ML18139C291

Evaluation of Online Boron Analyzers, Final Rept

ML18139C291
Person / Time
Site:	Surry, North Anna, 05000000
Issue date:	04/30/1982
From:	Lechnick W NUS CORP.
To:
Shared Package
ML18139C292	List: ML18139C291 ML18150A079
References
RTR-NUREG-0737, RTR-NUREG-737, TASK-2.B.3, TASK-TM NSAC-46, NUDOCS 8304010427
Download: ML18139C291 (90)

Text

• :* .. "'.

.  :*:*.{::

• An Evaluation of On-Line Boron Analyzers Prepared by NUS CORPORATION

• 1 .

.' . .  ; *: . : ..* ~

. * .. ; *.. *.*\ .

. .-~.; ****'.

• The Nuclear Safety Analysis Center Is operated for * * * *

• the electric utlllty Industry by the Electric Power Research Institute

- - -- ~~-- - - - -- - - -----

8304010427 830325 PDR ADOCK 05000280 p PDR

An Evaluation of On-Line Boron Analyzers NSAC-46 Final Report, April 1982 1*

i l I.

Prepared by NUS CORPORATION

-Park West Two

\ Cliff Mine Road ii Pittsburgh, Pennsylvania 15275 I'\

I/ Principal Investigator

!! W. Lechnick Prepared for Nuclear Safety Analysis Center Operated by Electric Power Research Institute 3412 Hillview Avenue Palo Alto, California 94304 NSAC Project Manager R. N. Kubik

ORDERING INFORMATION Copies of this report may be ordered from Research Reports Center (ARC), Box 50490, Palo Alto, CA 94303, (415) 965-4081. There is no charge to NSAC member utilities and certain other nonprofit organizations.

. I Keywords: Boron Analyzers. Boron Analysis, Postaccident Instrumentation. Qualification EPRl's programs receive federal financial assistance by virtue of sponsorship of the Institute by_ federally owned utilities, and their benefits are available to all eligible persons regardless of race, color, national origin. handicap, or age.

Copyright© 1982 Nuclear Safety Analysis Center. All rights reserved.

NOTICE This report was prepared by the organization(s) named below as an account of work sponsored by the Nuclear Safety Analysis Center (NSAC), operated by the Electric Power Research Institute, Inc. (EPA/). Neither NSAC, members of NSAC, the organization(s) named below, nor any person acting on behalf of any of them: (a) makes any warranty, express or implied. with respect to the use of any information, apparatus. method, or process disclosed in this report or that such use may not infringe privately owned rights; or (bl assumes any liabilities with respect to the use of, or for damages resulting from the use of, any information, apparatus, method, or pro-cess disclosed in this report.

Prepared by NUS Corporation Pittsburgh, Pennsylvania LiiEt..t ..% q ,.;; -0,Gi(:.... * - .

I

\

r j

j i

NSAC PERSPECTIVE PROJECT DESCRIPTION Boron dissolv~d in the reactor coolant is a primary means of reactivity control in pressurized water reactors and a backup means of reactivity control in boiling water reactors. Thus, boron concentration is a fundamental safety parameter and must be measured*

Under normal condit~ons the boron concentration is determined by analyzing a grab sample and in some cases by an on-line boron analyzer. However, under postacci-dent conditions grab samples may involve unwarranted personnel exposure and not all of the new postaccident sample systems provide rapid measurements* Conven-tional on-line boron analyzers are overwhelmed by the radiation expected during an accident* To ov~rcome these shortcomings, several new on-line boron analyzers are on the market. These. have been especially designed to function during an acci-dent.

In this service, high radioactive fluids will expose components of on-line analyzers to radiation levels which can be as high as 10 6 R/hr* Radiation of this magnitude can damage some types of electronic components and elastomers that are present in the instruments. Photoelectric devices and small solid state compo-nents are particularly sensitive to radiation damage. It is also possible that high radiation levels may temporarily affect sensing elements.

The market for on-line analyzers is limited. Because of this, NSAC was concerned that this equipment might not be thoroughly and independently tested. This test program was sponsored as a result of that concern. Three commercially available

. postaccident boron analyzers were tested in radiation fields up to and exceeding those that would be encountered in an accident.

PROJECT OBJECTIVE The three boron analyzers were tested under normal conditions and at radiation levels as high as 10 5 to 106 R/hr. The tests sought to determine the accuracy of iii

• the analyzers, their reliability under normal conditions, their susceptibility to radiation damage, and their accuracy when exposed to high radiation levels.

PROJECT RESULTS

~The Ionics Digichem analyzer as modified by Sentry, the Westinghouse Mark V boron analyzer, and the Combustion Engineering Boronmeter are.all suitable for postacci-dent service if properly installed and maintained. The testing did indicate improvements that could be made to some of this equipment. These suggestions were accepted by the manufacturers and are being incorporated into the product line.

Robert N. Kubik NSAC Project Manager iv

---

:- ~-*~

. ~-

ABSTRACT Testing has been performed to evaluate the performance of three on-line boron analyzers and determine the effect of a high intensity ga111Da field (10 3 _to 10 6 R/hr) on this instrumentation. The main objective of this work was to verify the applicability of the analyzers for boron analyses under post-accident conditions.

The on-line analyzers tested included an Ionics model (DigiChem Analyzer) as modified by Sentry, the Westinghouse Mark V Analyzer, and the Combustion Engineering High Radiation Boronometer System. Irradiation testing was also performed on elastomers, solid-state electronics, and pH probes. Results of this work indicate that the three on-line analyzers tested are suitable for boron determinations during accident conditions. Radiation exposure levels involved in determining boron concentration with these systems would be essentially zero.

Results from gamma irradiation tests indicate that teflon will remain serviceable 6 .

at 10 rads exp,osure. Other elastomers tested were more radiation resistant than 4

is teflon. Solid-state components tested showed radiation damage at between 10 5 .

and 10 rads exposure. A slight but. constant bias in readout was noted when pH probes we*re exposed to high radiation levels. This bias has no significant effect on boron analyses results obtained from titrating the boron-mannitol acid complex.

V J

ACKNOWLEDGMENTS We wish to note the cooperation of the Sentry Corporation, Ionics Corporation, Westinghouse Electric, and Combustion Engineering. These companies furnish~d the equipment used in this test program and provided the manpower required for initial startup of the equipment. In particular, we wish to thank Joe Leon of the Sentry Corporation, Dale Lueck of the Ionics Corporation, Rick Pod of Westinghouse Electric, and Joe Kowles of Combustion Engineering.

vii

CONTENTS Section 1 INTRODUCTION AND

SUMMARY

1-1 2

SUMMARY

OF RESULTS - SENTRY DIGICHEM ANALyzER 2-1 Background Information 2-3 Test Description 2-8 Test Results 2-13 Discussion of Results and Conclusions 2-35 3

SUMMARY

OF RESULTS - WESTINGHOUSE ANALYZER 3-1 Background Information 3-2 Test Description

. . . 3-7 Test Results 3-10 Discussion of Results and Conclusions 3-17 4

SUMMARY

OF RESULTS - CE BORONMETER 4-1 Background Information 4-2 Test Description 4-10 Test Results 4-13 Conclusions and Recommendations 4-19 ix

i

'~

' Figure ILLUSTRATIONS I 2-1 Simplified Flow Diagram of the DigiChem Analyzer 2-1 I

2-2 Typical Characteristics of Module H21A3 2-20 2-3 Typical Characteristics of Module MCA8 2-20 3-1 Westinghouse Sampler Tank Assembly 3-4 3-2 Westinghouse Mark V Boron Analyzer 3-5 3-3 Geometry of Irradiation Assembly for Westinghouse Mark V 3-8 Sampler Tank Assembley 53,000 Ci Total Radiation Source 4-1 Sampler for the CE Boronmeter 4-6 4- 2 CE' Boronmet er 4-7 4-3 Predicted Deiay Time Due* to Mixing for the CE Boronmeter 4-8 4-4 Geometry of Irradiation Assembly for Boronmeter Sampler 4-11 53,000 Ci Total Radiation Source 4-5 One Day Strip Chart Recording of 2960 ppm Boronmeter 4-18 Analyses Results (Background Radiation) xi

TABLES Tables 2-1 Compositions* of Matrix Solutions used 2-12 in Testing the DigichemAnalyzer 2-2 Radiation Testing of Various Elastomers 2-14 2-3 Irradiation Testing of Photo-Interrupter Cell MCA8 2-15 2-4 Irradiation Testing of Photo-Interrupter Cell H21A3

• 2-16 2-5 Effect of Radi~tion(~) on an Internal Reference 2-22 pH Probe (L & N Cat #117495) 2-6 .Effect of Radiation on External Reference Probes 2-23 (Fisher Cat #13-639-8 and 13-639-63) 2-7 Effect of Rgd1a~ion on External Reference pH Probes 2-24 with 5 x 10 Rads ~revious Exposure (Fisher Cat #13-639-8 and 13-639-63) 2-8 Effect of Long Term Radiation Exposure on pH Probes 2-26 Z-9 Boron Analysis Results.with the Sentry Modified Digichem 2-28 Analyzer in a 1.75 x 10 R/hr Radiation Field

• 2-10 _Boron Analysis Results yith Sentry Modified Digichem 2-29 Analyzer in a 8. 64 x 10 R/hr Radiation Field .

2-11 Boron Analysis Results ~ith the Sentry Modified J?igichem 2-30 Analyzer in a 1.7~ ~ 10 R/hr Radiation Field Beginning of a Week End RlU'l 2-12 Boron Analysis Results ~ith the Sentry Modified Digichem 2-31 Analy.zer in a I. 75 x 10 R/hr Radiation Field Near End of a Week End RlU'l 2-13 Standard Boron and Blan~ Analyses Result With the Production.Model Digichem Analyzer 2_:33 2-14 Matrix Solution Analyses Results With the Production 2-34 Model Digichem Analyzer 2~15 Boron Reproducibility Results 2000 ppm Standard 2-38 xiii

j I

Tables Page l

I J

l

,i 3-1 Base Level Fissioning Count Rate For the Westinghouse 3-11 1 Mark V Boron Analyzer (Pure Water Results) .,(

l 3-2 Effect of Irradiation With Pure Water in the 3-12 l Westinghouse Mark V Boron Analyzer j 3-3 Effect of Irradiation With 5140 ppm in the 3-14 ~*

l Westinghouse Mark V Boron Analyzer 1 3-4 Effect of Irradiation With 2570 ppm Boron in the 3-15 Westinghouse Mark V Boron Analyzer 3-5 Steady State Operation for I Second Count Periods With 3-16 2570 ppm Boron. in the Westinghouse Mark V Boron Analyzer (Background Radiation) 4-1 Performance Specifications for the Boronmeter 4-3 4-2 Performance Specifications for the Boronmeter Preamplifier 4-4 4-3 Performance Specifications for the Boronmeter Signal Processor 4-5 4-4 Fission Count Rate(!) as a Fi.mction of Radiation levels 4-14 for a 50 and 60 M. V. Discriminator Setting on the Custom.

Designed Preamplifier WL-24038 (2,960 ppm Boron Concentration) 4-5 Bororuneter Accuracy Results for Low Level Boron Concentrati~ns 4-15 4-6 Bororuneter Accuracy Results for High Level Boron Concentrations 4-17 xiv

Section 1 INTRODUCTION AND

SUMMARY

Four vendors of in-line boron analyzers were invited to participate in a program to test the ability of their equipment to withstand postaccident environmental conditions. Three vendors responded to this invitation as follows:

• Sentry and Ionics with the DigiChem Analyzer. This boron analyzer is manufactured by the Ionics Corporation and is modified by Sentry to withstand the high gamma radiation levels encountered in post-accident.application. The modified instrument is sold only through Sentry.

• Westinghouse with their Boron Concentration Monitoring System (BCMS) Mark V'Analyzer.

• Combustion Engineering with their High Radiation Boronometer System.

The Sentry Modi~ied ,Di.gichem Analyzer provides for boron determination by remote titration of the boron-mannitol acid complex. It is assumed that the boron solu-

.tion contains the normal isotopic concentration of lOB to provide for reactivity control of the system. The procedure followed is identical to the referee method used for normal laboratory determination of boron concentration. The Westinghouse and Combustion Engineering analyzers provide for boron determination by measuring t h e lO B concentration. or t h e neutron a b .

sorption ch aracteristics

.. of ht e system.

Since neutron absorption is determined directly, it provides for an absolute meas-urement of reactivity control.

Equipment provided by these vendors was tested under normal operating conditions and in the presence of high-level radiation. The high-level radiation testing was performed in a hot cell using 60 co as the radiation source. Energy level of the 60 co gammas are normalized so that the energy absorbed by the materials in test will be comparable to the accident case. Maximum radiation levels were on the order of 10 5 - 10 6 R/hr. Test description and results for each analyzer are described separately in the main body of the report.

For those who are interested in results on irradiation testing of elastomers, solid-state electronics and pH probes, your attention is called to the Sentry-Ionics report.

1-1

• The general conclus"ions derived from this overall study and the advantages of using these on-line analyzers are sumnarized below:

• ~entry Modified DigiChem Analyzer

--The Sentry modified DigiChem analyzer is acceptable for use to determine boron concentration under post-accident conditions.

. f Concerning its use for normal power operations, the accuracy is \

probably acceptable.

--All boron analyses operations can be pe~formed remotely. The exposure involved in determining boron concentration would approach zero.

--Sample volume requirements are on the order of 1-2 ml per analysis, thus shielding requirements would be minimal.

--Analyses results can be achieved within 10 minutes.after

.. , the sample line is purged to obtain a representative sample *

--Though not sealed gas tight, there would be little* tendency for release of gaseous activity to the atmosphere. This would be particularly true if the sample addition sequence is changed to add water prior to addition of the sample.

I .

I

• Westinghouse BCMS Mark V Analyzer

--The Westinghouse Mark V boron analyzer is acceptable for use under post-accident conditions. It should be possible.to obtain an analysis within 5 or 10 minutes with this system. Concerning its use for normal power operations, the accuracy is probably acceptable.

--Count rate increases, and thus the ppm boron readout decreases with increasing radiation levels, however, the effect is a predictable one and accuracy is still quite acceptable.

~-For maximum anticipated exposure levels of 5 x 10 5 R/hr (10 Ci/cc activity level), the fissioning count rate will increase by about 1-2

I I

5 percent. This 5 percent increase in count rate will result in

' ~

a small error relative to the accuracy required for post-accident conditio'!'ls.

--The increase in count rate from irradiation is essentially a constant (as percent of count rate) for the three conditions tested (pure water, 2570 and 5140 ppm boron). The increased count rate does not linger when the radiation field is removed.

• Combustion Engineering High Radiation Boronometer System

--The CE Boromneter is acceptable for use under post-accident conditions.

--Reproducibility of results is excellent as based on fission count rate, however, conversion of count rate to ppm is somewhat below the accuracy desired for daily operations. CE indicates, however, that the proper curve fit routine in the microcomputer will provide proper ppm indication.

--A 500 second count rate is recommended for determining boron concen.tratio?s below 1,000 ppm.

--The use of a strip chart recorder is recommended for use with the borofiometer. This will improve statistics and show trending.

--There is some incrgase in the standard deviation from radiation levels in the range of 10 R/Hr at the planned discriminator setting of 50 millivolts. The increase is not significant with respect to post-accident analyses requirements.

1-3

Section 2

SUMMARY

OF RESULTS - SENTRY DIGICHEM ANALYZER The Ionics DigiChem Analyzer, as modified by Sentry, performed properly at radiation levels. of 8.64 x 10 4 R/hr. Maximum radiation levels anticipated under credible accident conditions are on the order of 104 R/hr.

The analyzer operated at an integrated dose of 2.7 x 10 7 rads. This corresponds to about three months of operation at maximum dose rates _anticipated under accident conditions.

If this system is used, NUS recoUDDends that the analyses to determine boron concentration be performed titrating the boron-mannitol acid complex from a pH of about 5.5 to pH-8.5. Actual pH used for the low and high pH end points should be

• determined by titrating known boron standards after addition of mannitol to the boron solution. Titrafing from a low pH inflection point (pH 4-6) to a high pH inflection poi~t (pH 8-8.5) can also be used, however, results of previous testing

• performed by NUS indicates better precision can be achieved by titrating to specific pH end points. Either method of titration (pH end point or inflection point) is acceptable for post-accident use.

If the production model DigiChem analyzer is modified as indicated below it should perform properly at radiation levels of 104 - 105 R/hr and continue to operate at an integrated dose of l0 7 rads.

• Separate the rotary spin assembly and sample addition module so that only these components are exposed to high radiation fields.

• Replace the photon coupled modules H21AY3 and MCA8 with mechanical switches. Alternately. it would be possible to provide localiz5d shielding for these modules to limit exposure level*to about 10 rads.

• Move the solid state relay for the solenoid actuated valve on the rotary reaction cell to a location outside the high radiatiorr zone.

• Replace the two nylon pulleys used to drive the rotary reaction cell with metal pulleys.

2-1

• Replace the teflon with elastomers that are more resistant to radiation. The teflon does not have to be replaced if the integrated exposure is limited to 10 6 rads *

• 2-2

BACKGROUND INFORMATION TEST PURPOSE Testing was performed to determine if the Sentry modified DigiChem analyzer and selected compo~ents from an unmodified DigiChem analyzer would suffer radiation damage in analyzing for boron at radiation exposure levels anticipated under post-accident conditions. Jn addition. testing was performed to determine the accuracy that could be achieved with the DigiChem analyzer for boron determinations during normal operating conditions. Modifications made by Sentry to the DigiChem analyzer include replacement of selected components that would be in a high radiation field with components made of more radiation resistant material. The selected components tested from the unmodified system include all solid state components and.elastomers that would be exposed to high radiation fields, the rotary spin assembly, and sample burette assembly. Both the rotary spin and sample burette assemblies ~i:>~ld be* exposed to relatively high radiation levels if the system is used in fost-accident testing.

The system provides for boron determination by remote titrat.ion of the boron-mannitol acid complex. It is assumed that the boron solution contains the normal isotopic concentration of lOB to provide for reactivity control of the system. The procedure followed is identical to the standard method used for normal laboratory determi~ation of boron concentration.

SYSTEM DESCRIP.TION The DigiChem analyzer system consists of a microcomputer, a rotary reaction cell assembly, a measurement sensor (pH probe in this application), and up to five sample and reagent addition.modules. A simplified flow diagram of the system is shown in Figure 2-1. The microcomputer consists of a series of plug-in circuit boards and the keyboard control panel devices. -A motherboard of bus lines and con-nectors is spread along the inside rear for plugging in the circuit boards as needed. All boards are easily replaced.

2-3

To Drain 4 ' - - - - - - - - - - - - - S a m p l 1 Flow r - - - -t Cu 2 461 Di11ital Diaplar External

-L-----i lnout MICROCOMPUTER

~

Data Outout

- - - L - - - - - - - - - - - - _ - - ----;

L-- -..t:==l l.\__l Printer r-rseiisa-;- I I I I Si11nal 1

I I I '---- -----,----- I

~ ---ii ) II **"-

'r - - -

I Computer I lnt1rfac1 I

I

~ - - --Si11nal Tran1milt1r for Remote Readout (I) ID

z:* z a 0

a 0 0

z: .

It "'

a.

0 1-1 l

Drain Line Simplified Flow Diagram of th1 Digi Chem Analyzer FIGURE 2-1

The rotary spin assembly is of modular construction, located at the lower left side of the DigiChem analyzer enclosure. The reaction cell inside the spin assembly is fabricated from teflon. It forms the heart of the assembly. As programmed, the microcomputer controls a variable speed motor which spins the reaction cell to provide for mixing of the solution as reag~nts are added. A cover to the_spin assembly provides entrances for the sample and reagent addition lines and the pH probe. Reagent addition and sensing occurs below the surface of the sample.

The sample and reagent dispensing modules are located on the bottom right hand side of the DigiChem enclosure. All modules are interchangeable with each other. The sealed plug-in modules provide a dispensing capability for up to five fluids. such as samples, reagents, and buffers. Three reagent (acid, base, and mannitol) addition modules, one boron standard addition module, and one sample addition module are used in this application. The digital controlled module has a stepper-motor which pushes a plunger through a burette to dispense fluids in precise microliter increments *.

The DigiChem analyzer was designed for process control applications, providing on-line analyses.and control for continuous, semicontinuous, and batch processes. It automatically performs titrimetric, colorimetric or selective-ion analyses. The microcomputer cont~ols,the automatic functions of sample and reagent dispensing, solution mixing, and co~centration sensing through a programmed sequence of analyses. The instrument as it is normally used takes and measures a sample from an on-line stream and performs the following programmed operations automatically:

• A fixed but programmable volume of sample is forced into the reaction vessel. Sample volume required for boron analyses is on the order of 0.5-2 ml for boron concentrations in the range of 1000 to 6000 ppm. Low bqron concentrations require higher sample volumes.

• Next the instrument adds dilution water to flush the sample line and provide suffici~nt volume to cover the tip of the pH probe. If it is planned to use the instrument for post-accident analyses, the programming sequence should be changed to add water first. This will dilute the sample and thus reduce the potential for release of iodine gas which may be present. After the sample is added, a little more water (2-5 ml) is required to flush the sample addition tip.

• If the solution is basic, as could be the case during an accident

• the system can be programmed to add acid to neutralize the.

base. The manufacturer should be consulted concerning pro-gramming requirements.

• A programmed volume of mannitol solution is added to the reaction vessel. -Mixing is achieved by rotation of the reaction vessel.

2-5

• The solu.tion is titrated with NaOH to an end point pH of 8.5 Alternately, the volume of titrant used can be determined by automatic derivation of the change in slope of the pH line (inflection point), which occurs when the caustic titration of the boron-mannitol complex is complete.

• The microcomputer takes the information concerning sample size and NaOH titrant volume used and computes the boron concentration.

Boron concentration is printed out as ppm boron on a computer tape.

Digital readout of boron concentration can also be provided locally or at some distant point.

• • At the conclusion of each analysis, the rotary speed of the reaction vessel is increased to spin out the solution in the vessel. Water is added at this time to flush the vessel by centrifugal force. Waste solutions are gravity drained to a collection tank.

SYSTEM MODIFICATION FOR OPERATION IN A RADIATION ENVIRONMENT For operation in a radiation environment, it is necessary to separate the rotary spin assembly and the sample addition module from the mi~rocomputer section to provide for localized shielding of components containing primary coolant *. The microcomputer section and other components which are not ex~osed,to the primary coolant probably cannot withstand high radiation exposure levels. Separation poses no serious technical problem since the units are of modular. construction. However, this task should not be undertaken lightly since there are many* electrical lines which must be lengthened, three solid state components which must be changed or shielded and longer length tubing must be provided for sample and reagent feed.

Preamplification of the pH signal is also required.

SENTRY MODIFICATIONS TO THE DIGICHEM ANALYZER The Sentry approach in providing a system that is suitable for on-line boron analyses under post-accident conditions was to replace all elastomers with more radiation resistant materials where necessary. Specific change~ made to the DigiChem analyzer by Sentry prior to this test work are indicated below. Other changes have since been made to correct problems identified in the high level irradiation experiments.

• All teflon and Kel-F parts in the system were replaced with more radiation resistant elastomers.

• 0-rings in the radiation zone were replaced with 0-rings made of materials known to be more resistant to radiation.

• Solid state controls that will be in the high radiation.zone were replaced with mechanical switches.

2-6

• The rotary spin assembly, sample addition module and reagent addition modules were separated 25 feet from the control module.

Only the rotary spin assembly and sample addition module will be in the high radiation area. Shielding is provided for these components.

• A separate pH preamplifier was added.

2-7

---

- - ----- - ---, -- - --- - - ~--- - - - -- - --- - -- - - -

TEST DESCRIPTION DIGICHEM COMPONENTS PROVIDED BY IONICS The irradiation testing was performed at the hot cell test facilities* at Georgia Tech. Components tested were those from the DigiChem Analyzer which would be subjected to moderately high radiation levels during boron analyses under post-accident conditions. In selecting the components that will be exposed to high rad-iation levels, it was assumed that the rotary spin assembly and sample addition module would be located behind a lead shield to separate other components from the high radiation area. The components tested were in the form provided by the manu-facturer in their standard version of the DigiChem Analyzer. Th~se components are as follows:

• Rotary Spin Assembly - This was exposed to 10 7 rads.'

• Sample Addition Module - This was exposed to 10 7 rads~

• Separate photo-interrupter cells for the rota y spin gssembly and sample addition module were tested at 10, 4 105 and 10 rads. This additional testing wa.s performed to determine the failure point since the photo-interrupter cells included as part of the rotary spin assembly ,nd sample addition module failed totally after exposure to 10 rads. * .

, ( . )

• O-R1ngs Buna, Kalrez and V1ton - These were tested to 10 6 and 10 7 rads exposure.

6

• Delivery Tips (Kel-F) - These were tested at 10 and 10 7 rads exposure.

• Te 1on ubing - Two separate lots of teflon tubirig were tested at 5 6 10, 10 and 10 7 rads exposure.

• pH and Reference Electrodes - Testing was performed with two sets of pH probes with external reference cells of the type used by Ionics in their DigiChem analyzer. In addition, testing was performed on a pH probe with an internal reference cell. Four series of tests were performed at maximum radiation levels of 106 R/hr, as follows: *

--Testing was performed using the buffer solutions indicated below.

Buffer solutions were used to minimize the effect of CO 2 pickup from air on pH of the solutions. It was necessary to leave the

• solutions exposed to air during the course of this testing.

2-8

l Organic buffers were not used because these buffers will degrade under irradiation, resulting in a change in pH. This change in I ".;,.-

pH could be wrongfully attributed to radiation induced degradation of the pH probes.

I

....E!L.. Compound Concentration Comments 4.5 Potassium dihy~rogen 0.2 Molar Laboratory phosphate preparation 7.0 Monobasic potassium Commercial phosphate and preparation sodium hydroxide 10.0 Potassium carbonate, Commercial potassium borate and preparation potassium hydroxide

--One set of buffer solutions was exposed to the radiation field in the hot cell, checking the pH of each solution periodically during the course of the working day. The pH probes and .

reference cell were exposed to the same radiation field as were the buffer solutions.* The pH meter was installed outside the hot cell. A 10 foot lead was required for connection of the probe to the pH meter.

--The*temperature of the solution in the hot cell was monitored so that correction could be made for the temperature effect on pH. The hot cell lights were turned off when not in use so that.temperature inside the hot cell would remain relatively constant.

--The control buffer solutions were stored outside the hot cell.

checking the pH at the same frequency as were the solutions inside the hot cell.

MODIFIED DIGICHEM ANALYZER The Sentry modified Digi°Chem analyzer is programmed to determine boron concentration hr automatic derivation of the change (inflection point) in slope of the pH line which occurs when the caustic titration of the boron-mannitol complex is complete. After the sample is added to the rotary reaction cell, deionized water and mannitol are added to the sample. A pH determination is made at this point and if the solution is'basic, acid is added automatically to reduce the pH to the range of 2 to 2.5. Then a back titration is performed to neutralize the excess acid~ indicated by an inflection point at around pH 5-6 in the slope of ~he pH line. Titration of the boron-mannitol complex begins_ at this time and is complete at the high pH inflection point.

2-9

Analyses performed with the production model DigiChem analyzer followed the pattern indicated above except that a pH of 5.5 was used as a start point for titration of the boron-mannitol complex and a pH of 8.5 was used as the end point.

Initially the equipment was operated outside the hot cell using four standard solutions containing 60,600, 1200 and 3000 ppm boron to verify operation of the system. The equipment was operated with a 25-foot separation between the control unit and other components as it would be in post-accident conditions. Following the initial testing the rotary spin assembly, with its pH probe, the sample addi-tion module, and the 3000 ppm boron standard were installed inside the hot cell.

The other components, the preamplifier for the pH probe, and the 1200 ppm standard remained outside the hot cell.

4 Testing was performed at radiation levels of 1.75 x 104 R/hr, 8.64 x 10 R/hr and 1.57 x 10 5 R/hr. The central point for determining the radiatton level was adjacent to, and just above the top of the rotary spin assembly. -,Other areas may have been slightly higher or lower than the reported radiation level. Total radia-tion exposure for the Sentry modified equipment was approximately 2. 7 x 10 7 rads.

PRODUCTION MODEL DIGICHEM ANALYZER A series of boron standards and post-accident matrix solutions prepared by NUS were analyzed with the DigiChem analyzer at the Ionics, Inc., plant in Watertown, Massachusetts. The analyses were performed by a NUS represen*tative using a produc-tion model analyzer. Titration of the samples were performed with O.S!f and O.lN NaOH to determine if there is an advantage to using a more dilute titre. No radia-tion exposure was involved *in this testing.

2-10

The solutions analyzed are listed below. Concentration of the additives used to make up these solutions are shown on Table 2-1.

~--

• Boron standards based on the weight of boric acid used to prepare the solutions.

• Boron standards containing low concentrations of lithium hydroxide.

This was to simulate the buildup of lithium in the primary coolant during normal power operations.

• Post-accident fission product matrices containing known concentrations of boron.

• Simulated solutions that might be expected to develop in the reactar containment sump after a loss-of-coolant-accident and activation of caustic containment spray.

• Solutions containing calcium, this testing was performed to determine if calcium that is leached from the concrete during a loss-of-coolant-accident ~ould affect the boron analysis results.

2-11

~: .

TABLE 2-1 COMPOSITION* OF MATRIX SOLUTIONS USED IN TESTING THE DIGICHEM ANALYZER mg/1 mg/1 mg/1 mg/1 mg/1 mg/1 mg/1 mg/1 Sample _B_ Li OH  !:!...£!.3 Ba* N0 ...!L Cs c1 Ce (N0,4 (NH 4 ) 2 .£!£12 3 3 Matrix 1 60 6.9 4.0 15.4 51.0 312.9 20.7

, 0 Matrix 2 2000 6.9 4.0 15.4 51.0 312.9 20.7 0 Matrix 3 6000 6.9 4.0 15.4 51.0 312.9 20.7 0 Matrix 4 6004 0.7 0.4 1.5 5.1 31.3 2.1 0 Matrix 5 0 0.1 0.4 1.5 5.1 31.3 2_.1 0 Matrix 6 0 6,9 4.0 15.4 51.0 312,9 20.7 0 Matrix 7 600 1.4 0 0 0 .o 0 2018 Matrix 8' 60 6,9 0 0 0 0 0 0

• The ppm concentrations indicated are based on weighed amounts of salts dissolved in one liter of water, .The boron is indicatd as mg/1 of boron. The other salts are indicated as mg/1 of Li OH, La c1 3 and so forth *

. .y 2-12

TEST RESULTS IRRADIATION TESTING OF SELECTED COMPONENTS FROM THE DIGICHEM ANALYZER Prior to reporting results it should be noted that the maximum radiation 1.evel expected in the*DigiChem analyzer would be 10 6 R/hr to the teflon plunger of the sample addition module * . This considers both gamma and beta radiation levels.

Radiation levels in the other areas of the analyzer would be in the range of 104 -

10 5 R/hr. Radiation exposure for other components would be less than 10 5*rads.

Estimated radiation-exposures are based on the following assumptions:

• The first boron analysis will be performed in triplicate at one

• hour after the accident occurs. Approximately 25 minutes will be required to perform the triplicate analyses.

• The prim~ry coolant will contain a maximum activity concentration of 4 cur~es per ml during the first boron analysis performed.

• Total volume of primary coolant contained within the tubing. the one,ml s*ample addition module and rotary reaction cell will be on the order of 3 ml. This volume is assumed to exist as a point source*within an imaginary sphere of one foot diameter.

• The radioactive coolant will be flushed from the system with water when the triplicate analysis is complete. Flushing will require the use of manual commands to the DigiChem analyzer.

• There will be two additional triplicate analyses performed within the next 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />. Boron analyses performed on a once per day r basis after. this time will not add significantly to total radiation exposure.

Limited radiation ~amage was observed in the testing performed; however, this was to components which have been replaced with radiation resistant components in the Sentry modified system. So~id state components which were damaged were subse-4 5 6 .

quently tested at irradiation levels of 10, 10, and 10 rads to establish the

• threshold level at which damage occurs.

ROTARY SPIN ASSEMBLY After irradiation to 10 7 rads, the rotary spin cell assembly was installed in an operational DigiChem analyzer and the system was activated. The teflon reaction 2-13

• TABLE 2-2 RADIATION TESTING OF VARIOUS ELASTOMERS Exposure Item in Rads Material Results (1) 0-Ring Buna Control, 75 Durometer  ; Irradiated, 70-75 Duromet er O-Ring Buna Irradiated, 75-80 Durometer 0-Ring Kalrez Control, 84-85 Durometer(l); Irradiated, 80-85 Durometer

• 0-Ring Kalrez Irradiated, 83-89 Durometer (1) 0-Ring Viton Control, 78-80 Durometer  ; Irradiated, 75-80 Durometer 0-Ring Viton Irradiated, 75-80 Durometer Delivery Tips Kel-F No visible effect; material would still serve its intended purpose I

Delivery Tips 10 7 Kel-F Slight darkening noted;* material would still serve its inte~ded purpose Tubing (Lot 1)(3) 10 5 Teflon No irradiation effect Tubing (Lot 1) 10 6 Teflon Rupture pressure - 1600 psi( 2) for three specimens Tubing (Lot 1) 10 7 Teflon Severly embrittle?; tubing would break when bent Tubing (Lot 2)(3) 10 5 Teflon No irradiation effect Tubing (Lot 2) 6 Teflon Rupture press~e - 1600 psi <2 >

10 Tubing (Lot 2) 7 Teflon Longitudinal cracking occurred when the 10 tubing was bent Tubing 107 Tygon Rupture pressure - 300 psi( 4 ) for three irradiated specimens (1) Evaluation of results was based on change in hardness. There was no visual indication of damage, (2) Control samples from both lots ruptured at 1600 psi. The failure mode differed in that a bubble developed on the control sample prior to rupture. Pressure failure of the irradiated samples resulted from developnent of pin-hole cracks.

(3) Two lots of tubing from separate sources were tested.

(4) One control specimen ruptured at 270 psi and the other at 290 psi.

2-14

TABLE 2-3 IRRADIATION TESTING OF PHOTO-INTERRUPTER CELL MCAS Milliamp Irradiation Level Output *Comments 0 rads (Control Sample)301 cell tested 104 rads* 24 1 cell tested 10 5 rads 0.1 2 cells tested 10 6 rads 0 2 cells tested

• 20 ma source, 5V detector excitation 2-15

.i TABLE 2-4 IRRADIATION TESTING OF PHOTO-INTERRUPTER CELL H21A3 Irradiation Level Milliamps Out~t*

0 rads (control Sample) 17.5 0 rads (Control Sample) 14.3 104 rads 14.0 10~ rads 9.8 10 rads 10.4 6

10 rads 0.5 10 6 rads 0.9

• 60 ma source, lOV detector excitation

[

'l 2..;16

cell within the rotary spin cell assembly spun momentarily and stopped. Testing performed indicated that the photon coupled interrupter module had failed. Sentry has replaced this module with a mechanical system which is not sensitive to radiation. The photo interrupter module was replaced and the teflon reac-tion cell was operational. Also, a solenoid valve would not operate because the solid state relay which activates this valve had failed. Replacement of this relay was required to activate the valve. This relay can be located outside the radiation zone in the computer control system, without making any change other than installing longer connection wires. Output from this relay is 120 VAC.

A visual inspection was then made of the rotary spin assembly with the following results:

• All glass and clear plastics had darkened. This darkening does not detract- from the.physical properties of the material.

• The two nylon pulleys which provide the driving force to spin the teflon cell ha~ a myriad of cracks, however the pulleys held together when operated. It would be pointless to do any further test work with these nylon pulleys since they are easily replaced with metal pulleys which are not affected by radiation.

• No visual*indication of degradation (cracks, loss of elasticity) could.be foupd in_the elastomer belt which connects the nylon pulleys.

• The ~efl~n tubing which feeds reagents and sample to the assembly had becpme very brittle. Other testing performed with teflon tubing indica~es the ~hreshold damage indication for teflon tubing is between 10 and 10 rads. Considering radiation damage alone.

the safety factor involved with the use of teflon tubing in this application is several orders of magnitude.

• The teflon reaction cell suffered no apparent visual damage. No cracking occurred when the cup-shaped cell was spread apart and squeezed together with maximum hand pressure.

SAMPLE ADDITION*MODULE The teflon plunger of the sample addition module may see total radiation exposures in the range of 106 rads. This component is discussed separately because it is the high exposure item in the overall assembly. Testing performed, as discussed below, indicates that this plunger will be functionally adequate at 107 rads exposure.

However, the system failed at this exposure level for another reason as identified below.

After irradiation to 10 7 ra'ds exposure, the system was installed in an operational DigiChem analyzer and the system was activated. The module did not operate~

2-17

• Testing performed indicated that the photon_ coupled so1id state limit switch had failed. (Sentry has replaced this solid state switch with a mechanical limit switch which is not sensitive to radiation.) After the solid state limit switch was replaced, the sample addition module was operated continuously !Or about 1.5 hours5.787037e-5 days <br />0.00139 hours <br />8.267196e-6 weeks <br />1.9025e-6 months <br /> without problem. Around 100 samples could have been processed during this period. The operational test was terminated at this ,time.

A visual inspection was made of this module with the following results:

• The glass and clear plastics had darkened.

• There was no visual indication of degradation of the teflon plunger or leakage past this plunger as it was operated.

• The teflon tubing was severly embrittled *. However, actual irradiation level §hat will occur under post-accident conditions is on thg order of 10 rads or less. The tubing is still serviceable at 10 rads exposure.

0-RINGS AND OTHER ELASTOMERS The elastomers were tested at several different radiation levels with results as indicated in Table 2-2.

The result of the testing clearly indicates that all elastomers in the DigiChem analyzer will withstand 10 7 rads exposure except for the teflon tubing. Exposure 6

level for the teflon tubing should be limited to 10 rads. This is beyond the exposure levels anticipated under accident conditions. Heavi~r components, such as the teflon reaction cell and the teflon plunger in the reagent addition module, remained operational at 10 7 rads exposure. However, it would be desirable to limit all -teflon components in the system to 10 6 rads of cumulative exposure.

Data from the pressure tests performed on the irradiated and control samples of teflon tubing are somewhat unusual in that all specimens failed at exactly 1600

• psi. All specimens were pressure tested with compressed nitrogen in the same manner, slowly increasing the pressure while 111:onitoring a *pressure gauge till failure occurred. About one minute was required to increase pressure to the 1600 psi failure level.

Tygon tubing was tested even though none is used in the DigiChem analyzer to

• develop alternate materials in the event that the teflon tubing failed at some low irradiated exposure level. This material is very resistant to irradiation based on no indication of change or darkening of this material even at 10 7 rads exposure.

2-18

Results of the test work with teflon tubing reported here are consistent with results of testing performed by General Electric on their nuclear plane project.

In the General Electric work, teflon hose that was maintained under static pressure with a liquid at 1200 psig while under gamma irradiation, started to leak at s.- 6 slightly above 10 rads exposure. Five irradiation tests were performed-1n the temperature range of 100 to 350°F. Temperature had no effect on test results. The hose was pressurized with a liquid identified as MIL-L-7808C.

Observations mad~ indicate that failure of the elastomers tested ultimately occurs because of emb_rittlement. This failure mode does not present a problem with the DigiChem analyzer since there are no components in the system that are flexed on a constant basis. There may be some very minor flexing of the teflon tubing but this would occur very infreqllently.

PHOTO-INTERRUPTER CELLS Evaluation of results for.these solid state components is based on typical charac-teristic curves developed by the manufacturers. Typical curves for the unirradiated cells are shown in Figures 2-2 and 2-3. A comparison was made of output current vers~s input current for the control samples and for separate samples tested after 'irradiation. As indicated in Tables 2-3 and 2-4, testing was 4 . 5' 6 performed at 10, 10 *and 10 rads. Data reported in these tables indicate that

• . 4 5 . 4 threshold damage occ~rs between 10 and 10 rads for module H21A3 and about 10 rads for module MCA8. It cannot be concluded that the MCA8 module will withstand 104 rads on a consistent basis since only one module was tested at this exposure level. Slight damage resulted from the irradiation, however, the module was still operational.

IRRADIATION TESTING OF TJ{E pH PROBES A separate test wa~ performed to determine the effect of irradiation on pH probes because satisfactory performance _on their part while under irradiation is an absolute must to operation of the DigiChem analyzer. This topic is of additional interest because of NRC regulations concerning pH determination requirements for all nuclear systems under po~t-accident conditions. There is limited data a_vailable indicating that pH probes should perform satisfactorily under high level irradiation. However, additional testing was considered necessary to pr~vide direct experience.

TYPICAL CHARACTERIST_ICS OF MODULE H21A3 IO z

C 8

4 I/v

~

Z-W a:

..... 2 C C u

I V a

I

I

.8 I

, /

I NORMALIZED TO* .

l aa 0

0

.4

.2

,/

1,-*ZOIIIA Va;*5V l'Ul.5£1)

PW*ioo,..

C 2 JV PRR*IOOns i:a a: I ,

z 0 .oi,

.06

, ~I-

l. .04

/

.:! .oz / ~

/

.Cl I I 2 4

• 1!1 IO 2D 40 60 SOIOO ZOO 400 liDO IOOO lr'NPUT ClJRRIENT-...a OUTPUT CURRENT VS. INPUT CURRENT FIGURE 2-2 TYPICAL CHARACTERISTICS OF MODULE MCAS 100

--- Va*UV T,.

• 25'1:

1I 10

/

. ~

./

."~ MC, ~v 0

ti 1.0 I

I s

.JI I

I 0.1 I

1.0 10 100 a.--ltDCUIIIIINTO' UD--

. l FIGURE 2-3

• 2-20

The results of irradiation testing performed on an internal reference pH probe are presented in Table 2-5. *There is no effect on pH indication at a exposure level of 3 x 10 5 R/hr, however the pH showed a decrease of around 0.1 pH units for the pH 4 and 7 range at a radiation level of 9.77 x 105* R/hr. This differs from external ref;~ence probes response which showed a slight increase during irradiatio~-as is later discussed. The high level radiation had no significant effect on the internal reference probes for the pH ~O buffer solution.

Total exposure on the internal reference probe was about 2 x 10 6 rads at the time it was broken wl)ile being moved with the hot cell manipulators. There was no observed change w{th time in the behavior of the pH electrodes during the two hour period the probe was under test.

The results of irradiation testing performed on an external reference pH probe are presented in Table 2-6. An increase i.n pH of 0.14 units was observed for the pH 5

range of 4 through 10 at a radiation level of 3 X 10 R/hrs. A further increase of 5

0.21 pH units was observed when the radiation level was increased to 9.77 x 10 R/hr. The radiation effect is reversible based on data taken when the radiation level was reduced *and later eliminated. Note in the subject table that the pH of the neutral buffer i~creased from 7.06 to 7.20 at a radiation level of 3 x 10 5 R/hr. This pH increased to 7.28 at a radiation level of 9.77 x 105 R/hr and then dropped to 7.19 when the radiation level was reduced to 3 x 10 5 R/hr. The final pH reading at the end of the test was 7.10 for both the control and the irradiated sample. The pH of the control sample was taken with the irradiated probe and with a probe that had not been irradiated.

Some radiation degradation of the pH 10 solution was observed after exposure to an 6

integrated dose of 5 x 10 rads. Note that the measured pH of this.solution dropped from 10.06 to 9.6'0, however, the control sample outside the hot cell showed no change in pH level when measured with the* _irradiated probe. If the reduction in pH of the basic solution had resulted from radiation damage to the probe, the pH of the control sample should also have indicated a lower pH.

The results of irradiation testing performed on an external reference probe with a previous history of 5 x 106 rads exposure are presented in Table 2-7. Note that the effect of irradiation on pH is slightly enhanced over that previously~

experienced. The increase for exposure at a radiation level of 3 x 10 5 R/hr and 9.77 x 10 5 R/hr is 0.15 and-0.3 pH units respectively versus an increase of 0.14 and 0.21 pH units during the initial testing reported in Table 2-6.

2-21

TABLE 2-5 EFFEC! OF RADIATION{ 3 ) ON AN INTERNAL REFERENCE pH PROBE (L & N CAT #117495)

Type pH at pH at Buffer Initial pH 3 X 10 5 9. 77 X 10 5 Solution No Radiation R/Hr R/hr 4.50< 2 >

I KH P0 4.60 4.60(1) 2 4 KH 2_ PO&. + 7.05 7.06< 91 >. 6.93

~aOH Kf0 3 , K3Bo 3 10.10 10.13(1) 10.12

+ KOH""

(1) No change in pH from the instantaneous reading was noted over a 5-lOminute exposure period.

(2). After taking the initial readings, the probe was left immersed in this solution. Readout of the pH meter varied between ~.48 and 4.51 during a 90 minute exposure period. The probe was broken at this time when it was moved.

(3) Total exposure= 2 x 10 6 rads *

• 2-22

TABIE 2-6 EFFECT OF RADIATION ON EXl'ERNAL REFERENCE PROBES (FISIIER CAT I 13-639-8 and 13-639-63)

Types Initial pl! pl! At Rads pl! At Rade pl! At Rads pl! At S Rads Final pH(l) 3 X 10 5 5 Buffer No Cm,. 9.77 X 10 5 Cmi. 3 x 10 Cmi

• 3 X 10 Cmi. No Solution Rsdiation ....!l!!L. Ex R/Hr _!!_ R/Hr ...!!..... R/Hr ~ Radiation 5

KH P0 4.39 4.53 1.5 X 10 Spilled buffer 2 4 buffer aolution KB~ POI<

7.06. 1,2e l.S*x 10 5 7.28 5 X 10 5

7.19 10 6

i.19 5 X 10 6 7.10

+ aOH 10.06

• 10.19 1.5 X 10.

5 10.27 5 X 10 5 Not( 3 ) 106 Not( 3 ) 5 X 106 9.60C 2 >

KtOr K B KOH determined determined 3 3 (1) A pH determination was made after all sources were removed from the hot cell.

(2) A control sample that vas exposed to the same environmental conditions without radiation exposure had a pl! of 10.05.

(3) Readings were not taken because of the extreme difficulty in moving the pl! probe with the manipulators.

2-23

TABLE 2-7 EFFECT OF RADIATION ON EXTERNAL REFERENCE pH PROBES WITH 5 x 10 6 RADS PREVIOUS EXPOSURE (FISHER CAT #13-639-8 and 13-639-63)

Type pH ats , pH at Buffer Initial J>l\i) 3 X 10 9.77xl0 5 Solution No Radiation R/Hr R/Hr 4.45 4.60 4.75 7.08 *7.26 7.37 10.08 10.24 10.37 (1) The pH measurements were* taken on probes that were previously irradiated as indicated in Table 2-6. No change was noted after about 2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br /> exposure in the test indicated above *.

2-24

The results of long term irradiation testing performed on external reference probes are presented in Table 2-8. Note that there is an increase of 0.1 and 0.15 pH 3

units at a radiation level of I0 R/hr. Overall results differ somewhat from p~vious experiments in that there is little effect at the high radiation_level exposure (9.1 x 10 5 R/hr). This may be because the high radiation exposure was 3

preceded by 60 hours6.944444e-4 days <br />0.0167 hours <br />9.920635e-5 weeks <br />2.283e-5 months <br />. exposure at 10 R/hr. Other experiments did not have this low level exposure preceding the high level test.

IRRADIATION TESTiNG OF THE SENTRY MODIFIED DIGICHEM ANALyzER Checkout of the !]lOdified equipnent was performed outside the hot cell for nominal boron concentrations of 60, 600, 1200 and 3000 mg/1. The boron solutions used were obtained by known dilution of a 6000 mg/1 stock solution. The system was set up as it would be under accident conditions with 25 feet of separation between t_he control module and the components that will be exposed to irradiation. Multiple analyses were were. performed at each boron concentration. The end point of the titation was determined by automatic derivation of the change in slope of the pH line which occurs when titration of the boron-mannitol titration is complete.

Maximun deviation noted from actual boron concentration was I.I percent with an average deviation on.the order of 1 percent. Accuracy requirements for boron determination during no~al power operation, as specified by many utilities, are on the order of plus or minus 0.5 percent.

When checkout of the equipnent was completed, the rotary reaction cell assembly, the 3000 ppn boron standard, and the sample addition mo.dule were installed in the hot cell. The 1200 ppm boron standard remained outside the hot cell. The eight 60co frames (53,000 curies, total) were arranged around the rotary reaction cell assembly and the sample addition mo_dule to achieve a radiation level of 1. 75 x 104 R/hr, as measured near-the top center of the rotary reaction cell assembly. This value is comparable to the general radiation level that may be present during a post-accident condition, assuming a 4 Ci/cc activity in the coolant. Boron analyses results with the Sentry modified DigiChem analyzer in a 1.75 x 104 R/hr radiation field are presented in Table 2-9. This testing was performed in the hot cell where temperature was on the order of 95°F. Testing performed outside the hot cell to checkout the equipnent was performed at a temperature of 72°F. This change in temperature could have had some effect on results because of 2-25

fr

j t-11 r-ti' J-

}.*

TABLE 2-8 EFFECT OF LONG TERM RADIATION EXPOSURE ON pH PROBES

l

)I Two Fisher external refe!ence pH probes were irradiated "for 60 hours6.944444e-4 days <br />0.0167 hours <br />9.920635e-5 weeks <br />2.283e-5 months <br /> at 10 3

R/Hr while one probe was in ,a pH 3. 98 and the other probe in a pH 7 .00

.I buffer solution. One probe was new and the other probe had 5 x 10 6 rads IE previous exposure * . The probe with the previous exposure was not identified ll in the data that was taken. Results from this test are as follows:

lC 3

I.

'E pH With pH at 10 pH After 60 Jours

?E No Radiation R/Hr Expos~re a.t 10 R/hrs

e 3. 98 4.09 4.09

!~

1.00 7.15 7.15

E The probes were then restandarized with new buffer solutions :and the

!I radiation level was increased to 9.1 x 10 5 R/Hr. One probe was left in a pH

c 3.98 and the other probe in a pH 7.00 solution. Results from this test are

.s as follows:

i Radiatio'{l Level = 9,1 x 10 5 R/Hr lE pH With pH At pH At pH At pH At pH At Total No Radiation 5 Min 1 Hr 2, 33 Hrs 17.33 Hrs 20 Hrs Exposures*

I'll 3.98 3.99 4.00 7

3. 98 4.01 3. 98 1.8 X 10 rads

t 7.00 7.06 7.04 7.05 7.05 1,8 7 7.06 X 10 rads ll

• One probe which was not identifiedd'rior to performing the test had a previous exposure history of 5 x 10 rads.

,,n 2-26

evolution of gas bubbles from degasification occurring as the liquids were heated.

The boron analyses results had a higher error band and more scatter than was t.

observed in testing performed outside the hot cell or in testing performed at

~ IQnics with a production in model analyzer. However, the results observ~d were

~ totally acc~ptable for post-accident use.

Testing was then performed at a radiation level of 8.64 x 104 R/hr (factor of five higher). The higher radiation level was achieved by moving the 60 co frames closer to the test equipment. Results of this test work are presented in.Table 2-10.

Note that there is little or no change in variability from results shown in the last part of Table 2-9.

The final test phase was performed at a radiation level of 1. 75 x 105 R/hr. This

.was the maximum radiation level achievable at the top center of the rotary reaction cell assembly with the 53,000 curie source. This work was performed over the a weekend. The test was st~rted late Friday afternoon. Reasonable results*were achieved for the first few analyses, at which time, the test personnel departed for the weekend. The equipment started behaving erratically soon after the personnel departed and continued this behavior for most of the weekend. Results achieved at the beginning of .this w~ekend run are presented in Table 2-11. Note that the

.results of the 1200 ppm and 3000 ppm boron standards are unacceptable for post-accident use. However, equipment problems were identified that are responsible for this condition and*changes have been made to the equipment design to prevent repeat of this occurrence. This is discussed later in more detail. In any event, it should be noted that the radiation levels anticipated under post-accident conditions will not approach the radiation level used in the final testing of the

• equipment.

As shown in Table 2-12,*results improved near the end of the weekend run. Note, in particular, that all values in the 3000 ppm column, except two,. are within plus or minus 5 percent of actual. The two exceptions both indicate a boron concentration of 953 ppm (68.2 percent low). Improving results with increased time under exposure is not consistent with the behavior pattern expected from radiation damage. In particular, radiation damage would not be expected to result in a pattern where both exceptions to general results indicate a boron concentration of 953 ppm.

The test was terminated whe~ a nylon pulley broke on. the rotary reaction cell assembly. This nylon pulley is internally stressed with a press fit brass bushing.

2-27

t

,{

r,. ii TABLE 2-9 BORON ANALYSIS RESULTS WITH~ SENTRY MODIFIED DIGICHEM ANALyzER IN A 1.75 x 10 R/HR RADIATION FIELD Nominal 1200 ~E!!! Boron Solution(l) Nominal 3000 ~E!!! Boron Solution(l)

Percent Percent Steps( 2) of Indicated Deviation Steps <2 > of Indicated Deviation 0.1105 N ppm From 0.1105 N ppm From NaOH Boron Nominal NaOH Boron ....!5!!inal 524 1258 4.8 1243 2983 -0.57 537 1291 7.6 1263 3031 1.0

i. 546 1310 9.2 1233 2959 -1.4 535 1284 7.0 1198 2875 -4.2

? 506 1214 1.2 1201 2882 -0.39

.l 503 1207 0.6 1272 3053 1.8 3

498 1195 -0.4 1250 3000 0 505 1212 1.0 1276 3062 2.1 528 1267 5.6 1236

• 2966 * -1.1 529 1270 5.8 *1254 3009* 0.3 536 1286 7.2 1187 2849. -5.0 539 1294 7.8 1199 2878 -4.1 538 1291 7.6 1187 2849 -5.0 529 1294 7.8 1199 2878 -4.1 538 1291 7.6 1246 2990 -0.3 529 1270 5.8 1234 2962 -1.3 531 1274 6.2 1271 3050 1.7 529 1270 5.8 538 1291 7.6 X527 1267 5.6 1232 ,2957 +2.1 a+l3.9 +34.2 +2.8 +31.1 +74.6 +1.7 20'!27.8 +68.4 _!5.6 +62.2 +149.2 +3.4 (1) The historical samples could not be found at the Georgia Tech. test facility, so boron concentration cannot be verified. The indicated concentrations were obtained by kno!% dilution of a 6000 mg/1 boron stock solution.

(2) One step* 2.17 x 10 liters. .

The analyzer was programmed to alternately.analyze the 1200 ppm and 3000 ppm boron standards.

2-28

TABLE 2-10 BORON ANALYSIS RESULTS Wl:1'3. SENl'RY MODIFIED DIGICHEM ANALyzER IN A 8.64 x 10 R/HR RADIATION FIELD Nominal 1200 ppm Boron Solution Nominal 3000 ppm Boron Solution Percent Percent Steps (1) of Indicated Deviation Steps (1) of Indicated Deviation 0.1105 N ppm From 0.1105 N ppm From NaOH Boron Nominal NaOH Boron Nominal 538 1291 7.6 1250 3000 0 540 i296 8.0 1257 3017 0.6 536 1286 7.2 1256 3014 0.5 544 1306 8.8 1288 3091 3.0 544 1306 8.8 1260 3024 0.8 545 1308 9.0 1258 3019 0.6

• - 1166 2798 -6.7 i541 1299 8.2 1248 2995 1.7 a+3.7 +9.2 +0.7 +38.1 +91.6 +2.4 2a+7.4 !18.4 !1.4 !76.2 +183.1 +4.8 (1) One step* 2.17 x 10-6 liters The analyzer was programmed to alternately analyze the 1200 ppm and 3000 ppm boron standards.

2-29

TABU: 2-11 BORON ANALYSIS RESULTS WITH~ SENTRY MODIFIED DIGICHEM ANALyzER IN A 1, 75 x 10 R/HR RADIATION FIELD BEGINNING OF A WEEK END RUN Nominal 1200 21!! Boron Solution Nominal 3000 21!!!! Boron Solution Percent Percent Steps (1) of Indicated Deviation Steps (1) of Indicated Deviation 0.1105 N ppm From 0, 1105 N ppm From NaOB Boron Nominal NaOH Boron Nominal 538 1291 7.6 1265 3036 1.2 538 1291 7.6 1260 3024 0,8 364 874 -21.2 633 1519 -49,4 490 1176 -2.0 747 1793 -40.2 457 1097 -8.6 838 2011 -33.0 458 1099 -8.4 *1000 2400* -20.0 392 941 -21.6 1213 2911 -3.0 367 881 -26.6 824 197.8 -34.1 970 -19,2 . 1452 3485 16.2 404 440 1056 -12.0 1265

• 3036. 1.2 468 1123 -6.4 1257 3017. 0.6 538 1291 7.6 783 1879 -37.4 540 1296 8.0 1257 3017 0.6 528 1267 5.6 902 2165 -27.8 538 1291 7.6 1268 3043 1.4 528 1267 5.6 1246 2990 0.3 482 1157 -3.6 383 919 -69.4 521 1250 4.2 622 1493 -50.2 574 1378 14,8 619

• 1486 -50.5 561 1346 12.2 331 794 -73.5 542 1301 8.4 759 1822 -39.3 537 1289 7.4 684 1642 -45.3 541 1298 8.2 1209 2902 -3.3 527 1265 5.4 922 2213 -26.2 608 1459 21,6 937 2249 -25.0 4_81 1154 -3.8 1274 3058 1.9 i498 1196 10,4 960 2303 25.1 a+63.8 +153.0 +7.1 +309 +743 +22.8 20~127.6 . +306.0 +14.2 +619 +1486 +45.6 (1) One step* 2,17 x 10-6 liters The analyzer was programmed to alternately analyze the 1200 ppm and 3000 ppm boron standards *

• 2-30

F TABLE 2-12 BORON ANALYSIS RESULTS WITH~ SENTRY MODIFIED DIGICHEM ANALyzER IN A 1. 75 x 10 R/HR .RADIATION FIELD NEAR END OF A WEEK END RUN Nominal 1200 EE!!! Boron Solution Nominal 3000 EE!!! Boron Solution Percent Percent f; Steps (1) of Indicated Deviation Steps (1) of Indicated Deviation t: ppm From 0.1105 N 0.1105 N ppm From

!* NaOH Boron Actual NaOH Boron Actual r

f 531 1274 6.2 1289 3094 3.1

f:* 488 1171 -2.4 1292 3101 3.4 i:* 534 1287 7.2 1283 307f2> 2.6 r 564 1354 12.8 397 953 -68.2

' 530 1272 6.0 1269 3046 1.5 t\ 471 1130 -5.8 1287 3089 3.0 542 1301 8.4 1277 3065 2.2 t

} 473 1135 -5.4 1291 3098 3.3 570 1368

• 14.0 1269 3046 1.5 tf,: 534 1282 6.8 1295 3108 3.6 482 1157 -3.6 1284 3082 2.7 537 1289* 7.4 1287 3089 3.0 531 12.74 6.2 1298 3115 3.8 480 1152 -4.0 1292 3101 3.4 f 534 1282 6.8 1283 2.0 i'. 307f2) f 564 530 1354 1272 12.8 14.3 397 1269 953 3046

-68.2 1.5 471 1130 -5.8 1287 3089 3.0 542 1301 8.4 1277 3065 2.2 473 1135 -5.4 1291 3098 3.3 570 1368 14.0 1269 3046 1.5 534 1282 6.8 1295 3108 3.6

\ 482 1157 -3.6 1284 3082 2.7 537 1289 7.4 1287 3089 3.0 1293 3103 3.4 i521 1251 7.6 1285 3083 2.8 a+34 +82 +3.5 +9.0 +21.5 +0.75 2'+68 +163 !7.0 +18.0 +43 +1.5 (1) One step* 2.17 x 10-6 liters The analyzer was programmed ~o alternately analyze the 1200 ppm and 3000 ppm boron standards.

(2) Not within 3 STD Deviations of the mean, thus are not included in the calculations.

2-31

There was 2.7 x 10 7 rads exposure on the test equipment at this time. Testing sub-sequently performed with the Ionics equipment indicates that extensive cracking will develop on this nylon pulley with 10 7 rads exposure. While the pulley on the Ionics s~pplied equipment did not fail after operation at the 10 7 rads exposure level, its appearence was such that it could have easily failed. ~

Subsequent examination of the equipment performed by Sentry indicated that the nylon and Kel-F parts that were -not replaced had become severely embrittle.d *. This was expected, based on the total exposure levels involved. All metal and electronic components were fully operational.

TEST RESULTS FROM THE PRODUCTION MODEL DIGICHEM ANALYZER The DigiChem analyses results and laboratory analyses results for standard boron solutions are presented in Table 2-13. There is reasonable agreement between these results, however, there is more variation than was seen in previous testing per-formed with the DigiChem analyzer (Table 2-15). ~esults of this other work indicate that it should be possible to obtain a precision of:!:. one percent with the

• *.

• I DigiChem analyzer. Note in Table 2-13 that part of the t1trat1ons were perfomed with 0.5 N NaOH and part with 0.1 N NaOH solutions. A comparis.on of the data indicate that essentially equivalent results were achieved with either normality.

The analyses results for matrix solutions containing simulated fission product species and caustic solutions are presented in Table 2-14. These data indicate that the concentrations of fission product species.expected follo~ing an accident will not interfere with* boron analyses results.

The data also indicate that.boron analyses results will ~ot be affected by the caustic added to the primary coolant when the containment sprays are activated during a LOCA event. The limited testing perform~d concerning the effect of lithium alone on boron analyses results indicates this addition had no apparent effect on accuracy or precision.

2-32

he is

'UJIU: 2-13

,. STANDARD BORON AND BLANK ANALYSES RESULT

' WITH THE PRODUCTION M:>DEL DIGICl!EM ANAI.yzER

~:;

Nanimnal Nmiber Ke~ 2 ) Laboratory  % Error

!Se Boron Kean-Lab of Titrant Boron Analyst~)

S11111ple mgl l Analyses Normality .5L.!. Results Lab t-

~

Staodard Standard 6000 6000 4

6 0.5 o.5 6124 5962 6108 6108 1.90

-2.39

he ,,

, Standard 6000 s 0.1 5914 6108 -3.18

[

V Standard 2000 6 o.5 2082 2017 3.22 t

Standard 2000 4 0.1 2102 2017 4.21 Standard 1000 6 o.5 1002 1025 2.24 Standard 1000 3 o.5 1048 1025 2.24 Standard lOl>o 1 0.1 1065 1025 3.90 t Standard Stcndard I 60 60 6

6 0.5 0.1 57.25 59.05 61

-1.58 6.15 f Blank (1) 0 3 0.5 0.40 Cl) f Blmik Blank (1) 0 0

5 7

o.5 0.1

-1.12

-3.57 Cl) Deionized water (2) Analyses results vith the Digi.chem Analyzer (3) Ae determined by e&Wltic titration of the boron-mannitol CDllllplez l

2-33

. . . . . . . - ~. . . . * * * : , , - - . . - - . - . , . . ......... , . , . . , * * *~ . ,.. .. "T" . * . , * ..., _ _ , . ~ ** ~ - ... _ . . ~ - .:.~-. *T"* .. -"'-**~.,.,...~-.,...- :"'~*"'-' *.,r*--* ..c* ... ~""'7.........,.,~..........., . . .;?.":"~... !'P-."~- ~ --*:--* .~,.,... *.,. , ...... - ' .. -~~-_,.- ....., ....... -----....-*. ..__ .-....

TABLE 2-14 MATRIX SOLUTION ANALYSES RESULTS WITH THE PRODUCTION MODEL DIGICHEM ANALyzER Naninal Nmber Me~l) Laboratory % Error Boron of Titrant Boron

• Analysu Kean-Lab s-l!le mill Analyses Normality .5L!.. Results ) ~

Matrir-1 60 3 0.5 59.92 65 -7.82 Katrir-1 60 3 0.1 58.79 65 -9.55 Matrir-2 2000 3 0.5 2082 2022 2.97 Matrir-2 2000 3 0.1 2079 2022  :?.82 Matrir-3 6000 3 0.5 5897 6101 -3.34 Matrir-3 6000 3 0.1 5840 6101 -4.28 Matrir-4 6000 3 0.5 5995 6136 -2.30 Matrir-4 6000 3 0.1 5927. 6136 -3.41 Matrir-8 600 3 0.5 608.4 624 C Matrir-7 60 3 0.5 56.73 66 -14.0 Matrir-7

• 60 3 0.1 59. 76 66 -9.59 c Matrir-5 0 3 0.5 -3.61 E Matrir-5 0 5 0.1 -0.61 Matrir-6 0 7 0.5 -2.92 Matrir-6

• 0 3 0.1 -0.46 Boran + 0.4! 600 3 0.5 663:8 666 -0.33 I

NaaH Baran+ 0-4! 600 3 0.1 653.8 666 -1.83 E

NaaH Baran+ 0.4N 6000 3 0.5 5916 6056 -2.31 NaaH Boran + 0.4!! 600 3 0.5 626.8 666 -5.88 NaaH.

Boran+ 0.4! 600 3 0.1 624.4 666 -6.25 (1) Analyaes results with the Digichem Analyzer (2) Aa determined by cauatic titration of the baran-,nannital canplex 7

• 2-34

DISCUSSION OF RESULTS AND CONCLUSIONS DISCUSSION OF RESULTS Test results from the irradiation experiments clearly indicate that the critical components in the production model the DigiChem analyzer with respect to radiation damage are as follows:

• Photon coupled interrupter module (H21A3). This is a light activated speed control system for the rotary reaction cell.

• Solid state relay* for a solenoid actuated valve on the rotary reaction* cell.

• Photon coupled solid state limit switch (MCA8) in the sample addition module.

• Two nylon pulleys used to drive the rotary reaction cell

• 4 Threshold damage* level* for photon coupled interrupter module H21A3 is between 10 and 10 5 rads. I For *module MCA8 it is about 10 4 rads. Total radiation exposure for H21A3 and MCA8 could be at the 10 4 rad level in a accident condition, dependant on the overall design and operating philosophy of the sampling system. No conclusions 4

can be drawn that MCA8 will withstand 10 rads exposure since only one module was 4

tested at this level. The module suffered minor damage with 10 rads exposure, however, it remained operational. Threshold level for the solid relay (total 7

failure at 10 rads) was not determined since it is easier to locate this relay outside the radiation zone than it would be to determine the threshold damage level. The nylon_pulleys would almost* certainly remain operational at 10 6 rads exposure, however, should be replaced with metal pulleys since this change can be accomplished with little difficulty.

The solid-state components listed above that can be damaged by radiation have been replaced with mechanical switches in the modifications made to the DigiChem analyzer by Sentry. The nylon pulleys were replaced by Sentry with stainless steel pulleys as a consequence of the irradiation experiments performed at Georgia Tech.

2-35

Teflon tubing can withstand 10 6 rads exposure while the _heavier teflon components l

remained operational at 10 7 rads exposure. It is not expected that radiation b

damage would preclude the use of teflon components in a DigiChem analyzer during a 3

post-accident condition. However, the change made by Sentry to eliminate teflon in favor of more radiation resistant materials will add a higher degree..._of conservatism to the system, For example, the need for flushing the sample lines of r

highly radioactive coolant within a specified time period becomes less critical t

with the Sentry system since the teflon has been replaced with more radiation

\

resistant material.

Concerning pH probes the data indicate that high radiation levels (10 6 R/hr) will decrease the indicated pH by about 0.1 pH units for an internal reference probe.

ll Indicated pH will increase by about 0.1 or 0.2 pH units for external reference probes in a high radiation field. An initial effect is noted at 103 R/hr. The increase in pH is immediate. The effect is fully reversible when the radiation source is removed. The DigiChem system has an external reference pH probe.

l.

The shift in pH resulting from radiation should have a slight effect on accuracy of analyses with the DigiChem analyzer, however, the effect will not be significant as concerns post-accident requirements. During normal operating' c_onditions, the system will be titrating from pH 5.5 to 8.5 to determine boron co~centration.

Under high radiation conditions the system will still titrate from an indicated pH 5.5 to pH 8.5. However, in reality it may be titrating from say a pH of 5.3 to 8.3 i

because of the radiation induced shift in pH.

The erratic results noted in the high radiation level testing (Table 2-11) occurrec r

because of an electronic "loophole" created by the high radiation field. This r

resulted in the leakage of current, causing erratic pH electrode behavior. A t

design change has been made which includes a driven shield concept that will r

prevent radiation induced leakage in the cable shield to the pH electrode. This

(

driven shield will be a standard feature in all DigiChem analyzers. It should be emphasized, however, that the system tested without the driven shield operated satisfactorily at radiation levels anticipated under post-accident conditions.

Increased reliability can be anticipated with the addition of the driven shield.

Results of testing performed with the production model DigiChem analyzer are not equivalent to results previously achieved with this instrument. Compare for example, the data in Table 2-13 with the data from previous testing presented in

• Table 2-15. The difference between these results is not understood.

2-36 One posssi-

bility is that there may have.been some degassing of the titrant solutions or of the sample itself as system pressure is reduced when the plungers of the sample or titrant burettes are withdrawn to replenish system volumes.* *introduction* of n bubbles adds to the error because these bubbles are measured and computed as liquid vofome.

f

.; . It is apparent from the data presented in Table 2-15 that there is very little f' scatter to the results. Virtually all the results are low by about the same per-centage value. ~is pattern has appeared again and again, with some results con-sistently low ~nd others consistently high by some small percentage value.

Generally, the an~lyses results have been computed based on normality of the caustic solution used for titration. From examination of the data, it would appear that some improvment in accuracy can be achieved if results were compared directly to results achieved with a known boron standard. The computer can be progratmned to provide for such a comparision.

If the DigiChem analyzer is used for normal operation or post-accident analyses, it should be noted that the primary coolant must be degassed to a low level prior to introduction of the sample to the sample burette. With high concentrations of gas present,_as can occur;in an accident involving core damage, bubbles will be produced in the sample*stream when system pressure is reduced as the plunger in the sample burette i~ withdrawn. This would result in values which are lower than

.actual. The error ~ould be proportional.to the ratio of gas volume to liquid volume in the sample stream.

Another feature that would be desirable though not mandatory for this system, would be to inject a small volume of water to the ro_tary reaction assembly prior to injecting the sample itself. The reason for this is to dilute the sample itmnediately so that ther~- is less tendancy for radioactive iodine to escape from solution during 1_>os.t-accident conditions. It would also be necessary to inject a small volume of water after the sample addition to properly flush the sample tip.

The system can be progratmned to provide this sample addition sequence.

_The overall advantages of using the DigiChem analyzer for the boron determinations during accident conditions are as follows:

• All operations can be performed remotely. The exposure involved in determining boron concentration would approach zero.

• Sample volume requirements are on the order of 1-2 ml per analysis, thus shielding requirements would be minimal.

2-37

TABLE 2-15 BORON REPRODUCIBILITY RESULTS 2000 ppm STANDARD Analysis ppm  % Analysis ppm  %

Results DEV Error Results DEV Error 1975 -25 -1. 25 1975 -25 -1.25 1977 -23 -1.15 1975 -25 -1.25 1973 -27 -1.35 1975 -25 -1.25 1974 -26 -1. 3 1975 -25 -1.25 1074 -26 -1. 3 1974 -26 -1. 3 1977 -23 -1.15 1975 -25 -1. 25 1977 -23 -0.15

• 1977 -23 -1.15 1975 -25 -1. 25 1977 *-23 -1.15 1975 -25 -1. 25 1998 ,-2 I -0.l l 0 1973 -27 -1. 35 2000 0 1973 -27 -1. 35 1974 .  :

-26. -1. 3 1973 -27 -1.35 1978 -22 -1.l 1979 -21 -1.05 1981 -19 -0.95 1976 -24 -0.2 1977 -23. -1.15 1973 -27 -1. 35 1977 -23 -1.15 1974 -26 -1.3 1977 -23 -1.15 1973 -27 -1.35 1978 -22 -1. l 1973 -27 -1. 35 1981 -19 -0.95 1974 -26 -1.3 1982 *-18 -0.9 1975 -25 -1. 25 . 1977 -23 -1.15 1974 -26 -1.3 1978 -22 -1.1 1974 -26 -1. 3 1978 -22 -1.1 2002 2 0.1 1977 -23 -1.15 1974 -26 -1. 3 1981 -19 -0.95 1974 -26* -1.3 x1975 21. 9 1.1 1979 21. 2 0.6 cr+5.8 +8.9 +0.4 +6.5 +6.-5 +0.3 2CJ+ll. 6 +17.8 +0.8 +13.0 +13.0 +0. 6 Average error= -1.05%

Maximum error= -1.35%

,. 2-38 I

. *----*~-----.. --~--~-,......,~*.*=-.-,...,~--,.,.z,;.c_f',;?i"!:',~."J;l,> ..~~~. . ,~,.;~*~

• Analyses results can be achieved within 10 minutes after the sample line is purged to obtain a representative sample.

• Though not sealed gas tight, there would be little tendency for release of gaseous activity to the atomsphere. This would be-par-ticularly true if the sample addition sequence is changed to add water prior to addition of the sample.

The disadvantages of using the DigiChem analyzers tmder post-accident conditions are as follows:

• Waste solutions cannot be pumped back to the primary system since chemicals are added to the system in the analysis procedure.

• A small pumping system must be provided to pump the gravity drain waste solutions from the* analyzer to a waste disposal facility if the waste disposal system is above the level of the analyzer. Most plants using ~his equipment have gravity drain collection tanks.

CONCLUSIONS AND RECOMMENDATIONS

• The Sentry modified DigiChem analyzer is acceptable for use to determine' b9ron concentration under post-accident conditions.

Concerning' its use for normal operations, the accuracy is probably acceptable.

I *

• If the.D1giChem analyzer or Sentry modified system is used for boron determination during normal operations, results should be compared to known boron standards rather than computed solely on normality of the titrant solution. The system can be programmed to provide for such comparison.

• The primary coolant must be degassed to a low lever prior to introduction of the sample to the sample burette. This is to prevent introduction of gas bubbles in the sample stream.

• It would be desirable (but not necessary) to program the system to add *a ~all volume of water to the rotary reaction assembly prior to addition of the sample. This will further reduce an existing low potential for release of radioactive gas to the environment.

2-39

I *

~

~ ..

Section 3

SUMMARY

OF RESULTS - WESTINGHOUSE ANALYZER The Westinghouse ~ark V boron analyzer performed well, both at high radiation levels (3.45 x 10 5 R/hr) and under steady state conditions in the absence of radiation, There was an increase in fissioning count rate resulting from high level irradiation, however, the effect of this increase on accuracy of the boron analyses is not signific~nt. With operation at radiation levels anticipated under NRC post-accident reference conditions, the accuracy achievable is equal to1 or better than other methods of boron analyses available for use during post-accident conditions.

f The system can be used to monitor boron concentration during normal power operations. Accuracy expected at intermediate or high level boron concentrations should be suitable for normal requirements. Determination of low-level boron concentrations would ptobably require a 500 or 1000 second count rate period.

No problems of any ~ind were experienced in operation during a test period of about 15 days total, This is a relatively short period compared with duty in a power plant, however, we believe the analyzer will work for a long time in a power plant.

3-1

BACKGROUND INFORMATION TEST PURPOSE Testing was performed to determine if the prototype unit of the. Westinghouse Mark V boron analyzer would suffer radiation damage or reduction in accuracy when operated at radiation levels anticipated under post-accident conditions. Testing was also performed to establish reliability of the equipment when operated under normal conditions.

SYSTEM DESCRIPTION General

• I The Boron Concentration Monitoring System (BCMS) Mark Vis an electronuclear system that continuously measures the boron content in the primary'c;oolant of a pressurized water reactor (PWR) power plant and digitally displays the results in parts total boron per million parts of water (ppm). ln*a shielded tank, a sample of the primary coolant is positioned between a neutron source and a fission chamber. Neutrons originating at the source are thermal'ized, then pass through the

.boron solution (where some are absorbed) and impinge upon the. enriched uranium in the fission chamber. Fissioning occurs with the release of charged particles, resulting in voltage spikes in the fission detector that are translated into ppm boron. The charged particle population is directly proportional to the fissioning process~ and therefore proportional to the neutron population. This provides a measure of the boron concentration in the water since the fissioning rate and resulting charged particle population varies inversely as does the neutron absorp-tion characteristics of the primary coolant. The charged particle count rate is translated into ppm boron by an algorithm prog.rammed into the system's microcomputer which accounts for non-linear response and for temperature correction. Calibration is performed by determining the count rate for three known concentrations of boron solutions and entering this .information into the computer unit. The system is self-calibrating at this point. The microcomputer transmits this boron concentration data to local or remote displays.

The BCMS Mark Vis comprised of three major assemblies: the sampler tank, which 3-2

detects the charged particle count rate and coolant temperature; the electronic processor enclosure~ which contains the processing control and monitoring electronics for most of the system; and the remote display, which enables a remote indication of boron concentration. The sampler tank is shown in Figure .'.!-I, and the overall system is shown in Figure 3-2. Test equipment evaluated in this work did not include the use of a remote display unit. This equipment is not required for system operation. Also provide4 are an interrupt line output and serial data output which permit the processor enclosure to transmit data to the plant computer.

General descripti_ons of the three main BCMS Mark V assemblies are given below.

k V ted Sampler Tank Assembly o

The sampler tank assembly is a stainless steel cyinder, approximately 15.12 inches (38.4 cm) in diameter, 19 inches (48.3 cm) high, and weighing 100 pounds (45.3 kg),

which is secured to the mounting platform by four hold-down clips. The cylinder contains polyethylene* which functions as a neutron shield and moderator. The unit has two cavities, one neut;ron source well and one annulus assembly containing the fission detector. The neutron well is 1 inch in diameter by 7 inches deep in a em high density polyethylene epoxy resin. The neutron source is provided by one curie of americium/beryllium (Am-Be). The fission detector has 2 grams of enriched E uranium. The Am-Be*sour~e is _in the center of the tank in a vertical cavity which d is inserted on the end of a polyethylene rod. Surrounding the fission detector is a one liter stainles; steel annulus assembly. Coolant flow to and from this tank  ::t e is provided by 0.5 inch tubes with Swagelok fittings for connection to the plant ic piping.

, l The sampler tank assembly receives reactor coolant solutions from a sampling

o location such as the letdown heat exchanger or Boron Therm.al Regeneration System (BTRS). Reactor coolant samples are routed to the input port of the sampler tank.

A thermocouple inserted through the cover plate extends 8 inches into the polyethylene materi~l. Sample flow through the unit is determined by the pressure C drop between the inlet and outlet tube connections.

a Two electrical signals are derived from the sampler tank assembly: (1.) fission f

count rate from the fissioning detector and (2) thermocouple potential (in millivolts). Detector pulses are applied to the preamplifier in the processor enclosure via a coaxial cable attached to the detector. The thermocouple signal is applied to a digital thermometer in the processor enclosure via thermocouple wire.

3-3

---

15.IZ - - - - - - - - -

---

14,0APPROX, I

1' I I I I I l I I I I II Cl Aa-le I iSOUltCI

'-J IIOO I I I I I L_.,.

I .18,I I I I I I I WATEII I i

'-----J POLYETHYLENE 14.16 FIGURE 3-1

@) SAMPLER TANK ASSEMBLY 3-4

.. '. - ., ' ,., .. - *~ ... '" **. *.* ~.' ------~=-*-:*-*.

• -~

J 1 1'/

l j

i

• • j i

i j

SAMPLER

-PROCESSOR ASSEMBLY 1082010001

• J ASSEMBLY

• l 6082096001 PREAMP Al REMOTE DISPLAY 607500! ASSEMBLY DETECTOR 6085D68GOI Pl NOT REQUIRED COMPUTER /PRINTER DATA OUTPUT w

I 1.11 1111 VAC CABLE ASSEMBLY INPUT l08IID78GOI FIGURE 3-2 WESTINGHOUSE MARK V BORON ANALYZER ,1

Processor Enclosure The processor enclosure is a wall-mounted, louvered, NEMA 12 enclosure containing the compone_pts that control operation of the BCMS, Mark V analyzer. Operator

~ controls and indicators are contained on the control panel which is a~cessed by opening the hinged front door of the cabinet. Also contained in the process enclosure is the preamplifier with bias control to discriminate against detection of noise. For maintenance and troubleshooting purposes, the control panel is hinged to allow access to the microcomputer power supplies, preamplifier, card cage, terminal boards, and test point assemblies.

The electronic processor enclosure may be located hundreds of cable feet away from the sampler tank provided the preamp is removed and located within 20 feet of the tank. It receives the fission count rate and temperature from the sampler tank assembly, processes it, displays the calculated boron concentration (in measure mode) on the local display, and serial~y transmits the concentration data to.the remote display assembly and plant computer. The electronic processor enclosure contains a microcomputer made up of a single-board central processing

. , unit (CPU) board, complementary metal-oxide semiconductor random-access memory (CMOS RAM) board with battery backup: and input/output (I/0) expansion*board.*

Remote Display Assembly

)

The remote display assembly displays the boron concentration in ppm at a location

i (usually in the c_ontrol room). remote from the processor enclosure. Measuring t approximately 7.75 inches wide, 4.5 inches high, 9.62 inches deep, and weighing 10 i pounds, the unit can be installed up to 1000 feet from the processor enclosure.

t

[" Concentration data calculated by the processor enclosure is transmitted serially 2 over a twisted shielded pair. The remote display assembly contains the circuits

that receive, decode, and present the data on a four-digit light-emit"ting-diode (LED) display.

E e

3-6

a:.::::::;: _.......t- ..... _:'.....- . . -.. --- ,=

TEST DESCRIPTION GENERAL The irradiation testing was performed at the hot cell test facility at Georgia Tech. Testing to investigate reliability characteristics of the Mark V boron analyzer was performed at this same location. Testing to determine reproducibility of the boron analyzer was also performed *at Georgia Tech.

IRRADIATION TESTING The radiation source was provided by eight, 8 x 13 inch frame assemblies containing a total of 53,000 curies 60co (6,600 curies per frame). Radiation source from the one liter primary coolant sample tank under accident conditions will be around 40,000 curies for reactor coolant with activity of 4 Ci/cc. Radiation levels were increased or decreased ,by placing one or more of these frame assemblies around the sampler tank assembly as shown in Figure 3-3. The radiation level for maximum I

• radiation testing was.measured by ~lacing a dosimeter.at the detector location in a second sampler tank assembly. Geometry was held constant for the second sampler tank and the tested sampler tank assembly in the irradiation testing performed. A second tank was required to determine radiation dosage because the dosimeter was placed in the position that would have been occupied by the detector tube during irradiation testing. Testing was performed at a maximum radiation level of 3.45 x 5

10 R/hr, determined by ~osimetry. The level of 3.45 x 10 5 R/hr required the use of the eight frame radiation sources that were available. A radiation level with the second sampler was determined for only this one configuration because most of the irradiation experiments were performed at the maximum achievable level.

Estimated radiation levels for the Westinghouse boron analyzer for reactor coolant with an activity of 4Ci/cc are around the maximum radiation levels achieved in this test work.

The fission count rate was determined as a function of boron concentration and/or radiation level in the sampler assembly. Count rate was determined in the absence of radiation to determine a base level, *followed by testing with exposure to high and intermediate radiation levels *. The 60co frames were added or removed to change 3-7

rl i

t:

E GEOMETRY OF IRRADIATION ASSEMBLY FOR @) MARK 7. SAMPLER TANK ASSEMBLY S3,000 Cl TOTAL RADIATION SOURCE E

(

I I

I I

I I

I

~

__.I FIGURE 3-3 3-8 I

-.-......_..._~*-*"*n*~* .-*.--~,- .,.~..-.~._.~--*-***------- *--* . . ~ *., ... ..,..,--.--.**-* _...,.,,.,...,.~--*.

the radiation levels. Results are based on the fi~sioning rate rather than ppm readout because the main objective of this test was to determine the effect of high radia.tion levels on the detector equipment. Evaluation of this equipment can best b~_performed by monitoring the fission rate during testing under irradiation.

All testing involving radiation exposure was performed in a hot cell under no-flow conditions. The tank sampler tank was rinsed three times with the reference boron solution when concentration was changed.

3-9

TEST RESULTS IRRADIATION TEST RESULTS Prior to ins~allation of the sampler assembly in the hot cell, the boron analyzer was operated overnight with pure water in the sampler tank.* An average fissioning rate frequency of 476.83 counts/second was determined for the 30,000 second over-night run. This compares to an average count rate of 476.16 counts/second for a series of eighteen, 100 second count periods made prior to starting the irradiation tests. This series of 100 second count periods varied from a low of 473.56 to a high of 479.28 counts/second. The data are presented in Table 3-1.

Initial radiation under testing was performed with pure water in th~ sample tank.

The count rate increased by almost 3 percent from an averag.e*of 4?4.96 counts/second to an average of 487.96 counts/second when expo$ed to a radiation 5

level of 3.45 x 10 R/hr. Moving the connector cable so that is was further removed from the radiation source had no effect on count rate based on the average of 487.44 determined for six, 100 second count periods * . The count rate returned to the base level obtained in the preirradiation testing when all radiation sources were removed. Data obtained from the irradiation testing performed with pure water are presented in Table 3-2.

Testing performed with 5140 ppm boron solution in the sampler tank showed the same behavior as was observed with irradiation testing performed with pure water in the tank. In the absence of radiation, the count rate for the 5140 ppm boron solution was 125.98 counts/second for a 100 second count period. This increased by about 3 percent to 129.49 counts/second when exposed t~ a radiation level of 3.45 x 10 5 R/hr (eight 60 co frames). Four 60 co frames were removed leaving a total of four 60 Co frames around the sampler assembly. This reduced the count rate from 129.49 counts/second to an average of 128.19 counts/second or about 1.5 to 2 percent above the base level obtained in the absence of radiation. The count rate returned to the original base level when all radiation sources were removed. Data obtained with the irradiation testing performed with 5140 ppm boron solution are presented in Table 3-3.

3-10

• .i;.._ - - ~ * - * * .,s-2 . . . . :-c-v - ~ *

• _... ~*.*--t-*

• 7-'-!;! *- , ** -

• TABLE 3-1 BASE LEVEL FISSIONING COUNT RATE FOR THE WESTINGHOUSE MARK V BORON ANALYZER (PURE WATER RESULTS)

Run Counts Time Per Sec. Sec *

.30,000 476.83 100 474.91 100 473.83 100 475.86 100 475. 71 100 473.56

• 100 476.89 100 474.64 100 479.28 100 478.11 100 475.82 100 476.57 ,1 100 475.43 100 474.84 100 477 .36 100 479.11 100 478.61 100 475.53

.100 474.95 X = 476.17 a = +l.73 2a = ~3.46 3-11

L

l i

TABLE 3-2 t

0 EFFECT OF IRRADIATION WITH PURE WATER IN THE WESTINGHOUSE MARK V BORON ANALYZER*

(

r Zero Radiation 3.45'x 10 5 R/hr 3.45 X 10 5 R/hr(l)

Run Counts Run Counts Run Counts Time Per Temp. Time Per Temp. Time Per Temp.

Sec. Sec. DC Sec. Sec. DC Sec. Sec. DC

1 600< 2 > 475.69 100 490.93 25 1 100. 491.67 25 100 487 .8 3 26 100 474.20 23 100 486.93 25 100 488.47 26 100 476.45 23 100 487.93 25 100 ,485~49 26 100 475.02 23 100 487.07 25 100 *490.71 26 100 474.04 23 100 485.22 26 100 488.61 26 m

100 474.87 23 100 485,98 26 100 483,51 27 X 475 488 487 2

a + 0.95 2,45 2.55 2a + 1,90 4,90 5.10

.B (1) The cable which connects the sampler tank to the electronic processor enclosure was moved further away from the radiation source for this lC test sequence. This was to determine if count rate is affected by high t radiation level exposure of the cable, (2) Not included in standard deviation.

w, e

.r ir.

.e

, i.

• I I

l 3-12 i

I I

~

Testing was also performed with 2570 ppm boron solution in the sample tank. The s*ame behavior was observed as was noted with the pure water and 2570 ppm boron solutions. The count rate increased about 3 percent from 206._73 to 212.29 counts/second. These data are presented in Table 3-4.

Further testing was performed to determine if exposure of the connector cable to very high radiation levels would affect the count rate. The connector cable was clamped between two 60 co frames (1/2 inch gap) to obtain exposure level estimated to be in the range *of 10 6 to 10 7 R/hr. Radiation measurements made in connection with other irradiation experiments performed indicate that radiation levels between the two frames are* on the order of 10 6 R/hr with 3 to 4 inches gap between the two frames. Since the actual gap was about 1/2 inch, the radiation level would be well 6

over 10 R/hr. No effect'on count rate was noted from this radiation level based on count rates of 206.58, 209.35, 206~96 and 206.35 counts/second over four,- 100 second count periods.* The base level count rate for this system (2570 ppm boron) in the absence of radiation. was.206.73 counts/second. These data are consistent with the data presented in Table 3-2.

The testing performed indicates that the increase in count rate noted with the high radiation levels is an' instantaneous function of radiation levels. That is, the count rate changes as soon as the radiation level increases or decreases. There is I

no memory effect, nor .is there any indication of permanent damage suffered based on about 2 x 10 7 rads total exposure to the sampler assembly. This is equivalent to over 25 hours2.893519e-4 days <br />0.00694 hours <br />4.133598e-5 weeks <br />9.5125e-6 months <br /> operation with an activity level .of 4 Ci/cc in the primary coolant.

RELIABILITY TEST RESULTS After the irradiation testing was complete, the Mark V boron analyzer was operated under steady state conditions for a period of 13 days. This was done in the absence of radiation _with 2570 ppm o_f boron in the sample tank. Initially, data were taken every half hour during the course of an eight hour day. Later, the data were taken on an hourly basis or sometimes on a daily basis. This data is shown in Table 3-5.

The Westinghouse equipment operated very well during the reliability testing.

There were no outages or system malfunctions of any kind during this test perio'd.

Unfortunately, the data recorded in Table 3-5 represent 1 second count periods rather than the 100 second or 1000 second data intended. However. since the l

standard deviation is proportional to~-, n being the number of samples, we can n

infer a standard deviation for 100 second and 1000 second counting intervals of 1.02 and 0.32 respectively.

3-13

TABLE 3-3 EFFECT OF IRRADIATION WITH 5140 ppm IN THE WESTINGHOUSE MARK V BORON ANALYZER Zero Radiation 3.45 x 10 5 R/hr 2 X 10 5 R/hr(l)

Run Counts Run Counts Run Counts Time Per Temp. Time Per Temp. Time Per Temp.

Sec. Sec. oc Sec. Sec. oc Sec. Sec. oc 600< 2 > 125.97 32 100 130.20 29 100 127.79 30 100 129.87 30 *100 *

• 128.85 31 J

100 125.16 32 100 128.90 30 100 127.68 31 100 124.19 32 100 130.51 30 100 129.86 31 100 126.96 32 100 129.38 30 100 127.27 31 100 127.02 32 100 129.19 30 100 127.92 31-32 100 125.63 32 100 128.99 30 100 127.34 32 100 125.68 32 100 129.83 31 100 128.39 32 100 125.58 32 100 130.25 31 100 128.26 32

• X 126 129 128 a + 0.99 1.62 1. 72 21 + 1.98 3.24 3.44 (1) Estimated radiation level of 2 x 10 5R/hr (2) Not included in standard deviation *

• 3-14 I

. . *--=*-,,.*~-.... -=.-*, .... - - ............... ..., .. *- ---- -* ---"~---*"**** -*** .. -*~-- - ~-,- ... ---. ---- ., .............

---~-- , _*,.*.-... ~ -~-~ .* * ~~**.*. \-*--*-** **~--i-* __ *.._._;-.,,, ..._*.=..*v,.... ..,..**, .. -.->,,* .;.*.*,.._ * ....:-* ,_.,__ -*~-,-J,;~---*- *.;i'l.~.*.:,...:,a....i::...~,...._~--..,...-,.-.,.;;i.~~.......~.,+/-:SPc'>!IIIIJC_*. ::,z TABLE 3-4 EFFECT OF IRRADIATION WITH 2570 PPM BORON IN THE WESTINGHOUSE MARK V BORON ANALYZER Zero Rad,iation 3.45 x 105 R/hr Run Counts Run Counts Time Per Temp. Time* Per Temp.

Sec. Sec. oc Sec. Sec. oc 600(1) 207.02 32 600(1) 213.20 32 100 212.in 32 100 212.72 32 100 206.87

• 32 100 213.28 32 100 205.97 32 100 211. 99 32 100 I 205 *.61 32 100 213. 25 32 X 206 213 a + 0.65 0.49 2a + 1.30 0.98 (1) Not included in standard deviation.

3-15

• ~-

C TABLE 3-S STEADY STATE OPERATION FOR l SECOND COUNT PERIODS WITH 2570 PPM BORON IN THE WESTINGHOUSE HARK V ANALYZER (BACKGROUND RADIATION) 8-28-81 (l) 8-31-81 9-1-81 9-2-81 9-3-81 9-4-81 Count* Countl Count1 Count1 Count* Counts.

1 Per Temp. Per Temp. Per Tnp. Per Temp. Per Temp. Per Temp.

~ ~ ~ ....:L ~ ~ ~ ~ ~ ~ ~ ~

J 211 23 i 200 23 226 24 221 24 205 23 206 24 i 199 23 204 24 202 23 204 24 228 23 202 23 197 23 217 23 197 24 215 23 191 24 190 24 204 24 214 24 219 23 209 24 213 24 212 24 226 24 217 23 218 24 222 24 198 24 215 24 213 24 207 23 203 23 198 24 209 24 205 24 200 24 214 23 207 24 206 24 215 24 221 24 196 24 201 23 218 24 211 24 214 24 217 24 195 24 210 23 213 24 202 24 200 2.4 203 *24 210 24 219 24 200 24 199 24 191 24 197 24 199 24 215 24 198 24 189 24 204 24 212 24

! 233 24 201 23 205 24 207 24 202 24 209 24 187 24 209 24 217 24 217 24 , 196 24 191 24 i 193 24 214 24 193 24 190 24 194 24 213 24 212 24 212 24 186 24 206 24 196 24 215 24 c 208 24 208 24 180 24 211 24 199 24 205 24 206 24 217 24 190 24 209 24 211 24 193 24 I

!otal Count** 102

(

a .* 206.16 10.16 1 -

(l) The 1y1tea va1 operated, however, there va1 no data taken over the weekend.

i..

e

~

.E i.

,e 3-16

¥--'<+t' *~~~-.:._.-.~:t1!'>-.'.'f-~~-r.~.v~~:~~-**

DISCUSSION OF RESULTS AND CONCLUSIONS DISCUSSION Results of 'the irradiation tests indicate that the Westinghouse Mark V boron analyzer would perform well under post-accident conditions. Count rate increases, and thus the ppm boron readout decreases with increasing radiations levels, however, the effect is a predictable one. For exposure levels in the range of 2 x 10 5 R/hr and 3.45 x 10 5 R/hr (maximum achievable radiation level) the fissioning count rate increases by about 1.5 percent and 3 percent, respectively. Linear extrapolation of this data indicates that the fissioning count rate would increase by about 5 percent for a !ad,iation field of 5 x 10 5 R/hr. Extrapolation is based on results of other irradiation tests which indicate a linear relationship to radiation levels of 7.1 x 10 5 R/hr. A radiation level of 5 x 10 5 R/hr is antici-pated in the Westingh<_>u.se Mark V analyzer with a primary coolant activity of 10 Ci per cc.

An increase_ in cotint rate resulting from high radiation levels will not give an equivalent percent decrease in apparent boron concentration. The change in boron indication will be slightly less than the percent change in count rate. Even assuming a line_ar relationship between change and count rate and decrease in indicated boron concentration, the accuracy of this instrument is equivalent to~ or better than the accuracy that can be achieved with other methods.of on-line or wet-chemical analyses available for use during accident conditions. Consequently, no corrective factor would need be applied to results of this analyzer during operation in a high radiation environment.

The temperature correction system was not operated during this work. However, temperature was not a factor in the results since temperature did not vary by more than a few degrees in any one test. The intent was to determine the r'elative change that may result from high radiation levels rather than measure absolute values. Testing performed with the Combustion Engineering Boronometer indicate that a 5-10 degree change in temperature has no significant effect on count rate.

It is of interest that the increase in count rate resulting from irradiation effects is essentially a constant (as percent of count rate) 3-17

• It is of interest that the increase* in count rate resulting from irradiation effects is essentially a constant (as percent of count rate) for the three con-ditions tes~ed (pure water, 2570 and 5140 ppm boron).

The data indicat.e that the standard deviation for boron concentration is acceptable*

with respect to post-accident or normal conditions. High radiation levels have no significant effect on deviation as indicated below:

Boron Standard Deviation Standard Deviatio~

Concentration No Radiation 1go Sec. Count m&/1 100 Sec Count 1000 Sec Count 3.45 X 10 R/hr* 2 x 10 5 R/hr X a X a X a X a 0 476 1.73 0 475 0.95 488 2.45 0 487 2.55 2500 206 0.65 213 0.49 2500 206 1.02 206 0.32 5000 126 0.99 129 0.81 1_28' 0.86 CONCLUSIONS AND RECOMMENDATIONS

• The *Westinghouse Mark V boron analyzer is acceptable .for use under post-accident conditions. It should be possible to obtain an analysis within 5 or 10 minutes with this system. Concerning its use for normal power operations, the accuracy is probably accep-table.

• Count rate increases, and thus the ppm boron readout decreases with incr~asing radiation levels, however, the effect is a predictable

. one and accuracy is still quite acceptable.

• For maximum anticipated exposure levels of 5 x 10 5 R/hr (10 Ci/cc activity level), the fissioning count rate will increase by about 5 percent. This 5 percent increase in count rate will result in a small error relative to the accuracy required for post-accident conditions.

• The increase in count rate from irradiation is essentialiy a constant (as percent of count rate) for the three conditions tested (pure water, 2570 and 5140 ppm boron) *. The increased count rate does not linger when the radiation field is removed.

/

• 3-18

Section 4

SUMMARY

OF RESULTS - CE BORONOMETER The Combustion Engineering (CE) boronometer performed well, both at radiation levels of 10 6 R/hr .and under steady state conditions in the absence of radiation.

A 53,000 curie 6oCo radiation source was used to achieve the 106 R/Hr levels in the high radiation level test work. The boronometer operated at integrated dose of about 2 X 10 7 rads. This corresponds to 20 hours2.314815e-4 days <br />0.00556 hours <br />3.306878e-5 weeks <br />7.61e-6 months <br /> of operation at maximum radiation levels anticipated under radiation conditions., It is expected that the system would remain operational at higher exposure levels based on known characteristics of the system. However, prudent considerations would dictate that radiation exposure be minimized by flushing the sample vessel when the required boron concentration information has been obtained during post-accident conditions.

The system can also be:used to monitor boron concentration during normal power operations. The instrument provides readout of the fission rate of the enriched uranium in the fission chambers. Fission rate is inversely proportio~al to the boron concentration in the sample tank surrounding the neutron source. The boron concentration is derived from the fission count rate by a mathematical_ curve fitting routine performed by a microcomputer. Use of the boronometer would not eliminate the need for periodic check analyses performed using the boron-mannitol titration. However, it would provide a continuing check against sudden changes in boron concentration and would reduce exposure to personnel *

• 4-1

---

~----- -

BACKGROUND INFORMATION .

TEST PURPOSE Testing was performed to determine if the CE boronometer would suffer radiation damage or reduction in accuracy when operated at radiation levels anticipated under post-accident conditions. Testing was also performed to establish accuracy and reliability of the equipnent when operated under conditions as anticipated during normal operations. Testing was performed on* a preproduction model in the latter stages of developnent.

SYSTEM DESCRIPTION General The boronometer consists of a sampler, preamplifier and sign~l'processor. The system used in this test included a strip chart recorder. This is ~ot part of the normal equipnent package, however, its use is recommended to improve statistics and show trending. Perfonnance specifications for- these components are listed in Tables 4-1, 4-2 and 4-3. Schematic design of the sampler which contains the neutron source and fission chambers is shown in Figure 4-1. Only those components shown in this figure are in the high radiation field. Overall schematic system design is shown in Figure 4-2. Predicted delay time due to'mixing is shown in Figure 4-3.

Operation of the boronometer is based on the principle of neutron absorption by 10 B. A small flow of primary coolant containing boron passes through a tank which holds an americium-beryllium source in the center of the tank. Neutrons from this source are thennalized and pass through the primary coolant to cause fissioning of the 93 percent enriched uranium contained in the four fission chambers. Location of the fission chambers relative to the neutron source is shown in Figure 4-1. The counting rate of the fission chambers is inversely proportional to the 10B concentration in the primary coolant, due to the neutron absorption characteristics of lOB. Signals from the fission chambers (neutron detectors) are accepted by the preamplifier box which amplifies and transmits the signals to the signal processor

• 4-2

l !

TABLE 4-1 PERFORMANCE SPECIFICATIONS FOR THE BORONOMETER Neutron Detectors Four fission chamber neutron detectors l*

TheT1D.istor Contains one theT1D.istor for temperature compensation control 1 Pressure Drop 0.04 psid at 1.0 GPM, 0.01 psid at 0.5 GPM, 0.0004 psid at 0.1 GPM Construction Designed to ASME B31.l power piping code, rated at 200 psig and 250°F.

.I All wetted parts are 300 series aus*tenitic stainless steel.

Standard inlet and outlet connection are 1/2 inch, Schedule.

40 butt weld.

Volume 0.9 gallon Dimensions Approximately 12 inches in diameter and 19 inches high.

Weight Approximately 35 pounds Neutron Source 2 curie AmBe, double encapsulated, with source handling tool, DOT approved shipping container and vessel padlock.

Ambient Operating Temperature Range 40 to 250°F Finish The sampler is constructed of 300 t series stainless steel. No finish is applied. g i

4-3

TABLE 4-2 PERFORMANCE SPECIFICATIONS FOR THE BORONOMETER PREAMPLIFIER POWER REQURIED Low Voltage+ 15 VDC 100 mAmps High Voltage:- maximum 800 Volts MAXIMUM RATINGS Preamplifier Operating Temperature 122°F Pressure 70 psig Relative Humidity 95%

High Voltage + 800 Volt(

Maximum Output Signal Cable Length 500 feet

• TYPICAL CHARACTERISTICS Conversion Gain Input 800 mV/pC Rise Time each Input (maximum) 50 nSec Fall Time each Input 200 nSec_

Equivalent Noise Charge 2.5 X 10_ 15 C (rms)

High Voltage in Leakage Current (maximum) 1.4 X 10 4 Amps Enciosure All electronics are contained in a 14 gauge steel 20 X 20 X 8 inch NEMA 4 box. The box is finished in gray enamel over phosphatized surfaces.

Cabling to Signal Processor l - RG-59/u 1 - 3 conductor No. 16 AWG 1 - 8 conductor No. 16 AWG consisting of four twisted shielded pairs.

4-4

TABLE 4-3 PEFORMANCE* SPECIFICATIONS FOR THE BORONOMETER SIGNAL PROCESSOR Digital Displays Sample Temperature - °F Detector Count Rate - counts/second Boron Concentration - ppm natural Boron Analog Outputs One of the following:

4-20 ma into 0-600 ohms 1-5 ma

• 1 into 0-2400 ohms 10-50 ma into 0-

'-l 1 200 ohms Oto 10 VDC into 500 ohms Full scale for the above signals can be switched to either 3,000 or 6,000 ppm.

Alarms High and low alarms, front panel adjustable with indicator lights.

Each alarm utilizies a relay with SPDT contacts rated at .1 amp at 120 VAC. Relays deenergized on .

alarm.

Digital Output Serial, teletype compatible Front Panel The front panel is brushed aluminum with a clear anodized finish.

,Ambient Operating 40 to 122°F Temperature Dimensions 8-3/4" H X 19" W X 16" D, designed for 19" rack mounting Weight 35 pounds 4-5

di;

• 1'/

FIGURE 4-1 SAMPLER roa THE CE BORONOMETEll tlNLET' PLAN VIEW SECTION A-A

• FIGURE 4-2 CE BORONOMET!R To Microcompat* .

/ 2 Curie Neutron f Soun:o In Well Temp. Sensor Fission Chambers Primary Coolant PREAMPLIFIER Located Near Sampler Outside) SIGNAL PROCESSOR

( the High 'Radiation Area (Located In Control Room)

---

.-..ji---."'""" POWER HIGH VOLTAGE SUPPLY LOW VOLTAGE POWER SUPPLY CHARGE PULS£ AMPLIFIER SHAPER MICRO- PPM BORON COMPUTER INDICATOR POWER SAMPLE

. SUPPLY TEMPERATURE COUNT RATE ANALOG OUTPUT Ht ALARM LO ALARM 4-7

FIGURE 4-3 PJIEDICT!D D!:LA.Y TIM! DU! TO MUING FOB. TB! C! BOROHOM!'?!Jl

• a, II&

I-

~

z 2

II.I 2

1.0 I-0.1 ,___ _...__ _..____.....,1_.i.....a..-.1....1.....,__ _.....,_.....,_.....,,,....._ _ _. _ _ _ . . . . _ _ . . . _.....-..1~1--,1""'""

0.10 1.0 Fl.OW RAT£, GPM

• 4-8

The preamplifier is located remotely outside the high radiation area. The signal processor continuously monitors the signal rate, and through an algorit~ stored in a microcomputer in the instrument, converts the count rate to parts per million of bor-on. Count rates are normally averaged over a time period which can be_ adjusted over the range of 1 to 999 seconds.

4-9

TEST DESCRIPTION

• GENERAL The irradiation testing was performed at the hot cell test facility at Georgia Tech. Testing to investigate reliability characteristics of the boronometer was performed at this same location. No radiation exposure was involved with the reli-ability t~sting. Testing to determine accuracy and reproducibility of the borono-meter was performed at the CE test facility and witnessed by NUS. Check analyses of the boronometer results was performed by CE and NUS using a boron-mannitol titration to determine boron concentration.

IRRADIATION TESTING The radiation source was provided by eight, 8 X 13 inch frame. as,sembl ies containing a total of.53,000 curies 60Co (6,600 curies per frame). Radiation levels desired were achieved by placing one or more of these frame assemblies around the sampler assembly as shown in Figure 4-4. Radiation levels were measured with dosimetry at the center of the assembly at a point just above the neutron source and estimated elsewhere. Testing was performed at an estimated maximum radiation level of l X 10 6 R/hr at the detector tubes. Maximum radiation levels at the center reference f point as determined by'dosimetry were 7.1 X 10 5 R/hr. The detector tubes were 5 several inches closer to the radiation source than the central reference point, therefore are in a higher radiation level area than is the reference. point.

The fission count rate was determined as a function of boron concen~ration and/or radiation level in the sampler assembly. Count rate was determined in the absence of radiation to determine a base level, followed.by testing with exposure to low, intermediate and high radiation levels.

All testing involved radiation exposure was performed in a hot cell under loop flow conditions. Only the sampler assembly was exposed to the radiation source. The remainder of the equipmeqt which includes the preamplifier and the signal processor were installed outside the hot cell. This is the manner in which the equipment would be installed for post-accident or normal operation.

4-10

FIGURE 4-4 GEOMETRY OP' IRRADIATION ASSEMBLY P'OR BORONOMETER SAMPI.Ell 53,000 Ci TOTAL RADIATION SOURCE 8 in. Wide by 13 in High 60co Frame

_____.j-,,~

Detector Sampler Wells Enclosure Radiation Leve I Neutron Source Measured at Center r

4-11

• RELIABILITY AND ACCURACY TESTING In the reliability testing, a boron solution was circulated through the boronometer for a perio..d of seven days while monitoring the fission count rate.

- performed under normal background radiation levels.

This work was Accuracy testin~ under loop flow conditions was performed at CE using boron solutions containing about 100, 600, 1,800, 3,200 and 5,000 ppm boron.

r.

L t

E h

a a

E E

LI" 4-12

~ - --------

TEST RESULTS IRRADIATION TEST RESULTS Note that the radiation levels noted are measured at the center reference point.

Actual radiation levels at the detector tubes which were affected by this radiation, were about 25 to 50 percent higher than the reference point measure-ments.

Typical results for the fission count. rate as a fooction of boron concentration and radiation levels for discriminator settings of 50 and 60 millivolts are presented in Table 4-4. The results *indicate that virtually total discrimination against radiation noise can be achieved. There is no memory effect nor is there any indi-catiqn of permanent damage based on about 2 X 10 7 rads total exposure to the detector tubes. This *is above the exposure levels anticipated under post-accident conditions.

RELIABILITY TEST RESULTS After the irradiation testing was complete, the boronometer was operated ooder steady state conditions for a periqd of seven days. Water containing about 2,960 ppm boron was circulated through the sampler and the fission count rate was recorded on a strip chart recorder. Some noise pickup was evident as is shown in Figure 4-5, demonstrating results of a one day roo over this period. However, the system was found to be completely free of noise when the development model preamp-lifier was replaced with a production model preamplifier.

ACCURACY TEST RESULTS The boronometer test results ~or low level boron concentrations are presented in Table 4-5. Note that the data are presented in terms of ppm.boron for an approxi-mate curve fit that was used when the data was recorded. This curve has been refined subsequent to the testing reported here to provide the proper ppa indi-cation. Accuracy results for high level boron concentrations are presented in Table 4-6. the deviation from results .indicated are acceptable for post-accident analyses.

4-13

TABLE 4-4 FISSION COURT RATE(l) AS A FIJHCTIOR OF RADIATION LEVELS FOR A SO AllD 60 M.V. DISCRIMINATOR SETTING OH THE CUSTO!f DESIGNED PREAMPLIFIER WL-24038 (2,960 PPM BORON*CONCEHTRATION)

SO M. V. Diocriminator Setti!YI 60 M.V. Diacriminator Setting Background 7.1 Xd9 s Background 4 X f25, 7 .1 Xd_9 5

2 X f2~ 4 X f2~ 2 X f25 R/Br ) R/Br )

6 X f25 R/Br )

Radiation !ili_ !l!!!.._ R/Br bdiation

• R/Br I

208 210 210 204 124 120 120* 119 110 214 214 211 219 121 121 121 121 111 210 211 214 218 121 124 120 119 114 209 212 214 225 120 121 119 120 115 211 205 122 118 118

__lli.....

Average i 210 212 212 214 121 121 120 120 114 a :I: 2.3 1.7 2.1 9.3 1.7 2.0 .B .9 3.2 2a :I: 4.6 3.4 4.2 18.6 3.4 4.0 1.6 1.8 6.4 (1) 100 oecond t i - interval (2) bdiation leveb at the detector tubeo wre higher by an eatiaated value of 25 to SO percent than indicated here *

• 4-14

TABLE 4-5

.BORONOMETER ACCURACY RESULTS FOR LOW LEVEL BORON CONCENTRATIONS

. (1) 99 El?! Boron 620 El?! Boron(l)

Count Period Count( ) PPM Count Period Count( 2 ) PPM Seconds Rate 2 DisJ:!lal Seconds Rate DisJ:!lai 100 278 504 100 275 603 100 275 653 100 314 89 100 275 679 100 315 68 100 275 697 100 314 56 100 275 688 100 310 76 100 274 690 100 314 72 100 274 710 100 311 77 100 275 697 100 I 314* 61 500 275 684 100 315

• 58 500 275 695 ioo 310 74 500 278 671 500 314 70 500 .llL.. 667 500 314 71 275 664 JC 500 315 62 a +/- 1.5 55.2 500 .lli..._ 65 2a :t: 3.0 110.4 JC 313 69 a +/- 1.8 9.0 2a :t: 3.6 18.0 (1) As determined by chemical analyses (2) Combined count rate from four fission chambers 4-15

• At the conclusion of the test, the boronometer was operated briefly while increasing temperature of the solution from 80°F to 117°F.

~temperature -~as controlled at 80°F.

In the half hour testing performed, there was no change in ppm readout beyond the spread noted when Admittedly, this was a very brief test period, yet it does indicate that minor fluctuation in temperature will have little if any*

effect on boron *readout.

The CE wide range boronometer with BF detectors was also tested by KWU in Germany 3

at GKN for a period of about eight months. They reported that the "measured values, compared to other chemical measurement and evaluation methods were within the specified accuracy of :!:. one percent ( +5 mg/1)." They further recommended the boronometer for use at the KWU site.

The advantages of operating a boronometer for determining boron concentration during post-accident conditions are as follows:

• All operations can be performed remotely. The exposure involved for determining boron concentration would approach z~ro.

• No chemicals are added to the sample. Sample flow can be pumped back to the primary system reducing the lpad on t!te radwaste system,

• It provides a direct measure of boron-10 or neutron poison concentration in the system.

• 2

• The system is sealed, thus preventing release of gaseous activity to the envirorunent,

• Analyses results can probably be achieved within a matter of 15 to 30 minutes dependent on flow rate through the sampler, The disadvantages of operating a boronometer during post-accident conditions as follows:

• The sampler system would have to be shielded since a relatively large volume of coolant is required. About 15,000 curies of activity would need to be transported outside the primary contairunent to operate this system. *

• The system has not been proven under long-term use. However, there is not reason to assume that it would not be reliable, Individual components within the system are off-the-shelf items, Most of the electronics are identical to those used on CE's wide range boronometer and CE has been shipping these units since 1977,

• 4-16

• - TABLE 4-6 BORONOMETER ACCURACY RESULTS FOR HIGH LEVEL BORON CONCENTRATIONS 1825 El!!!! Boron(l) 2904 EEm Boron(l) 4928 El!!!!

BoronH) Count PeriodCount PPM Count PeriodCount PPM Count Period Count PPM ( )

Seconds Rate 2 DisElal'. Seconds Rate <2 >nisElal'. Seconds Rate <2 >nisElal'.

100 234. 1921 100 215 2996 100 191 5309 100 234 1915 100 215 2946 100 191 5317 100 234 1921 100 218 2844 100 190 5394 100 235 1904 100 215 2912 100 195 5004 100 234 1861 100 214 2901 100 194 5001 100 231 1899 100 215 2917 100 194 5005 100 234 1897 100 215 2944 100 191 5033 100 234 1915 100 215 2937 100 195 4968 500 234 1915 100 215 2958 100 194 5217 500 230 1~51 100 215 2908 100 194 5101 500 231 ._ilii_ 100 215 2884 100 190 4954 233 1913 100 211 2981 100 194 4951 X

a :t 1.7 24.0 100 214 2934 100 194 4864 20 :t 3.4 48 0 100 215 2943 100 195 4834 100 211 3012 100 194 4875 500 215 2991 100 191 4999 500 214 .1Q!!_ 100 195 4921 215 2943 500 194 5077 i"

a :t 1 6 46.7 500 194 5045 21 :t .3 2 93.4 500 191 12!L.

r 193 5046 i

l a :t 1.8 154.1 i

21 :t 3.6 308.2 F

(1) Aa deteriained by ~hemical analy*e* (2) Combined count rate from four fi*aion chawber*

i i.

4-17

$$JS.,3f! , .. PW*

r

• 1'/

May be 1l1clronlc nol11 pickup j

]

1 j

I 1

FIGURE 4-5 1' ONE DAY STRIP CHART RECORDING OF 2960 ppm BORONOMETER ANALYSES RESULTS (Background Radiation)

/

....: .. -~*.- ~.:... ....... ~,-

CONCLUSIONS AND RECOMMENDATIONS

'

• The CE Boronmeter is acceptable for use under post-accident conditons.

• Reproducibility of results is excellent as based on fission count rate; however, conversion of count rate to ppm is somewhat below the accuracy desired for daily operations. CE indicates, however,

.I.

that the proper curve fit routine in the microcomputer will provide proper ppm indi_cation.

• A 500 second count rate is recoUDDended for determining boron concentrations below 1,000 ppn.

• The use of st~ip chart recorder is recommended for use with the boronometer. This will improve statistics and show trending.

• There is some increase in the standgrd deviation (Table 4-4) from radiation levels in the range of 10 R/Hr at the planned discriminator setting of 50 millivolts. The increase is not significant *with respect to post-accident analyses requirements.

I.

4-19