-

Module 15

Neutron Detectors And ReactorInstrumentation

7.1 MODULE OVERVIEW 3

7.2 MODULE OBJECTIVES 3

7.~ ..EtJrlrRO" IC)I:r1rE~O~ •.••••.•••••.•.•••........••••....••••..•...••••••....••••...••••••3

Neutron Detection Methods •••••••••••••••••••••••••••••••..••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 4

Nucleer Reectlons Used In Neutron Detection 5

7.4 THERMAL NEUTRON DETECTORS 6

Boron Proportlonel Counters ••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 7

fission Reectlon Detectors••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 7

7.5 SELF-POWERED NEUTRON DETECTORS 8

Electrons•__ _ _ _ 0:••_ 12

July, 1997 (R·1)l:\authtmg\bruce b\Inltial tralnlng\sclence fundamentals\22106 nucleertheory\masterslmanual\22106-trn-15-r1.doc

lIodule15 Page 1

7.6 NEUTRON DETECTOR APPLICATIONS' 14

Startups and Low Power Neutron Detectors 17

Reactor Control at the High Power Range 19

Transfer of Reactor Control 19

Neutron Flux Mapping 20

Platinum Detectors calibration Requirement 20

7.7 COMMON PROBLEMS OF NEUTRON DETECTORS•.•••••••••••••••••••••••••••••22

Ion Chamber Detector Failures 22

In-core Detector Failures 24

ASSIGNMENT .................................................................................. 26

July, 1997 (R-1) Module'S Page2I:\authtmg\bruce bIlnltlal trainlng\science fundamentals\221 06 nuclear theory\masters\manuall22106-tm­15-r1.doc

-

7.1 MODULE OVERVIEW

This module will cover the different types of thermal neutrondetectors and their applications including various failure modesof these detectors when used in reactor instrumentation andcontrol.

7.2 MODULE OBJECTIVES

After studying this module, you should be able to:

L1C 7.1 State why different detectors are used in reactor instrumentation.

L1C 7.2 State the principle of operation of a neutron flux detector.

L1C 7.3 Identify the different failure modes of reactor Instrumentation.

L1C 7.4 State why in-core neutron flux detectors require continuous thermalcalibration.

7.3 NEUTRON DETECTION

Neutrons, like gamma-rays, have no charge and therefore cannotinteract in matter by means of the coulomb force whichdominates energy loss mechanisms for charged particles andelectrons. Neutrons can also travel through many centimeters ofmatter without any type of interaction and thus can be totallyinvisible to a detector of common size. When a neutron doesinteract it is with a nucleus of the absorbing material. Thisinteraction may either cause the neutron to disappear totally and

. be replaced by one or more secondary radiations, or change theenergy or direction of the neutron significantly.

July, 1997 (R-1)1:\alithtrng'bNce b\lnitial traInlng\sclence fundamentals\22106 nucleartheory\masters\manual\22106-tm-15-r1.doc

Course 22106

Notes & References

Module 16 Pllge3

Course 22106

Noles & References

Secondary radiations resulting from neutron interactions arealmost always heavy charged particles. These particles may beproduced either as a result of neutron-induced nuclear reactions,or they may be the nuclei of the absorbing material itself whichhave gained energy from the neutron collisions. Most neutrondetectors utilize some means of converting the incident neutroninto secondary charged particles that can then be detecteddirectly.

7.3.1 Neutron Detection Methods

Neutrons are detected through nuclear reactions which createenergetically-charged particles such as protons, alpha particles,and so on. For slow (thermal) neutrons, significant interactionsinclude elastic scattering with absorber nuclei and a large set ofneutron-induced nuclear reactions. Because slow neutrons have asmall amount of kinetic energy, very little energy can betransferred to the nucleus in elastic scattering. Consequently, thisinteraction cannot be used to detect slow neutrons. Severaldifferent neutron reactions have been used to detect slow

. neutrons. These include: (1) prompt capture of neutrons resultingin charged particle emission, (2) delayed activation reactionswhere an activated nucleus emits some form of radiation within aconvenient half-life and energy, and (3) fission reaction resultingfrom neutron capture.

Because the capture cross section for thermal neutrons is high,many neutron detectors use thermal neutron capture reactions asthe basis for their measurements. Any type of neutron (thermal,intermediate, or fast) can be measured with this kind ofinstrument as long as the neutron is in the thermal energy range(below 0.5 eV) when it enters the sensitive volume of thedetector. This range is conventionally called the slow neutronregion and is distinguished from intermediate and fast neutronswith energies above this value.

July. 1997 (R-1) Module 16 ,..,,,l:\aulhlmglbruce b\inltlal trelnlng\sclence fundamentals\22106 nuclear theory\masters\manuall22106-tm­15-rl.doc

7.3.2 Nuclear Reactions Used in NeutronDetection

It is important to point out that all common reactions used todetect neutrons create heavily-charged particles, although veryrare specialized detectors use gamma rays produced in thereaction. Possible reaction products are listed below:

Course 22106

Notes & References

target nucleus + neutron ~

recoil nucleus (fast neutron)

proton

alpha particle

fission fragments

The boron-IO reaction is the most popular way to convert slowneutrons into directly detectable particles. The reaction may bewritten

The Li* indicates that the reaction product 7L may be left in itsground state or in its excited state. When thermal neutronsinduce the reaction. about 94% of all reactions lead to the excitedstate and only 6% directly to the ground state. The energy valuein either case is very large (2.310 or 2.792 MeV). When theexcited lithium returns to its ground state, it emits gamma rays,which always escape and do not contribute to the response of thedetector. The two reaction products are emitted in oppositedirections, and always share the energy of the reaction in thesame manner.

This is a widely-used reaction due to: (1) its large cross sectionfor thermal neutrons (3,838 barns), (2) the abundance ofnaturally-occurring boron-10 (19.9%), (3) its large release ofenergy per reaction making discrimination against gamma easy,and (4) the ease with which boron-10 gas can be incorporatedinto gas-filled detectors or used in solid form to coat the detectorelectrode.

July, 1997 (Ro1)I:\authtmg\bruce b\lnltiaJ traInIng\science fundamentals\22106 nucleartheory\mllsters\manuall22106-tm-15-r1.doc

Module 15 Pages

Course 22106

Notes & References

Thermal neutrons can be detected by using'35U, 23"Pu or otherfissionable materials coated on the detector wall. When neutronsare captured by the fissionable material on the detector walls anda fission reaction takes place the fission fragments move inopposite directions. One enters the sensitive volume of thedetector and the other strikes the wall. The advantage of adetector using the nuclear fission reaction is the large energyrelease that allows simple discrimination against gamma rays(which cause ionization that transfers much less energy).

The remaining two slow neutron detectors measure beta orgamma decay current produced after the neutron is captured aswell as the secondary electrons produced by gamma decay.These detectors are discussed in more detail in the self-powereddetectors section of this module.

7.4 THERMAL NEUTRONDETECTORS

A detector that is often used for slow neutrons is the proportionalcounter which uses boron as a fill gas, in solid lining, in an ionchamber, or in proportional counters coated with fissile materials(fission chambers).

7.4.1 Boron Proportional Counters

BF, (boron trifluoride) gases used in proportional counters areenriched to about 96% in IlIB which results in a level of efficiencyfive times greater than the natural boron gas. BF, tubes candiscriminate against gamma rays that are often found with theneutron flux being measured. Gamma rays interact primarily inthe wall of the counter and create secondary electrons that mayproduce ionization in the gas.

duly. 1997 (R-1) Modu~.16 ,..6I:\authtmg'bruce bllnltlal tralnlng\sclence fundamentals\22106 nuclear theory\masters\manuall22106-lm­15-r1.doc

-Because the stopping power for electrons in gases j3 quite low, atypical electron will only deposit a small fraction of its initialenergy in the gas before reaching the opposite wall of thecounter. Thus, most gamma ray interaction should result in low­amplitude pulses that can simply be eliminated by an electronicamplitude discrimination circuit without sacrificing detectionefficiency.

Alternatively, boron can also be introduced into a conventionalproportional tube by applying boron as a solid coating on theinterior wall of the proportional tube. The advantage of thisconfiguration is that a more suitable gas than BF, can now beused. When introduced as a wall coating, the boron is applied asboron oxide (B.o,) or boron carbide (BC.). A detector using aboron wall coating is inferior to one that introduces the boron as afill gas (BF,) because a significant amount of the alpha particlesare absorbed by the boron coating or the wall resulting in asmaller number of electrons being collected by the central anode.On the contrary, when BF, gas is used, a much larger amount ofsecondary electrons are collected by the anode. Thus, the mostsensitive neutron detectors use BF, gas. This type of detector isused to measure neutron flux during the initial startup of thereactor.

7.4.2 Fission Reaction Detectors

The detectors that use the fission reaction to detect neutrons arecalled fission chambers. These small ion chambers are typicallymade of stainless steel walls and electrodes, with an operatingvoltage from SOV to 300V. The chamber walls are usually linedwith highly enriched uranium to enhance the ionization current.Argon is the common choice for the chamber fill gas and it isused at a pressure of several atmospheres. The elevated pressureensures that the range of fission fragments within the gas doesnot exceed the detector's small dimension. The pulses producedby fission fragments entering the sensitive volume of the detectorare large, and because the ion chamber does not produce large

. current flows, the output from the fission chamber is a series ofpulses that can be counted.

July, 1997 (R-1)I:\authtmg'bruce b\lnltlal tralning\science fundamentals\22106 nucleartheoryVnasters\rnanual\22106-lm-15-r1.doc

Course 22106

Notes & References

Module 15 PIIt/tI 7

Course 22106

Notes & References

When the detector is operating, the fissionable material on thedetector walls is being consumed (by fissioning). To help slowthe rate of depletion, a fast neutron absorber such as mU issometimes added to the fissionable material on the wall. WhenmU absorbs a fast neutron, 23~ is created after beta emission.The 23~ is itself a fissionable material.

Miniaturized fission chambers can be tailored for in-core useover any power range likely to be encountered in reactoroperation. These detectors can be used as travelling detectorsand as a reference point to calibrate self-powered detectors.

7.5 SELF-POWERED NEUTRONDETECTORS

A unique type of neutron detector that is widely used in reactorcores is the self-powered detector. Self-powered detectors arealso known by a variety of other names. In recognition of theearly pioneering work done by G.W. Hilborn, they aresometimes called Hilborn detectors. They are also referred to asbeta emission detectors, collectrons, electron emission detectors,PENA (primary emission neutron activation), and self-poweredneutron (SPN) detectors. However most CANDU stations usethe term in-core detector (lCD) for this family of devices.

The advantages of self-powered detectors when compared toother neutron sensors, include size, low cost, and the relativelysimple electronics required to use them. Disadvantages stemfrom the low level of output current produced by the devices, arelatively severe sensitivity of the output current to changes in theneutron energy spectrum, and for many types, for instancevanadium detectors, a rather slow response time.

July, 1997 (R-1) lIodu1e 15 Page B1:\authtmglbR1C8 b\Inltlal tralnlng\science fundamentals\22106 nuclear theory\masters\manuaJ\22106-tm­15-r1.doc

-7.5.1 Self-Powered Neutron Detectors

Based On Beta Decay

In its simplest fonn, the detector operates by directly measuringthe beta decay current after the neutrons have been captured.This current is proportional to the rate at which neutrons arecaptured in the detector.

Figure 7.1 illustrates the possible sequences which contribute tothe measured current. The operation is very simple: The currentthat corresponds to the beta rays given off by the emitter ismeasured between the emitter and an outer shell, called thecollector. The intervening space is filled with an insulator whichis selected to withstand the extreme temperature and radiationenvironment typically found in a reactor core. Insulators areoften made of metallic oxides, the most common beingmagnesium oxide. Collectors are generally made of high-puritystainless steel or Inconel. When these detectors are beingfabricated, great care must be taken to keep the material as cleanas possible to avoid contamination by substances that might alsobecome radioactive and add an interfering current to themeasured signal. Vanadium is the most popular emitter material.

July, 1997 (R-1)I:\authtmg\bruce b\lnltlal ttalning\sclence fundamentals\22106 nuclearlheory\masters\rnanuaIl22106-tm-15-r1.doc

Couru22106

Notes & References

lIoduie 15 PIIgfI.

CoUI'N 22106

Noles & References

-I ,

I!!...,;.J..Figure 7.1: Events that occur in neutron detector based on beta

decay

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

The name self-powered detector derives from the fact that thebeta decay current is measured directly, and no external voltageis required. Figure 7.2 shows a sketch of a typical SPN detectorbased on beta decay. The heart of the device is the emitter, madefrom a material chosen for its relatively high cross section forneutron capture leading to a beta-active radioisotope. The rest ofthe detector should not interact strongly with the neutrons, andconstruction materials are typically those with relatively lowneutron cross sections.

July, 1997 (R-1) Module 15 Page 10i:lauthtmglbruce b\lnltlal training\science fundamentals\221 06 nuclear theory\mastel1l\manuaJl22106-tm­15-r1.doc

-

Figure 7.2: Cross section view of a typical detector based onbeta decay

The choice of emitter material is the key to how well the detectorperforms. When selecting an emitter, it is important to considerits neutron capture cross section together with the energy andhalf-life of the resulting beta activity. The capture cross sectionshould not be too high or too low because a very low crosssection will lead to detectors with low sensitivity, whereas anexcessively high cross section will result in rapid burnup of theemitter material in the high neutron fluxes associated with reactorcores. Sufficiently high-energy beta rays should be produced toavoid excessive self absorption in the emitter or insulator. Aswell, the half-life of the induced activity should be as short aspossible to permit the detector to respond quickly to rapidchanges in neutron flux.

July, 1997 (R-1)l:\authtrng'bruc b\Inillal trainlng\sc:lence fundamentals\22106 nucleartheory\mastel'8\manua1\22106-tm-15-r1.doc

Course 22106

Notes & References

Module 15 Page 11

CQUrtle 22106

Notes & References

Vanadium detectors have a prompt gamma ray response that isgenerally a few percentage points of the neutron responsecomponent of the signal. The ratio of prompt to delayed signal incommercial vanadium detectors is reported to be about 6.5%.

7.5.2 Self-Powered Neutron DetectorsBased On Secondary Electrons

The primary disadvantage of self-powered detectors based onbeta decay is their relatively slow response time. Some effortshave been made to remedy this situation by electronic or digitalsignal processing, but it is preferable to speed up the inherentresponse of the detector itself. One way to accomplish thisobjective is to rely on the secondary electrons produced byprompt capture gamma rays resulting from neutron capture in theemitter. Figure 7.3 illustrates the sequences which contribute tothe measured current.

- ..•

Figure 7.3: Events that occur in detector based on secondaryelectrons

July, 1997 (R-1) 1Iodu,. 1$ P.,12i:lauthtmg\bruce bllnitial tralning\science fundamentll/s\22106 nuclear theorylmasters\manual\22106-lm­15-r1.doc

Some fraction of these gamma rays will interact to.formsecondary electrons through the Compton, photo electric, andpair production mechanisms. The current of the secondaryelectrons can then be used as the basic detector signal. Thesecapture gamma rays are emitted within a small fraction of asecond, as opposed to the much slower decay of typicalvanadium detectors.

The most commonly-used detector to optimize the signal fromgamma rays coming from neutron capture is the platinum emitterdetector (only 80-88% prompt when responding to flux change).With some exceptions, the neutron sensitivity of prompt detectorsis substantially less than detectors based on beta decay. It shouldbe noted that their faster response can make them the detector ofchoice for reactor power regulation and control.

The detector will also respond to external gamma rays. Externalgamma rays on the detector produce secondary electrons whichyield a detectable signal as illustrated in Figure 7.3. The externalgamma-ray induced signal can be either positive or negative, inthat the net flow of the current may be in the same or the oppositedirection to the neutron-induced current. Which polarity prevailsdepends on whether more gamma-ray-induced electrons flowfrom the emitter to the collector or vice versa. Either situationcan exist, and this depends on the specific construction of the.detector. Because they are sensitivity to gamma rays, thesedetectors are unusable at low neutron flux.

The most common variant of the platinum in-core detector usedin CANDU reactors is the platinum-clad detectors. This type ofdetector is shown in Figure 7.4.

JUIy,1997 (R-1)l:\a~ruce b\Initlal tralnlng\sclence fundamentals\22106 nucleartheory\muters\manuaIl22106-tm-15-r1.doc

Cours.22106

Notes & References

1Iodu,. 16 PI/fI8 13

Cou,..e22106

Noles & References

-Figure 7.4: Cross section view of platinum-clad detector

._------------------

7.6 NEUTRON DETECTORAPPLICATIONS

Power produced by thermal nuclear reactors is generated throughfission induced by slow neutrons. Nuclear sensors that are part ofreactor control or safety systems are generally based on detectorsthat respond primarily to slow neutrons. The detector typesdiscussed earlier can be modified for reactor neutronmeasurement.

There are two general categories of reactor instruments: in-coreand out-of-core. In-core sensors are those that are located withinthe reactor core and are used to provide detailed knowledge ofthe flux shape within the core. These sensors can be either fixedin one location or provided with a movable drive. Obviouslythey must be small to fit in between the pressure tubes or fuel

.channels in the reactor core.

.July, 1997 (R-1) lIodu" 15 P.14l:\authlmg\bruce b\lnitlaJ tralnlng\sclence fundamentaJs\22106 nuclear theory\masters\manuaJl22106-tm­15-r1.doc

-Out-of-core detectors are located outside the core ~d thusrespond to properties of the neutron flux integrated over theentire core. They measure leakage flux at their location. which,by calibration can be made representative of the whole core. Thedetectors may be placed outside the pressure vessel and normallywill be located in a much less severe environment than that of in­core detectors. Size restrictions are also less of a factor in theirdesign.

The neutron detectors used in a typical thermal nuclear reactormust satisfy the following requirements:

• fast response,

• monitoring over a wide range from low to high power,

• monitor the rate of increase of neutron power at all powerlevels,

• accurate power measurements at high power,

• control of spatial power distribution on medium & highpower levels.

• SDS instruments must be able to limit peak local power(channel or bundle powers and high rate with powerincrease).

• RRS signals must represent average power in the regulatedregion.

In choosing specific detector types. consideration must be givento the expected neutron signal level compared with noise sources,the speed of response of the detector. and the ability todiscriminate against gamma-induced signals. Each of thesecriteria assumes different importance over various ranges ofreactor power. and as a result multiple detector systems areusually provided. each designed to cover a specific subset of thepower range.

July, 1997 (R-1)I:lauthtmglbruce b\Inltlal traInIng\sclence fundamentals\22106 nucleartheory\masterslmanuaI\22106-tm-15-r1.doc

Course 22106

Notes & References

Modu,. 15 PIIge 15

CoUrtHI 22106

Notes & References

Protection and regulation for a nuclear reactor require thecapability of fast response over the complete range of neutronflux. Figure 7.5 gives a typical value of the flux for power level10-14 fraction full power and 150 % full power. Figure 7.5illustrates the three sets of sensors with overlapping operatingrange to cover the entire range of the reactor as no singleinstrument gives adequate coverage for all flux conditionsencountered.

••

..",.

..",.

..",.

..",.....

..",.------.---•Figure 7.5: Range sensitivity of nuclear instrumentation for

reactor power measurement and control

July, 1997 (R-1) Modu1e 15 Pilge 16l:\authtmglbruce b\lnltlal tralnlng\science fundamentals\22106 nuclear theory\mastel8\manuall22106-tm­15-r1.doc

-7.6.1 Startups And Low Power Neutron

Detectors

Neutron sensors used for startups and low power operations aregas-filled. Their advantages in this application include:

• inherent gamma-ray discrimination properties (sensitiveonly to neutrons),

• give immediate response to flux change,

• sensitive over wide range (at high power log rate signalused for SDS),

• can be accessed on power (installed out of core),

• easy testing of device.

The lowest range, usually called the initial startup range, isencountered during the initial startup of the reactor power. Atstartup, there is not enough neutron flux and the neutroninteraction rate is very low. Good discrimination against gammarays is essential in this range, and it can be accomplished byusing either fission chambers or BF3 proportional counters.

As the power level increases to the low power range, the neutronflux is usually so large that gamma-ray-induced currents in ionchambers are no longer significant, and simple uncompensatedion chambers are commonly used as the principal neutron sensor.Because these instruments are often part of the reactor safetysystem, there is a premium placed on simplicity which alsofavours the use of uncompensated ion chambers. Figure 7.6shows a simplified illustration of an ion chamber neutron detectorsystem that is commonly used in reactor control.

July, 1997 (R·1)1:\aulhtmg'bRJC8 b\lnltlal tralnlng\science fundamentals\22106 nuclearlheory\masters\manuall22106-tm-15-r1.doc

Course 22106

Notes & References

1Iodu,. 16 ,.. 17

Course 22106

Notes & References

...._ _ __.- _....•.............._...•...._ .

•r~

Figure 7.6: Ion chamber neutron detector system

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

The presence of high voltage on the HV electrode of the chamberis detected by the tell-tale alarm circuit. This tell-tale connectiondetects the loss of the applied high voltage.

The linear amplifier produces a voltage output that isproportional to the ion chamber current. This output is then usedas input to the reactor regulation system.

The log amplifier is capable of producing output voltageproportional to the logarithm of the ion chamber current. Thevoltage signal produced by the log amplifier is used for reactorcontrol at lower power levels. As the power level increases thethermal neutron flux signal is supplied by the linear amplifier.

.Iu1y, 1997 (R-1) Module 15 PIItJtI1B1:\aUthtmg\bruce bllnitiaJ training\sclence fundarnentals\221 06 nuclear theorylmasters\manual\22106-tm­15-r1.doc

-The differentiator produces a log rate signal (whic~ is a timederivative of the log signal), that it supplies for the safety systemto monitor how fast the log signal is changing. The log ratesignal remains accurate and continues to be used right up to, andbeyond, full power.

After a shutdown, reactor restart monitoring is done by the ionchambers. Initial startup instruments are required only duringinitial startup or after a shutdown during which the ion chambersystem has gone off-scale.

7.6.2 Reactor Control At The High PowerRange

At the higher end of the power range the ion chambers are notsufficient for power control. Although the ion chamber and thelinear amplifier are accurate to 150% full power, the leakage fluxthey monitor is not a good representation of the flux in the reactorcore for the simple reason that these detectors are normallyinstalled outside the reactor. The leakage flux will be stronglyaffected by moderator poison, fueling in adjacent channels, fluxtilts, movements of nearby reactivity devices, etc., and it will berelatively insensitive to power changes in the central core or thefar side of the reactor.

In these high power ranges specially designed self-poweredplatinum detectors are used because of their much faster responsethan vanadium detectors.

July, 1997 (R-1)I:\authtmg'bruce b\Inltial tralning\science fundamentals\22106 nucleartheory\masters\manuaI\22106-tm-15-r1.doc

Course 22106

Notes & References

Modultl 16 PIIge 19

Cou,..e22106

Notes & References

7.6.3 Ttansfer Of Reactor Control

During reactor restart, the RRS ion chamber monitors reactorpower at low power and the rate of power increase at all powerlevels. At low power when ion chambers are used for reactorcontrol, the requirement for accurate thermal powermeasurements can be dropped, but a good indication of the rateof change of fission power is needed. At high power, the in-coreplatinum detectors are used and these are calibrated to givethermal power level. However, thermal calibration of the ionchambers is required so that control passes smoothly from one setof regulating instruments to the other as power increases ordecreases.

7.6.4 Neutron Flux Mapping

Flux mapping is the method of monitoring the three dimensionalpower distribution within the reactor so that th~ individual fuelbundles and fuel channels can be operated within their designlimits.

Flux mapping uses a large number of in-core vanadium detectorslocated throughout the reactor. The response of these detectors,unlike those of the platinum ICDs, is all due to thermal neutronabsorption in the emitter. This means the signal output is exactlyproportional to the thermal neutron flux at the detector. Theiraccuracy, neutron sensitivity and very localired "point" fluxreadings make them ideal for providing information on in-corespatial variations of neutron flux (flux mapping). Thisinformation is necessary for fuel management and as a referencein calibrating the other detectors used for reactor control.Vanadium detector readings must be corrected by independentthermal power readings.

Ju!V. 1997 (R-1) IIoduIe 16 PtIgtItlOI:\authtmg\bruce b\inltlal tralnlng\sclence fundamenlals\22106 nuclear theorylmasters\manuaJl22106-lm­15-r1.doc

7.6.5 Platinum Detectors CalibrationRequirement

The reactor regulating system's platinum ICDs monitor bulkpower and zone average flux at high power. Accuratemeasurement must be provided to prevent overrating the fuel.However, platinum ICDs will respond to both neutron andgamma flux, and not directly to fuel thermal power. The actualICD signal will not be an accurate indicator of fuel power for thefollowing two reasons, and it therefore requires continuousthermal calibration.

• Platinum ICD signals, although very fast, do not reliablyindicate average power in the zone because they onlysample the flux in their region. They are thereforesusceptible to localized flux changes such as refueling orreactivity mechanism movement.

• Platinum ICDs have delayed components, approximately15% to 17% of the total signal lags power change by atime constant varying from a few seconds to hours. So thesignal produced by the detector during a power changemay not accurately represent the true fuel power.

Corrections to these measurements are made by using bulkthermal power measurements and/or using spatial informationderived from the set of in-core vanadium detectors that are part ofthe flux mapping system.

Thermal measurements are capable of providing accurateindication of the actual power output from each channel.Thermal measurement is done by measuring the power outputfrom some of the channels in the inner zone of the reactor whereaverage power output can be obtained that is independent of thelocal flux variations. These thermal power measurements areused for calibrating the platinum ICDs and vanadium detectors.This calibration will ensure accurate measurement by theplatinum detector for prompt power indications over theoperating range.

July, 1997 (R-1)I:\authtmg\bn.rce b\lnltial training\science fundamentals\22106 nucleartheory\masters\manual\22106-~15-r1.doc

Cou,..e22106

Notes & References

Module 15 Pagetl1

Course22106

Notes & References

Although thermal power measurement is accurate, it is not usedfor reactor power regulation for the following reasons:

• Slow response time because of the thermal lag of the fuel,coolant and temperature detectors. This lag is in the orderof 20 seconds or more. Precise control of bulk powerwould be difficult with this lag in the control system.

• Thermal measurements can only measure the sum of thebundle power in the channel. Since the fuel channels passthrough inner and outer zones of the reactor, thermalmeasurements cannot assess the individual contributionsof the two zones to the total channel power. For zonecontrol the power in each zone must be known and sothermal measurement alone are not suitable.

7.7 COMMON PROBLEMS OFNEUTRON DETECTORS

Faulty detector operation will be apparent from erratic, high orlow signals. The operator may be able to determine this bycomparing the signal with the other detectors installed in thesame location. Comparison can also be made with fuel channeltemperature profiles, channel power measurements, or signalsfrom other in-core detectors of SDSI and SDS2.

7.7.1 Ion Chamber Detector Failures

Proper operation of ion chamber systems is essential to avoidunnecessary shutdowns, especially at low power when the in-coreflux detector system is not capable of providing accurate data andtherefore is not used. To achieve high reliability of the ionchamber system, the following features are incorporated in theirdesigns:

July, 1997 (R-1) lIodule15 Pelletl2l:\authtmg\bruce b\initial tralnlng\sclence fundamentals\22106 nuclear theory\masters\manual\22106-tm­15-r1.doc

-• the use of three independent channels,

• the use of two computers for one signal,

• computer intercomparison checks between channelsignals, and irrationality checks by using the upper andlower signal margins,

,• alarm system for loss of ion chamber polarizing voltage,

• use of guard shields to bypass leakage current,

• vibration resistant design and construction.

Even with these features, problems can occur:

• electrical breakdown of the ion chamber or cables,

• noisy output signals,

• failure of polarizing voltage,

• low or negative log signal.

Electrical breakdown will cause the high voltage supply todecrease or fluctuate. Ion chambers can also fail due to brokenconnections or the amplifiers could fail due to high humidity withcondensation producing erratic signals.

A leak into the chamber volume will result in a reduction of theinsulation resistance between the guard shield and reactorground. This could result in high leakage current andcontamination of the ion chamber's sensitive volume and upset itssaturation.

Noisy output can also be cause by faulty (or incorrect) grounding,vibration of the ion chamber, pickup from a strong source ofnoise, or failure of the amplifiers.

July, 1997: (R·1)I:\authtmg\bruce b\inltial tralnlng\sclence fundamentals\22106 nucleartheory\maslers\manuaIl22106-tm-15-r1.doc

Course 22106

Noles & References

Module 15 Page23

Course 22106

Notes & References

An indication of loss of polarization voltage can be due to thefollowing failures:

• power supply failure (broken fuses).

• faulty tell-tale circuit or incorrect setting.

• lost connection in the interconnecting cables.

• electrical breakdown of the ion chamber.

A lower or negative signal can be caused by either the reactorpower being off scale (that is. less than 10% to 7%). or a failurein the system. The other two channels will determine whether thepower is off scale. A failure can be caused by a faulty amplifieror a broken conductor in the ion chamber or interconnectingcable.

7.7.2 In-core Detector Failures

There are two failure modes of self-powered detectors: open orshort-circuit faults. and deterioration of insulation resistance.The open or short-circuit fault appears as an absence of all or partof the signal. or the appearance of intermittent transients. Theother fault is not easy to detect without making periodicmeasurements of the insulation resistance. unless erratic signalsappear. The usual cause of insulation failure is either a failure ofthe sheath or the epoxy end seal. leading to the ingress ofmoisture into the hygroscopic MgO. Another possible cause oflow insulation resistance is deterioration of MgO insulation dueto radiation effects. This should not occur if MgO of the properpurity is used in fabrication of the detector.

JuIa41997 (R-1) 1Iodu»15 Pagd4I:\authlmglbruee bllnltlal trainlng\sclence fundamentals\22106 nuclear theory\maslerslmanuaJl22106-tm­15-r1.doc

Detectors used for flux mapping are designed to~ tolerant offaults in individual detectors and associated amplifiers, cabling,etc. Failure of one or a few detectors is not normally expected toresult in noticeable degradation of mapping accuracy. Howeverin operating in off-normal states, (for example, adjusterswithdrawn or mechanical control absorbers inserted), mappingerrors due to failed detectors may become relatively large(several percent or more), depending on the location of the faileddetectors. In particular, failure of detectors located close to aremoved reactivity device will tend to result in larger than normalerrors in detecting the associated reactivity device modes. Thismay cause comparatively large "local effects" errors in mappingbundle fluxes and channel powers located near the reactivitydevice in question.

If a significant number of signal are "failed" (about 5% or moredepending on detector location), convergence of the correctionloop, and mapping accuracy in general, will begin to showconsiderable degradation. Under this circumstances it becomeincreasingly important to correct the fault condition or, if thefault is in the detector themselves, to replace the defectivedetectors.

July, 1997 (R-1)1:\authtrng\bnJce b\initial training\sclence funclamentals\22106 nucleartheory\maste18\manual\22106-tm-15-r1.doc

Course 22106

Notes & References

Module 15 pageS

Course 22106

Noles & References

Assignment

1. Identify two applications wherin boron is used in ionchambers for neutron flux detection.

i) _

ii) _

2. In low neutron flux where gamma rays are present, BF3 caneffectively measure neutron flux by eliminating (gammapulse/electrons) through the use of an electronicdiscrimination circuit.

3. Neutron induced pulses have (lesser/greater) amplitude thangamma induced pulses.

4. At higher power levels, neutron flux is usually so large thatgamma pulses have no effect in the signal produced.(truelfalse)

5. Identify two common failures of in-core detectors.

i) _

ii) _

6. Vanadium detector sensitivity is not affected by the reactorflux changes. (true/false)

7. A vanadium detectoris 100% sensitive to neutron flux.(truelfalse)

8. Give examples of two applications of vanadium detectors.

i) _

ii) _

July, 1997 (R-1) Module 16 Page26l:lauthtmg\bruce bllnltlal tralnlng\sclence fundamentals\221 06 nuclear theory\masters\manuall221 06-tm­15-rt.doc

-9. A platinum detector (is, is not) suitable for low power

neutron flux measurements because incident gamma cancause (accurateJfalse) neutron flux signal.

10. At higher power levels, ion chambers (are/are not)sufficient for reactor control.

11. Ion chambers can give an accurate representation of theneutron flux in the reactor core at high power levels.(trueJfalse)

12. Three factors that can affect the ion chamber neutron fluxmeasurement at high power are:

i) _

ii) _

iii) _

13. Platinum detectors have a (faster/slower) response thanvanadium detectors.

14. Self-powered detectors based on beta decay have a much(faster/slower) response than detectors based on secondaryelectrons.

15. Give two reasons why thermal power calibration ofplatinum detectors is required when they are used as neutrondetectors during high power operation.

July, 1997 (R-1)I:\authtmglbruce b\initial trainlng\science fundamentals\22106 nucleartheory\masters\manuaJ\22106-tm-1S-r1.doc

Course 22106

Notes & References

Module 16 PI/(/tl tlT

Coursetl2106

Notes & References

16. Identify the method used to detect ion chamber detectorfailures while in operation.

17. Give the special feature used by ion chambers to detectfailure of high voltage supply.

18. Ust four ion chamber problems that can be encounteredduring operation:

i) _

ii) _

iii) _

iv) _

JvIIt. 1997 (R-1) lIodu1tl15 PIIgtI28l:\aulhtmg\bruce b\lnltial tralnlng\sclence fundamentals\22106 nuclear theory\inasterslmanua!\22106-tm­15-r1.doc