CHAPTER ONE

INTRODUCTION

1.1         INTRODUCTION

Fuel burn-up was shown to be linearly dependent on the reactor thermal power (Podvratnik, 2011). It is therefore obvious that the reactor thermal power calibration is very important for precise fuel burn-up calculation. The reactor power can then be determined from measuring the absolute thermal neutron flux distribution across the core in horizontal and vertical planes (Musa et al., 2012). It has also been established that flux distributions can be measured with activation of cadmium covered and bare foils irradiated at steady reactor power (Souza, 2002). But Shaw (1969) demonstrated that this method consumes a lot of time and is not accurate. It can therefore be said that the foil activation method is most suited for zero power reactors and seldom applied to bigger reactors. In the case of high power reactors in which a temperature rise across the core is produced and measured then a heat balance method is the most common and accurate method of determining the power output of the core (Mesquita et al., 2007).

Accurate reactor thermal power calibration is important for: safe monitoring and evaluation of reactor dynamic behavior, determination of fuel burn-up and normalization of neutron fluxes and dose rate. (Mesquita et al., 2007; 2009; 2011, Podvratnik, 2011). Power excursion of any reactor is of great concern to reactor physicist for safe operation reasons. As power is related to the neutrons population and to the mass of fissile material present, its measurement is essential to the safe control and operation of the reactor as well as the reliability of the research reactor (DOE, 1993, Podvratnik, 2011). It therefore became imperative to undertake power measurements and calibration from time to time to establish

1

the stability of the reactor and evaluate its fuel burn-up. Reactor power calibration helps in maintaining and improving safety (Beamex 2014). A lot of literatures show how safety is based upon never exceeding established operation limits such as reactor power (Lopez, 1961, Zagar et al., 1999, Mesquita et al., 2007, 2011, Beamex 2014). A byproduct of improved calibration is an improvement in safety. (Beamex 2014).

As the reactor is operated, energy is released through the fission process. The majority of this energy appears as energy carried by fission fragments, gamma rays, neutrons and beta particles emitted (Jevremovic, 2008). When these particles interact with the surrounding materials, heat is produced. This process heats up the fuel meat and starts the chain of heat transfer The heat released from nuclear fuel is transferred to the coolant through the heat conduction of fuel and cladding, and the heat convection between cladding outer surface coolant. Then, the heat is carried out of the reactor core by the coolant.

The thermal-hydraulic design of Nigeria Research Reactor-1 (NIRR-1), intended as with no reactivity insertion, is closely dependent on the structural design of the reactor. The core is cooled by natural convection which is established through the heat generated by fission occurring in the core. The reactor coolant is drawn through the inlet orifice by natural convection flow through the channels within the fuel elements. The coolant moves up through the core and exits through the core outlet orifice at temperatureT2 to the upper part of the tank where the temperature isT3. The coolant inside the core passes through the aperture surrounding the upper ends of the fuel. New colder coolants is substituting the hotter one in the core causing the coolant in the downcomer (with temperature T1) to move

2

downwards and maintain its temperature so that the coolant enters the reactor core at temperature T1. In the simplified model, the velocity of the coolant in the downcomer is supposed to be equal to the velocity of the coolant in the core. Heat transfer from core to the reflector is by conduction and convection and also heat transfer from the tank to the pool (T0) across the wall of the tank is again by conduction and convection mechanisms

While there has been a great deal of research on the behavior of NIRR-1 power resulting from changes in core-coolant temperature (Ahmed et al., 2008), this study focuses on the heat transfer in the system as a whole. The fuel rods in the NIRR-1 core heat the surrounding water which flows through the fuel channels then up and over the rods in a natural convection manner. The heated water then can either evaporate from the open top of the reactor pool or the heat is transferred to the environment. Transferring the heat to the environment can be either through the top surface which is open to the air or through the aluminum tank wall, concrete barrier and outer steel tank wall. Currently, through one of these two methods is the only way the pool water is cooled back to ambient temperature.

SAR, 2005 pointed out that the decay heat in NIRR-1 is removed by natural circulation of the reactor vessel water in a manner similar to that during operation. It has been established in SAR, 2005 that a number of thermal hydraulic tests and calculations, especially from the dynamic experiments, have shown that the natural circulation of the prototype MNSR (which is comparable to NIRR-1) has the following characteristics:

i)          Negative Feedback Effect:

3

This occurs when the temperature difference between the inlet and outlet coolant of MNSR increases; the floating force and circulating head will increase to make the volume flow rate and flow velocity high, which in turn will limit the rise in temperature.

ii)          Insufficient Circulation:

Because of the small size of the core, the distance from the inlet orifice to the outlet orifice is small. The water, after being heated in the core goes out through the upper part of the core. Part of the water does not get sufficiently cooled before it sinks down, resulting in part of outlet water being carried back into the core due to siphoning effect. This direct re-circulation of the part of hot water causes a rise of the inlet water temperature. This phenomenon is called insufficient circulation, which speeds up the rise of the coolant temperature in the core and shorten the function time of the temperature effect. It is thus not possible to cause the inlet water temperature to rise in such a short time by heating the core only, but by the coupling action between the inlet and the outlet coolant. Consequently this offers some benefit to the reactor safety

For safe operation of NIRR-1, it is imperative to measure its power accurately. There are different methods reported for measuring the reactor power such as thermal hydraulic methods (Ahmed et.al, 2006; 2008), neutron flux measurement (Ahmed et. al., 2008; Musa et al., 2012), Cherenkov radiation intensity measurement (Arkani et al., 2009) and 16N gamma activity (Hamid et al., 2011).

4

The neutron flux method of measuring reactor power can be handled through the foil activation method, calorimetric (slope) method and heat balance method. Earlier, power measurement has been conducted using the foil activation method (Musa et. al., 2012). The present work therefore, to explored the calorimetric (slope) and heat balance methods which were hitherto not used to measure and calibrate NIRR-1 power.

1.2          AIM AND OBJECTIVES

The aim of this research work is to use calorimetric and heat balance methods in the thermal power calibration of Nigeria Research Reactor-1 (NIRR-1).

The following objectives would be achieved.

1                    To analyze the accuracy of calorimetric and heat balance methods in NIRR-1 thermal power calibration.

2                    To Calibrate the thermal power of NIRR-1 using calorimetric and heat balance methods

3                    To evaluate the dependency of power as a function of temperature-rise rate.

4                    To Determine NIRR-1 power using flow rate and coolant temperature difference.

5                    To compare between the results obtained and flux parameters on Console to determine best method.

5

1.3         STATEMENT OF RESEARCH PROBLEM

The most important factor in reactor control is the precise information about reactor power at any moment. Accurate reactor power measurement is required not only for safety purposes but also by research reactor users for experiments. The knowledge of online reactor power level has been a problem to MNSR reactor operators. The Integrated Safety Assessment for Research Reactor (INSARR) mission 2009 has recommended for power display in NIRR-1 control console to have redundant systems or components that perform the same safety function by incorporating into the systems or components different attributes, such as principles of operation, operating conditions and manufacturers for reactor safety and international standard.

The need for accurate power level determination is a safety criteria and operational limiting condition. In the operational limits and conditions (OLCs), we have the power limit which said that the power should not be more than 20% of the full power (SAR, 2005). However, no power is displayed in NIRR-1 control system. Research reactors are often involved with experiments which require an accurate determination of the reactor power. Normalization of experimental measurements cannot be possible unless the statistical variation of the reactor power is known.

1.4         JUSTIFICATION

Thermal power calibrations of low power research reactors (up to 1MW) are normally performed during the initial startup and their results are used for many years (Mesquita et

6

al., 2007; Ahmed et al., 2008; 2011). However, heat capacity usually change when experimental installations are made, changes in the collimator or other mechanical modifications like the case of NIRR-1 installation of cadmium-lined irradiation channel (Ahmed et al., 2013). Consequently, the need to calibrate the system with the present installations in and around the core is very crucial.

From the time when NIRR-1 was commissioned in 2004 till date attempts were made to measure its flux stability using Activation foils method (Musa, et al., 2012, ),and also to make certain its flux variation (Ahmed, et al., 2008), maximum operable time (Ahmed et. al., 2011). No comprehensive effort was made to calibrate NIRR-1 power despite its usefulness in utilization and experimental work. Available literature (Mesquita et al., 2007; 2009; 2011) showed that slope and heat balance methods are less cumbersome and straight forward with good results. This work explored these methods in the calibration of Nigeria Research Reactor-1 (NIRR-1) power. And recommend ways to adopt one of the methods for routine calibration of NIRR-1 power.

1.5         PREVIOUS WORK

The need to use more than one method to calibrate the power of a reactor became very important as ascertained by (Shortall et al., 1954) who used three independent methods (Gold Foil Activation, Fission Product Determination and Calorimetric Methods) in the calibration of CR&D water boiler reactor. They observed that though the three independent methods gave answer to within 13 percent, the accuracy of Fission Product Determination Method was estimated to be within  10 to 15 . They reported that the calorimetric

7

method just like its counterparts is the most inherently accurate, straightforward and presumably the simplest method for measuring the power of CR&D water boiler reactor.

The behavior of reactor power level and flux with changes in core-coolant temperature for a MNSR facility was investigated earlier by (Ahmed et al., 2008) at two power levels using NIRR-1. The method used in the measurements was a thermal hydraulic by exploiting core coolant temperature values to predict the reactor power. Their results show that the semi-empirical relationship based on thermal hydraulics and neutronics parameters could be used to predict the reactor‟s operating power. Also, the data measured indicated that the reactor‟s operating power level can be estimated from the preset thermal neutron flux value. The measured data shows that there is a strong dependence of the reactor power on coolant temperature in agreement with the design of MNSR. In an effort to extend the earlier work to other methods available for measurements and calibration of NIRR-1 power, we used the calorimetric (slope) and heat balance methods.

Much earlier works like (Lopez 1961) calibrated the power of Ksutmil reactor by noise analysis. He determined the zero power reactor transfer function via frequencies ranging from 2cps to 1000cps. He also determined the break frequency by a least squares fit of a

function of the form A+  . The value obtained was 19.3  was said to compare favorably with the value obtained by other methods. Knowledge of transfer function was used in the noise analysis to calculate detector efficiency and charge transferred per neutron absorbed and therefore calibrated the reactor power at several levels. He reported that at low power levels, the effects of instrument noise was important

8

enough and overrides the pile noise while at high powers, feedback temperature effects are present and that the calculated power level was compared with the power indicated by the linear recorder of the reactor console, the results agreed within 20%. He finally concluded that the determination of zero power reactor transfer function is possible although it involves a considerable amount of effort and that the determination of reactor power by noise analysis seems rather cumbersome and probably of no application to power and research reactors.

Recently, (Podvratnik, 2011) preformed calculations to support absolute thermal power calibration of Slovenian TRIGA Mark II reactor. He found that thermal power calibration with an absolutely calibrated fission chamber proves to be viable. He reported that knowing the accurate mass of the uranium contained inside the fission chamber and accurate calibration of the fission chamber is crucial in correct thermal power calibration. This informs the earlier use of absolute method in the determination of NIRR-1 power (Musa et al., 2012). The work which we seek to extend here is by utilizing the calorimetric and heat balance methods in the calibration of the NIRR-1 reactor power.

Recently, (Ahmed, et al., 2011) measured the available excess reactivity of NIRR-1 using the method of control rod critical depth of insertion and a method in an effort to measure the reactor‟s power and obtain its maximum operable times. Their result shows that there is a strong correlation between the reactor‟s operating time and its coolant temperature with the core excess reactivity. At 3.77mk excess reactivity it was observed that for a full-power flux with the control rod position unlimited, the reactor operated for five continues hours. At half-power and under the same excess reactivity condition, the reactor reaches 8 h

9

before the temperature effects sets-in. However, re-measurements done in 2009 shows that excess reactivity of the reactor has reduced to 2.80 mk, the operable time at full flux dropped to 3.5 h while that of half-power became 7 h. They also observed that the reactor‟s fluence depletion over the period under review was directly related to the excess reactivity reduction. They therefore conclude that for a clean core excess reactivity of 3.77 mk, 5 and 8h are the optimum times for the operation of miniature neutron source reactors under the two flux conditions (full and half power, respectively). As demonstrated in the study, a deviation from these optimum times is an indication of a drift in the reactor‟s core excess reactivity and hence the need to add beryllium shims to compensate for the lost. The present work therefore ascertains if this need exist using calorimetric (slope) and heat balance methods.