Response to U.S. EPR Design Certification Application RAI ...Additional information on the AMS (e.g., AMS overview) is provided in the Response to RAI 194, RAI 205, and in the AREVA

EVA

March 22, 2010NRC:10:017

Document Control DeskU.S. Nuclear Regulatory CommissionWashington, D.C. 20555-0001

Response to U.S. EPR Design Certification Application RAI No. 308, Supplement I

Ref. 1: E-mail, Getachew Tesfaye (NRC) to Ronda Pederson, et al (AREVA NP Inc.), "U.S. EPRDesign Certification Application RAI No. 308 (3791), FSAR Ch .4," October 21, 2009.

Ref. 2: E-mail, Martin Bryan (AREVA NP Inc.) to Getachew Tesfaye (NRC) "U.S. EPR DesignCertification Application RAI No. 308, FSAR Ch .4," March 1, 2010.

In Reference 1, the NRC provided a request for additional information (RAI) regarding the U.S.EPR design certification application (i.e., RAI No. 308). A schedule for responding to this RAIwas provided in Reference 2. Technically correct and complete responses to all questions inRAI No. 308 are enclosed with this letter.

The enclosed response consists of the following:

Question # Start Page End PageRAI 308 - 04.04-59 2 24RAI 308 - 04.04-60 25 42

AREVA NP considers some of the material contained in the enclosure to be proprietary. Asrequired by 10 CFR 2.390(b), an affidavit is enclosed to support the withholding of theinformation from public disclosure. Proprietary and non-proprietary versions of the enclosure tothis letter are provided.

If you have any questions related to this submittal, please contact me by telephone at434-832-2369 or by e-mail at sandra.sloanaareva.com.

Sincerely,

Sandra M. Sloan, ManagerNew Plants Regulatory AffairsAREVA NP Inc.

Enclosures

cc: G. TesfayeDocket No. 52-020

7-7

AFFIDAVIT

COMMONWEALTH OF VIRGINIA )) ss.

COUNTY OF CAMPBELL )

1. My name is Sandra M. Sloan. I am Manager, New Plants Regulatory Affairs

for AREVA NP Inc. and as such I am authorized to execute this Affidavit.

2. I am familiar with the criteria applied by AREVA NP to determine whether

certain AREVA NP information is proprietary. I am familiar with the policies established by

AREVA NP to ensure the proper application of these criteria.

3. I am familiar with the AREVA NP information contained in letter NRC:10:017,

the enclosed "Response to Request for Additional Information No. 308, Supplement 1" and

referred to herein as "Document." Information contained in this Document has been classified

by AREVA NP as proprietary in accordance with the policies established by AREVA NP for the

control and protection of proprietary and confidential information.

4. This Document contains information of a proprietary and confidential nature

and is of the type customarily held in confidence by AREVA NP and not made available to the

public. Based on my experience, I am aware that other companies regard information of the

kind contained in this Document as proprietary and confidential.

5. This Document has been made available to the U.S. Nuclear Regulatory

Commission in confidence with the request that the information contained in this Document be

withheld from public disclosure. The request for withholding of proprietary information is made in

accordance with 10 CFR 2.390. The information for which withholding from disclosure is

requested qualifies under 10 CFR 2.390(a)(4) "Trade secrets and commercial or financial

information".

6. The following criteria are customarily applied by AREVA NP to determine

whether information should be classified as proprietary:

(a) The information reveals details of AREVA NP's research and development

plans and programs or their results.

(b) Use of the information by a competitor would permit the competitor to

significantly reduce its expenditures, in time or resources, to design, produce,

or market a similar product or service.

(c) The information includes test data or analytical techniques concerning a

process, methodology, or component, the application of which results in a

competitive advantage for AREVA NP.

(d) The information reveals certain distinguishing aspects of a process,

methodology, or component, the exclusive use of which provides a

competitive advantage for AREVA NP in product optimization or marketability.

(e) The information is vital to a competitive advantage held by AREVA NP, would

be helpful to competitors to AREVA NP, and would likely cause substantial

harm to the competitive position of AREVA NP.

The information in the Document is considered proprietary for the reasons set forth in

paragraphs 6(b) and 6(c) above.

7. In accordance with AREVA NP's policies governing the protection and control

of information, proprietary information contained in this Document has been made available, on

a limited basis, to others outside AREVA NP only as required and under suitable agreement

providing for nondisclosure and limited use of the information.

8. AREVA NP policy requires that proprietary information be kept in a secured

file or area and distributed on a need-to-know basis.

9. The foregoing statements are true and correct to the best of my knowledge,

information, and belief.

SUBSCRIBED before me this ,

day of March, 2010.

Kathleen A. BennettNOTARY PUBLIC, COMMONWEALTH OF VIRGINIAMY COMMISSION EXPIRES: 8/31/2011

cms.~L-slow e wo Ii 9U,

I

Response to

Request for Additional Information No. 308, Supplement 1

10/21/2009

U.S. EPR Standard Design CertificationAREVA NP Inc.

Docket No. 52-020.SRP Section: 04.04 - Thermal and Hydraulic Design

Application Section: 4.4

QUESTIONS for Reactor System, Nuclear Performance and Code Review (SRSB)

AREVA NP Inc.

Response to Request for Additional Information No. 308, Supplement 1U.S. EPR Design Certification Application Page 2 of 42

Question 04.04-59:

AMS Commissioning Testing

The Aeroball Measurement System (AMS) software corrects the raw activation measurementfor a number of factors to infer the neutron flux at each AMS stack location. Some of thesefactors include: (1) wait time for the ball stack at positions B, C and D, (2) bleed over fromadjacent AMS detector locations, and (3) corrections for detector dead time, which affect thedetection efficiency at the large expected count rates. The accuracy of some of these correctionalgorithms will be verified during AMS commissioning testing.

During presentations to the staff, the applicant suggested that these correction algorithms canbe verified by the following tests:

1. Altering the measurement order of the subsystems; i.e., measure first A, B, C, D, wait for along enough period to allow for decay, and perform a similar measurement, but alter theorder, i.e., D, C, B, A of the measurements. Both measurements should result in the samepower shapes.

2. Perform a standard AMS measurement for -3 minute irradiation and compare the powershape with the one estimated for a -15 minute irradiation. Since the Vd lifetime is of theorder of 3 minutes, the -15 minute irradiation will have a significantly higher count rate:.Thus, this measurement will provide verification that the detector dead-time correction iscorrectly applied.

3. Perform measurements of bleed over from one measurement location to another using acalibration source.

Provide a testing plan to verify the accuracy of the correction algorithms that are applied to theraw AMS activation measurements. This question is a follow-up to a public meeting on the AMSsystem held at the NRC headquarters on September 16, 2009.

Response to Question 04.04-59:

AREVA NP has test information which can be used to verify the accuracy of the correctionalgorithms that are applied to the raw AMS activation measurements. This additional testinformation and how it verifies the accuracy of the correction algorithms is describedcb1low.

Additional information on the AMS (e.g., AMS overview) is provided in the Response to RAI194, RAI 205, and in the AREVA NP presentation to the NRC on the AMS (presentedSeptember 16, 2009 (ML092680278)).

For the AMS, each detector mounting beam of the measuring table contains 36 (32 KONVOI)silicon surface barrier (PIPS - Passivated Implanted Planar Silicon) detectors (see Figure04.04-59-1 and Figure 04.04-59-2). Each detector measures the characteristic gamma rays (viaCompton Effect) irradiated by the activated aeroballs. The amount of the characteristic gammarays is proportional to the neutron flux in the active core. The measurement is based onmeasuring the count rate due to gamma rays emitted by the activated aeroballs. A signal canbe measured only if the aeroballs are positioned under the detectors, and thus themeasurement does not have a memory effect. The position of the aeroball measuring probes

AREVA NP Inc.


within the detector mounting beams does not affect the measured signal. This providesmeasurement reproducibility.

There are several tests that can be performed on the AMS to evaluate system performance andcalibration. The AMS testing capabilities are described below:

Test requirements:

Functional testing of the AMS shall be performed.

System status is continually provided to the operator via a terminal. There are permanent testsfor nitrogen pressure, moisture, and power supply. Detected failures are identified on both thePICS and local indications.

The calibration of the PIPS detectors, as well as the inspection of the transportation system, isperformed at least once during a plant operational cycle. All other tests are performed ondemand. The additional on-demand tests are used to confirm that the instrumentation fulfills the.functional and performance requirements specified for it (see System Supporting Functions,shown below).

The AMS PIPS detectors are calibrated (via computer assistance) with a gamma source.

System Supporting Functions:

For verification and surveillance of the system the following functions are required:

" Residual Activity Measurement - measurement of residual activity.

" Calibration of the Detectors - calibration of individual semiconductor detectors to adjust thesensitivity of each detector.

" Zero Rate Measurement - locate defective semiconductor detectors from their increasingnoise signal level.

" Discrimination Threshold Monitoring - surveillance of the detector signal noise threshold(i.e., discriminator threshold).

" Single Measurement - count rate registration of one selected measuring section (e.g.,adjustment of the discrimination threshold or performance of an error diagnostic).

" Protocol - plotting of relevant system data on demand.

" Subsystem Locking/Pre-selecting - single operation (e.g., locking and pre-selection) ofsubsystems or single probe fingers.

" Transport Time Measurement - measurement of the transport time of the ball stacksbetween the activation position in the core and the measuring position in the detectormounting beam array for monitoring the mechanical properties of the transportation system.

Decay Time Measurement - logging decay curves of different ball stack segments to getinformation about the ratio of nuclides in the ball material. (This information is used forcalculation of activation values during aeroball measurement.)

AREVA NP Inc.


Special Technical Functions - Calibration of the- PIPS Detectors:

The PIPS detectors will potentially have different sensitivities due to manufacturing tolerances.These differences in sensitivity are measured and taken into account by the AMS computer by acorrection factor specified for each detector. This correction factor is used during eachsuccessive aeroball measurement and residual activity measurement.

The-calibration procedure is performed manually. Measurements, data analysis, and recordingof the results are automated.

The calibration source is positioned in a calibration device that is placed next to the detector-mounting beam array. The PIPS detectors are moved one after another into the calibrationposition. The radiation source generates pulses which are counted by the AMS calibrationprogram. The resulting count rate provides a measurement of detector channel sensitivity, -

which is used during calculation of activation values (correction program). The results of eachcalibration are displayed at the operator panel. The measured data are stored-in a file and canbe printed or displayed on the operator panel at any time.

Special Technical Functions - Transport Time Measurement Program:

Inductive sensors are used to monitor the movement characteristics of the ball stacks. Thesensors are mounted close to the measurement table on each ball guiding tube. During balltransport, a voltage is induced to the sensors. This voltage is measured and recorded. The .results, which provide information about the mechanical status of each guiding tube, can bedisplayed and plotted. The transport time can be measured, which provides indication of theballs in a ball stack being together (i.e., evidence that a ball stack was transported correctly anddid not become stuck).

Different modes of the program provide information that can be evaluated independeritly:

* Transportation'Time Measurement from the core to the table: For a selected system, ballstacks are transported from the activated position in the core via the wait position (below-...magnetic ball stops) to the measuring table while the ball stops are opened.. --

* Correlation measurement: This mode is similar to Transportation Time Measurement, but,the magnetic ball stops are opened consecutively. The measured signal allows the correctcorrelation between ball stops and beam position to be checked (i.e., after reconnecting thetubes following fuel load or other maintenance or activity requiring AMS tube disconnection).

* Manual Transportation Time Measurement: This mode allows Transportation TimeMeasurement in each direction (i.e., core - table or table - core), after moving the ballstacks in the appropriate position by using the ball transport mode (see below).

" Ball transport: This mode allows any kind of ball transport by controlling the single valves.

The results are stored and can be displayed or printed.

Testing and Inspections - Pre-operational Tests:

After the system is installed, pre-operational tests are performed to confirm proper function ofthe AMS and correct support of the interfaces. This includes testing:

AREVA NP Inc.


* Electrical functions, power consumption, error indications.

* Measurement functions of detectors and amplifiers.

* Transport functions.

* Operator interfaces.

* Data exchange with PTX (POWERTRAX, and system platform for Operational I&C).

These tests are completed during first startup of the plant with integral tests including PTX toverify the measured values.

Testing and Inspections - Periodic Tests:

Two different periodic tests confirm the correct function of the AMS. These tests contain thefollowing steps:

* Calibration of the PIPS detectors can be performed at any time during power operation orduring an outage. PIPS detectors are calibrated using a radiation source. The time intervalbetween the calibration tests is approximately the fuel cycle time, but the test can beperformed on-demand if the accuracy of the measured values is decreasing.

* Inspecting the transport system including transportation time measurement and assignmentcheck (assignment of activation position in the core and measurement position on the table).This test is performed after disconnection and reconnection of AMS components.

Testing and Inspections - Maintenance:

Different error indications (i.e., abnormal system operating conditions, AMS transport failure)and several test programs (see Special Technical Functions above) can identify problems.Maintenance by qualified personnel is, therefore, only required if errors are detected.

AMS Uncertainty:

The AMS uncertainty has been evaluated. This analysis uses KONVOI experimental data (i.e.measured count rates, from 46 subsequent aeroball measurements under normal operatingconditions taken from the first half-year of a burn-up cycle). Based on these measured countrates, the vanadium activation rates are calculated under application of a number of corrections.These corrections have also been taken into account for the uncertainty analysis.

The final uncertainty of the AMS is calculated via the Gaussian error propagation of individualstandard deviations from the individual error contributions. The final AMS standard deviationsare obtained at each measurement position of the AMS.

E ]Based on the similarities between KONVOI and U.S. EPR (see Figure 04.04-59-3), this analysismethodology is applicable to the U.S. EPR.

AREVA NP Inc.


To determine a general value for the AMS uncertainty, the normalized number of occurrences(i.e., probability density distribution) of the relative standard deviations (1 a error values) of thevanadium activation rates has been analyzed (Figure 04.04-59-4 and Figure 04.04-59-5 for theaxially integrated activation rates). The data was normalized using the mean value of thevanadium activation rates. Forty-six aeroball measurements (41216 single measurementpositions) have been considered. The normalization of the number of occurrences has beencalculated so that the integral is equal to one. Thus, a probability density is obtained.

The distribution of the relative standard deviations (1 (a error values) is globally representative ofthe distribution of local neutron fluxes in the core. The overall error is determined, so that it isapplicable to 95 percent of the AMS positions (i.e., 95% of the individual standard deviations arelower than the overall determined error). The value that is applicable to 50 percent of the pointsin the core (median value) is computed accordingly.

The overall relative AMS uncertainty of the vanadium activation rates calculated for theindividual measurement positions based on 46 aeroball measurements measured at a KONVOIplant under normal operating conditions ((99.6 ± 0.3)% of full power) has been determined tobe:

ElThe overall relative AMS uncertainty of the axially integrated vanadium activation rates basedon the same aeroball measurements has been computed to be:

E JIn general, it can be concluded that the AMS uncertainty depends on the local neutron flux.(i-e,,.the local power density). Positions subjected to a high neutron flux show a reduced standarddeviation. Thus, the usage of the standard deviation applicable to 95 percent of all AMSpositions (comprising positions with low neutron flux) is a conservative estimation of the AMSuncertainty especially at the positions with high neutron flux.

Response to specific AMS tests mentioned in Question 04.04-59:

1. The measurement order of the subsystems is automatically changed in a cyclic order duringnormal operating conditions of the AMS (order of subsystems: first 1, 2, 3, 4, then 4, 1, 2, 3,then 3, 4, 1, 2, then 2, 3, 4, 1) (not the order of the quadrants) - this is the general

measurement procedure. As such, there is no specific need to create a test to verify thatchanging the order does not alter the flux distribution curves of the AMS, as demonstratedby more than 35 years of AMS plant operating experience worldwide (AMS Presentation,ML092680278). This meets the requirements of the first proposed test.

For a given AMS measurement, subsystems are activated in the reactor core at the sametime, thus, they are subject to identical physical conditions in the core. The count ratemeasurements take place consecutively at the measuring table, regardless of the order -subsystem measurement. The measurement order of the subsystems only influences theirwaiting time. The difference in waiting time is completely corrected, because the time

AREVA NP Inc.


difference between end of activation and start of measurement is measured and the decayconstant of vanadium is well known. After the correction, the final count rates of allsubsystems, independent of their measuring order, are correlated to core neutron flux at the"end of activation" time point.

Therefore, the change of measurement order of the subsystems does not influence theresults (only the accuracy).

The transport times (from the core to the measuring table) of the aeroballs are onlymeasured for the first measured subsystem. However, the wait times for the othersubsystems are measured and included in the correction. To continuously monitor thetransport times of all subsystems, the measurement order of the subsystems is changedcyclically.

The first measured subsystem always has the lowest error because it has the shortestwaiting time. Changing the measurement order of the subsystems allows the same generalmeasurement precision for each subsystem (after four measurements).

Each activated Aeroball decays exponentially with the material specific decay constant A.Thus, it is important to determine the exact time difference between the end of the activationin the core and the start of the measurement. This time difference should be as short aspossible to obtain maximum count rates. The corrected count rate for vanadium is obtainedby extrapolation. The count rate corrections depend on the measuring section k (k=l,kmax), the detector mounting beam i 0=1,... ,imax) and the subsystem m (m=1 ,..,mmax).

For the U.S. EPR: kmax = 36, imax = 10 and mmax = 4

For a KONVOI plant: kmax = 32, imax 7 and mmax = 4.

(Ik,i,m )VK = (Ik,,,m )V " exp(4 5 2 " 5Tk,.m)

Where, (Ik,im)V -- Count rate corrected for decay time (unit: 1/s).

(Iki,m)V j Count rate due to vanadium activation (unit: 1s).

4V52 -, Decay constant of V52 (unit: 1/s).

b7'k,m Effective decay time, time difference between measurement

time Tkm and "end of activation time" To (unit: s).

Example AMS Measurement Data:

For the example AMS measurement data shown below from a single AMS measurement set(Figure 04.04-59-6 to Figure 04.04-59-9), the measurements have been performed in thefollowing order of the subsystems: first subsystem 4, then subsystem 1, followed bysubsystem 2 and 3. In general, the measurement order of the subsystems is automaticallychanged cyclically during normal operating conditions of the AMS. This is done to

AREVA NP Inc.


continuously monitor the transport times of all subsystems because the transport time canonly be determined for the first measured subsystem.

In addition, the first measured subsystem always has the lowest error because it has theshortest waiting time. Changing the measurement order of the subsystems allows the samegeneral measurement precision in turn for each subsystem.

All aeroball stacks (all subsystems) are simultaneously activated (i.e., they are irradiated atthe same time at different positions within the core). Due to the AMS design, their activitycannot be measured by the detectors simultaneously. Consequently, the aeroball stacksare divided into four subsystems which are measured one after the other.

The measurement order of the subsystems only influences the waiting time (time betweenend of activation and start of measurement). Thus, the accuracy of the measurementdecreases slightly (Figure 04.04-59-6 through Figure 04.04-59-9) for the later measuredsubsystems. The decay of the activated vanadium during the waiting time is corrected. Thetime difference between end of activation and start of measurement of each section of thespecific subsystem is measured. Because the decay constant of vanadium is a knownvalue, the decay of the aeroball stack activity is determined with high precision. Themeasured count rate of all subsystems is calculated relative to their activity at the time point'.end of activation" (as shown above). Although the subsystems are measuredconsecutively, the corrected count rate refers to the same time point ("end of activation")and therefore, the same core state for each subsystem. Thus, the order of themeasurements of the four subsystems does not influence the measurement results (only theaccuracy). Comparison of the data shown in Figure 04.04-59-6 through Figure 04.04-59-9supports the assumption that the count rates have not decayed to a point at which they cannot be accurately measured (i.e., too small to discriminate).

A set of four consecutive AMS measurements were performed over a 72 minute time-span.Figure 04.04-59ý1 0 and Figure 04.04-59-11 show the normalized vanadium activation ratesfor four consecutive measurements (of all subsystems) at four arbitrarily chosen -_measurement positions. In Figure 04.04-59-10 the color code indicates the chronologicalorder of the measurements. Where as, in Figure 04.04-59-11 the color code indicates themeasurement order-of the different subsystems for each of the four measurements, not thechronological order.

The last measurement set was performed 72 minutes after the first measurement. The co rehad not been stabilized;. consequently there were fluctuations resulting from the powerhistory and xenon fluctuations. Understanding that the core was in a transient conditionwhen the measurements were performed is important in interpreting the data, shown inFigure 04.04-59-10 and Figure 04.04-59-11.

Figure 04.04-59-10 shows the normalized vanadium activation rates at four measurement-positions within the core in their chronological order of these measurements. Thisrepresentation demonstrates that within the 72 minute time-span between the firstmeasurement set (10:42 h) and the fourth measurement set (11:54 h), the core changesfrom a rather flat distribution to a more "peak-at-top" pronounced power density distribution(systematic change). Thus, the change in the axial power density distribution can beconsidered as real. Additionally, these measurements where taken at the beginning of afuel cycle at full power.

AREVA NP Inc.


Except for these real differences in the power density distribution of the core between themeasurements, no systematic effects were observed for the measurement order of thedifferent subsystems (Figure 04.04-59-11). Thus, the order of the measurements of the foursubsystems does not influence the measurement results. Table 04.04-59-1 shows therelationship between the measurement number and the chronological measurement order(e.g., 1. corresponds to first measured subsystem, i.e., first measured position in the core)for selected measurement positions in the core (N02, B11, G10, L10).

Figure 04.04-59-12 shows the relative standard deviations at all AMS positions of 46aeroball measurements taken at a KONVOI plant (corresponding to N'kmaJxmax'mmax=(46)'(32)'(7)'(4)=41,216 single measurement positions). All measurement positions areincluded in this analysis. Forty-six subsequent aeroball measurements of the first half-yearof the burn-up cycle were evaluated for this uncertainty analysis. These aeroballmeasurements were taken under normal operating conditions ((99.6 ± 0.3) percent of fullpower).

The standard deviations at the AMS measurement positions depend on the neutron flux atthe regarded measurement positions. On positions at the edge of the active core zone (top,bottom and side edge positions), the neutron flux (vanadium activation rate) and the localpower density is reduced. Thus, the AMS standard deviations are increased at thesepositions.

In addition, the AMS uncertainty increases depending on the elapsed time between end ofactivation and start of the measurement due to the vanadium decay in the aeroballs. Thisimplies that the AMS standard deviations are increased for the subsystems that have beenmeasured at a later time. In Figure 04.04-59-12, the measurement sequence of thesubsystems is indicated by a color gradient: the first measured subsystem is plotted in red,the second and third measured subsystems in shades of violet, and the last measuredsubsystem is plotted in blue.

2. Using 35 years of operational plant data (ML092680278), a model was developed tosimulate different activation durations. The activation duration is a parameter which isadjustable. The standard value is 3 minutes, but it can be increased if desired. Thedependence of the AMS uncertainty on the activation duration was analyzed by simulatingactivation durations of 6 and 12 minutes based on original data with activation duration of 3minutes. The measured count rate was multiplied by a factor according to the requiredactivation duration. The simulated activation data was modeled using the followingrelationship:

(lk,i,m )sim " ('k,i,m) 1 - exp(---V52 g Tsa)

1- exp(-2AV5 2 - .3.

Where, (Ik,).' Simulated measured count rate.

o5Trsl -- Simulated activation duration.

Q0, ) -- Measured count rate for an activation duration of H7Ta.

AREVA NP Inc.


AV52 Decay constant of V52 (unit: 1/s).

Generally, during normal operating conditions (100 percent full power), it is not necessary toincrease the activation duration - three minutes is the optimum. If an accurate measurementis required at 30 percent full power, it is possible and reasonable to increase the activationduration. This would result in an increased count rate - but this count rate would beapproximately comparable with the count rate obtained at 100 percent full power and threeminutes activation duration. Consequently, the dead time correction and effects resultingfrom detector saturation would not play an important role for the measurement results.

For testing purpose, measurements have been performed at 100 percent full power withactivation duration of 15 minutes. No problems with the detectors or processing electronicshave been observed. Thus, an increase of the activation duration is possible.

In general, an optimum count rate exists. If the count rate is too high, the uncertainty willincrease, because of the dead time correction. On the other extreme, if the count rate is toolow, the statistical error of the count rates will dominate the overall uncertainty.

The relative standard deviations (AROm I Rkim ), plotted as a function of the normalized

vanadium activation rates (R norm ) for 46 measurements, are shown in Figure 04.04-59-13

and Figure 04.04-59-14. The data for the activation durations of 6 and 12 minutes weresimulated based on the measured data for 3 minutes activation duration. The data for thefirst measured subsystem is shown in Figure 04.04-59-13 and the data for the lastmeasured subsystem is shown in Figure 04.04-59-14.

For increased activation duration, the (simulated) measured count rates are'increased. Thisreduces their Poisson error and thus, their uncertainty is reduced. This is shown in Figure04.04-59-13 and Figure 04.04-59-14 for the first and the last measured subsystem.Especially at the edge positions of the core (low vanadium activation rates), the Poissonerror is the main contribution to the AMS uncertainty.

The increase of the activation duration decreases the uncertainty most effectively at theseedge core positions (i.e., top, bottom and side edge positions). The uncertainty at relevantpositions with high vanadium activation rates (high neutron fluxes) is less improved,especially for the first measured subsystem.

An increase of the activation duration of more than three times of the half life of vanadium(more than 11 minutes) is not reasonable because the vanadium is almost completelysaturated after an activation duration of approximately three times of its half-life. This can'also be observed in Figure 04.04-59-13 and Figure 04.04-59-14, especially for the firstmeasured subsystem.

In general, it is not possible to reduce the AMS uncertainty ad infinitum by simply increasingthe activation duration, because other factors also contribute to the final uncertainty.

In practice, the activation duration can be adjusted accordingly. [

AREVA NP Inc.


] This covers the intent of the secondproposed test.

3. During a PIPS detector calibration procedure, the bleed over to adjacent measuring beamsis corrected. This meets the requirements of the third proposed test. This correction isbased on data from actual measurements. The scatter factors from adjacent measuringbeams are measured during initial reactor plant testing and then stored within the AMScomputer as configuration parameters. The scattering from adjacent measuring beams isthen automatically subtracted from the measured count rate for each measurement.

The calibration of the detectors is performed with an external calibration source. Duringcalibration, each detector (one after another) is taken from its position at the measuringtable and placed onto the calibration source. The calibration source is a Co6° sourcepositioned in a shielded container. The calibration source geometry is identical to the,geometry of the aeroball tubes in the measuring table (see Figure 04.04-59-15). Becauseonly one detector is measured at the same time during calibration, and this detector ispositioned at the calibration source (not at the measuring table), scattering and bleedingeffects do not exist during detector calibration. No aeroballs are located at the measuringtable during detector calibration. Thus, there is no additional radioactive source in themeasuring table room which could disturb the calibration measurement.

Scattered radiation and bleed over radiation between neighboring detectors of adjacentmeasuring beams (see Figure 04.04-59-16) only has to be considered during an aeroballmeasurement. When performing an aeroball measurement, the aeroballs of only onesubsystem at a time are located at the measuring table. For the U.S. EPR, one subsystemcorresponds to 10 radioactive ball stacks that are measured at the same time (i.e., onesubsystem per measuring beam), see Figure 04.04-59-1 and Figure 04.04-59-2.

For protection against scattered radiation from the aeroball stacks positioned in neighboringdetector mounting beams, the PIPS detectors are mounted in shielded tapering ports(collimation apertures). Radiation-shielded material is also mounted between the detectormounting beams. Because of the high energy of the gamma radiation emitted by theaeroballs, a measurable component is still scattered to the detectors of the neighboringdetector mounting beams. [

Within one mounting beam, the scattered radiation from neighboring aeroballs (not directlypositioned under the detector tapering port) causes an increase of the effective detectorcollimation aperture. This effect is the same for all detectors and mounting beams and istherefore eliminated by the normalization of the measured activation values (not theabsolute values, but the normalized values are used for power density calculation).

The compensation of scattered radiation from the aeroballs positioned in neighboringdetector mounting beams considers the position of the tubes under the shielded collimationaperture (Figure 04.04-59-16). [

AREVA NP Inc.


FSAR Impact:

The U.S. EPR FSAR will not be changed as a result of this question.

AREVA NP Inc.


Table 04.04-59-1-Relation between the Number of the Measurement andthe Chronological Measurement Order

N02 B11 G10 L1010:42 h 4. 3. 1. 2.11:16 h 3. 2. 4. 1.11:33 h 2. 1. 3. 4.11:54 h 1. 4. 2. 3.

AREVA NP Inc.

Response to Request for Additional Information No. 308, Supplement 1U.S. EPR Desiqn Certification Application Page 14 of 42

Figure 04.04-69-1-AMS Measurement Table Use In KONVOI (Type) Plants

123

40

Figure 04.04-59-2-AMS Measurement Table According to EPR Design

1__. - 1 - -. .. . .

ITT f 1-1-1 1U1-11. 1'fm

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

AREVA NP Inc.

Response to Request for Additional Information No. 308U.S. EPR Design Certification Application Page 15 of 42

Figure 04.04-59-3-Core Scheme for KONVOI (left) and EPR (right) with the Positions of the AMSInstrumentation

p

0

N

L

K

H

0 us

- - I 1-4 4141 t - r -

x

- - 'lii

N A

13

12

11

10

9

a

7

6

5

4

3

2

U] 1

- , - * -l-4-4 4-4-4--f -

112 1 5 1 I ° I I6 I7 I8 6 P'° 1 12 113 11 115 Istem 1 Q Subsystem 2 Subsystem 3

IAl I 1CI D I El I F I GI 1J I IL I MINI I RI SI TISSubsystemn 4 / Instrumentation Lance

AREVA NP Inc.


Figure 04.04-59-4-Normalized Number of Occurrences of RelativeStandard Deviations '.ks,m IRkim (1 a error values) for 46 Aeroball

Measurements under Normal Operating Conditions

AREVA NP Inc.


Figure 04.04-59-5-Normalized Number of Occurrences of RelativeStandard Deviations AR /R•,, (1 a error values) of the Axially Integrated

Activation Rates for 46 Aeroball Measurements under Normal OperatingConditions

AREVA NP Inc.


Figure 04.04-59-6-Normalized Vanadium Activation Rates R non' for the

First Measured Subsystem at 99.9 Percent Full Power

Figure 04.04-59-7-Normalized Vanadium Activation Rates R nor for the

Second Measured Subsystem at 99.9 Percent Full Power

AREVA NP Inc.

Response to Request for Additional Information No. 308U.S. EPR Design Certification Application * Page 19 of 42

Figure 04.04-59-8-Normalized Vanadium Activation Rates R',°, for the

Third Measured Subsystem at 99.9 Percent Full Power

Figure 04.04-59-9-Normalized Vanadium Activation Rates R no= for the

Last Measured Subsystem at 99.9 Percent Full Power

AREVA NP Inc.


Figure 04.04-59-10-Normalized Vanadium Activation Rates R n°r for Four

Measurements Taken Within 72 Minutes

Figure 04.04-59-11-Normalized Vanadium Activation Rates Rk", for the

Same Four Measurements Taken Within 72 Minutes

AREVA NP Inc.

Response to Request for Additional Information. No. 308U.S. EPR Design Certification Application Page 21 of 42

Figure 04.04-59-12-Relative Standard Deviations A.Rkim/RkIm as a

Function of the Normalized Vanadium Activation Rates R nom for 46

Measurements

AREVA NP Inc.


Figure 04.04-59-13-First Measured Subsystem, Relative StandardDeviations

Figure 04.04-59-14-Last Measured Subsystem, Relative StandardDeviations

i

AREVA NP Inc.


Figure 04.04-59-15-Tapering Port with its Detector and Arrangement of theBall Tubes under the Collimation Aperture to Explain the Geometry

Correction

AREVA NP Inc.

Response to Request for Additional Information No. 308U).S. EPR Design Certification Application Page 24 of 42

Figure 04.04-59-16-Scheme of a Detector Section to Explain the Correction of Scattered Radiation fromNeighboring Detector Mounting Beams

AREVA NP Inc.


Question 04.04-60:

Methodology to remove the 60Co background from SPND measurements

The 59Co SPNDs measurements can be divided in at least four components:

(1) Prompt response to neutron flux by neutron capture

(2) Prompt response to prompt fission gammas

(3) Prompt response to delayed fission gammas (decay heat)

(4) Delayed response by buildup of 60Co in the detector.

In discussions with the staff, the applicant described that the 60Co background is removed in Europeanplants based on measurements during refueling outages and an extrapolation algorithm that is afunction of burnup and is based on plant measurements. In principle, the 60Co background could beupdated every time a new AMS measurement is performed and new Cij calibration parameters aregenerated. The 60Co background should be updated in both, the PowerTrax calculation as well as inthe protection system where the measured current from the SPND is converted to a power level.

Provide-the methodology to remove the 60Co background from SPND measurements and how will thisbackground will be treated by the Protection system and the AMS PowerTrax calculations. Thisquestion is a follow-up to an audit on SPNDs held at the AREVA Rockville Office on September 17,2009.

Response to Question 04.04-60:

The method of background C060 signal removal described in this response uses a two step processwhereby theoretical flux models and actual operating history from cobalt detectors are used todetermine a factor (i.e., K-60 factor) for each Self-Powered Neutron Detector (SPND). This factor isthen used in the plant as part of an iterative algorithm to obtain the background signal correction(CUCO). This correction is updated in the protection system (PS) software for each SPND at theSPND's calibration interval. The main advantage of this method is that no continuous monitoring of theSPND currents or the local neutron flux is needed.

The U.S. EPR delivers continuous measurements of the local neutron flux density at given positions inthe core by using 72 SPNDs in 12 detector fingers or measuring probes. The 12 detector fingers will beequally distributed among 4 divisions (see Figure 04.04-60-1). The SPNDs will be n, 13 detectors withcobalt emitters and will require no polarization voltage power supply during operation. Cobalt, Co5 9 isused for emitter material because of its ability to promptly generate signals which rapidly follow thechange of neutron flux and low gamma sensitivity.

The neutrons and gammas (generated in the fuel by nuclear fission) produce Compton electrons in theCo 59 emitter material, insulation, and emitter conductor of the detector; therefore, the SPND is self-powered. The emitter conductor used for transmitting the detector signal is subject to the same gammaradiation as the emitter material and produces a current over its entire length. The effect of thisinduced current on the measurement signal is compensated for by running a compensating conductoralongside the emitter conductor. The emitter conductor and the compensating conductor use mineralinsulated metal sheathed cable. The main interactions leading to the formation of electrons contributing

AREVA NP Inc.

Response to Request for Additional Information No. 308U.S. EPR Design Certification Application Page 26 of 42-

to the measured signal are shown in Figure 04.04-60-2. The radial location of the detector strings isshown in Figure 04.04-60-3.

AREVA NP Inc.


AREVA NP Inc.


K

AREVA NP Inc.


)

AREVA NP Inc.


FSAR Impact:


AREVA NP Inc.

Response to Request for Additional Information No. 308U.S. EPR Design Certification Application

Table 04.04-60-1--Neckar II K60 Factor Data

Page. 31 of 42

AREVA NP Inc.


Table 04.04-60-2-Neckar II Data for Figure 04.04-60-5

AREVA NP Inc.


Figure 04.04-60-1 -Allocation of Co5 9 SPNDs within U.S. EPR Core

0.5 MM0.4 mmSh,,t

InsulatoI Wre

wireI finger perinstrument

1119IWFF0

I:1Th41Th4:flTh

* , . * ' 4;: 01

0j

0

One tube only Is infsns m eddependng on th. locaftn of Meinstrumented assembly In te core

IA I * @1 *I&Ir I@I~l~ lilt liii, liii lii

* lW&bnm~ aSWIMbNW

72 SPNDs in the U.S. EPR Core

AREVA NP Inc.


Figure 04.04-60-2-Co5 9 SPND I Neutron Flux Interactions

Neutron flux

The Emitter gets proportionally & positively charged to the neutron flux

j Co60+n.-,,, Ni60÷ +÷ + Y

Co6l + n -- ioNl61 + 0 ry Cmtn and-ray Photo- electrons

Emitter Is additionally positively charged (Co60/61 backaroun

Reactor y - flux

rCompton- andcable Photo- electrons 0

Emitter and cable are additionally positively chargc(compensation of the cable signal)

+ •' Total

Q+ , - DetectorS ~CurrentT+

AREVA NP Inc.


Figure 04.04-60-3-Position of SPND Fingers in KONVOI Core

JKS71 Jo . JKS82

JKS65JK2

JKS52 JKS12

JKS42 • *JKS31

0 POD probeBoundaries of surveillance zones

AREVA NP Inc.


Figure 04.04-60-4-EFPY Evolution of SPND Signal with Typical EPR Neutron Flux (12 month fuel cycle)

AREVA NP Inc.


Figure 04.04-60-5-Neckar II Burnup Data for Eight SPNDs

AREVA NP Inc.


Figure 04.04-60-6-Co 60 Current Evolution over Time

AREVA NP Inc.


Figure 04.04-60-7-C_UNON Slope and 160 Evolution at given Calibration Intervals

AREVA NP Inc.

Response to Request for Additional Information No. 308U.S. EPR Design Certification Application Paqe 40 of 42

Figure 04.04-60-8-Algorithm of Calibration Correction, CUCON

AREVA NP Inc.


Figure 04.04-60-9-Co6 ° Current with CUCO Compensation (12 month Fuel Cycle)

AREVA NP Inc.


Figure 04.04-60-10--Co6 0 Current with CUCO Compensation (24 monthI Fuel Cycle)

FSAR Impact:


Documents

Response to U.S. EPR Design Certification Application RAI ...Additional information on the AMS (e.g., AMS overview) is provided in the Response to RAI 194, RAI 205, and in the AREVA