as the SCALE T6 depletion calculations, except for lattice cell data (not required for continuous energy libraries). Modifications of the SCALE input to adapt the depletion calculations to support criticality calculations include:

" Specifying continuous energy libraries

" Removing depletion-specific parameters (lattice data, burndata, OPUS

" Revising unit materials to single identification for non-fuel materials (since self-shielding calculations are not required for continuous energy library data)

" Simulating clean critical calculations with fuel temperatures of 300'C

" Developing fresh and burned fuel and fission product material files for each configuration

" For the current configuration, developing material data at intervals corresponding to control rod worth calibration data The initial CASAS6 file (as modified) simulates all rods fully withdrawn. Four additional calculations are performed, with a single control rod inserted for each calculation (regulating rod, shim 1, shim 2, and the transient rod). Where / is the effective delayed neutron fraction, kARO is keff (transport calculation) with all control rods out, and kRio is keff with control rod i out:

Excess reactivity with all control rods removed is calculated as:

$= k =k O- "6 k TkAROsr

" Total integral control rod worth (with a single control rod inserted) is calculated as:

$=(Ak)ýa8 k 41?0 -kRO.

12 Table 11: Calculated Reactivity Values CORE DATE MWD RR + SH1I SH2 +/- TR - EXCESS +

87 03/16/92 0 $3.57 0.010 $2.52 0.010 $2.90 0.010 $3.06 0.010 $9.20 0.018 87 10/26/99 31.07 $3.79 0.012 $3.21 0.011 $3.10 0.012 $2.96 0.012 $7.88 0.017 90 03/04/99 31.07 $4.58 0.013 $4.09 0.013 $2.94 0.014 $2.56 0.014 $7.60 0.016 90 10/26/99 31.31 $4.09 0.013 $2.85 0.013 $2.03 0.013 $2.25 0.014 $7.00 0.017 89 04/27/00 31.31 $4.17 0.013 $3.18 0.012 $2.26 0.012 $2.63 0.012 $8.04 0.016 89 01/31/01 36.82 $3.82 0.012 $2.88 0.012 $2.31 0.011 $2.81 0.011 $6.31 0.014 92 06/29/00 36.82 $4.92 0.014 $3.31 0.015 $2.40 0.015 $1.87 0.015 $7.46 0.015 92 07/18/01 36.82 $4.16 0.013 $2.95 0.012 $2.86 0.012 $2.16 0.013 $7.18 0.015 95 07/22/02 45.72 $4.55 0.015 $3.58 0.014 $2.67 0.014 $1.61 0.013 $7.02 0.015 95 11/07/02 81.19 - $4.46 0.015 $3.48 0.014 $2.68 0.014 $1.31 0.014 $6.11 0.015 103 11/22/02 81.19 $3.90 0.011 $2.78 0.011 $3.37 0.012 $2.62 0.011 $7.36 0.014 103 03/17/04 106.12 $3.95 0.012 $2.48 0.012 $3.08 0.013 $1.63 0.012 $3.89 0.009 17

PART I Table 11: Calculated Reactivity Values CORE DATE MWD RR + SH1 +/- SH2 +/- TR +/- EXCESS +

102 07/15/04 106.12 $3.26 0.010 $3.45 0.010 $2.38 0.010 $3.02 0.010 $6.56 0.014 102 06/30/05 121.82 $3.12 0.009 $3.35 0.010 $3.24 0.009 $3.32 0.010 $8.42 0.018 104 07/13/05 121.82 $3.75 0.012 $3.22 0.011 $2.97 0.011 $2.81 0.012 $8.87 0.018 104 07/02/07 186.53 $3.27 0.003 $3.18 0.007 $2.62 0.007 $2.86 0.008 $7.64 0.005 108 07/24/07 186.53 $2.51 0.008 $1.46 0.008 $3.10 0.008 $2.30 0.008 $7.72 0.017 108 06/04/08 204.88 $3.25 0.010 $2.06 0.011 $2.66 0.010 $2.25 0.010 $7.70 0.017 110 06/11/08 204.88 $4.33 0.004 $2.49 0.004 $2.95 0.004 $1.85 0.004 $8.06 0.005 110 06/14/10 226.17 $4.09 0.004 $2.74 0.004 $2.74 0.004 $1.83 0.004 $7.54 0.005 114 06/23/10 226.17 $2.79 0.002 $2.39 0.003 $2.60 0.003 $2.52 0.003 $8.50 0.005 114 07/25/11 237.26 $3.11 0.002 $2.43 0.002 $2.67 0.002 $2.37 0.002 $8.33 0.002 114 08/02/12 250.7 $3.20 0.003 $2.43 0.003 $2.81 0.003 $2.60 0.003 $8.28 0.005 114 10/14/14 261.79 $2.41 0.004 $2.05 0.004 $2.74 0.004 $2.43 0.003 $5.38 0.006 3.3.2 Reactivity Measurements Control rod calibrations are performed experimentally under surveillance procedure SURV-6, and excess reactivity determinations are documented with SURV 3 (Table 12). Control rod installation began 1/28/1992, and standard core fuel loading began 2/10/1992. The initial control rod reactivity worth calibration was completed on 03/31/1992, although excess reactivity was not determined until completion of initial testing in July when both SURV-3 and SURV-6 were performed.

Table 12: Reactivity Surveillance Data DATE MWD Excess RR SH1 SH2 TR 07/01/92 0.00 $5.57 $4.08 $3.03 $3.17 $3.26 08/10/93 3.98 $5.61 $3.97 $3.00 $3.18 $3.24 10/25/94 6.44 $5.56 $3.99 $2.93 $3.19 $3.21 08/10/95 8.70 $5.46 $4.02 $2.98 $3.18 $3.21 03/05/96 9.65 $5.48 $4.02 $2.98 $3.18 $3.21 07/23/96 10.62 $5.50 $4.02 $2.98 $3.18 $3.27 01/29/97 11.73 $5.49 $4.02 $2.98 $3.18 $3.27 09/11/97 12.79 $5.44 $4.08 $3.00 $3.18 $3.22 01/23/98 14.48 $5.40 $4.08 $3.00 $3.18 $3.22 07/23/98 17.97 $5.40 $4.06 $3.06 $3.20 $3.23 07/02/99 26.00 $5.00 $4.05 $3.01 $3.22 $3.23 04/27/00 31.31 $5.53 $4.50 $3.48 $2.73 $2.36 06/30/00 34.76 $4.56 $4.50 $3.48 $2.73 $2.36 09/07/00 34.91 $5.50 $3.90 $3.02 $3.24 $3.17 07/30/01 45.81 $4.59 $4.19 $3.24 $2.94 $2.41 07/24/02 67.32 $4.09 $4.08 $3.16 $2.84 $2.47 11/14/02 81.29 $5.69 $4.30 $3.34 $2.75 $2.51 07/24/03 90.33. $5.20 $3.88 $3.31 $2.74 $2.46 07/29/04 106.23 $5.77 $3.33 $2.78 $3.25 $3.33 18

PART I Table 12: Reactivity Surveillance Data DATE MWD Excess RR SH1 SH2 TR 07/18/05 121.93 $5.55 $3.07 $2.94 $3.14 $3.28 07/19/06 145.21 $4.97 $3.09 $2.89 $3.02 $3.29 01/25/07 165.51 $4.47 $3.09 $2.89 $3.02 $3.29 07/25/07 186.65 $5.04 $2.84 $2.75 $3.30 $3.32 06/19/08 205.04 $4.45 $3.65 $2.35 $3.27 $2.04 06/25/09 214.05 $4.75 $3.99 $2.45 $3.36 $2.04 06/29/10 226.30 $5.79 $2.90 $2.54 $3.11 $3.14 06/29/11 236.82 $5.56 $2.83 $2.52 $3.07 $3.01 07/13/12 260.14 $4.83 $2.76 $2.47 $3.01 $3.04 07/16/13 281.57 $4.20 $2.75 $2.45 $2.91 $3.00 07/22/14 286.66 $4.70 $2.50 $2.74 $3.15 $3.17 3.3.3 Reactivity Calculations and Reactivity Surveillance Data Excess reactivity was calculated using keff data with all rods fully withdrawn, and independently using the keff data from all rods fully inserted and the integral rod worth for each control rod.

6 SDM = kARI +

i=1,4* 13 The limiting shutdown margin does not credit the most reactive control rod, and is therefore calculated as:

SDM = 15kARI + I R~i0 - 8kR,maX! 14 i=1,4 INTEGRAL CONTROL ROD WORTH COMPARISONS Reactivity measurements and core configuration changes do not have a one-to-one correspondence.

Calculated values and measured reactivity values were therefore compared based on burnup values (CORE, MWD and MEAS, MWD respectively) assumed in calculations and the burnup values corresponding to measured data between the start and end of the core configuration, indicated in Table 13, "CORE" column.

Table 13, Comparison of Reactivity Calculations to Surveillance Data CORE MWD DATE Burnup (MEAS) Control Rod Worth EXCESS (CALC) (CALC) MWD AMWD RR SH1 SH2 TR SUM B 87 0 07/01/92 0.00 0.00 12.5% 16.7% 8.5% 6.2% 11.0% -65.1%

08/10/93 3.98 3.98 10.1% 15.9% 8.8% 5.6% 10.0% -63.9%

10/25/94 6.44 6.44 10.5% 13.9% 9.1% 4.8% 9.5% -65.4%

08/10/95 8.70 8.70 11.2% 15.3% 8.8% 4.8% 10.0% -68.4%

03/05/96 9.65 9.65 11.2% 15.3% 8.8% 4.8% 10.0% -67.8%

07/23/96 10.62 10.62 11.2% 15.3% 8.8% 6.5% 10.4% -67.2%

19

PART I Table 13, Comparison of Reactivity Calculations to Surveillance Data CORE MWD DATE Burnup (MEAS) Control Rod Worth EXCESS (CALC) (CALC) MWD AMWD RR SHI SH2 TR SUM B 01/29/97 11.73 11.73 11.2% 15.3% 8.8% 6.5% 10.4% -67.5%

09/11/97 12.79 12.79 12.5% 15.9% 8.8% 5.1% 10.6% -69.0%

01/23/98 14.48 14.48 12.5% 15.9% 8.8% 5.1% 10.6% -70.3%

07/23/98 17.97 17.97 12.1% 17.6% 9.4% 5.3% 11.1% -70.3%

07/02/99 26.00 26.00 11.9% 16.2% 9.9% 5.3% 10.8% -83.9%

89 31.31 04/27/00 31.31 0.00 7.3% 8.7% 17.1% -11.6% 6.3% -45.3%

06/30/00 34.76 3.45 7.3% 8.7% 17.1% -11.6% 6.3% -76.2%

09/07/00 34.91 3.60 -7.0% -5.2% 30.1% 16.9% 8.1% -46.1%

92 36.82 07/30/01 45.81 8.99 0.8% 9.0% 2.9% 10.4% 5.2% -56.4%

95 45.72 07/24/02 67.32 21.60 -11.6% -13.3% 5.9% 34.9% 1.1% -71.7%

95 81.19 11/07/02 81.29 35.57 -3.7% -4.1% 2.5% 47.9% 7.5% -7.4%

.03 81.19 07/24/03 90.33 9.14 -0.5% 16.0% -23.1% -6.5% -2.3% -41.5%

102 106.12 07/29/04 106.23 0.11 2.0% -24.2% 26.7% 9.4% 4.5% -13.6%

104 121.82 07/18/05 121.93 0.11 -22.0% -9.4% 5.5% 14.3% -2.5% -59.8%

07/19/06 145.21 23.39 -21.2% -11.3% 1.7% 14.6% -3.7% -78.5%

01/25/07 165.51 43.69 -21.2% -11.3% 1.7% 14.6% -3.7% -98.4%

108 186.53 07/25/07 186.65 0.12 11.5% 46.9% 6.1% 30.7% 23.2% -53.1%

.10 204.88 06/19/08 205.04 0.16 -18.7% -6.1% 9.8% 9.4% -2.8% -81.2%

204.88 06/25/09 214.05 9.17 -8.6% -1.8% 12.2% 9.4% 1.8% -69.8%

.14 226.30 06/29/11 236.82 10.52 -9.9% 3.5% 13.1% 21.1% 7.4% -49.9%

.14 237.26 07/13/12 260.14 22.88 -10.1% 3.6% 13.4% 20.9% 7.5% -57.4%

.14 261.79 07/13/12 260.14 -1.65 12.7% 16.8% 9.0% 20.0% 14.6% -11.3%

07/16/13 281.57 19.78 12.4% 16.1% 5.9% 18.9% 13.3% -28.0%

07/22/14 286.66 24.87 3.6% 25.0% 13.0% 23.3% 16.6% -14.4%

The difference between the burnup assumed in calculation and the burnup associated with the measurement is also noted (AMWD). For deviation in excess reactivity, calculations were made based on all rods fully withdrawn (column label Excess - A) and independently balancing the reactivity for all rods fully inserted against the sum of the integral worth of all control rods (column label Excess - B). The deviation (E) of the calculated reactivity values (p"o"C) from the surveillance data (p"?"aS)was calculated in Table 13 as:

meas calc Prod - Prod

= meas 15 Prod As previously noted, configurations with 3 element facilities usage prior to the 102 core involved intermittent configuration changes, but were approximated as a single burnup interval for calculations.

The reflector and beam ports partially flooded prior to the 102 configuration then intentionally flooded to stabilize internal reflector pressure. A new reflector and grid plate were installed prior to establishing the 102 element configuration. The calculation for the current core configuration was conducted at three burnup values, corresponding to surveillance data. Although the regulating rod error is higher than desired, the total integrated rod worth matches surveillance data well and the calculated values for 20

PART I the regulating rod exceed the values in surveillance data. Therefore calculation of the limiting shutdown margin (with the most reactive rod fully withdrawn) is conservative.

3.4 Fuel Temperature Observations of current core operation in 2015 (Table 14) provides fuel temperature associated with the instrumented fuel elements (in core positions B03 and B06) and reactivity data, correlated to operation at specific core power levels.

Table 14: Observed Data CORE Fuel Temp 1 B03 Fuel Temp 2 B06 REACTIVITY PWR 0 0 INSERTED kW cc cC K 1 21 294 21 294 $7.72 100 84 357 91 364 $8.07 250 159 432 173 446 $8.62 500 243 516 263 536 $9.46 750 300 573 320 593 $10.04 950 340 613 363 636 $10.41 Core peaking factors are developed from SCALE physics calculations. Fuel temperatures as a function of element power is developed from TRACE thermal hydraulic calculation data. An average fuel element temperature is developed from core peaking factors (that characterize the fraction of core power in the element) and the average fuel element temperature associated with the element power.

3.4.2 SCALE and TRACE Calculations for Power Distribution SCALE calculations were performed to determine the fission distribution across all elements in the core (Fig. 8). The distribution across elements is used to calculate the power produced in individual elements. The distribution within a fuel element is used as input to thermal hydraulic calculations to determine the radial and axial power profile. The distribution within a fuel element is 2-dimensional.

The fuel element geometry is segmented into equal volume partitions, but variation along single axis at selected locations are provided in Fig. 9 and 10.

21

PART I 0.70 0,68 0,67 0.67 0.80 0.60 0.65 0.68 0.68 0.67 Figure 8: 114 Core Peaking Factors Radial Fission Distribution 1.80

....

.........

.. ............... iii7 1.60

.... ...... ................... .......

..... -0 1.50 0

1.40 . X 0

T 1.30 0 1.20 1.10 U,

6 w-a90 I

.XO ýX X X~Z 0.80 0.70 0.6 0.50 0.3 065 0.8 1.05 1.3 1.55 1.8 Distance from Fuel Element Center (cm)

3.81cm 016.51cm A19.OScm X34.29cm XAVE Distance from Bottom of Fuel Matrix (cm)

Figure 9: Radial Peaking Factor 22

PART I Axial Fission Distribution

........

..... .....i.......

1.7 ....... i.....

... ....i 1'6 n .. ......

1.4 I ............

...... ..

"]G iT . 1 .2 - ....... . .

0.

o 1.3 x 1.2 x X 41 1.2, 1.1 .... ......... ... X

.............

x X 01.0 .- ........ .. X

A o0 0.9 XA U0.8 i.....

60.7 u.b .....

...... : .............i............ ............. . ............

..........

........... ...... .............

÷.....+............i.......................

0.5 S ........

....

32 34 36 38 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 Distance from Bottom of Fuel Matrix (cm)

0.64cm L,1.22cm )1.66cm E31.73cm xAVE Distance from Fuel Element Center (cm)

Figure 10: Axial Peaking Factor Data from thermal hydraulic analysis provides a relationship between element power, monitored, maximum and average element fuel temperature (Fig. 11) for a single fuel element. The monitored temperature is the temperature at the location of the center thermocouple. The maximum fuel element temperature is taken as the maximum temperature in the zirconium fill rod. The average fuel temperature includes the zirconium fill rod and the fuel matrix.

FUEL TEMPERATURES AT CHANNEL POWER 1100 1000 900

)

800 ...._n.........

.................

E 700 600 LL 500 -

400 300 0 5000 10000 15000 20000 25000 30000 Channel Power (W)

Tave(fuel) -0 T(IFE) .*. T(MAX)

Figure 11: Monitored and Average Fuel Temperatures for Channel Power 23

PART I 3.4.3 Core Fuel Temperature Based on Measurement The power level of each fuel element in the 114 element core was calculated from the peaking factor and the core power level (Fig. 8). The average core fuel element temperature for a specified core power is taken as a linear average of the temperature (including the zirconium fill rod and fuel matrix segments) for all equal volume segments in the fuel elements. The temperatures in the positions corresponding to FT1 and FT2 were calculated based on the power level in the installed instrumented fuel element positions (B03 and B06) at each power level. Core average temperature at power is therefore directly correlated to measuring channel indication (Fig. 12).

(CALCULATED) AVERAGE & MONITORED TEMPERATURES TEMPERATURES AT CORE POWER 600 550 ........... ........ ........

!......................

........... ....-......... ... ..........................

................ ..... ...

................

5-00 .............. .......................... ....... .......................

..........

...................... ...........

.......................................

450 ...................

400..,.

3 00 ... ... .. . ............................ .. ....

.................................

200 400 600 800 1000 Core Power (kW)

-A B03 FT1 -*. BOG FT2 CORE AVE Figure 12: Calculated Core Average and FT1/FT2 Temperatures, 114 Element Core 3.4.4 Comparison Based on the correlation from Fig. 11, the FT1 and FT2 measurements are used to calculate the average fuel temperatures at each power level used in measurements (Table 15). Using the measuring channel data as the basis for comparison (observed values and average fuel temperature based on the observation), the calculated values were compared to the values from observation. The FT2 measuring channel data has good agreement with the calculated values (Table 16). The FT1 channel agreement is qualitatively similar, but agreement is less than for FT1; this issue will be discussed in greater detail as part of the thermal hydraulic analysis. The results indicate modeling provides a reliable means of predicting reactor behavior.

24

PART I Table 15: Average and Monitored Fuel Temperatures (°K)

Power AVERAGE FUEL TEMPERATURE B03 FT1 B06 FT2 Excess kW SCALE FT1 FT2 SCALE Meas. SCALE Meas. Reactivity 1 305 300 298 306 298 306 294 $4.29 100 337 320 331 368 335 368 357 $3.93 250 380 344 372 446 380 446 432 $3.39 500 440 373 420 546 432 546 516 $2.55 750 487 392 453 621 465 621 573 $1.97 950 520 407 477 673 491 673 613 $1.60 Table 16: Comparison Measuring Channel to Calculated Temperatures Average Fuel Temperature Measuring Channel FT1(SCALE) FT2(SCALE) FT1(SCALE) FT2(SCALE)

-1.6% -2.3% -2.7% -4.1%

-5.4% -1.8% -9.7% -3.0%

-10.5% -2.2% -17.3% -3.2%

-18.0% -4.7% -26.5% -5.7%

-24.4% -7.5% -33.6% -8.4%

-27.9% -8.9% -37.0% -9.8%

3.6 Fuel Temperature Reactivity Fuel temperature coefficient is the change in change in reactivity with respect to the change in fuel temperature. The fuel temperature measuring channels are related to average fuel temperature, as previously indicated. Excess reactivity as a function of core average fuel temperature (Fig. 14) exhibits a linear relationship, with the slope representative of the fuel temperature coefficient for the observed data in Table 15. The slope of the line representing excess reactivity as a function of core average fuel temperature based on the FT1 measuring channel shows a reactivity coefficient of -0.0169 $/'K, with FT2 at -0.0162 $/*K.

Independent calculations using SCALE were developed to determine excess reactivity response to fuel temperature changes. Calculations were performed with SCALE using ENDF/Vll the continuous energy cross-section libraries corresponding to 300 °K and 600 *K. Temperature correction is not available to continuous energy libraries, so a second set of calculations was performed using the 238 energy group library. The fuel temperature coefficient based on SCALE calculations is -0.0157 $/°K for both the continuous energy and the group energy data.

The maximum difference between all of the fuel temperature coefficients is 7%. The agreement between the fuel temperature coefficients using measuring channels and SCALE calculations provides confidence that the SCALE model is capable of simulating the UTTRIGA reactor behavior as a function of fuel temperature changes.

25

PART I EXCESS REACTIVITYAS A FUNCTION OF CORE AVERAGE FUEL TEMPERATURE

$$9 0 . . . . .. . . . .... .. . . . .. . .

y= -0.0157x + 12,.581

$8.0 R1 0.9854 4E 8$ .0 0 .... ..* ........................ ..... 9...

............

$ 7.00 ...... 1................

.....

y = -0.0157x + 12.882 S6.00

$4.00 S $3.00 x

S$

2.00 .......................................... ... ...... . .0162x + 9.5601 Y -

R- = 0.9992Z

$1.00 y = -0.0169x + 9.6919 R' = 0.9994

$0.00 ........... .*i 250 350 450 550 650 750 850 950 Average Fuel Temperature (°K) o FT1 Data o FT2 Data + ENDF/VII CE X V7-238 Figure 13: Reactivity and Average Fuel Temperature 3.7 Pulsing Power and temperature data for the most recent pulses performed using the current core (32 total measurements, Table 17) was compared to calculations utilizing SCALE and TRACE data (Figs. 16 and 17).

A correlation was developed for the core fuel element average temperature and the monitored fuel temperature for the current 114 core.

Table 17: Pulse Data DATE CALC MAX OBS SCALE/TRACE CALC PULSE POWER FT2 FTave POWER FTB FTave 368 05/10/12 2.269 916 338 204 937 346 208 369 05/21/12 2.324 992 343 207 1020 358 215 371 05/21/12 1.840 411 263 161 425 252 155 370 05/21/12 2.316 999 343 207 1006 356 214 359 05/21/12 2.324 1002 344 207 1020 358 215 372 07/20/12 1.840 411 263 161 425 252 155 373 07/20/12 1.884 441 275 168 468 261 160 374 08/22/12 2.332 1014 348 210 1032 360 216 375 08/28/12 2.392 1096 357 215 1123 373 224 376 08/31/12 2.340 1033 353 212 1042 362 217 378 11/01/12 2.660 1616 404 242 1595 431 258 380 11/02/12 2.699 1672 408 244 1670 441 262 381 11/12/12 2.219 834 330 199 867 335 202 382 11/13/12 2.660 1591 403 241 1595 431 258 383 11/13/12 2.660 1607 405 242 1595 431 258 384 11/20/12 1.899 470 276 168 483 264 162 26

PART I Table 17: Pulse Data MAX OBS SCALE/TRACE CALC PULSE DATE CALC POWER FT2 FTave POWER FTB FTawe 385 28 nOv 12 2.324 992 343 207 1020 358 215 386 12/18/12 2.357 1052 351 211 1070 365 220 387 02/20/13 1.910 476 275 168 495 267 163 388 02/20/13 1.918 486 286 174 503 269 164 389 04/08/13 1.918 486 279 170 503 269 164 390 04/09/13 2.047 628 297 180 647 297 180 391 04/29/13 1.438 118 192 121 132 164 105 392 04/29/13 1.918 489 274 167 503 269 164 393 06/12/13 1.895 470 274 167 480 264 161 394 06/12/13 1.918 493 278 170 503 269 164 395 06/28/13 1.938 502 278 170 523 273 167 396 07/19/13 1.955 513 281 171 542 277 169 397 10/22/13 2.726 1722 421 251 1727 446 266 398 11/22/13 1.161 21 125 82 29 110 73 399 11/22/13 1.914 472 270 165 499 268 164 400 11/22/13 2.634 1519 397 238 1538 426 254 The peaking factors for each element were used to determine the power in each element for a range of power levels from 100 kW to 1100 kW (in steps of 100 kW). The core average fuel temperature for each power level was taken as the average temperature of all the fuel elements. The temperature of the B03 element at the thermocouple location was then correlated to the average fuel temperature (Fig. 15).

The formulation of the in-hour equation by Johnson, Lucas, and Tsvetkov (INL/EXT-10-19953, Modeling of Reactor Kinetics and Dynamics, 2010) was implemented in Matlab. Precursor group constants were taken from Reactor Dynamics and Controls, Weaver (1968).

do _ p(t) + af 0P(TF(t) - TFo) + a.m (Tmn(t) - Tmn0 ) - flmixt(t) 6 16 dt A i=1 With the precursor rate of change as:

dC= flx0) 17 dt A 27

PART I MONITORED VERSUS AVERAGE FUEL TEMPERATURE, 114 ELEMENT CORE 500 ............................................ . ............ ... .. . .....

..........................

450 y = 1.7513x- 19.05

) 400

.4-,

R2 = 1 350 _ ........

. .__ ............

CL ..................................

E 300 Q) i 250

-)nn

'a o --- ---- - -- ------

0 15 0

~~~~0 ~ ~ --- -----.- -_ __ - - - _ _ _

0 _ ..~ ..

._

.... _...._..

50 100 150 200 250 300 Average Fuel Temperature (°C)

Figure 14: Correlation of Current 114 Element Core Average Fuel Temperature to Measuring Channel Temperature And the fuel temperature rate of change as:

dTF P_ef 18

- Y' (TF-Tm) dt m *cP Where 0 is the neutron flux, proportional to power p is the pulsed reactivity a is the temperature coefficient, subscripted F for fuel, m for moderator T is the temperature, subscripted F for fuel, m for moderator

!3mnix is the effective delayed neutron fraction for the 235/238 fuel A is the prompt neutron lifetime, taken as the generation time

.1, is the decay constant for the groups of delayed neutron precursors C, is the concentration of the groups of delayed neutron precursors Q is the contribution from the neutron source Simplified by assuming adiabatic conditions (i.e., water temperature does not change) and neglecting source neutrons:

p(t) + a d (f E.c dt - TF,O) - flmix(t) 6 19 dt A * (t)+ i=1t"C 1 (t)

The temperature feedback coefficients are evaluated as:

dp 20 d Tx 28

PART I The rate of temperature change of the fuel is calculated as:

dTF Peff 21

- Y(TF -- T.)

dt m.cp

-

The prompt neutron lifetime (51.9 ps) was taken from the average of SCALE the value from SCALE calculations supporting fuel temperature reactivity calculations. Fuel density was calculated by the methodology of Simnad (op. cit.). The effective delayed neutron fraction was calculated using keff data generated by the MCNP model for the UT TRIGA, using the expression:

kp fleff = 1 22 kp+d The effective delayed neutron fraction is 0.007036 +/- 0.000729, compared to the nominal value of 0.007.

Calculations of peak pulse power were compared (Fig. 16) to the measured values from surveillances (SURV7) recorded in Table 17.

PEAK POWER FOR PULSED REACTIVITY INSERTION 1 ,60 0. ............

....

..

MATLAB 1,400 y= 567.56x' - 1128x + 577.23 - .

RI=r/R2 1200 ---

-=

S1,000--.___

0 o 800 . ..

...

SURV7 Data 400 -. y =601,33X2 - 1258.5x + 682.72 R20,9995

$1.0 $1.5 $2.0 $2.5 $3.0 Pulsed Reactivity ($)

0MATLAB o SURV Figure 15: Pulsed Power Levels The core average temperature calculated for each pulse using MATLAB was adjusted by the correlation (Fig. 17) to simulate the fuel temperature measuring channel response.

29

PART I FUEL TEMPERATURE FOR PULSED REACTIVITY INSERTION 450 ---

400* £ 300 ............ ...............

.............

............ ......................

. . . ... .. .

........................... ......... ..........................

......!.........

.................................... -* .....................

T.........................

Z- 2 00 .. ............ ...

U-100 ....

..........................................

10 0 -. * - .. . . .. .. .__I

$1.0 $1.5 $2.0 $2.5 $3.0 Pulsed Reactivity(S)

-e SURV7 -O-MATLAB Figure 16: Pulsed Fuel Temperatures The agreement between calculated and measured data provides confidence that the UTTRIGA reactor models are capable of simulating reactor behavior resulting from sudden reactivity insertions.

4.0 RESULTS Nuclear data for a limiting core configuration are calculated. The total integral worth of each control rod, the total integral worth of all control rods, and excess reactivity for calculation were compared to surveillance measurements.

4.1 Nuclear Data The prompt neutron lifetime for the 80 element core is taken as the generation time. The generation time is calculated by SCALE to be 4.71405E-05 +/- 4.36711E-08 s at 300'K and 4.80039E-05 +/- 4.56268E-08 s at 6000K.

Using methodology previously described, the fuel temperature coefficient of reactivity is evaluated to be -0.0153 $/°C (Fig. 18).

30

PART I EXCESS REACTIVITY AS A FUNCTION OF LCC AVERAGE FUEL TEMPERATURE

$6

$4 '~

Y = -0.0152x + 9.2716 2

2R =1

.. .. ..

........... ...

..

y =-0.0154x+9.0278i "-. ..---

, , -$ 24 .. R2 . . ....... ..... ........ .............

.. ...... ....

-$4

................

--66 ..................

... .......

............ ............

..........

......... ............

....... ...................... ................

....... ......................

.................

......... ...

300 400 500 600 700 800 900 Average Fuel Temperature (°K) 0 ENDF/VII 0 ENDF/VI Figure 17: LCC Excess Reactivity and Fuel Temperature 4.1 Peaking Factors The fission density calculated by SCALE was used to determine the ratio of power produced in each fuel element to the average power per element in the core (Fig. 19).

0-60 070 0.77 0-78 0-71 0.61 0.65 0.79 0.93 0.98 0.94 0.81 0,68 0371 086 119 0.90 0.71

0. 1.30 1-37 1.32 0.71 08808 0.88 1-32 1 a.85 O-g8 .11-813 0-90 0.73 0.71 0.86 0.91 0.73 0.65 0.79 0.93 0.98 0.94 0.82 0.70 0,59 0.70 0.78 0.78 0.72 0.62 Figure 18: LCC Peaking Factors 80 Element Core The distribution of power within a fuel element in the B ring was determined for a 15X15 cylindrically segmented fuel element, with radial and axial dimensions of each segment established to bound equal volume segments. The fission density for the sum of segment in each radial position was compared to the average of all radial positions (Fig. 20), and the fission density for the sum of all segments at specific axial positions similarly compared to the average of all axial positions (Fig. 21). Because of self-shielding effects, tallies at radial locations near the center of the fuel have large errors.

31

PART I 80 ELEMENT CORE AXIAL POWER PROFILE 1.3 1.21. i................... .. .....

...............

'*...............

..- ...........

...............

0 i */ ............................

..................

..............

.................

V...... . ..................................

.......................

...........

U M 0 .9 ................ .......... ... . ............ .-.......

U-tw 1.0. /

1.0-

. ..............

... ...................................

........

.......... .................................

...................................

................ ....... .....

......................

........ .. ...

0 .8 ........ ....... .................

.

0.7 0 5 10 15 20 25 30 35 40 Axial Displacement from Bottom of Fuel (cm)

Figure 19: LCC Axial Peaking Factors 80 ELEMENT CORE RADIAL POWER PROFILE 1.4 1.3 .......

i-

......

U- 1.2 0- /

1.1 0.50 .0.75 1..0.

an Cr 1.0 0.9 0 .8 ............

. . . ..

... ......,,..

.....

,,. ... ......... ........1* ! . -

.....

0.00 0.25 0.50 0.75 1.00 1.25 1.50 1.75 2.00 Radial Displacement from Center (cm)

Figure 20: LCC Radial Peaking Factors 4.2 Fuel Temperature Using thermal hydraulic data generated by TRACE, with input assumptions for the minimum permitted pool level and the maximum permitted pool temperatures, functional relationships were developed for power produced in an individual fuel element and (1) the average fuel temperature, (2) the temperature at the location of the center thermocouple, and (3) the maximum temperature in the fuel element (Fig.

22).

32

PART I LCC FUEL ELEMENT TEMPERATURES AT ELEMENT POWER 1,100 5 3 2 T,=9.937E-20x - 8.701E-15K4 + 2.859E-10x - 4.349E-06X + 5.311E-02x + 3.217E+.02 R'=9.999E-01 1,000- Tm 3 2

-1.578E-15,c' + 1.072E-i0X - 2.504E-06x +~4.660E-02x + 3.274E+02 R'=9.998E-01U O 900

~23-V 800 4-,

'U 700 -

V a

E 600 cu I-500 U-3 400 300 T,_

1 ,=-1.071F-15o'

. .........

+ 7.622E-11x - 1.912E-06x + 3.471E-02x - 3.248E+02 R3= 9.999E-01

.. ....... -

0 5,000 10,000 15,000 20,000 25,000 30,000 Fuel Element Power (W)

FUEL&ZRAVE O-lZr,max OFTB Figure 21: LCC Correlations of Fuel Temperatures and Element Power The power produced in each fuel element was calculated by distributing the core power (for a range from 100 kW to 1210 kW) and then multiplying the average power by the peaking factor associated with element. The maximum temperature was calculated using the correlation in Fig. 22 with the fuel element producing the highest power. The fuel channel measuring temperature was simulated by applying the Fig. 22 correlation for the temperature at the thermocouple location to the B03 and B06 positions. The average fuel temperature was determined as the average of the temperature of each fuel element as determined by the correlation of Fig. 22. The results for fuel temperature at core power are provided in Fig. 23.

LCC FUEL TEMPERATURESAT CORE POWER 500 .

400 ...............

4UU

-- 300 E

200 100 Y .6E-08X 3 - 1.75E-04X2 + 3.98E-Olx + 5.48E+01 R2 = 1.OOE+00 0 1 0 200 400 600 800 1,000 1,200 1,400 Core Power (kW)

--e FT AVE Figure 22: LCC Fuel Temperatures at Core Power for the 80 Element 33

PART I Core at Limiting Pool Level and Temperature 4.3 Control Rod Worths Criticality calculations were performed assuming all control rods fully inserted, all control rods fully withdrawn, and control rods individually inserted. Excess reactivity for each condition was calculated as:

p($) =__1 --I_Pi 23 Pi "'feff Table 18 values are noted ARI for all rods fully inserted, SHI for shim 1 , SH2 for shim 2, RR for the regulating rod, and TR for the transient rod (with the remaining elements removed).

Table 18: LCC Reactivity Values Config. keff + $ Worth ($)

ARI 0.91680 0.00065 -12.964 17.697 ARO 1.03426 0.00061 4.732 NA RR 1.00580 0.00065 0.824 3.908 SH2 1.00943 0.00064 1.335 3.398 SH2 1.01074 0.00063 1.518 3.214 TR 1.01132 0.0007 1.599 3.133 The worth of each condition (p($) ) relative to the all rods out configuration (PARO) was calculated as:

M($ = PARO - Pi 2z PARO 'Pi

fleff Shutdown margin with all control rods fully inserted is $8.9, calculated:

25c5 PSDM = PARO - PRR - PSH1 - PSH2 - PTR Shutdown margin with the most reactive rod fully withdrawn is $5.0, calculated:

2E5 PSDM = PARO - PSH1 - PSH2 - PTR 4.4 Pulsed Reactivity Response Using the methodology previously described, the average fuel temperature in the core was correlated to the maximum temperature in the core and the fuel temperature measuring channel (Fig. 24).

Using a correlation (for 100 kW to 1,210 kW) between the core power and the average fuel temperature developed from Fig. 23, average fuel temperature is related to core power by:

TF,ave = 5.26

10-8 . p3 - 1.75. 10-4 . p2 + 0. 398

P + 54.8 2; 7 34

PART I The fuel temperature deficit from the temperature associated with operating at power was calculated from the fuel temperature coefficient for the 80 element LCC. The maximum pulse possible was determined by calculating excess reactivity at power as the reactivity deficit at power and the maximum allowed excess reactivity.

The MATLAB code for solving the in-hour equation was modified for the fuel mass and the prompt neutron lifetime (based on SCALE results for the 80 element core) associated with 80 fuel elements. The pulsed reactivity, initial fuel temperature, and initial power level for a range of values from 100 kW to the power where a $1 pulse is possible were used as input to determine reactor response to pulsing at power (Table 19, Fig. 25).

Table 19, LCC Response to Pulsing from Power Core Temp Peak Fuel Peak Power Power Deficit Tem kW °C $ $ °C MW 50E-3 25 0.00 3.00 408 1763 100 93 -1.42 2.58 430 1232 150 111 -1.70 2.30 405 864 200 128 -1.96 2.04 382 578 250 144 -2.21 1.79 359 354 300 160 -2.45 1.55 337 188 350 175 -2.68 1.32 316 78 400 190 -2.90 1.10 299 19 450 203 -3.11 0.89 304 2.2 RESPONSE TO PULSING ALL AVAILABLE REACTIVITY FROM CORE POWER 450 1 1,300

... ..

........ . ..... ..... -J. 1 1IM 410 .....................

I.......................... .

..L... . .......... ............

...... .................... ..

3 70 ........................ 900

..... .. 0 Q3) .... ........ . . .. ......................

. ...........

...........................

....................................

. ............................

.

0-,

3 7 0= ...............

. .....

.....

............................. .. ......... ..... 700 'ý E

500 :ý LL Ile 330 .........

................ .........

............... .................................... 300 CL 310 ...........

................................ . ..

......... * ......

....... 100

.....

290 -100 100 150 200 250 300 350 400 Core Power (kW)

[I PEAK FUELTEMP -0.PEAK POWER Figure 23: LCC 80 Element Core Response to Pulsed Reactivity Insertion 35

PART I Although the initial fuel temperature is higher while operating at power, temperature associated with the operation limits the maximum pulse and therefore the maximum peak fuel temperature. At 100 kW core power, the maximum fuel temperature is approixmately 40°C greater than the temperature generate by a three-dollar pulse, and well within any limiting value. At slightly greater than 150 kW core power, the maximum power and temperature are lower than generated by a three-dollar pulse.

4.5 Control Rod Speed The methodology for evaluating power and temperaure response to continuous control rod withrawal analysis by the DOW TRIGA reactor 14 using SIMULINK (a MATLAB product) was implemented for the UTTRIGA reactor using the LabVIEW Design and Simulation Suite. The UT TRIGA implementation was modifed to include (1) compensation for the initial neutron generation time based on reactivity, (2) speicific heat capacity from Simnad (op cit), (3) the use of the LabVIEW Runge-Kutta45 variable time solver, terminates rod withdrawal at scram (the DOW code continues rod withdrawl psot-scram), and adds reactivity with a ramp function based on rod speed (the DOW code uses a step function).

Table 20: Simulation of Continuous Control $5 Rod Withdrawal Reactivity Peak Temperature Max Excess Max Temp rate) Power at Peak Reactivity [60 s after]

%Ak/k-s MW °C $ °C 0.1 5 35 1.050 128 0.2 12 40 1.133 258 0.3 20 44 1.189 364 0.4 29 47 1.235 394 0.5 37 50 1.274 408 0.6 45 56 1.309 417 0.7 56 56 1.341 423 0.8 62 62 1.369 428 0.9 74 61 1.399 432 1.0 81 66 1.423 435 In this analysis the scram time was set long enough after intiiation of the rod withdrawal to prevent the scram from terminating the transient. Therefore, no credit was taken for actuation of the power reactor trip in determining maximum power, total reactivity, and maximum fuel temperature during the rod withdrawal. The maximum control rod worth since the installtion of the UTTRIGA was slightly larger than $4; for reactivity additon rates from 0.002 to 0.1 %Ak/k-s of a $5 (integral worthl) control rod, the maximum power, temperature, and total reactivity are much less than the results of pulsing to $3 (Table 19), realted to the introduciton of temperature feedback on a time scale capable of regualting power.

14 Analysis of the Thermal Hydraulic and Reactivity Insertion behavioroa the DOW TRIGA Research Reactor, Submitted to the NRC in Support of the DTRR License Renewal, M. R. Hartman (03/12/2011) 36

PART I 5.0 Conclusion Modeling the UTTRIGA reactor with SCALE and TRACE provides data consistent with experimentally determined values. Therefore, the models are expected to provide a reasonable estimate of reactor behavior.

When the models are applied to an 80 element core under limiting conditions of pool level and temperature, thermal hyraulic and temperature limts are preserved with a large margin. The current limit on pulsed reactivity insertion in TRGIA reactors restricts fuel temperature to less than 830°C, which does not occur for pulsing less than about $5.

37

PART I Appendix 1: Fuel Element Core Locations Element Utilization in Core ELEMENT 87 90 89 92 95 103 102 104 108 110 114 2899 46 46 46 46 46 46 15 15 15 15 15 2902 47 47 47 47 47 47 86 86 84 84 84 2903 48 48 48 48 48 48 116 116 116 116 116 2904 49 49 49 49 29 29 114 114 114 114 114 2905 50 50 50 50 50 50 39 39 39 39 39 2906 51 51 51 51 51 51 35 35 30 30 30 2908 52 52 52 52 52 52 101 101 101 101 101 2910 53 53 53 53 80 80 28 28 28 28 28 2911 54 54 54 54 54 54 97 97 97 97 97 2912 55 55 55 55 55 55 84 84 86 86 86 2913 56 56 56 56 98 98 122 122 122 122 122 2915 57 57 57 57 57 57 76 76 76 76 76 2918 58 58 58 58 58 58 72 72 72 125 125 2925 59 59 59 59 59 59 24 24 24 24 24 2927 60 60 60 60 27 27 108 108 108 108 108 2928 61 61 61 61 61 61 34 34 34 34 34 2929 62 62 62 62 62 62 43 43 43 43 43 2930 63 63 63 63 63 63 118 118 118 118 118 2931 80 121 20 20 20 20 20 2932 64 64 64 64 64 64 36 36 48 56 48 2935 65 65 65 65 65 65 117 117 117 117 117 2938 66 66 66 17 17 111 111 107 107 107 2939 67 101 67 119 119 119 30 30 35 35 35 2940 68 102 68 100 100 100 89 89 89 89 89 2941 69 69 69 69 69 92 25 25 25 25 25 2943 112 123 123 40 40 40 40 40 2944 70 70 70 70 70 70 19 19 19 19 19 2946 71 71 71 71 71 71 77 77 77 77 77 2947 72 72' 72 72 72 72 96 96 96 96 96 2948 73 73 73 73 28 28 121 121 121 121 121 2950 74 74 74 74 74 74 42 42 42 42 42 2951 66 75 75 43 43 103 103 103 119 119 2952 76 76 76 76 11 11 112 112 2954 77 77 77 77 26 26 123 123 123 123 123 2955 78 78 78 78 78 78 44 44 44 44 44 2957 79 79 79 79 97 97 106 106 106 106 106 2958 111 80 80 53 53 113 113 80 80 80 2959 81 81 81 81 81 81 29 29 29 29 29 2960 82 82 82 82 82 82 95 95 95 95 95 2962 83 83 83 83 83 83 32 32 32 32 32 2964 84 84 84 84 84 84 27 27.27 27 27 2965 85 85 85 85 85 85 17 17 17 17 17 38

PART I Element Utilization in Core ELEMENT 87 90 89 92 95 103 102 104 108 110 114 2968 86 86 86 86 96 96 102 102 102 115 115 2969 87 87 87 87 87 87 91 91 91 91 91 2970 88 88 88 88 45 45 110 110 110 110 110 2971 89 89 89 89 89 89 90 90 90 90 90 2974 90 90 90 90 90 90 38 38 38 38 38 2975 91 91 91 91 91 91 45 45 45 45 45 2976 92 92 92 92 92 69 107 107 111 111 111 2977 93 93 93 93 93 93 37 37 37 37 37 2979 94 94 94 94 94 94 26 26 26 26 26 2980 105 105 105 23 23 23 23 23 2983 95 95 95 95 95 95 21 21 21 21 21 2984 96 96 96 96 86 86 18 18 18 18 18 2985 97 97 97 97 79 79 11 11 11 11 11 2992 109 109 109 31 31 31 31 31 3013 114 114 114 14 14 14 14 14 3384 12 12 3496 72 3504 73 3513 99 99 99 99 99 99 93 93 93 93 93 3700 102 3703 74 5198 98 98 98 98 56 56 12 12 12 5283 26 5844 28 28 28 28 73 73 58 58 58 58 58 5845 29 29 29 29 49 49 46 46 46 46 46 5846 30 30 30 30 30 30 52 52 52 52 52 5902 31 31 31 31 31 31 63 63 63 63 63 5903 32 32 32 32 32 32 53 53 53 53 53 5904 34 34 34 34 34 34 64 64 64 64 64 5911 127 127 71 71 71 71 71 5912 35 35 35 35 35 35 55 55 51 51 51 5913 36 36 36 36 36 36.54 54 81 81 81 5914 37 37 37 37 37 37 67 67 67 67 67 5915 38 38 38 38 38 38 49 49 49 82 49 5916 17 17 17 17 66 66 88 88 88 88 88 5917 18 18 18 18 18 18 81 81 54 54 54 5918 39 39 39 39 39 39 48 48 36 36 36 5919 40 40 40 40 40 40 60 60 60 60 60 5920 42 42 42 42 42 42 73 73 73 104 104 5921 11 11 11 11 76 76 61 61 61 61 61 5922 12 12 12 12 12 12 98 98 98 98 98 5981 13 5982 13 13 13 39

PART I Element Utilization in Core ELEMENT 87 90 89 92 95 103 102 104 108 110 114 6142 43 43 43 43 75 75 68 68 68 68 68 6143 14 14 14 14 .14 14 87 87 87 87 87 6886 15 15 15 15 15 15 50 50 50 50 50 6889 16 26 26 26 77 77 66 66 66 66 66 6923 44 75 44 115 115 115 57 57 59 59 59 6924 19 19 19 19 19 19 78 78 78 78 78 6925 45 45 45 45 88 88 59 59 57 57 57 6926 20 20 20 20 20 20 92 92 92 92 92 6927 21 21 21 21 21 21 62 62 62 62 62 6928 23 23 23 23 23 23 69 69 69 69 69 6929 24 24 24 24 24 24 51 51 55 55 55 6930 25 25 25 25 25 25 65 .65 65 65 65 6931 83 103 6932 27 27 27 27 60 60 47 47 47 47 47 10146 33 33 33 33 33 33 33 33 33 33 33 10147 41 41 41 41 41 41 41 41 41 41 41 10148 22 22 22 22 22 22 22 22 22 22 22 10699 120 120 120 10700 124 124 124 10701 105 105 105 105 10702 109 109 109 109 10703 127 127 127 10704 100 100 100 10708 16 16 16 16 16 16 16 16 16 16 10810 101 74 74 74 126 126 10811 103 94 94 94 94 94 10812 104 79 79 79 79 79 10813 110 85 85 85 85 85 10814 111 99 99 99 99 99 10815 113 80 80 113 113 113 10816 120 75 75 75 75 75 10817 124 70 70 70 70 70 10878 13 13 13 13 13 13 13 40

PART II ANALYSIS OF THERMAL HYDRAULIC PERFORMANCE OF THE UNIVERSITY OF TEXAS TRIGA MARK II NUCLEAR RESEARCH REACTOR

PART II THERMAL HYDRAULIC ANALYSIS OF THE UTTRIGA REACTOR 1.0 Introduction This report documents analysis of the thermal hydraulic characteristics of the UTTRIGA in support of renewal of the U.S. Nuclear Regulatory Commission facility operating license.

The UT Austin TRIGA Research Reactor (UTTRIGA) is a TRIGA Mark-Il nuclear research reactor licensed to The University of Texas at Austin for operation up to 1.1 MW thermal power level. The geometry of the UTTRIGA core is based on seven concentric hexagons (designated as rings) that fix locations for fuel elements, graphite filled elements, and various experimental facilities. The core is surrounded by a modified cylindrical annulus in an aluminum container filled with graphite (neutron reflector), a rotary specimen rack (RSR), four beam port penetrations, and void spaces accommodating the RSR and beam port facilities. The core and reflector are located in an aluminum tank (pool) filled with high-purity water.

The water acts as a neutron moderator, coolant, and radiation shield.

Thermal hydraulic modeling of the UTTRIGA was performed with TRAC/RELAP Advanced Computational Engine (TRACE) using the Symbolic Nuclear Analysis Program (SNAP) interface. Thermal hydraulic characteristics were developed from classical methods and corrections for UTTRIGA geometry using the computational fluid dynamics code FLUENT. Distribution of fission activity was developed from transport calculations in SCALE, a comprehensive modeling and simulation suite for nuclear safety analysis and design.

The thermal hydraulic codes TRACE are designed to perform best-estimate analyses of operational transients and accident scenarios by modeling physical geometry and thermodynamic conditions. TRACE was developed for commercial nuclear reactors applications, and RELAP has been widely used in characterizing research reactor thermal hydraulic performance. TRACE is the NRC's flagship thermal-hydraulics analysis tool consolidating and extending the capabilities of NRC's 3 legacy safety codes - TRAC-P, TRAC-B and RELAP. The Symbolic Nuclear Analysis Package (SNAP) is a graphic user interface that standardizes input and interaction for supported analysis codes.

NRC guidance' defines a "limiting core configuration" as the core that would yield the highest power density using the fuel specified for the reactor, with all other core configurations demonstrated to be encompassed by safety analysis for the limiting core configuration. The guidance references an "operational core." Analytical methods used to define the limiting core configuration are applied to the operational core, providing confidence that the model adequately supports limiting core configuration analysis.

2.0 General Description of Heat transfer at the UTTRIGA Heat is generated in the fuel by the fission process. Cooling is required to maintain fuel temperature low enough to prevent challenges to cladding integrity. The UT TRIGA reactor operates in a natural convection-cooling mode. Heat transfer from fuel to the coolant in the core area is developed by generation of heat in the fission process, conduction of the heat to external surface of the fuel element, and heat transfer by convection from the fuel element surface to water in the core area.

Temperature increase of the water in the core area develops buoyancy forces that drive flow. The flow is diminished by momentum changes and friction (across the gird plates, fuel element end fittings, and fuel I NURGE 1537, Guidelines for Preparing and Reviewing Applications for the Licensing of Non-Power Reactors, Format and Content 2

PART II element cladding surfaces). Above a "critical" heat flux, coolant flow will not be adequate to prevent thermal hydraulic conditions from exceeding limits. This analysis demonstrates that operation at the maximum licensed power level has adequate margin to the critical heat flux.

3.0 Power Distribution The distribution of heat generation across the fuel elements in the core is affected by the core configuration. The amount of heat generated in a specific fuel element can be characterized as a "peaking factor," the ratio of the power produced in that element to average (total heat distributed equally over all fuel elements). A larger number of fuel elements tend to exacerbate the peaking factor of higher power fuel elements. However, the average power per element is reduced by a larger number of fuel elements so that maximum power produced by a fuel element tends to decease. Determining the maximum power produced by a fuel element requires evaluation of peak to average power ratios for the core configuration.

Distribution of heat production within a fuel element also varies spatially, affecting the distribution of fuel temperature in the element as well as localized heat transfer.

3.1 General "Core power" refers to the total power produced by all fuel elements, and "average power" (per element) is the core power distributed uniformly across all fuel elements. The ratio of a specific fuel element power to the average power per element is referred to as core peaking factor. The hot channel is the fuel element producing the maximum power (the fuel element with the largest peaking factor) and the surrounding cooling flow. The fuel element and cooling channel geometry is reduced for thermal hydraulic calculations to a "unit cell" (repeatable geometry that can be used to replicate the geometry of the fuel in the core). Acceptable thermal hydraulic performance of the UTTRIGA is based on the heat generated in the unit cell corresponding to the hot channel.

Neutron flux has a spatial distribution across the core, causing variations in the rate of fission reactions in fuel elements. The variation is influenced by fuel element location, local geometry, and fuel element materials. SCALE transport codes calculate the fraction of total fissions generated in each element, allowing core peaking factors and the fuel element producing the most power to be identified directly.

Neutron flux also varies within fuel elements, creating spatial variations of heat production within the fuel matrix. More discretized fuel element modeling is used in SCALE to calculate the fraction of fissions occurring in segments of the fuel element. Segmentation allows development of a mesh of radial and axial fuel elements to define the heat-generation structure of the unit cell. SCALE reports the fraction of fission occurring in the segments, used to evaluate spatial variation in power production. These 2-dimensional distributions can be used explicitly in TRACE, while RELAP assumes the distribution can be decomposed into independent axial and radial factors. Analysis is core-specific to the extent that the power distribution specified for a specific fuel element varies with core configuration and burnup.

Fuel element material compositions were calculated for each element in neutronic analysis of the UTTRIGA SCALE model. The SCALE calculations used to develop fuel element material inventories are based on uniform fuel composition. However, burned-material and neutron-flux distribution are not independent; since neutron flux varies spatially, the products of neutron reactions are expected to vary.

Limits on microprocessor capabilities prevent discretizing of the internal fuel element structure. The effects of this assumption are mitigated in thermal hydraulic analyses since maximum fuel burnup is correlated directly to maximum power production; higher burnup regions are likely to have lower power and lower local temperature compared to calculations with uniform material composition. Fuel in the center an element that has burnup will consequently generate less power in comparison to SCALE calculations that assume uniform material composition. Consequently the effect on fuel temperature 3

PART II calculations is assumed to be conservative, and small.

Similarly, the initial SCALE calculations assumed a uniform fuel temperature for all fuel elements consistent with full power operation. However, fuel temperature and fission rate are not independent as elevated temperatures lower the fission cross section. Higher temperatures near the core midplane during reactor operation are likely to reduce the local fission rate. Consequently, the assumption of uniform temperature in SCALE calculations is expected to result in higher element peaking factor compared to actual reactor operations. Therefore, the effect on fuel temperature calculations is assumed to be conservative, and small.

Limiting pool water level (5 feet above the core) and pool water temperature (49 °C) bound the limiting core configuration, while nominal pool water level (6.25 feet above the core) and pool water temperature

(-25 'C) apply to the remainder of the analyses.

3.2 Criticality Calculations SCALE calculations were performed to determine first the minimum number of close packed fuel elements at ambient temperature (300'K) for criticality, and second the minimum number of close packed fuel elements required for operation at an assumed full power operating temperature (600'K).

The minimum number of fuel elements required for criticality at power is the lowest number of fuel elements possible for the limiting core configuration. The actual limiting core configuration is selected by calculating margin to thermal limits for the single fuel element generating the highest power. Three assumptions are used to calculate the minimum number of fuel elements required.

(1) Calculations were performed with graphite dummy rods and then water voids in all positions which do not contain fuel.

(2) Calculations were performed using material specifications for fresh fuel. Since reactor operation reduces fissionable material and introduces fission product poisons into a fuel element, the number of fuel elements required to support full power operations with fresh fuel is the minimum. As the number of fuel elements increases, the distribution of heat generation over more fuel elements reduces the heat generated in the hot channel.

(3) Fuel material specifications are assumed to be the average initial (unirradiated) values of all TRIGA fuel elements possessed under the UTTRIGA reactor license.

Results of reactivity calculations from 40 to 89 fresh fuel elements at ambient (300'K) and an assumed uniform temperature consistent with full power steady stare operations (600'K) are provided in Fig. 1, with excess reactivity calculated as:

_ k~ff -1 I

~EX-1 Minimum Number of Fuel Elements for Criticality and Operation Criticality at ambient temperature requires a minimum of 55 fresh fuel elements. The minimum number required for criticality at full power operating temperature is 69 fuel elements. Although a maximum of 78 fuel elements can be loaded in a close packed core with graphite rods in the remaining core spaces and remain within the maximum excess reactivity limits, replacing graphite rods with water voids reduces excess reactivity and allows more fuel to be loaded. Actual loading that meets reactivity limits is 4

PART II determined and validated experimentally.

Excess Reactivity and Average Rod Power Vs. Core Configuration (No. of Fuel Elements)

$4 Maximum Excess o

'4

$2 - 4l 20*'II 0

M -Critical 4- *42 m ' 14w 402 a

-44 17 U' 44-4 N-. 14 IA-110

-014 - X-14

-$12 -

10 U-

-$14 10 20 42 Il o ?1 K2 $

Number of Fuel Elements in Core Figure 1, Excess Reactivity and Average Rod Power Hot Channel Selection Analysis (using SCALE) was performed for core configurations to evaluate peaking factors and the heat generated in each B ring element (Table 1 and Fig. 2) at the nominal core power of 1210 kW (1100 kW licensed power with a maximum potential instrument error of 10%).

MAX ROD CHANNEL POWER, 1210 KW CORE AT VARYING CORE CONFIGURATIONS 40

- 35 30 0

0-c 25 M

-C U

~20' y = -3.866E-05X 3 + l.25SE-02-X2 - l.514E+00x + 8.524E+01 R2 = 1.000E+00 40 50 60 70 80 90 100 110 120 Number of Fuel Elements Figure 2, Maximum Channel Power at 1210 kW The average fuel element power was calculated distributing the core power over the number of fuel elements in each configuration. The power generated in each element is the product of the applicable 5

PART II peaking factor and the average power. The hot channel power values for the remaining cores demonstrate a definite and decreasing trend in the maximum hot channel power. The 80 element core is selected as the minimum number of elements in the limiting core configuration.

3.3 Power Distribution within Fuel Elements Interactions in the outer radial reduce neutron flux in the inner radial segments of a fuel element, and statistics associated with fission rates near the center of the fuel element are challenging. The Monte Carlo calculations in areas of lower neutron flux in smaller dimensions require a significantly larger number of histories to reduce noise (piecewise variation). Consequently the SCALE model modification segmenting the fuel element axially and radially is based on equal volume segments (Table 1).

Distribution fractions as used are the fraction of power in a specific segment to average fraction of power across 225 equal volume segments. Although fission distributions are calculated for each segment, distributions are provided in this report only for projections (1) near the radial and axial extremes, (2) at the respective centers, and (3) fuel element averages (to reduce complexity).

Table 1, Geometry for Fuel Segments Axial Segments Radial Segments Z Z2 Zave ri r 2 rave 1 19.05 16.51 17.78 0.3175 0.5603 0.4389 2 16.51 13.97 15.24 0.5603 0.7260 0.64315 3 13.97 11.43 12.7 0.7260 0.8604 0.7932 4 11.43 8.89 10.16 0.8604 0.9764 0.9184 5 8.89 6.35 7.62 0.9764 1.0801 1.02825 6 6.35 3.81 5.08 1.0801 1.1746 1.12735 7 3.81 1.27 2.54 1.1746 1.2621 1.21835 8 1.27 -1.27 4E-15 1.2621 1.3439 1.303 9 -1.27 -3.81 -2.54 1.3439 1.4210 1.38245 10 -3.81 -6.35 -5.08 1.4210 1.4941 1.45755 11 -6.35 -8.89 -7.62 1.4941 1.5638 1.52895 12 -8.89 -11.43 -10.16 1.5638 1.6306 1.5972 13 -11.43 -13.97 -12.7 1.6306 1.6947 1.66265 14 -13.97 -16.51 -15.24 1.6947 1.7564 1.72555 15 -16.51 -19.05 -17.78 1.7564 1.8161 1.78625 Fission generation data for each segment was used to calculate the fraction of fissions in the fuel element that occurred in each segment. Peaking factors in the axial or radial direction can be calculated by normalizing the sum of the fractions on the axis to the average of all summations on the axis.

The analysis for the core peaking factors of the current 114 element core is provided in in Fig. 3. In this analysis, the fuel element in position B02 produces the maximum power. Analyses for axial and radial variations in B04 for the 114 element core are provided in Fig. 4 and 5.

Axial Peaking Factors Axial peaking factors were developed for cores representing slightly less than the minimum number of fuel elements required for full power operation, the initial UTTRIGA core, and the current core configuration. While there is some variation in the axial power distribution,'the analyses reported in 6

PART II Table 2 indicates suggest that axial power distribution for the specified locations is similar regardless of the core configuration. The results show that normalized axial power distribution is relatively stable over varying core configurations. Although specific power distributions were used to develop input for thermal hydraulic analysis the effect on axial distribution, and any associated errors, should be insignificant. The major effect on variations in power distribution is on core-wide, fuel-element peaking factors.

0.63 0.65 0.77 0.56 0.65 0.61 0.76 0.88 0.97 1.12 0.84 0.89 H0 0.79 0.95 1.08 1.21 1.27 1.22 0.97 0.94 0.71 0.67 0.83 1.181.3 08E I 0.61 0.84 0.94 1.21 1.25 1.35 1.21 0.91 0.54 0.66 0.7 1.03 125 1.40.941.0 0.00 0.62 0.2 1.12 1.32 1.49 1.20 0.78 0.88 0.63 H2 0 2H 1.27 1.4 1.27 1.34 1.07 0.85 0.95 1H, 0.97 1.12 1.23 0.97 0.93 0.72 0.74 0.92 1.01 1.14 1.12 1.04 1.05 0.77 0.53 0.76 0.64 0.92 0.87 0.81 0.77 0.78 0.59 0.72 0.70 0.74 0.66 Figure 3, Peak to Average Power for 114 Element Core at 600 °K 114 Element Core Radial Fission Distribution Selected Elevations 0.O D7o...... .-i 0.0075 ..........

.......

. . .. .......... ..... - . ] ...

.................

.........................

...

.....

.... ............

w* 0.0065

.... . ........ - x

..... ... ....... ...........

0.00 65 E 0.0055 .. .... . ... . X 0 0.0050 X i.. ..... ,- _

"- ................ .........

0 .o o M 5s " ..................

_ -°X 0.0040 0.0035 LI.

0 0.0025 ir 0.0020 . .

0.32 0.52 0.72 0.92 1.12 1.32 1.52 1.72 1.92 Distance from Midptane/Center(cm) 0 LOWER 0 MIDPLANE A UPPER X AVERAGE Figure 4, 114 Elements 600 °K, B04 Radial Power Distribution 7

PART II 114 Element Core Axial Fission Distribution Selected Radii u 0.007 E

- 0.006 -

0

U 0

.P 0.004

'4 C 0.003 0

0.002

-20.0 -15.0 -10.0 -5.0 0.0 5.0 10.0 15.0 20.0 Distance from Fuel Center (cm)

O CENTER 2 OUTER SURFACE A 1.3 cm XAVERAGE Figure 5, 114 Elements 600 'K, B04 Axial Power Distribution Table 2, Fuel Element Axial Peaking Factors No. of Fuel Elements in Core AVE DEV Segment 70 87 114 1 0.65 0.63 0.66 0.64 1.83%

2 0.76 0.76 0.78 0.76 1.28%

3 0.91 0.90 0.91 0.91 0.52%

4 1.04 1.04 1.04 1.04 0.27%

5 1.14 1.14 1.14 1.14 0.26%

6 1.22 1.22 1.22 1.22 0.28%

7 1.26 1.27 1.26 1.26 0.40%

8 1.27 1.28 1.27 1.27 0.39%

9 1.25 1.25 1.24 1.25 0.65%

10 1.20 1.20 1.19 1.20 0.30%

11 1.12 1.12 1.11 1.12 0.47%

12 1.01 1.02 1.00 1.01 0.52%

13 0.87 0.87 0.87 0.87 0.36%

14 0.72 0.71 0.72 0.72 0.45%

15 0.60 0.59 0.59 0.59 0.76%

4.0 Thermal Hydraulic Modeling, Unit Cell Geometry and Thermal Hydraulic Characteristics The flow channel unit cell cross section is based on the fuel element geometry, as illustrated in Fig. 6 (unit cell and the surrounding fuel elements). As illustrated, the unit cell is a fuel element and the surrounding flow area (end fittings have more complex geometry) circumscribed by a hexagon with an inner radius of 1/2 of the pitch. The cooling flow channel is modeled as a heated pipe with thermodynamic characteristics based on physical dimensions and properties of the coolant around the fuel elements. A large fraction of the unit cell is occupied by the fuel element, leaving a relatively small flow area. The complex geometry of the fuel element end fittings are approximated as hydrodynamic characteristics.

8

PART II Since a regular hexagon can be decomposed into six equilateral triangles, a triangular unit cell is the smallest possible unit cell. However, RELAP and TRACE heat structures (described in a following section) have limited options for temperature analysis of sold structures; a cylinder can be used to develop a heat source, but a half-cylinder is not possible. This does not limit fluid analysis in thermal hydraulic calculations with a triangular unit cell, but limits the ability to calculate temperatures in the fuel element since the geometry of a triangular unit is Y2of the heat contribution from a single fuel element. Intrinsic properties used to calculate thermal hydraulic conditions are fully represented, but total heat for the cylindrical fuel element (used in material temperature calculations) in a triangular unit cell is not.

Fuel Dia. = 1.4748 Pitch = 1.7149 in.

COOLING WATER CDING Inner Radius = 0.857 in.

Outer radius = 0.990 in.

FUEL Zr FILL ROD/

Figure 6, Flow Channel for UT TRIGA The volume of the flow channel is calculated as the product of the flow area and length. The length of the TRIGA flow channel is defined for the heated (adjacent to fuel) and unheated surfaces of fuel element cladding. The heated length is divided into smaller sections for analysis, consistent with the axial segmenting indicated in Table 1. The geometries and thermal hydraulic parameters of the upper and lower gird plate/fuel element are calculated through equations 2-10, with results summarized in Table 3.

4.1.1 Unit Cell Geometric Parameters The area of a regular polygon is calculated using the interior radius (r,) and perimeter (P) as:

1 A=- r,P 2 2'

The unit cell is a hexagon (i.e., 6-sided perimeter) with each side one leg of an equilateral triangle; the height of the triangle is the hexagon's interior radius. The hexagon/triangle dimension (a) in terms of the internal radius is calculated:

2 a =T . 3 Substituting 4.1 into 4.2, the cross sectional area of the hexagonal unit cell (A g_)using the interior radius is therefore:

14 rl 1C 6r2= -. ~ l=~

~

'J. r~

2 2 (6. 3) 2 9

PART II The inner radius of the unit cell is /2 the distance between two fuel elements or Y of the fuel element pitch (Pe) so that:

2 5 The cross sectional area of a fuel element (AF) is calculated:

A," 6 The area of the flow channel in the unit cell (Alc) is the difference between the unit cell area (eq. 4.1) and the area occupied by fuel (eq. 4.2). Since the interior radius is Y of the pitch, the area of the flow channel is calculated by:

A FC ,=i p2 7 Non-circular pipes are approximated as a pipe with an equivalent hydraulic diameter (Dh) with a wetted perimeter (Ply), where the hydraulic diameter is calculated as:

4 AF(., 8 PDV The wetted perimeter is the length of the flow channel in contact with channel wall surfaces (i.e., the perimeter of the fuel element):

P,, = ;r"- D,ý. 9 Substituting equations 4.3 and 4.4 for flow area and perimeter into equation 4.5, hydraulic diameter is:

_2 -4 2"3P_ _ Pe 1 D 22 F 2 DF 10 Asumm ry, an c a parpi A summary of primary and calculated parameters is provided in Table 3.

9 3, Summary of Principle Thermal Hydraulic Values Description Var. Value Fuel Element Pitch P 1.714 in 0.142833 ft 4.35356 cm 0.043536 m Fuel Element Diameter Dfuel 1.4784 in 0.123200 ft 3.755136 cm 0.037551 m Wetted Perimeter P~, 4.64453058 in 0.387044 ft 11.79711 cm 0.117971 m Fuel Cross Section/Area At 1.7166185 in2 0.011921 ft 2 11.07494 cm 2 0.001107 m2 Unit Cell Area A cel 2.54420597 in2 0.017668 ft 2 16.4142 cm 2 0.001641 m2 AFC 0.82758747 in2 0.005747 ft2 5.339263 cm2 0.000534 m2 Flow Channel Area Hydraulic Diameter Dh 0.71274154 in. 0.059395 ft. 1.810364 cm 0.018106 m 10

PART II 4.1.2 Unit Cell Thermodynamic Loss Factors Pressure drops (head loss) across hydraulic components are the product of the fluid flow and factors such as the coefficient of friction between the fluid and the pipe wall, changes in flow area and diameter, flow channel surface roughness, and/or flow channel length. Within limits, the factors (Kfactors) are constant, the sum of the pressure drops in linear flow is additive. This analysis provides a traditional approach to evaluating the loss factors and loss factors reported by analysis and experiments conducted at the UT reactor, followed by the results of analysis and experiments conducted for the UTTRIGA facility.

Traditional Loss Factor Calculations The impact of sudden expansion or contraction is principally in velocity changes. Bernoulli's equation applied to non-compressible fluids relates area and velocity. The Kfactors for sudden expansions or contractions are based on the ratio of the flow areas (Equation 7).

Ký = [I_11]=[IA Other K factors are based on the magnitude of the direction change, the pipe surface roughness, and flow mode (turbulent, laminar, etc.). Calculations are simplified by using the Darcy-Weisback friction factor (1) as a multiplier on applicable aspects of system geometry. The friction factor is a function of the Reynolds number, wall surface roughness, and flow channel. The relationship is described in the Colebrook formula:

=-2.0. log,,- + 2.51 12 3.7.0D Re. -f In practice, the Moody chart (Fig. 7, a parametric representation of the friction factors) is frequently used to determine the friction factor. For reasonable and expected flow rates at the TRIGA reactor, the Reynolds number is between 1X10 4 and 3X10'. Over this range, convergence exists for wall surface roughness values between 5X107 to 1X10-3. The broad range of surface roughness values indicates a very low sensitivity for roughness, and that any surface roughness within this range can be used without 2

affecting the friction factor significantly. For comparison RELAP analysis conducted for DOW Chemical reactor used surface roughness of 2.13E-6.

For losses in a straight pipe:

L K 13 D

For a 45* turn:

K 45 = f"16 14 For a 90' turn:

K 0,= f. 30 15 2 ANALYSIS OF THE THERMAL HYDRAULIC AND REACTIVITY INSERTION BEHAVIOR OF THE DOW TRIGA RESEARCH REACTOR, Submitted to the NRC in support of the DTRR License Renewal (M. H. Hartman, 03/12/2011).

11

PART II Table 4: Channel End Geometry Location Component Eff. Area Bottom Entrance Lower grid plate 1.2 cm 2 Bottom Exit Lower End fitting/Channel 3.9 cm 22 Top Entrance Upper End Fitting/ Channel 3.9 cm Top Exit Upper Grid Plate 1.2 cm 2 The K factor for elevations above the flow channel is based on a 450 turn out of the main channel (0.344) and sudden contraction at the upper grid plate (0.43). The Kfactor below the flow channel is based on a sudden expansion exiting the grid plate (0.9) and a 45' turn (0.344) into the main channel. Therefore the Kfactors are 1.244 at the inlet and 0.844 at the outlet. The results of calculations for Kfactors associated with the hydraulic parameters in Tables 3 and 4 and are provided in Table 5.

Table 5: Classical K factors Location Characteristic K Factor 450 Turn 3 0.344 Inlet Expansion 0.9 TOTAL lower 1.244 45' Turn 0.344 Outlet Contraction 0.43 TOTAL Upper 0.774 Correlation of K factors and flow are based on historical, experimental measurements with cylindrical pipes. Additional work validated this approach for rectangular ducts. In practice, non-circular cross sections are reduced to a flow area and hydraulic diameter with the length as measured for the pipe.

However, the complexity of the TRIGA inlet and exit flow channel geometry is challenging. As fluid interacts with non-circular structures (or components), non-uniform surfaces can result in forces leading to secondary and/or internal flow paths that affect head loss/pressure drops. This suggests two potential issues using K factors calculated classically in analyzing thermal hydraulic response of the TRIGA reactor.

The actual entrance and exit to the flow channel between the grid plates is directed by fins mounted on a conical shape that terminates in cylindrical alignment (bottom end fitting) and handling (upper end fitting) structures. The wetted perimeters and flow areas vary continuously from entrance and exit for each end fitting.

" The interface between adjacent fuel channels is not separated by a physical boundary.

Differential pressure between adjacent flow channels at interfaces can support cross-channel flow.

Therefore thermal hydraulic analysis to support relicensing was developed 4 to:

(1) Model the UT TRIGA reactor using TRACE (2) Develop an independent solution tool using MATLAB to calculate thermal hydraulic performance based on mass and energy balance and K factors, 3 Friction factor times 16 4 Development of Thermal Hydraulic Correlations for the University of Texas at Austin TRIGA Reactor Using Computational Fluid Dynamics and In-Core Measurements, A. D. Brand 12

PART II (3) Develop a computational fluid dynamics model using FLUENT, and (4) Conduct experiments to develop a UT TRIGA specific heat transfer correlation Traditional Moody Diagram Z -- :

... .... ..

T'01 Figure 7, Moody Diagram These methodologies were used to independently model thermal hydraulic performance from (1) first principles, (2) TRACE thermal hydraulics code, and (3) FLUENT computational fluid dynamics code. The results of experiments in the TRIGA core were used to evaluate UTTRIGA-specific K factors based on actual fuel element geometry. A summary of K values determined from both the traditional/classical method and the UT analysis is provided in Table 6, with a fractional deviation between factors provided.

For comparison, RELAP work5 performed for DOW Chemical facility used Kfactors of 2.26 and 0.63 for the lower and upper channels.

Table 6, K Factors APPLICATION CLASSICAL FLUENT6 DEVIATION Lower Channel 1.244 1.63 23.7%

Upper Channel 0.844 1.12 33.6%

The values determined from the UT research program were used in modeling for TRACE calculations.

4.2 Physical UTTRIGA Thermal Hydraulic Model Standard TRACE components are structured to simulate physical characteristics of flow loop components.

Descriptions of the TRACE components required to characterize the UTTRIGA hot channel are provided below, followed by the specific facility application.

s ANALYSIS OF THE THERMAL HYDRAULIC AND REACTIVITY INSERTION BEHAVIOR OF THE DOW TRIGA RESEARCH REACTOR, Submitted to the NRC in support of the DTRR License Renewal (M. H. Hartman, 03/12/2011).

13

PART II 4.2.1 Fluid System Component Modeling TRACE analysis is based on modeling a set of representative, defined components where component characteristics are specified by the user to model the system. The UTTRIGA model uses Break, Pipe, Heat Structure, and Power Components. Heat structure material properties are used to calculate temperature distribution for fuel element components (zirconium fill rod, U-ZrH matrix, gas gap, and cladding).

a. Break: A break component is a boundary component normally used to provide a sink for liquid flows exiting the system. TRACE also uses a "Fill" as a similar component for inlet flows, but the fill flow rate is specified by the user while flow rate in a break is developed in calculations, and therefore not constrained. Since flow rates in the UTTRIGA model are developed by convection during reactor operation, the flow rate is not specified as an input. Therefore the use of a fill is precluded and breaks are used to specify both the entrance and exit conditions.

b. Pipe: The pipe component is a cylindrical volume containing water flow with various geometric and hydrodynamic properties. Analysis of the flow loop requires the flow across changes in elevation balance. Analysis requires limits on the magnitude of changes in adjacent flow areas.

b. Heat Structure: TRACE defines heat structures as rigid components that absorb, transfer, or radiate heat. A heat structure is specified by geometry, inner and outer radial boundary conditions, and material information. These attributes are specified in the "general" section. Power distribution is specified in the "Power Component" section of TRACE.

Heat structure cells are axially uniform. Geometry is specified in the "Radial Geometry" section.

Geometry includes both radial data which is constant along the axial length of the structure, and axial data that identifies surface areas for heat transfer. Initial conditions for the surfaces at each axial location are specified in "Initial Temperature." The gas gap heat transfer coefficient is explicitly specified the TRACE in the "Gas Gap HTC." Boundary conditions for heat transfer are specified for axial nodes/surfaces, linking the heat source to the heated lengths of the pipe to represent the active (fueled) part of the fuel element.

e. Power Component: The power component specifies how power is provided to the heat structure.

The shape of power distribution in TRACE is managed by specifying the fraction of power supplied between the inner and outer radial boundaries at each axial node. The "power shape" section of the power component is used to specify a 2 dimensional distribution. The power distribution fractions are specified based on axial and radial locations that segment the fuel element.

The Numerical solutions to the heat transfer equations are determined iteratively in TRACE. Iterative calculations with large step changes may lead to instability in solutions. A Power Table in the "general" section of the power component allows time based changes to simulate steps in calculation leading to a final power level.

f. Materials: TRACE has a limited set of material characteristics applicable to nuclear power plants. The default set of materials can be augmented by the user.

User defined materials are defined either in a data table or a functional fit table. The functional fit is a 5 th order polynomial in temperature, although setting the coefficients to 0 can reduce the order of the polynomial. Properties are specified over a range of temperature, and include (1) density, (2) specific 14

PART II heat, (3) thermal conductivity, and (4) emissivity.

4.2.2 UTTRIGA Application TRACE components were assembled to model the thermal hydraulic performance of the unit cell flow channel as shown in Fig. 8. User supplied values for the source, downcomer, connecting pipe, fuel element, and sink simulate the thermal hydraulic characteristics of the components.

BreakLt 10 181eak Dowcoinwr Connecting Pipe TRACE Model Figure 8, TRACE Model

a. Break Components applications:

The TRIGA hot channel pressure and temperature specifications are based on local environmental conditions (barometric pressure, confinement pressure regulation) and the pool (level and water temperature), specified as in the TRACE break component.

The NETL building is approximately 240 m above sea level, corresponding to 96 kPa at standard atmospheric conditions. The reactor bay confinement system is designed to control differential pressure to 0.06 in. (14.9 Pa) below atmospheric (minimal compared to atmospheric pressure). Total pressure at the top of the core is therefore:

PA = 96{KPa} +p,,, 16 Pool water is a minimum of 5.25 m above the core, nominally 7.25 m. Constant pressure is established by setting the "rate of change" variable to zero in the break. Pool water temperature is limited to less than 49 °C, nominally 25-27 *C. Where g denotes the gravitational constant (9.8 m's- 2),

the pressure (PH2o) exerted by a column of water (at density p in kg-m 2and height h in m) is given by:

pH2,o = p .g .h 17 A second break is connected to the exit of the core, simulating exit from the flow channel.

Parameters associated with the exit are the same as the entrance. Pressure boundary conditions for 15

PART II the limiting and nominal cases are provided in Table 7.

Table 7, Pressure Boundary Condition Temp Density Height Hydrostatic Pressure Pressure Condition TPressure oC 2 kg'm m kPa kPa psia Limiting 49 988.4881 5.25 50.9 146.9 21.3 25 997.0479 7.25 70.8 166.8 24.2 27 996.5162 7.25 70.8 166.8 24.2

b. Pipe applications:

Three pipes are used in modeling. One pipe represents movement of cooling flow from the top to the bottom of the flow channel. A second pipe moves flow to the entrance of the flow channel, connecting the down comer to the third pipe, the flow channel.

Down Comer/Cold Leg Conservation requirements for calculations require balanced elevation changes, with a "downcomer" at the same length and area as the fuel element region. Instabilities can occur in TRACE calculations if adjacent volumes are sufficiently different, and the downcomer is segmented to meet the ratio criteria (for convenience, segmenting has equal lengths). Dimensions for the downcomer pipe are provided in Table 8.

Table 8, Down-comer Pipe Length (segments) 0.09985 m Length (total) 0.5991 m Flow area 5.39E-4 m2 Volume (segments) 5.38E-5 m33 Volume (total) 3.23E-4 M Hydraulic diameter 0.0183 m Height Change (segments) -0.09985 m Height Change (total) -0.5991 m Connector A pipe with two elbows (Fig. 9) connects flow from the downcomer to the unit cell flow channel.

Dimensions of the connecting pipe are provided in Table 9.

Table 9, Connecting Pipe FLOW HEIGHT AREA HANGE SEGMENT VOLUME LENGTH AREA CHANGE mi3 m mi2 m 0.1m 1 5.38E-05 0.01 5.39E-04 -5.OE-3 Figure 9, Cold Leg to Flow Channel Connector 2 5.38E-05 0.01 5.39E-04 0.0 3 5.38E-05 0.01 5.39E-04 5.OE-3 Unit Cell Flow Channel/Fuel Element Region 16

PART II The flow channel for the fuel element region in the unit cell is modeled as a pipe. Specifications for the simulated fuel element cooling channel are provided in Table 10. Inlet and outlet geometry are reduced to loss factors (previously discussed K). The K factors as previously described are applied to the 2 nd and the 1 9 th segments.

Table 10, Specifications for Unit Cell Flow Channel VOL LENGTH FLOW AREA Az 2

m3 m m m 1 5.14E-06 0.01905 2.70E-04 0.01905 2 2.43E-05 0.09 2.70E-04 0.09 3 6.86E-06 0.0254 2.70E-04 0.0254 4 6.86E-06 0.0254 2.70E-04 0.0254 5 6.86E-06 0.0254 2.70E-04 0.0254 6 6.86E-06 0.0254 2.70E-04 0.0254 7 6.86E-06 0.0254 2.70E-04 0.0254 8 6.86E-06 0.0254 2.70E-04 0.0254 9 6.86E-06 0.0254 2.70E-04 0.0254 10 6.86E-06 0.0254 2.70E-04 0.0254 11 6.86E-06 0.0254 2.70E-04 0.0254 12 6.86E-06 0.0254 2.70E-04 0.0254 13 6.86E-06 0.0254 2.70E-04 0.0254 14 6.86E-06 0.0254 2.70E-04 0.0254 15 6.86E-06 0.0254 2.70E-04 0.0254 16 6.86E-06 0.0254 2.70E-04 0.0254 17 6.86E-06 0.0254 2.70E-04 0.0254 18 2.43E-05 0.09 2.70E-04 0.09 19 5.14E-06 0.01905 2.70E-04 0.01905 Total 1.62E-04 0.5991 5.13E-03 0.5991

c. Heat Structure Application The heat structure consists of 15 axial cells connected to the heated section of the flow channel (cells 3 through 17 of the unit cell flow channel pipe). "Outer surface boundary" conditions are connected to the unit cell pipe segments, with "Inner Surface Boundary Conditions" of 0.

Heat structure cells simulate the zirc fill rod at the center of the fuel element, ZrH-U fuel, the gap between the fuel and cladding, and the cladding. The UTTRIGA model includes:

zirconium from a radius of 0 cm to 0.3175 cm (3.175E-3 m)

zirconium-hydride from a radius of 0.3175 cm to 1.74117 cm (0.0174117 m), subdivided into 15 equal volume segments

gap gases from a radius of 1.74117 cm to 1.8161 cm (0.018161 m)

" stainless steel 403 cladding from a radius of 1.8161 cm to 1.8263 cm (0.018263 m)

The gas gap heat transfer coefficient 7 of 2840 W m2 K' is specified in TRACE as "Gas Gap HTC."

Power distribution is accomplished in the heat source geometry. The axial segments are divided 7 Reference for the gas gap heat transfer coefficient 17

PART II radially to provide equal volume segments. As previously discussed, heat generation is distributed in the heat structure based on SCALE transport calculations, based on the fission rate in each segment.

The SCALE model provides data for each of the 225 segments of the radial and axial boundaries, and the complete distribution is used in TRACE (Power Component section).

d. Power component: Two sections of the power component module are used in the UTTRIGA model, "General," and "Power Shape."

General Large changes in power can cause instability in calculation; the "Power Table" allows incremental steps at user specified times form a minimum to maximum power, allowing the calculation to stabilize. This function is accomplished in RELAP through a data table in the Control Systems section.

Power Shape The 2-dimensional fission density profile as described previously is used in the TRACE power shape.

e. Materials: Material data is specified in the Thermal section of TRACE Materials section. A library of reactor material characteristics are provided, but only "gap gases" and "Stainless 304" apply to TRIGA fuel; characterization of ZrH-U fuel and zirconium is required.

The thermal conductivity of TRIGA fuel is noted to be 0.042 cal.s.-lcm-i.°C-1 (17.573 W.m1 .°K1 ) 8, insensitive to temperature. The volumetric heat capacity calculated (Cp, referenced to temperature Tin °C) as:

Ci,((-rH,,.6,)=2.04+4.17x .T Ws3.{} 18 Specific heat capacity is calculated by normalizing the volumetric heat capacity by the density (p),

with the density of the fuel in the matrix (PL-zrHI.6) calculated as:

PU-ZrH,1.6 { C 3 -- 141--/ZrH r 19

+

PA PZrH With subscripts indicating Uranium and Zirconium-Hydride, the weight-percent of the components represented as w, the density of ZrH 1.6 (PZrH.6 ) is reported as:

PZrHl1 6 {CM 3} =0.1706+/-+O.0042-1.6 =5.63 95 20 Calculations of U-ZrH density and heat capacity (specific Cp - as determined by Cp,v normalized to the density) at a wide range of temperatures were performed (Table 4.10).

Thermal conductivity for the zirconium fill rod at the center of the fuel element was taken (even 100 temperature values) from the Journalof Physical and Chemical reference Data (Volume 3, 1974, 8Simnad, The U-ZrHx Alloy: Its Propertiesand Use in TRIGA Fuel (August 1980) 18

PART II Supplement 1, Table 184), with intermittent values interpolated. Volumetric heat capacity data was taken from a compilation 9, with data interpolated by a curve fit. Mass-specific heat capacity used in TRACE is calculated as the ratio of the volumetric heat capacity to the density. Zirconium data is provided in Table 11.

Table 11, TRIGA Fuel and Zirconium Material Properties TRIGA Fuel (ZrH 1 6 -U) Zirconium (Fill Rod)

T Cpv p Cp C ,V p Cp Conductivity Emiss.

°K J .m. 3 .K.1 kgm.3 1.K-1 Ws-kg- J* m3 . K 1 gm33 kg-M 1.K-1 W-skg- Wmm 1 .K W

200 1.73E+06 6000.507 2.89E+02 1.71E+06 6520 261.7 25.2 0.8 300 2.15E+06 6000.507 3.59E+02 1.76E+06 6520 269.62 22.7 0.8 350 2.36E+06 6000.507 3.93E+02 1.81E+06 6520 276.91 22.1 0.8 400 2.57E+06 6000.507 4.28E+02 1.83E+06 6520 280.56 21.6 0.8 450 2.78E+06 6000.507 4.63E+02 1.85E+06 6520 284.21 21.3 0.8 500 2.99E+06 6000.507 4.98E+02 1.88E+06 6520 287.86 21 0.8 550 3.19E+06 6000.507 5.32E+02 1.90E+06 6520 291.51 20.85 0.8 600 3.40E+06 6000.507 5.67E+02 1.92E+06 6520 295.16 20.7 0.8 650 3.61E+06 6000.507 6.02E+02 1.95E+06 6520 298.81 20.8 0.8 700 3.82E+06 6000.507 6.37E+02 1.97E+06 6520 302.46 20.9 0.8 750 4.03E+06 6000.507 6.71E+02 2.OOE+06 6520 306.11 21.25 0.8 800 4.24E+06 6000.507 7.06E+02 2.02E+06 6520 309.76 21.6 0.8 850 4.45E+06 6000.507 7.41E+02 2.04E+06 6520 313.41 22.1 0.8 900 4.65E+06 6000.507 7.76E+02 2.07E+06 6520 317.06 22.6 0.8 950 4.86E+06 6000.507 8.10E+02 2.09E+06 6520 320.7 23.15 0.8 1000 5.07E+06 6000.507 8.45E+02 2.10E+06 6520 322.53 23.43 0.8 1050 5.28E+06 6000.507 8.80E+02 2.11E+06 6520 324.35 23.7 0.8 1100 5.49E+06 6000.507 9.15E+02 2.14E+06 6520 328 24.3 0.8 http://www.efunda.com 19

PART II 5.0 Model Validation TRACE calculations were performed for flow channel/fuel element power levels from 200 watts to 29 kW.

Fuel element and cooling temperatures are calculated as the TRACE heat structure. The temperature response to power generation was evaluated using the UTTRIGA model. Fuel temperature observations (operating data) at varying power level are provided (Table 12). The relationship between power level and (fuel element) component material temperature was evaluated from TRACE thermal hydraulic calculations. This methodology is based on flow channel analysis, while the temperature data is correlated to core power. Observations were compared with the two independent measuring channels.

Temperatures calculated by TRACE consistent with the nominal thermocouple locations were compared to observations.

There is no measurement available for flow rates or heat fluxes that would allow comparisons with calculations of mass flow rate or critical heat flux ratio, CHFR (the ratio of fuel element local heat flux to the heat flux that could result in departure form nucleate boiling). Model validation is supported by comparing results to data associated with accepted reference work. Thermal hydraulic data from TRACE is used to calculate the mass flow rate and the ratio of heat flux at each power level to critical heat flux.

5.1 Temperature Calculations Temperatures of materials in the TRACE heat structure for flow channels/fuel element operation at selected power levels are provided in Figs. 10 and 11 for selected power generation levels. Locations bounding the zirconium filler at the center of the fuel element, thermocouples, and cladding are marked for reference in Fig. 10.

FUEL ELEMENT TEMPERATURE PROFILE 1200 "r_.. . ... IGp 110 0 :. Z : :: .! .: * : ,.. , ...*..:::

.............. ...................... Ga A .. ...... ....... i Gap..........

1000 ___ -~ ~Cladding v 9 00 _. - . ".." ...I..

..o

....--...

..

..............

8 00 -*................. ....

..............................

............................ ................................

................................. . ...............

700 - - ..- .-1.

.............. .....

...................

.....

. .................

E 600~ .......................... ................... .............

- 500 " Zirc FilIRod .............. Therm ocouple Position .......................

. ......................

I 4 00 ..-. . ...... G- Q-- -

0.00 0.25 0.50 0.75 1.00 1.25 1.50 1.75 2.00 Radial Position (cm) 20oW -8 12.5kW -0.22kW ..*.29kW -- Clad Figure 10, Fuel Element Temperature Radial Profile 5.1.2 Operating Data Instrumented fuel elements (IFEs) are located in the B ring (currently B03 and B06). Power level, fuel temperature, and control rod position data are routinely recorded in nuclear engineering exercises at 20 kW, 60 kW, 100 kW, 500 kW, 750 kW and 950 kW. Data (taken in 2013) is provided in Table 12 20

PART II AXIAL TEMPERATURE PROFILE 12.5 KW ELEMENT 700 650

.-

........... .. --

........-- .. .. .. ...

i..........

600 ArN 550 All

~*0~ __ __

500 E ___ ___ __"~~~~

=__

45 *0 .. . ...... . ........

................

. . .............

.... ... .................

........... .. .............

.....................

.*...

40 !- ,*-i * " - -O' - 4 -F I--- - -

0 .......- * .-.......-.-..

35 0.000 0.050 0.100 0.150 0.200 0.250 0.300 0.350 0.400 Axial Elevation (m)

-8 CL -- FTTC -e Gap/Fuel Gap/Clad ..& Clad/Water Figure 11, Fuel Element Temperature Axial Profile Table 12, Operating Data Core Pwr FT1 FT2 Pool (kW) °C °C oC 20 27 32 19 60 47 60 19 100 68 86 19 250 136 171 19.4 500 221 264 20.7 750 278 331 23 950 319 370 23.4 Confirmatory measurements during the summer of 2014 showed minimal deviation from the 2013 measurements. Based on historical values and confirmatory measurements, temperature response to the heat generated during reactor operation is well characterized by the values in Table 12 and Fig. 12.

Core power is distributed non-uniformly across all fuel elements bas on the ratio of the power in an individual element to the average across all elements, or peaking factor. Fission data from SCALE burnup calculations modeling the fuel materials form the initial 1992 core to the current 14 element core was used to determine the peaking factors for B03 and B06. The B03 peaking factor was calculated to be 1.49, with the B06 peaking factor 1.52.

5.1.3 Comparing FT1 and FT2 Measurements to UTTRIGA Model Calculations Pool temperatures varied during operation for measurement data points, while calculations are performed using a single value for pool temperature. Therefore, values from thermocouple readings were corrected to allow direct comparison between observed and calculated temperatures. Observed temperatures were modified by subtracting pool temperature and adding a 300°K correction to agree with pool temperature in calculations and perform unit conversion (Table 13, Fig. 11).

21

PART II Table 13, Fuel Temperature Data Corrected Corrected B03 Power ForeTem B06 Power FTrremp Core Power FT1 Temp FT2 Temp (kW) (W) °K (W) °K 20 261 308 267 313 60 784 328 800 341 100 1,307 349 1,333 367 250 3,268 417 3,333 452 500 6,535 500 6,667 543 750 9,803 555 10,000 608 950 12,417 596 12,667 647 ELEMENT POWER LEVEL & MEASURING CHANNEL RESPONSE (ADJUSTED FOR POOL TEMPERATURE, CALCULATED & OBSERVED) 700 1 650 600 550 500 a) 450 0-E a) 400 F-350 300 _______ -~

0 2000 4000 6000 8000 10000 12000 14000 Power Level (W)

FT1, B03 -B FT2, B06 -FTA ...... FTB --- FTC Figure 12, FT2 Temperature Response to Core Power Level The FT1 data response is qualitatively similar to calculations (and FT2), but the FT1 indication diverges by about 12% from calculated temperatures.

The FT2 data is remarkably consistent with calculated temperatures, with some divergence at higher power (Fig. 12). The strong correlation between observed and calculated fuel temperatures for FT2 begins to deviate at core power levels greater than 750 kW, calculated B06 power greater than 1000 W.

Calculations indicate conservative values (i.e., higher temperatures). There are several possible explanations.

As previously noted, flow channels are not well isolated from adjacent flow channels. Significantly different hydrostatic or hydrodynamic conditions at the boundary could affect flow rates. However, flow and local pressures calculated at power levels associated with adjacent fuel elements are not significantly different from values for the B03 position.

During full power operations, there is frequently a vigorous stream of bubbles entrained in coolant flow from the core to the pool surface. The onset of nucleate boiling occurs in circular TRIGA cores as low 22

PART II as 210 kW'. However, subcooled nucleate boiling effects are extremely short range and cannot survive distance required to exit the flow channel. These bubbles have the potential to affect system dynamics.

Nitrogen solubility decreases as depth and temperature increase 2. The change in depth as water passes through the core causes a decrease of approximately 6% in nitrogen gas solubility. At 750 kW, the maximum water temperature result in an additional 41% decrease in solubility as cooling water passes through the channel. With a nearly 50% decrease in nitrogen gas solubility occurring over 38.1 cm of heated water, some nitrogen degassing is possible. Nucleate boiling from the fuel element surfaces may provide nucleation sites for evolution of nitrogen bubbles.

These effects are not modeled, but could contribute to a complex flow with additional mixing action. Potential for these mixing effects is minimized at the low water temperature changes associated with low power levels. Higher flow associated with better mixing would cause lower fuel temperatures in the B ring fuel elements as local water temperatures increase and affect nitrogen solubility. This mechanism is consistent with the observed deviation between fuel temperature measuring channels and thermal hydraulic calculations that occurs at power levels greater than about 750 kW.

5.1.4 FT1 and FT2 Comparisons The close agreement between FT2 and calculated data in conjunction with the lower agreement between FT1 and calculated data prompted closer evaluation of FT1 and FT2. Both measuring channels qualitative response to power was similar (see Fig. 12), indicating the two thermocouples were responding in a similar manner, but the temperature indicated by the FT1 measuring channel was much lower than expected.

The FT1 and FT2 fuel elements (B03 and B06, respectively) are functioning at some fraction of the core power determined by the peaking factor. Fundamentally, heat transfer (Q) is based on temperature difference (Tc.L as maximum fuel temperature, Tb as bulk cooling temperature, A as the heat transfer area) driving heat transfer though a system with an overall heat transfer coefficient (h):

Q = h. A. (TcL- T) 21 The thermocouples in B03 and B06 are nominally located 0.762 cm from the center of the fuel element, with one installed an inch above the mid-plane, a second at the mid-plane, and a third an inch below the mid-plane. Consequently, the monitored temperature is slightly below the centerline temperature.

However, the difference between the temperature at the nominal IFE position and the maximum centerline temperatures over a wide range of power levels is small, ranging from virtually no difference to a few per cent.

Factors implicit in the overall heat transfer coefficient depend on the magnitude of heat generated.

Different power levels could have different heat transfer coefficients, but the small difference in power between B03 and B06 (indicated by the ratio of the peaking factors) would result in comparable heat transfer coefficients.

The difference in temperature between fuel measuring channel and coolant temperatures is therefore approximately proportional to power, where the heat transfer from B03 and B06 have approximately the 1ThermohydraulicsAnalysis of the University of Utah TRIGA Reactor of Higher PowerDesigns, P.M. Babitz, University of Utah, December 2012 2 EIFAC. 1986. Report of the working group on terminology, format and units of measurement as related to flow-through and recirculation system. European Inland Fisheries Advisory commission. Tech. Pap., 49. 100 pp. & Multiphase Flow Dynamics.

http://dx.doi.org/10.1007/978-3-642-20749-5 11 Springer Berlin Heidelberg 2012-01-01 A Kolev, Nikolaylvanov P 209-239 23

PART II same constant of proportionality (a factor of h-4). At a specific average core power, the ratio of the power produced in the two fuel elements with peaking factors PFBo3 and PF 30 6 is:

PFB°6 (TFT1 - TO) 22 PFBo3 (TFT2 - Tb)

The ratio of the peaking factors derived from SCALE calculations B03 and B06 in the 114 element core at nominal operating temperature is 0.98. However, the data in Table 13 shows ratios of values derived from FT2 and FT1 (in absolute temperature) to be approximately 0.91. The FT2 agreement with calculated data suggests that B06 the peaking facto is appropriate. Agreement can be forced by adjusting the FT1 peaking factor, but agreement-is achieved only with values approaching 1.1, which is not considered likely.

Portable thermocouple instrumentation showed the two functional spare thermocouples in B06 to agree within a few degrees to the fuel temperature measuring channel at 950 kW operations. The unmonitored thermocouple in the instrumented fuel element of B03 was found to be significantly higher than the installed FT1 channel. The spare thermocouple in B03 was therefore installed in to the FT1 fuel temperature channel, and a test operation showed FT1 indication at a 950 kW test operation of 343°C (611-K).

Since application of the peaking factor for B06/FT2 resulted in agreement between calculated and measured data, and since an unrealistic peaking factor would be required to bring the B03 data into agreement, the transport calculations were not considered a likely source of the disagreement. Since the characteristic response to changes in power level indicated on FT1 and FT2 were qualitatively similar, neither the instrument nor the thermocouple was considered the likely source of the disagreement.

If the radial position of the thermocouple previously in the FT1 channel is different than the nominal value, the channel response would be qualitatively similar but biased to temperatures associated with the different positon. Inspection of calculated temperature data for 12500 W indicates fuel temperature is 605°K at the lower thermocouple axial position (1 in. below mid plane) at a radial position of 1.495 cm.

This approaches the measured value of 596°K for the B03 power of 12417 W. Therefore as a test case, temperatures at a radial positon of 1.495 cm as a function of power were compared to measured data for with good agreement.

Although the non-nominal positioning of the thermocouple in B03 is a plausible explanation for the disagreement between FT1 and temperature calculated data, the thermocouple position cannot be directly measured or observed. Given the uncertainty in the FT1 data and the loss of confidence in historical temperature data (replacement of the thermocouple), FT1 data was disqualified for model validation.

5.1.5 Summary Data from an installed fuel measuring channel (FT2) is used in comparing model calculations to observed data. The comparison of TRACE temperature calculations to the FT2 measuring channel gives confidence that the models predict reasonably accurate values for fuel temperature calculations. The calculations are less accurate, but conservative, at power levels exceeding about 750 kW.

5.2 Comparison to Reference Thermal Hydraulic Analysis Benchmark data is not available to validate the use of TRACE for thermal hydraulic analysis,,except to the extent that the model is capable of predicting fuel temperature where the fuel element is cooled by water 24

PART II flow. Nevertheless, RELAP has a long history in analysis of TRIGA systems, and TRACE is the current platform endorsed by NRC (incorporating RELAP and TRAC codes). A comprehensive review of methods for predicting power at which critical heat flux occurs in TRIGA fuel is documented in ANL/RERTR/TM-07-01 3 ;

the report provides thermal hydraulic data for comparing the UTTRIGA calculations to a generic hexagonal TRIGA core:

Mass flow rates calculated by the UTTRIGA model using TRACE are compared to the reference document (RELAP) values.

" Axial distribution of critical heat flux ratios from the UTRRIGA model using TRACE is compared to the reference document values.

5.2.1 Coolant Flow Rates Data from ANL/RERTR/TM-07-01 (Fig. 4) for mass flow rate is used as reference for comparison of values calculated with the UTTRIGA TRACE model. The UTTRIGA cooling water inlet was assumed to be 30°C to be consistent with reference-data inlet cooling water temperature. Power levels considered in the reference extend considerably higher than the UTTRIGA data used in temperature comparisons.

The reference document was based on RELAP calculation, with a different algorithm for flow calculations.

Calculated TRACE UTTRIGA model flow rates are within a factor of 2 of the reference calculation flow rates (Fig. 13). The difference in flow rates between the TRACE and the reference RELAP calculations is reasonable.

MASS FLOW RATE AT FLOW CHANNEL POWER 0.18 ...

0.20

..... -.-

.... .........

..... ...... .:-.............

- ............

...

.........

..... . ...

.....................

_

..........

..

. ............... ...... ..-..-

.................

__-

'.....

0.168..... ...........

.i.........

... ....

.!

....... . ...*............

......... ... ...........

..:............

................ . ...

..........

.....

00.18

.16 . . ... . .. - -- ------------

Cr 0 .10 .... i i " .

0 .12 ....................

............

........ ..

" ... .0 " !

.._.. _. ........ ........... ._.i ...........

i ............-....

_.. .. . i....

i.... * *:......... ..

..........

i !...................

0.00 ...... ...

... ............

. .....................

..................... .....

' ...............

!...........................

!..................

................................ :.................

....................

..

0 5,000 10,000 15,000 20,000 25,000 Channel Power (W)

-0 .TRACE - - ANL/RETR/TM-07-O1 Figure 13, Comparison of Calculated Flow Rates for UTFRIGA and Reference Calculation 5.2.2 Critical Heat Flux Ratio Calculations Critical heat flux ratio (CHFR) is the ratio of the critical heat flux (CHF) to the actual heat flux. The correlation for critical heat flux developed by Bernath is recommended for evaluating TRIGA fuel 3 ANL/RERTR/TM-07-01, Fundamental Approach to TRIGA Steady-State Thermal-Hydraulic CHF Analysis (E.E. Feldman, Nuclear Engineering Division, Argonne National Laboratory) 2007 25

PART II performance 4. The Bernath correlation (where CHFBo is the heat flux that results in burnout hBo is the convection heat transfer correlation at burnout, TYBO is the temperature of the cladding surface at burnout, and V is the fluid velocity, Tb is the cooling water bulk temperature, and dimensional variables as previously described determines the critical heat flux that results in burnout as:

CHFB° = hBo . (TV,Bo Th) 23 Where the heat transfer coefficient for burnout conditions is calculated:

hBo 10890 DýDe + +V.

D 48o.°--2 24 The formula predicting wall temperature at burnout is:

T".Bo = 57. In P - 54 P V 25 P+15 4 Substituting equations 14 and 15 into equation 13 results in:

CHFo = 10890 C 1 D

D,, + Di 48 D eO6P (57.InP P

+ 15

-V Tb 26 2

The Bernath formulation is in "pound centigrade units," converted to BTU h-1 ft-2 by a factor of 1.8.

WCHF = 1.8- 10890. D, +D, +V- 48 ].([57-InP-54

-be , P+15 4 -Tb 27 The results of calculations of CHFR calculations at 29 kW based on TRACE shows reasonable agreement from the reference document (based on RELAP) at 30 kW. There is a slight shift in the location of the minimum CHFR that would be expected at higher flow rates as calculated for the UTTRIGA model).

5.2.3 Comparison to Reference Values Critical heat flux calculations using the Bernath correlation and RELAP data were performed by Feldman (op. cit.) for typical TRIGA reactor configurations. The UTTRIGA TRACE ran stably up to 29,000 W, while the maximum RELAP calculation ran at 30,000 W. There is some difference in flow calculations are performed (previously noted). Nonetheless, the TRACE model shows results similar to the reference (Fig. 14),

providing confidence that the model provides reasonable results in thermal hydraulic calculations.

4 ANL/RERTR/TM-07-O1, 26

PART II ANF REFERENCE AND UTTRACE CHFR 6 - . .... . . .. . 4 .. ... . . . . . . . . ..i..........

... .............

...... ................. .... ...... .......

.....*..

.......

UJ............

............

..........

. ............* ': ... ..........

...... . . . ......

.............

.........*... .......... .. .. . . . ."

LL. %

5 I- .

0 5 10 15 CALCULATION NODE

- UTTRACE (29 kW) - - ANL REF (30 kW)

Figure 14, Comparison CHFR for Reference and UTTRIGA Model 6.0 Results A limiting case and a nominal case were defined for analysis based on Technical Specifications and normal operating conditions in Table 7.

Calculations performed with TRACE are based on heat generation (or power) in an individual element.

Power in an individual element is related to total core power by the number of fuel elements in the core and the individual element peaking factor.

Transport calculations were performed to evaluate the minimum number of fuel elements that could be loaded within reactivity limits and to determine peaking factors for each core configuration. The peaking factor data was used to determine the maximum power generated in a single fuel element assuming the maximum core power and maximum instrument error (1210 kW). Thermodynamic analysis with the UTTRIGA model was used to determine heat flux for the limiting channel.

6.1 Fuel Element Power at Maximum Core Power A series of MCNP and SCALE (KENO) calculations were performed to determine kff and the maximum peaking factor for UTTRIGA cores loaded with varying numbers of fuel elements. Material specifications for the fuel were assumed to be the average of all fresh fuel present at the facility. Calculations were made assuming ambient temperature and 600'K (assumed to be consistent with full power operation).

Calculations with water voids and graphite "dummy" rods in non-fuel spaces showed that the maximum peaking factors occurred with water voids.

6.1.1 Core Configurations and keff Transport calculations with SCALE resulted in lower keff values than MCNP, therefore using the data from SCALE will reflect the minimum number of fuel elements to achieve conditions. Operating a power (nominal 600°K) requires a minimum of approximately 77 fresh fuel elements (Fig. 15).

27

PART II KEFF AND CORE CONFIGURATION (NUMBER OF FRESH FUEL ELEMENTS) 1 .1 0 .................

..........

.

...* .............

1.051.0 ... ....... ... ..... ................... . .... .. ..... -,11............

.. . .r -. -

...........

. .... .... ' . ...i.!...............; .. .. ...... *....

...

... .. ........... ........ , . . .......

-i ..

.......... .......9............ .......

1.00... ........ ... ....

.."......0 f .......

............

i i.......

' .......

....1.111..1111.....

...........

......... .....

.....

.I..

I......

............

I]....... I.....

..

.. .....

i....... .....I..... .

!........ ...

!........ . ..I..........

F .....i........*...... ...............

0.95 .

...............

...

....

......... I

. . . . . . ......

..... ...... ... ...

...1 ...........

100 10.. ".. 1 ................

70 75 80.5............ 90.. .....95 0.90 50 55 60 65 70 75 80 85 90 95 100 105 110 115 120 Number of Fuel Elements

-e.600K -- 300K Figure 15, Criticality Considerations At 80 elements, keff at cold clean critical conditions is the maximum permitted of 1.04. Therefore the limiting core configuration is expected to be in the range of 77 to 80 fresh fuel elements.

6.1.2 Maximum Fuel Element (Hot Channel) Power The fission distribution data from the transport calculations was used to evaluate peaking factors for each fuel element. The maximum fuel element (hot channel) power was calculated assuming the maximum core power (license limit of 1100 kW with a 10% potential instrument error, or 1210 kW), the number of fuel elements in the core, and the maximum fuel element peaking factor. The keff values were not considered in calculations of fuel element power levels, i.e., not all configurations could achieve criticality at the specified temperatures. Values for maximum fuel element power for each configuration fell within a narrow range in MCNP and SCALE for both hot and cold calculations (Fig. 16).

ROD CHANNEL POWER AT 1210 CORE POWER AT VARYING CORE CONFIGURATIONS 40 ý _m 35 .....

30

........... ..... .......

0 CL U

W)20

..........

.................

........!.. i ..........

......... .. ............. ......... i ...........

. ...........

15 45 50 55 60 65 70 75 80 85 90 95 100 105 110 115 120 Number of Fuel Elements

-4a -SCALE600 --e -SCALE293 -&- MCNP 293 MCNP 600 Figure 16, Maximum Element Power for 1210 kW Core (MCNP & SCALE at 300 & 600°K) 28

PART II With a potential range of 77-80 fuel elements in the limiting core configuration, the power produced in the hot channel is expected to be approximately 23000-25000 W.

6.2 Fuel Temperature The maximum fuel temperature is limited to prevent long term fuel degradation and short term cladding failure. Long term, stead- state operations with fuel temperatures greater than 750'C (1023 0 K) have resulted in morphological changes that expanded to the limit of the axial gas gap 5 . This is a long-term process, requiring extended intervals of operation at temperatures exceeding 750'C to develop the growth.

Historical temperature limits for TRIGA reactors prevent precipitous cladding failure from internal pressure.

The limits are 950°C (1223°K) for cladding temperature greater than 500°C (7730 K) , and 1150 0 C (1423°K) for cladding temperature below 500 *C. Following an operational event at a TRIGA conversion-fuel reactor, and additional limit of 830°C (1103 0 K) for pulsing operations has been imposed.

The maximum fuel temperature occurs in the fuel element that is generating the most heat, or the fuel element with maximumrpeaking factor. Reactor power is limited so that the fuel element with the maximum peaking factor does not exceed temperature limits.

The TRACE heat structure analysis calculates the fuel element temperature in axial and radial segments from the center of the fuel to the cooling water surrounding the fuel element. A series of calculations was performed to determine the maximum and average fuel temperature as a function of the power generated in a fuel element (Fig. 17). The maximum fuel temperature exceeds 1023°K (750 0 C) at approximately 26500 W.

FUEL TEMPERATURE REPONSE TO FUEL ELEMENT POWER 1, 10 0 ..........

.l* llll lll

..........

. ........... ........... .............

........

lll . ........llll;lllll*llll;.................

.......... . .................

lll*...........

........ ................

......... .l ll...l....llll 1,050 1,0 00 .+.................. ... ........

.........................

..... .......

................ .. .

.................

.............................. T ; T . .TT

..............

....... .T

..'.

950 .'.. .........

..................... .. ........ .................. . .......

95 0 ....

..........

600 -- ,-- --- ...... .....

. - .........

.. .

- -

450 300 .. .. *+

4500....

300- - -..-- .......

- - . -

.......

--. .--. . .

..........

.

.................................

- .........

" - - - - . . -.- - .

. . . .......

-.-

--.......

. . ........-.... . . .....

. . . . ..L . .: . = . . . . +. . .

......

. . =. .

0 5,000 10,000 15,000 20,000 25,000 30,000 Fuel Element Power (W)

-.6MAX -e -AVE ,

Figure 17, Fuel Temperatures as a Function of Element Power Level Although there are two cases for consideration (nominal and limiting pool level and temperature), pool parameters are shown to have negligible effect on the maximum temperature and the temperature profile within the fuel element for a specified fuel element power level (Fig. 18).

5 Need reference 29

PART II FUEL ELEMENT TEMPERATURE PROFILE 10 0 0 . ...... ........

i. ..... ... ................. ................. ................... .................. ....... ........ "..'........

Ga p 9 00 -. . ................

..........

i I

I ~ l Cla d d in g

-~800 7*0 0 [ . . . . . . .. ... .......

..... ........... . . .........-. ....... . _......... i __-

600 7 O 0. * ..A i ,

....

5 00 * . ........... ..........................

............................ .................................

.. .... ... . ................... ......- .* ......

-- Zirc FillRod I 400 .* ........... I )hrtooulePosition

30C, ... ...... ........................

......................

1 ..... 1 ... ... ..................

...........................

=..-.................

0.00 0,25 0.50 0.75 1.O0 1.25 1.50 1.75 2.00 Radial Position (cm)

.,NOM 12.5 kW -. NOM 22 kW --e LCC 12.5 kW -f-B-LCC 22 kW Figure 18, Fuel Element Temperature Profiles for Selected Element Power Levels The temperature calculations demonstrate that the limiting core configuration will prevent exceeding fuel temperature limits for a fuel element power less than approximately 29000 W.

6.2.2 Fuel Temperature Measuring Channel Protective Action As previously described, fuel elements with embedded thermocouples are installed to monitor fuel temperatures. Temperature monitoring by instrumented fuel elements provides assurance that reactor operation is terminated if limits are exceeded. However, the instrumented fuel element may not be in the position that produces the greatest power (i.e., the maximum temperature) in the core. Therefore the trip set point for the fuel temperature measuring channel is established at a level that initiates action to prevent the maximum fuel temperature in the B ring from reaching the limit regardless of the instrumented fuel element location. Since the peaking factor is calculated:

-- = PE28 PC The power generated in the instrument fuel element is a fraction of the power in the maximum heat element:

PFF Using SCALE and MCNP data at operating temperature as previously described, the minimum power that is being generated in a B ring fuel element is 93% (MCNP) or 94% (SCALE) of the maximum power being generated in any element in the core. The minimum power that is being generated in a C ring fuel element is 74% (MCNP) or 75% (SCALE) of the maximum power in the core. The minimum power in the core at full power is 31% (SCALE) or 33% (MCNP) of the maximum power being produced by the maximum element.

Temperature response to power for representative factions were prepared and plotted against the power produced by the maximum heated fuel element. Fig. 19 demonstrates that if the instrumented fuel element power level is at least 70% of the maximum power level, a set point of 550'C is adequate to terminate operations if the maximum power element reaches 1023°K (750°C).

30

PART II FUEL TEMPERATURES CORRELATED TO THE MAXIMUM ELEMENT POWER 1,100 1,000 900 800 700 CL 600 500 400 300 0 5,000 10,000 15,000 20,000 25,000 30,000 Maximum Element Power Level (W)

- R=100% - - R=90% --- R=80% ...... R=70% - R=60% - "R=50%

Figure 19, Maximum & Monitored Fuel Temperatures at Maximum Element Power Table 14, Limiting Temperatures Application oK °C Safety Limit for Cladding <500°C 1423 1150 Safety Limit for Cladding <500°C 1223 950 Limit for pulsing operations 1103 830 Limit for fuel growth 1023 750 Temperature Trip 823 550 The disagreement between FT1 and FT2 that lead to disqualification of the channel for model validation could be explained by a radial displacement from the nominal position., The effect of radial displacement of a thermocouple is similar to the effect of changes in peaking factor. Fig. 20 illustrates that data representing a thermocouple position at 1.6306 cm brings the calculation into close agreement with the measurement data, as does adjusting the temperature response data to reflect the IFE simulation operating at 70% of the power produced in the maximum power fuel element.

31

PART II B03 POWER LEVEL AND FT1 RESPONSE 650 600 550

...... ..........

.... ... .. ......

.......... ...........

.. ................

500 450 CL 400 E

0) 350 .....

.... -*..................

............

.....

........ .................. .......................

..................... ............

........... ...................

300 0 2,000 4,000 6,000 8,000 10,000 12,000 14,000 16,000 B03 Power Level (W)

-*- FT1, SCALE PF -B. FT1, MCNP PF -- 1.6306 cm - - "B01:B03=0.7 Figure 20, B03/FT1 Effects 6.3 Critical Heat Flux Limiting the critical heat flux ratio (CHFR) in the hot channel assures that departure from nucleate boiling will not occur in the core. TRACE calculations were performed were performed using the limiting values for pressure and water temperature over a range of fuel element heat generation/power levels. Water temperature, pressure, and mass flow rate at each elevation were used to calculate critical heat flux using the Bernath correlation (Eqn. 5.8). The minimum ratio of heat flux to critical heat flux as a function of power level is provided in Fig. 21. For power levels less than about 30,000 W in a fuel element, critical heat flux ration remains above 2.0.

CRITICAL HEAT FLUX RATIO FOR FLOW CHANNEL POWER LEVEL (LIMITING POOL CONDITIONS) 9 . ........ ...................-

- ........

....

.......

.... ...........

.................

..................................

T.....................

.........................

89 ......* * ........... ...........

...

i.. ............... T...........

7 "................... ... ..

.................... ...............

. .. . .... . ..............

...

,, 6 ....

2 1lyl 7'454E'17x4 -i6"864E-12x3R~ +=2"364E-O7x29'9E-l- 3.663E-O3xl " + 2.426E+01i.. -

RK2 =49.997E-01 6000 8000 10000 12000 14000 16000 18000 20000 22000 24000 26000 28000 30000 Flow-Channel Power (W)

Figure 21, TRACE Critical Heat Flux, Limiting Pool Conditions 32

PART II 7.0 Limiting Core Configuration Critical heat flux considerations limit the maximum power produced by a fuel element to less than approximately 30,000 W. Fuel temperature considerations limit the maximum power produced by a fuel element to less than approximately 27,000 W.

Criticality considerations (Fig. 21) show that excess reactivity reaches the maximum (keff of 1.04) with 80 elements. The maximum power produced by a fuel element by the minimum core size supporting operations at the limit of excess reactivity is on the order of 23,000-25,000 W (Fig. 22). For a more precise evaluation, the highest power level for all calculated cases was used to develop a relationship between the hot channel power and the number of fuel elements. Sensitivity of the power to changes in the number of fuel elements is taken as the derivative.

PHC = -3.866x10 -5

N3 + 1.255X10- 2

N 2 - 1.514

N + 85.24 30 dPHC - -1.160xlO-

N2 + 2.5 10X10- 2

N - 1.514 31 dn At 80 elements, the hot channel power is approximately 24,600 W with sensitivity to a change in the number of fuel elements of 2,480 W.

MAX ROD CHANNEL POWER, 1210 KW CORE AT VARYING CORE CONFIGURATIONS 40 ................

... ...

.....

30 ..

- 25

.....

... ...

20

~E~J 15 ... ........

40 50 60 70 80 90 100 110 120 Number of Fuel Elements Figure 22, Highest Calculated Hot Channel Power Calculations were performed in TRACE assuming limiting pool conditions of level and temperature at a hot channel power level of 24664 W. The minimum critical heat flux ratio remained well above 2.0 for all positions in the channel.

33

PART II HOT CHANNEL CRITICAL HEAT FLUX RATIO FOR 24664 W 6 .0 .. . ..... .. ..................

... ... ... .....................................

. ....................... .........

.....................................................

...........................................................

............

........................

5.5 j.................................................... .....

.. ...............

. ........................ ..... ........ ...... .... ..................

..................

5 .0 .........! .............. .. .................................

. ... .. .. . .. ........................... .

.......................................................................................

0 5.0 '

4 .5 . * ..... .......*... ............................................................................... .................

4-1 4.0 -

I 3.5 -- X

.........

...

/

3.0 ....

......

2.5 x... .........................

2.0 .........................

0 10 20 30 4C3 Elevation Along the Heated Length (cm)

Figure 23, Limiting Core Configuration Hot Channel Critical Heat Flux Ratio The radial temperature profile for materials in the fuel element and the (average) cooling water (Fig. 24) shows that the maximum fuel temperature remains within limits at all locations in the fuel channel. The axial profile (Fig. 25) shows the response across the radius of the fuel element is consistent with previous work that demonstrated the fuel temperature monitoring system capable of providing protection with an adequate margin to temperature limits.

HOT CHANNEL RADIAL TEMPERATURE PROFILE FOR 24664 W LCC SELECTED LOCATIONS 10 000 i 000 .OD-. - C .E ,............... v ................

..... .... ............... ..oi. . . . ...o. . .....

..j......... ........

.....

li

..................... ......................

...........

900 .. 79 ....

800 ------- ~---- - __

700 600 E

500 400

............ ........

.........

300 .

0.0 0.5 1.0 1.5 2.0 2.5 Radial Position (cm)

-u-ENTRANCE -o-MIDPLANE -e.EXIT Figure 24, LCC Hot Channel Radial Temperature Profiles (Upper, Lower & Mid Heated Length) 34

PART II HOT CHANNEL AXIAL TEMPERATURE PROFILE FOR 24664 W LCC 1000 .......

00.........

,......... ......

.... *.........

i .- ..... 1*.... i

.................. ........... ......... -.-.-.............

900 . ................ ........................................................... i ...

, . i.............. .. .

7 0 ..... . .... ...

=oo~ ~~ ".i ..

.....

~~ .ii....]i..

. ....

.. --.,.-

7oo~~~~~~~~~~ -*+ .->. -!-. --. * . ......!.......... ..... ..........

..

600 . ... ' .. .* . .. . .......... ....

T ...........

........................ = .' **.

. .......

t......................

............

.................... t... ...:... ......

500 ..... .*...

!.........

..i.!... ........... i......[

.. . .....i..........

i... .... ..i.. .. . ;- i .:

500 i i ....i ;..... . .. .......

.................. *............

....... .-........

.... - ..... .......

400*'=-I,

...... --. * - -A ,T&- .- * - -A--4-- - A- -A-i-A *- -* -1 .......

.... .......

.........................

". ' .- ....- . - *' E .... " " W"e`... '*1 ' f-=' ' Z... .. :

X- . K ..... .. .

300 - ' _ "

0 5 10 15 20 25 30 35 40

-- TC --a.CL -6 FUEL/GAP -*-GAP/CLAD -e-CLAD/H20 --.- H20 Figure 25, Limiting Core Configuration Hot Channel Axial Temperature Profiles (Heated Length)

Conservative factors in UTTRIGA analysis:

(1) Peaking factors were calculated based on a uniform average fuel temperature; the uniform average fuel temperature assumption was shown to result in higher peaking factors than a more realistic description of fuel temperature distribution.

(3) Calculated temperatures are higher than observed temperatures above approximately 750 kW, and actual thermal hydraulic conditions are therefore not as severe as calculated.

(4) Modeling a flow channel does not account for mixing with adjacent flow channels; mixing flow from adjacent channels reduces hot channel cooling temperatures.

Summary Based on TRACE calculations at power levels up to 1210 kW (maximum measuring channel error), a minimum CHFR of 2.0 is assured in the limiting core configuration (minimum pool water level, maximum pool temperature) for all cores from 75 elements to the maximum number of fuel element locations in the core. Cores with greater than 83 elements maintain CHFR greater than 2.0 up to 1500 kW.

Therefore the minimum core configuration is selected to be 80 elements.

Hot channel power for a fuel element operating in an 80 element core at 1210 kW is approximately 24 kW.

The peak centerline temperature for a 24 kW hot channel under limiting pool level and temperature is calculated by TRACE at 687 TC and by RELAP as 695 'C.

The instrumented fuel elements are capable of initiating a reactor trip prior to exceeding maximum permitted fuel temperatures in the hot channel. A trip setpoint of 500 'C accounts for differences between the sensor location and the maximum temperature of the hot channel.

35

v t e Letter Sequence Request
TAC:ME7694, (Approved, Closed)
Initiation Request, Request, Request, Request, Request, Request, Request Acceptance... Supplement, Supplement, Supplement Administration Questions 25 June 2012 31 July 2015 13 April 2016 27 January 2017 Answers 17 September 2012 22 March 2013 22 December 2015 5 February 2016 2 May 2016 15 March 2017 Results Approval Other: ML120050106, ML14002A317
MONTHYEAR2013-08-21 [Table View]

ML15211A638

Text

SUBJECT:

RAI 7

Response

Response

RAI 9

Response

RAI 10

Response

RAI 11

Response

RAI 13

Response

RAI 14

Response

RAI 15

Response

RAI 35

Response

Navigation menu

Search