PSU Neutron Beam Group (NBG)

MODELLING OF EXISTING BEAM-PORT FACILITY AT PENN STATE UNIVERSITY BREAZEALE REACTOR BY USING MCNP

Participants:	Kostadin Ivanov, Associate Professor of Nuclear Engineering
	Yousry Azmy, Professor of Nuclear Engineering

	Jack S. Brenizer, Professor of Mechanical and Nuclear Engineering
	Kenan Ünlü, Professor of Mechanical and Nuclear Engineering

	F. Alim, Graduate Student
	B. Sarikaya, Graduate Student

Services Provided:	Neutron Beam Laboratory

Sponsors:	U.S. Department of Energy-Innovations in Nuclear Infrastructure and Education (INIE)

Introduction

The Penn State University (PSU) Code System that has been developed to perform computer simulations of the beam port facility at the Reactor Science and Engineering (RSEC) at PSU consists of two major parts, the core model and the beam port model. The core model is needed to compute the flux at the reactor core – beam port facility boundary so that it can be used to compute the flux at the end of beam port where the experimental data were taken. Core calculations are performed by using the nodal diffusion code ADMARC-H [ 1]. The few-group cross section library needed to perform the diffusion calculations with ADMARC-H has been generated by the lattice physics code HELIOS [ 2]. The beam port calculations require very detailed geometrical definition of the system. This complexity of the geometrical model and the number of materials used in modeling the beam port facility prevent the use of deterministic codes in this study. Therefore, because of its geometrical flexibility a general Monte Carlo N-Particle Transport Code, MCNP5 [ 3], is used in this study to model the beam port facility. The source distribution used in beam port model is taken from the ADMARC-H code. An interface program has been developed at PSU to link the diffusion code to the neutron transport code. This interface reads the ADMARC-H output then computes the source term for MCNP and finally prepares the necessary MCNP input card in the requested format. The core model, the interface module, and the beam port model were described in RSEC 2003 annual report. Verifications of these models and applications of the tools developed for these models will be described in this report.

Verification and Application of the PSU Code System

Verification

The results of the overall PSU Code Package, which contains the core model, the beam port model that consists of the D2O tank and the beam port tube models, and the interface module to link these two models are verified by comparing these results with the available experimental data [4] as shown in Figure 1.

Figure 1: Comparison of the results of the PSU Code Package with the experimental data for the exit of the beam port #4. The solid curve represents the experimental data and the dashed line represents the model predictions with the associated error bars (the dotted vertical lines).

Applications

After establishing the tools for the beam port design study, these tools were used to model beam port #7. Two different MCNP generated plots of the model are given in Figures 2.a and 2.b since beam port #4 and #7 are axially located at different elevations.

Both figures show the top view of the system at different axial locations. Since both beam ports and the D 2O tank are cylindrical in shape, in both plots, their sizes differ at different elevations. The same methodology used in the BP #4 calculations was applied for the BP #7 calculations. The results of the D 2O tank model for 9x10 8 neutron histories with the associated statistical errors are shown in Figure 3.

Figure 3: The D 2O tank results for BP #7 (all angles) . The solid line represents the model predictions and the dotted lines represent the associated error bars (statistical error in the calculations)

This result represents the spectrum at the tallied surface for all angles. However, in order to prepare the input for the beam guide tube, a highly collimated neutron flux was needed. Therefore, the D 2O tank model was designed to tally for this very highly collimated flux, and this study is still underway and has been running.

Even though the verification of the system was made for the neutron analysis, we turned on the gamma analysis option in the MCNP calculations and performed an analysis for the neutron induced gammas within the system (not core gammas). Figure 4 shows the plot of the gamma spectrum at the beginning of the beam port guide tube (end of D 2O tank model) for all angles.

Conclusions and Future Work

In this study, the existing beam port facilities at PSU Breazeale Reactor were modeled by using a code package developed at PSU, which consists of two major steps, the core calculations and the beam port facility calculations. The core calculations were performed by using the diffusion code ADMARC-H, which utilizes a few-group cross section library developed with HELIOS. MCNP5 was used to perform the beam port facility model calculations. The link between the core calculations and the beam port calculations was established with an interface program specifically prepared for this study.

Figure 4: Gamma spectrum at the entrance of the beam port guide tube (BP #4)

The beam port #4 model was used to verify the PSU code package since the experimental data for this beam port and core configuration was available. The results presented for the beam port #4 model showed that the prediction of PSU code system agrees well with the available experimental data [5,6].

The same tools and methodology were used to simulate the beam port #7. Since beam port #7 is located 5” lower than beam port #4, which is exactly located at the axial centerline of the reactor core and since there is more D 2O and an extra graphite block in between the beam port #7 and the reactor core, the number of neutrons that can reach up to the beam port #7 is much less than that of the beam port #4. Therefore, the statistics of the beam port #7 calculations were worse than that of the beam port #4 calculations for the same number of histories. Hence, beam port #7 calculations were performed for much longer histories.

References

1. K.N. Ivanov, N. Kriangchaiporn, “ ADMARC-H Manual”, the Pennsylvania State University, 2000.

2. “ HELIOS Methods”, StudsvikScandpower, 2000

3. X-5 Monte Carlo Team, “ MCNP – A General Monte Carlo N-Particle Transport Code, Version 5, Volume I: Overview and Theory”, LA-UR-03-1987, April 24, 2003

4. J.H.J. Niederhaus, “ A Single-Disk-Chopper Time-of-Flight Spectrometer for Thermal Neutron Beams”, M.S. Thesis, the Pennsylvania State University, University Park, PA (2003)

5. B. Sarikaya, “Modeling of Existing Beam Port Facility at Penn State Breazeale Reactor by Using MCNP5”, M.S. Thesis, the Pennsylvania State University, University Park, PA (2004)

6. B. Sarikaya, F. Alim, K. Ivanov, J.Brenizer, K. Ünlü, and Y. Azmy, “Modeling of the Existing Beam Port Facility at the PSU Breazeale Reactor by Using MCNP”, Proceedings of PHYSOR 2004, Chicago, April 2004, USA.