ML20236Q956

Requests That Proprietary Rev 0 to TR ER-80-P, Improving Thermal Power Accuracy & Plant Safety While Increasing Operating Power Level Using LEFM Check Sys, Be Withheld from Public Disclosure,Per 10CFR2.790(b)(4)

ML20236Q956
Person / Time
Site:	Comanche Peak
Issue date:	07/16/1998
From:	Hastings C External (Affiliation Not Assigned)
To:	NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM)
Shared Package
ML20236Q951	List: ML20236Q948 ML20236Q956
References
CAW-98-01, CAW-98-1, NUDOCS 9807210149
Download: ML20236Q956 (6)

Text

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

{

/0 C (X%Ch[h D

=

.f=E. E ' =EEEE ~EE=:5 '

O c':- l.~ s.rg " + ',u 7: -'G u m : 5.1 n' 'M (: '.*i 2 - Mr T 1 Y.'

' ' ' ',M l

Caldon,Inc.

1070 Banksville Avenue j

Pittsburgh, PA 15216

{

412-341-9920 Tel 412N9951 Fax July 16,1998 www.caldon. net Document Control Desk U. S. Nuclear Regulatory Commission

. Washington, DC 20555 -

... Attention: Mr. Tim Polich APPLICATION FOR WITHHOLDING PROPRIETARY INFORMATION FROM PUBLIC DISCLOSURE

Subject:

Topical Report," Improving Thermal Power Accuracy and Plant Safety '

While Increasing Operating Power Level Using the LEFMV System",

Caldon, Inc. Engineering Report ER-80P (Proprietary), March 1997.

Dear Mr. Polich,

The proprietary information for which withholding is being requested in the above-referenced report is further identified in Affidavit CAW-98-01 signed by the owner

- of the proprietary information, Caldon, Inc. The affidavit, which accompanies this letter, sets forth the basis on which the information may be withheld from public disclosure by the Commission and addresses with specificity the considerations listed in paragraph (b)(4) of 10 CFR Section 2.790 of the Commission's regulations.

Accordingly, this letter authorizes the utilization of the accompanying Affidasit by TU Electric.

Correspondence with respect to the proprietary aspects of the application for withholding or the Caldon affidavit should reference this letter, CAW-98-01, and should be addressed to the undersigned.

.Very truly yours, 15

.J 7 Calvin R. Hastings l

t President and CEO-3

[.

. Enclosures.

l 9907210149 990716

^

PDR ADOCK 05000445/

P PDR (

i

l July 16,1998 CAW-98-01 AFFIDAVIT COMMONWEALTli OF PENNSYLVANIA:

l ss COUNTY OF ALLEGliENY:

Before me, the undersigned authority, personally appeared Calvin R. Hastings, who, being by me duly sworn according to law, deposes and says that he is authorized to execute this Affidavit on behalf of Caldon, Inc. ("Caldon") and that the averments of fact set forth in this Affidavit are true and correct to the best of his knowledge, information, and belief:

Wh

& inet

(/

Calvin R. Hastings, I

President and CEO Caldon, Inc.

. Sworn to and subscribed before me this

/4-day of L

Ju t.-u 1998 I

A J cb

['

d Notarial Seet Kathryn J. Headley, Notary PutAic Pittsburgh, Allegheny County My Commission Expres March 6.2001 Member, Pennsylvania Association of Notaries l.

l l

1. I am the President and CEO of Caldon, Inc. and as such, I have been specincally delegated the function of reviewing the proprietary information sought to be withheld from public I

disclosure in connection with nuclear power plant licensing and rulemaking proceedings, and am authorized to apply for its withholding on behalf of Caldon.

2. I am making this AfGdavit in conformance with the provisions of 10CFR Section 2.790 of the Commission's regulations and in conjunction with the Caldon application for withholding accompanying this Af6 davit.

3. I have personal knowledge of the criteria and procedures utilized by Caldon in designated i

information as a trade secret, privileged or as conndential commercial or fmancial infom1ation.

4. Pursuant to the provisions of paragraph (b)(4) of Section 2.790 of the Commission's regulations, the following is furnished for consideration by the Commission in determining I

whether the information sought to be withheld from public disclosure should be withheld.

(i)

The information sought to be withheld from public disclosure is owned and has been held in con 6dence by Caldon.

(ii) The information is of a type customarily held in con 6dence by Caldon and not customarily disclosed to the public. Caldon has a rational basis for determining the types of information customarily held in confidence by it and, in that connection, utilizes a system to detennine when and whether to hold certain types ofinformation in confidence. The application of that system and the substance of that system constitutes Caldon policy and provides the rational basis required.

i Under that system, information is held in confidence ifit falls in one or more of several l

1 types, the release of which might result in the loss of an existing or potential advantage, as follows:

1 2

J l

l (a) The information reveals the distinguishing aspects of a process (or component, I

structure, tool, method, etc.) where prevention ofits use by any of Caldon's competitors without license from Caldon constitutes a competitive economic advantage over other companies.

(b) It consists of supporting data, including test data, relative to a process (or component, structure, tool, method, etc.), the application of which data secures a competitive economic advantage, e.g., by optimization or improved marketability.

(c) Its use by a competitor would reduce his expenditure of resources or improve his competitive position in the design, manufacture, shipment, installation, assurance of quality, or licensing a similar product.

(d) It reveals cost or price information, production capacities, budget levels, or commercial strategies of Caldon, its customer or suppliers.

(e) It reveals aspects of past, present or future Caldon or customer funded development plans and programs of potential customer value to Caldon.

(f) It contains patentable ideas, for which patent protection may be desirable.

l There are sound policy reasons behind the Caldon system which include the following:

i (a) The use of such information by Caldon gives Caldon a competitive advantage over its competitors. It is, therefore, withheld from disclosure to protect the Caldon competitive position.

j (b) It is information which is marketable in many ways. The extent to which such information is available to competitors diminishes the Caldon ability to sell products or services involving the use of the information.

i (c) Use by our competitor would put Caldon at a competitive disadvantage by j

reducing his expenditure of resources at our expense.

(d) Each component of proprietary information pertinent to a particular competitive advantage is potentially as valuable as the total competitive advantage. If J

competitors acquire components of proprietary information, any one component may be the key to the entire puzzle, thereby depriving Caldon of a competitive advantage.

3

l

\\

(e) Unrestricted disclosure would jeopardize the position of prominence of Caldon in the world market, and thereby give a market advantage to the l

competition of those countries.

(f) The Caldon capacity to invest corporate assets in research and development depends upon the success in obtaining and maintaining a competitive

\\

advantage.

(iii) The information is being transmitted to the Commission in conDdence, and, under the provisions of 10CFR Section 2.790, it is to be received in confidence by the Commission.

(iv) The information weght to be protected is not available in public sources or available information has not been previously employed in the same manner or method to the best of our knowledge and belief.

(v) The proprietary information sought to be withheld in this submittal is that which is appropriately marked in the Topical Report entitled," Improving Therrnal Power Accuracy and Plant Safety While Increasing Operating Power Level Using the LEFMV I

System", Caldon, Inc. Engineering Report ER-80P, (Proprietary), March 1997 being transmitted by TU Electric letter and Application for Withholding Proprietary Information from Public Disclosure, to the Document Control Desk, Attention, Mr. Tim Polich. This proprietary information as submitted for use by TU Electric for the Comanche Peak Nuclear Plants is expected to be applicable in other license submittals forjustification of the use of the Caldon Leading Edge Flow Meter (LEFMV) to increase reactor plants' thermal power.

This information is part of that which will enable Caldon to:

(a) Demonstrate the design of the LEFMV and accuracy of the LEFMV flow and temperature measurements, as well as the improved calorimetric thermal power accuracy based on the LEFMV measurements.

(b) Demonstrate the reliability of the LEFMV based on design features and on compiled field experience data.

l l

4 l

L

l l

(c) Establish technical and licensing approaches for the application of the improved accuracy of this method toward increasing thermal power.

(d) Assist customers in obtaining NRC approval for increases in thermal power based on appropriate use of the LEFMV for calorimetric power measurement.

Further this information has substantial commercial value as follows:

)

(a) Caldon plans to sell the LEFMV and use of similar information to its l

customers for purposes of meeting NRC requirements for operation at increased thermal power.

(b) Caldon can sell support and defense of the technology to its customers in the licensing process.

Public disclosure of this proprietary information is likely to cause substantial harm to the competitive position of Caldon because it would enhance the ability of competitors to provide j

similar flow and temperature measurement systems and licensing defense services for f

commercial power reactors without commensurate expenses. Also, public disclosure of the inform ition would enable others to use the information to meet NRC requirements for licensing documentation without the right to use the information.

The development of the technology described in part by the information is the result of applying the results of many years of experience in an intensive Caldon effort and the expenditure of a considerable sum of money.

In order for competitors of Caldon to duplicate this information, similar products would have to be developed, similar technical programs would have to be performed, and a significant manpower effort, having the requisite talent and experience, would have to be expended for developing analytical methods and receiving NRC approval for those methods.

Further the deponent sayeth not.

5

o, ~,

g

. m

. - c j -

4 4

~,,4.

pW, g

a c..

e,

4 4

9-

' gg

)

S C'

/*

b s

,.s

,.- L, <

7

> ^

j_-

+

f.a e

/

4 y

+

w

.. ~.

2 4 s-

. ~.

_., _.m m-

.,, m m.m

x. ,v J

n' A

4..

'M 4

ER R 1

/

~

,.. ~ u, c.

+

. Revision 0.-

. a

_, s C.f >..

l March,1997

, j

m l'

'. } l t

/

C'

l _.

.c s '.T_ ' g j

, ?

,.f g :..

.'6&

i s

t r

c 1

- e p y

,v

,--f, E

.-l.',.

E

.'-j P

p g g g

<v I.e s-s 1

W s

r J '

n,

.s c

i[

,-[.D*

,s

i e

m 4

W v c

y

+,

s I #

~3 4

p m?n:E ' "."{,

? <

CALDON,..INCE, 1

D r

' l.'

<y t*

i >

J c

s+

'.?.

(

r 4

.~

~

i s f. '

. : x..

~

~

, -+..,.

> ! ENGINEERING RE' ORT-800 M i',.,

=..

P

>. c a-

~.; a."

.,o ".. ' '

f'-

5

. I"

[.

.'s.

..'f6

4. '

g

i b T.

,c N,1 d 7 TOPICAL REPO$tT-

[M' f; ' ' (fIriiproving Thermai PowcfAccuracy, and Plant Safety ' r

( f.. a. : i L

m

[

. ~

m 7

t" g.,%l, v,o ^

While Increasing Operating Power Lhvelk c "-

?

,,, ' Using t,he' LEFM?

. Systemi

' ' ' ' w.2 E

- r.

S-s o-

s. <

vs,

'I T

,g,

-.4,,

f 4

.s wi w

tg

'1 8

4 I

t

. r..

, n.

g 4+ +

..o

. r

.c u..

s R...s.evisio. n 0 -.-

,i 3

T

..o, 1,

-o e

<, f s

t

~

e

~

v.. r,

.i

),

- e g

5

,y. f,- ',,

,a

,,, c ~.-

.r

.g-

..-.-?.. ;,

s

\\

Y ~_ 4 g

4' v

f-'

Y b"

e-

• e-3 s

n e

e o

v

.L

- b g ;...

s x

/ ;

' %.. 3 7

.l.l -

m

'A

- ~

..+;

,1

'; 4,,

,-. i.

s e

y *,,.

s

,x.

b

[ --

t

- d

~ys

. ?.

t

+'

e 3

g' J

J e

~

a.

h,...

3.

O.,'

, e

,-.,4

-.s

',a g.

g s

s b

3.'

ii I #

J,-

t 3 s

4 f

i F

j j

.'s 4

i U

\\

'g.

e3 6

b i

f.

T a..f

.4

.. +

s<cg, s

sm

.s

=

+

..e-

,?,

+L

't

.(.,

.v

,1

.,r..

+

s't I

i g).m, 3

IP

,s I i 4

i-J*',_*.

\\

• L k'

4 r

e r

i.

t

e 5-e t.j.

i

.g.

s:

.?

og y

s s:

' 'O Cal' don, Ini.1997

~

r 4

+1 :

A IL f

2

,.b hy.

k, 4e 4

c g

A

, -h s

.s y

s',

\\.

.7 4f _ ) '

/)' ) ry .

a4 j.

_ '.. _;.. w'P 2);, bp.

r

__ _e s

jd-35EFEi ER-80 Revision 0 March,1997 CALDON, INC.

ENGINEERING REPORT-80 TOPICAL REPORT Improving Thermal Power Accuracy and Plant Safety While Increasing Operating Power Level Using the LEFM /

System Revision 0 Prep.ared By: Jennifer Renan Reviewed By: Herb Estrada, Jr.

C Caldon,Inc.1997 LEFM and LEFM / are registered trademarks of Caldon,Inc. All rights reserved.

SUMMARY

This report describes an improved system for the measurement of feedwater flow and temperature to determine reactor thermal power in nuclear power plants. The system is based on the use of chordal LEFM" technology that has been applied to flow measurements in nuclear power plants for almost 25 years. The system proposed for this application is referred to as the LEFM/". The LEFM/ is superior to the 1970's vintage flow nozzi:-based instrumentation currently in use on two counts:

1. The elements of LEFM/ accuracy can be verified on-line, and

2. The LEFM/ is demonstrably more accurate. The LEFM/ is accurate to 0.6% of thermal power,95% confidence limits, versus 1.4% representative of current instrumentation.

This Topical Report may be referenced by licensees currently using Part 50 Appendix K ECCS evaluation models in applications for thermal power uprates up to 1% based on the installation and use of the LEFM/ technology. Approval of uprates is subject to the c.onditions set fonh in this report.

This report includes analyses and other information in the body of the report and in the appendices supporting the performance capabilities of the LEFM/ and comparisons with systems based on prior technology. Technicaljustification is provided for increasing thermal power by 1% without reanalysis of ECCS performance based on the use of the LEFM/. This topical report is intended to be referenced in support of applications for license amendments by licensees who install and use the LEFM/ technology pursuant to the conditions set forth in this report.

LEFM and LEFM/ are trademarks of Caldon. Inc. All rights are reserved.

TABLE OF CONTENTS Summary

1. Introduction...

1-1

2.

Background:

Calorimetric Detennination of Thermal Power..

.... 2-1 2.1 The Requirement to Determine Reactor Thermal Power 2.2 nermal Power Calorimetric Calculations in PWR's 2.3 Thermal Power Calorimetric Calculations in BWR's 2.4 Method for Calculating Calorimetric Uncertainty

3. Measurement of Feedwater Flow and Temperature With a Chordal LEFM/...... 3-1 3.1 Summary System Description 3.2 An Overview of LEFM/ Principles 3.3 The Chordal LEFM/ Algorithm 3.4 Accuracy of Calorimetric Power Determinations With the LEFM/

4. Current Calorimetric Power Determinations......................

... 4-1 4.1 The Mass Flow Algorithm for Nozzle-Based Instrumentation 4.2 Accuracy of Calorimetric Power Determinations With the Current Instrumentation

5. Basis for The hermal Power Uprate With the LEFM/......

.............5-1 5.1 Comparison of the Flow Algorithms for LEFM/ and Current Instrumentation 5.2 Improvements in Calorimetric Accuracy: Basis for the Power Uprate 5.3 Benefits of On-Line Verification 5.4 Using the LEFM/ to Control nermal Power

6. On-Line Verification of LEFM/ Accuracy.........

.............6-1 6.1 Elements of the LEFM/ Measurement 6.2 Bounding LEFM/ Elemental Errors 6.3 On-Line Verification of Accuracy 6.4 Quality Measures in Design, Fabrication and Factory Acceptance Testing of the LEFM/

7. Reliability........................

7-1 7.1 History of LEFM Technology 7.2 Nuclear Operating Experience With the Chordal LEFM and Component Reliability 7.3 Nuclear Plant Operation With LEFM/ Out of Service 7.4 LEFM Operating Experience in Other Applications i

TABLE OF CONTENTS 7.5 Survey of Experience Reported in Licensee Event Reports (LER*s) and Information Notices

8. Conditions for Use of LEFM/ With 1% Power Increase.

. 8-i

9. References....

.. 9-1

10. Appendices A. Basis for Uncertainties in Existing Calorimetric Instrumentation.

A-1 B. Principles of Chordal LEFM/ Flow And Temperature Measurements........ B-1 C. Verification of LEFM/ Temperature Correlation......

.........C-1 D. Identifying and Bounding the Uncertainties in LEFM/ Flow and Temperature Measurements...

... D-1 E. Basis for Uncertainties in Determining Thermal Power With the LEFM/.... E-1 F. A Survey of Hydraulic Performance of Chordal Leading Edge Flow Meters......

......... F-1 G. Estimates of the Probability of Exceeding Power Margins of up to Five Standard Deviations.........

............G-1 H. Caldon Flow Measurement Applications............

........ H-1 ii

Section 1 INTRODUCTION The LEFM/ described in this report is an improved system for use in accurately determining and monitoring thermal power in nuclear power plants. He LEFM/ provides increased safety by providing on-line verification of the accuracy of the feedwater flow and temperature measurements upon which thermal power detenninations are based. In addition, the LEFM/

provides a significant improvement in accuracy and an increase in reliability of flow and temperature measurements. These improvements permit licensees who currently use Appendix K ECCS models to increase licensed thermal power levels up to 1% without any reanalysis of accidents or transients. At the same time, the LEFM/ provides a significant improvement in the probability that the power level and peak specific power for which the accident and transient analyses were performed will not be exceeded.

Currently,10CFR50 Appendix K,(for ECCS analysis) requires an initial power level assumption 1

of at least 2% above the licensed power level. Furthermore, Regulatory Guide 1.49 and the Standard Review Plan, NUREG 0800 (for other design basis accidents and transients) recommend an initial power level at least 2% above the licensed power level to provide margin to account for inaccuracies in thermal power determination. A study of the uncertainties in thermal power determinations demonstrates that, with an LEFM/ system and a 1% increase in licensed thermal power, the odds of exceeding the thermal power at which accidents are analyzed by 0.5% are virtually nonexistent (less than one in three million). Coincidentally, this is the same probability of exceeding the same analyzed power level by 1.5% for thermal power determinations based on the prior technology with no increase in licensed power. In other words, the safety margin is improved with the LEFM/ in use, even with a /% power increase.

These probabilities are only valid if the instruments are performing as designed. The on-line verification features of the LEFM/ provide the ability to assure on-line that performance is consistent with the design basis. No such means can be provided by the prior technology.

The LEFM/ system described in this report is an ultrasonic flow meter that measures the transit times of pulses of ultrasonic energy traveling along chordal acoustic paths through the flowing fluid. The LEFM/ system consists of a spool piece in each of one to four feedwater line(s).

Each spool piece has four parallel (chordal) acoustic paths, for measurement of fluid velocity and sound velocity. An electronic unit serves up to four spool pieces, and is specifically designed to provide flow and temperature measurements and on-line verification of these measurements.

This technology provides significantly higher accuracy and reliability than flow instruments which use differential presste measurements and temperature instruments which use conventional thermocouple or ri. ' stance thermometers.

This report includes analyses and other information in the body of the report and in the appendices supporting the performance capabilities of the LEFM/ and comparisons with 1-1

l systems based on prior technology. Justification is provided for increastng thennal power by 1%

without any reanalysis of ECCS perfortnance based upon the use of the LEFM/. This topical report is intended to be referenced in support of applications for license amendments by licensees I

who install and use the LEFM/ technology pursuant to the conditions set forth in this report.

l t-2

Section 2 BACKGROUND:

CALORIMETIUC DETEIGUNATION OF THEIDIAL POWER 2.1 The Requirement to Determine Reactor Thermal Power U.S. nuclear power plants are licensed to operate at power levels up to a specified thermal power rating. Safety analyses and evaluations are performed at conditions selected to account for uncertainties in determining thermal power. The NRC provides guidance regarding the amount of margin needed to account for such uncertainties in Regulatory Guide 1.49, Rev.1, December 1973 (Ref.1). The guides states:

Analyses and evaluation in support oflicense applications should be made at an assumed core power level equal to 1.02 times the proposed licensedpower level.

for (a) normal operating conditions. (b) transient conditions anticipated... and (c) accident conditions necessary to evaluate the adequacy ofstructures, systems and components providedfor the prevention ofaccidents and the mitigation ofthe consequences ofaccidents.

Regulatory Guide 1.49 states that the reason that analyses should be performed at a

" slightly higher power level (is) to allow for possible instrument errors in determining the power level. The Regulatory staff has determined that a margin of 2 percent of the licensed power level is adequate for this pugose...".

This margin is also called out in the Standard Review Plan and was originally incorporated in 10CFR50.46 and Appendix K to Pan as the required margin betwe' n the licensed power e

level and the level at which Loss of Coolant Accident (LOCA) and Emergency Core Cooling System (ECCS) analyses are performed. The current acceptance criteria for ECCS in 10CFR50.46, paragraph (a)(1)(i), promotes the use of models for the calculation of ECCS performance which are more realistic than altemative models permitted by paragraph (a)(1)(ii) which must satisfy the requirements of Appendix K. The only prerequisite for approval of more realistic models under paragraph (a)(1)(i) is that the more realistic model be supported byjustification of the uncertainties involved so that there is a high level of probability (about 95% probability with 95% confidence) that the calculated results will not exceed the criteria set fonh in 10CFR50.46(b).

Most licensees have chosen to continue to use Appendix K ECCS evaluation models, assuming that the reactor has been operating continuously at 1.02 times the licensed power, at the maximum peaking factor allowed by the technical specifications, to allow for uncertainties in determining thermal power.

In most plants, operators obtain a continuous indication of core thermal power from nuclear instruments which provide a measurement of neutron flux. In PWR's the instruments used 2-1

are external to the reactor vessel; in BWR's they are distributed throughout the core. For both types of plants, the nuclear instruments must be calibrated periodically to counteract the effects of flux pattem changes, fuel bumup, and instrument drift To provide a basis for this calibration, reactor operators determine the core thermal power at which they are operating by steam plant calorimetry, that is, they calculate reactor thermal power by performing an energy balance around the nuclear steam supply system.

In a few plants, operators use a direct indication of thermal power, as calculated &om a steam plant energy balance, as a basis for control. They adjust power by manipulating turbine steam demand so as to maintain the thermal power at orjust below the licensed value. In such cases, nuclear instruments are used as an anticipatory indication in transients, such as when reactivity adjustments are made with the control rods.

The steam plant calorimetry by which thermal power is currently established is described below for both PWR's and BWR's. It entails calculating the heat input to the nuclear steam supply system, the heat output of the system, and heat losses, in an energy balance. The accuracy of the resulting thermal power value depends primarily upon the accuracies with which feedwater flow and temperature are determined.

Licensees calculate thermal power using equations based on an energy balance around the nuclear steam supply system, and this calculational method will not be changed with the proposed use of the LEFM/. The differences in calorimetric power calculations among plant designs are all in small components of this energy balance.

In the remainder of this section, thermal power calculations will be described in detail for PWRs and BWRs. It will also describe the method for calculating calorimetric uncertainty to ensure that thermal power is known to within t2% and hence consistent with the basis for the 1.02 factor required by Appendix K. The remainder of the report is dedicated to justifying the use of a more realistic model pursuant to 10CFR50.46(a)(1)(i) and allowing the licensed power to be increased by 1% (by license amendment) without reanalysis of ECCS performance based on use of the LEFM/ system.

l i

l 2-2

{

l t

2.2 Thermal Power Calorimetric Calculations in PWR's In general, the energy balance around a PWR nuclear steam supply system is determined as illustrated in Figure 2-1. It can be described by the following equation:

CTP = Wr(h,- hs)- BD (h - heo)- NHL (2-1)

Where:

CTP

= Core Thermal Power (btu /hr)

Wr

= Feedwater Flow (lbm/hr) h,

= Enthalpy of steam exiting the steam generator (btu /lbm) hs

= Enthalpy of feedwater entering the steam generator (btu /lbm)

BD

= Blowdown Flow (Ibm /hr) (not applicable to B&W units]

h84

= Enthalpy of blowdown flow exiting the steam generator (btu /lbm)

NHL = Net rate of heat addition to/ loss from the reactor coolant system (btu'hr) i The net rate of heat addition to the reactor coolant system accounts for the heat added by the coolant pumps and pressuri7er heaters, heat losses through the makeup / letdown / purification system (s) and control rod drm cooling, and heat losses due to radiation and conduction / convection into the containment.

In Equation 2-1 above, the dominant component in determining net core thermal power is steam supply, i.e., steam generator thermal power, SSTPpwn :

SSTPewa = Wr (h. - hs)

(2-2)

In Equations (2-1) and (2-2), the feedwater mass flow measurement is commonly used to represent both feedwater mass flow and steam mass flow. The feedwater mass flow measurement is used because it is generally more accurate than the steam flow measurement and, over the long term, these two flows and the blowdown flow are in equilibrium (A correction is made separately for the blowdown flow if blowdown is not isolated when the calorimetric calibration is performed).

Feedwater enthalpy is determined from measurements of feedwater temperature and feedwater pressure, typically using a correlation based on the ASME Tables for Compressed Water. Steam enthalpy is based on a measurement of steam pressure and, except for B&W plants, an estimate of steam generator moisture carryover. The estimate may be based on steam generator test results, on conservative assumption of zero carryover, or on some estimate between the results and zero carryover.

The blowdown heat loss and other net heat losses from the reactor coolant system are small corrections to the energy balance. Thus, the accuracy with which the thermal power is calculated is primarily dependent on the accuracy of the feedwater flow measurement and the accuracy with which the steam generator enthalpy rise is determined.

23

l l

l i

l l

l l

I 2.3 Thermal Power Calorimetric Calculations in BWR's In general, the energy balance around a BWR nuclear steam supply system is determined as shown in Figure 2-1. It can be described by the following equation:

)

CTP = Wr (h, - hrw) + CRD (h - here) + Qr.4-Qp + Qci anup (2-3) l Where:

CTP

= Core Thermal Power (btu /hr)

Wr

= Feedwater Flow (Ibm /hr) h,

= Enthalpy of steam exiting the reactor vessel (btu /lbm) hrw

= Enthalpy of feedwater entering the reactor vessel (btu /lbm)

CRD = Control Rod Drive Flow (Ibm /hr) hero

= Enthalpy of control rod drive flow entering the reactor (btu /lbm)

Q,4

= Radiative thermal losses to the containment environment (btu /hr)

Qp

= Reactor recirculation pumping power added (btu /hr) i Qcie.nop= Net thermal loss from the reactor water cleanup system (btu /hr)

In Equation 2-3 above, as in Equation 2-1 for PWR's, the dominant component in determining net core thermal power is the steam supply, i.e., in this case, the reactor thermal power, SSTPawn:

SSTPawa = Wr(h - hr.)

(2-4)

In Equations 2-3 and 2-4, as with PWR's, the feedwater mass flow measurement is commonly used to represent both feedwater mass flow and steam mass flow because the feedwater flow measurement is more accurate than the steam flow measurement in BWR's and the two flows in combination with the CRD flow are in equilibrium over the long term (a correction for CRD flow is made). The steam and feed enthalpy calculations are similar to PWR's.

2-4

2.4 Method for Calculating Calorimetric Uncertainty In addition to determining thermal power via calorimetric calculation, licensees must also show that the thermal power is known to within 2%. The uncertainty with which thermal power is calculated also determines the probability of exceeding the power level at which design basis transient and accident analyses were performed. The overall thermal power uncertainty depends on the methods and bases of the calculation; the methods used to combine the uncertainties in the individual measurements required for the determination of power and the bases for these individual measurement uncertainties.

The method used by most licensees to estimate the thermal power uncertainty typically involves; (1) calculating the sensitivity of thermal power to each process variable and measurement parameter required to determine it; (2) calculating or otherwise determining the instrument error associated with measurement of each variable on a consistent probabilistic basis; and (3) multiplying these errors by their respective sensitivity coefficients to determine the error contribution of each in terms of calorimetric power. After all contributing errors are calculated, they are combined in accordance with standard methods for determining measurement uncertainty (for example, the methods described in ASME PTC 19.1 (Ref. 2), and ISA RP 67.04, (Ref.3)) to calculete total thermal power uncertainty within a selecter' confidence interval. The errors are combined by algebraic sum if they are mutually depend it (i.e., systematic), and by the root-sum-square if they are independent.

Many licensees determine instrument errors on the basis of a 95% confidence interval. This is sometimes referred to as a "two standard deviation" basis because, if the probabilities of error magnitudes for an instrument are normally distributed, there is a 95% probability that the error will lie in a 2 standard deviation band *. Suppose, for example, that a 1.5%

thermal power uncertainty was calculated, and the instrument uncertainties used represented 95% confidence intervals. If the plant operator controls steam demand to maintain power exactly at the licensed value, the 1.5% thermal power uncertainty means that the probability is 0.95 that the true power lies between 0.985 and 1.015 of the licensed power.

The uncertainty elements for thermal power determinations using current and LEFM/

systems are discussed in detail in Sections 3 and 4, respectively. In discussing the uncertainty elements of thermal power determinations, both existing and proposed, this topical report will follow the practice of the AShE and ISA standards nreviously cited. For example, instrument performance will be described on a 95% confidence interval basis.

• Some licensees estimate thermal power uncertainty on a one, rather than a two standard deviation basis. This is equivalent to a 68% confidence interval.

25

i m

h s

t h

ne t

t n

n mCo e

t Wy no 3C

, g Wy wre o

,g l

n wr FE e

o n e e r r lFE t ta a mm ww s

aa dd e

e e e e s

t t e e SS FF so l

y r

te s

m n

ir i

o a

la g

C t

e n

l e

a c

r l

u P

y t

c a

a r

i t

e V

n

+

p em m

r

)

e e

e w

T w

o ru d

s rI P

e a

h e

l e

F a

M m

6 r

w 3

e 2

o h

l h

w F

(

T o

f l

w F

o r

d n

e i

W e

o i

F ta

=

n i

r m

e re r

w o

te t

r a

o mr o

D t

Ra e Rc P

nr a

Wt eoWe e

1 l

PSG BR 2

a e

m ru r

g e

F i

hT 7

l l

I g

l

\\

l

- )

R

- W P

I

\\

/

f l

I I

g

-(lI 1

r g

l I

I

\\

]

l l

Section 3 MEASUREMENT OF FEEDWATER FLOW AND TEMPERATURE WITH A CHORDAL LEFM/

1 3.1 Summary System Description The LEFM/ is an ultrasonic flow meter that measures the transit times of ultrasonic energy traveling along chordal acoustic paths through the flowing fluid. A schematic and photos of the LEFM/ system are shown in Figure 3-1. The LEFM/ system consists of one to four spool pieces in the feedwater line(s), each having four parallel acoustic paths across the spool piece, and an electronics processing unit specifically designed to provide flow and temperature measurements and on-line verification of the key elements of these measurements.

As shown in Figure 3-1, the digital electronics unit performs three principal functions: (1) the acoustic processing unit manages ultrasonic timing and control. It transmits, receives and detects ultrasonic pulses and measures their transit times; (2) the central processor performs all flow and temperature calculations from the transit time data, and performs on-line verification checks; and (3) the control and display panel and associated output devices provide the operator with flow and temperature indications and on-line verification of the measurements via data display screens and alarm indicators. It also provides system set-up and diagnostic capability. These functions are all performed digitally, and are shown in more detail in Figure 3-2.

Figure 3-3 illustrates the LEFM/ spool piece and a typical transducer well (also refened to as a housing) as it fits in a spool piece boss. As shown in Figure 3-4, eight transducers are mounted in eight wells in each spool piece to fonn the four acoustic paths. The [.

] transducer wells are mounted in the bosses such that the wells do not protrude into the flow stream, but are wetted by the fluid medium. The distance between the faces of a pair of wells forming an acoustic path is closely controlled since this dimension directly affects the LEFM/ flow determination. Likewise, the angle formed by the centerline of each pair of housings and the axis of the spool piece is closely controlled since this parameter (the path angle) also directly affects the flow determination. Certain other dimensions (e. g., internal diameter) are similarly controlled, for the same reason.

3.2 An Overview of LEFM/ Principles A transit time ultrasonic flow meter operates on the principle that the time required for a pulse of ultrasonic energy to transit between two transducers immersed in a fluid medium is given by the quotient of the distance between the transducers and the ultrasound propagation velocity. Consider the two ultrasonic transducers in the pipe illustrated in Figure 3-3. They are placed to define a diametral diagonal acoustic path through the fluid in the pipe. The ultrasound propagation velocity is the algebraic sum of the ultrasound propagation velocity of 3-1

the fluid at rest and the velocity of the fluid medium itself. Hence the transit time of a pulse 1

traveling from one transducer to another in the direction of flow is shonened, while the transit time of a pulse traveling against the direction of flow is increased. The fluid velocity I

and the ultrasound propagation velocity of the fluid at rest can be determined simply by measuring the transit times of the two pulses in the fluid and the distance between the two transducers. The algebra for these determinations is given in Figure 3-3.

It is impractical to measure the transit times in the fluid directly. The actual transit times measured include delays associated with pulse generation, the transit delays of the cables connecting the electronics to the transmitting and receiving transducers, the transduction delays of the transducers themselves, the delays through the [

]

transducer wells, and the delays of the pulse detection electronics. These delays are measured and subtracted from total transit time before the sound velocity and fluid velocity calculations of Figure 3-3 are carried out.

j Given measurements of transit times, non-fluid delays, and distance, then, the velocity of the flowing fluid along a straight line between the two transducers and the ultrasound velocity in the fluid at rest is determined, with an accuracy limited only by the accuracies of the transit time, time delay, and distance measurements. In a configuration like that in Figure 3-3, however, what has been measured is the fluid velocity projected along an acoustic path (as well as the at-rest sound velocity along that path). If an accurate ultrasonic flow measurement is to be effected, the velocity along the acoustic path or paths must be related to the axial fluid velocity; it is the axial fluid velocity, integrated over the pipe cross section that determines the volumetric flow through the pipe. As will be seen, the LEFM/ determines mass flow from volumetric flow and from fluid properties based on the sound velocity measurement. The four path chordal LEFM/ described in this topical report integrates the fluid velocity numerically, in what can be thought of as a three step process:

1) The fluid velocity along each path is used to determine the axial velocity that

)

produced it. This is accomplished by dividing the fluid velocity along the path by the sine of the angle between the path and a normal to the spool piece axis, as illustrated in Figure 3-4.

2) The axial velocities projected onto each of the four paths of the LEFM/ are multiplied by the respective projections of the chordal paths onto the spool piece cross section, forming an axial velocity / length product for each of the paths. These four velocity / length products are weighted, summed, and multiplied by the intemal diameter of the spool piece. This process follows the rules of a numerical integration technique [.

~

.] The result is a close approximation to the true volumetric flow.

I

3) 'Ihe result of the numerical integration is multiplied by a profile factor. The profile factor is determined in a full scale test of the spool piece in a certified hydraulic facility. The test is roughly analogous to the calibration test of a flow nozzle, except 3-2 L

i that the LEFM/ profile factor test typically includes a full scale model of the upstream feedwater system hydraulic configuration. Such modeling is rarely performed for nozzle tests. Although the profile factor correction to the numerical integration is small, it is important in an instrument whose overall accuracy is in the range of 0.5% ofindicated flow. The profile factor accounts for To determine the mass flow, the volumetric flow is multiplied by the average density of the fluid. Density is determined in a three step process:

1.

The sound velocity for each acoustic path is determined from the mean transit time less the non-fluid delay for that path and the path length. The mean sound velocity for the fluid flowing through the spool piece is determined by combining the sound velocities measured for each path, using appropriate weighting factors.

2.

The LEFM/ determines the fluid temperature to within 1* Fahrenheit. As shown in Figure 3-5, temperature is determined using a correlation relating it to the mean ultrasound propagation velocity of the fluid and the pressure (pressure is a constant input to the LEFM/). The correlation is supported by the literature (Reference 4), and by data collected in numerous LEFM installations (summarized in Appendix C, Verification of LEFM/ Temperature Correlation).

3.

The LEFM/ determines density from the fluid pressure and the mean fluid temperature, using a correlation based on the ASME Steam Tables, Reference 5.

i 3-3

3.3 The Chordal LEFM/ Algorithm Appendix B, Principles ofChordal LEFM/ Flow and Temperature Measurements, uses the principles outlined above to derive the LEFM/ mass flow algorithm. This algorithm is as follows:

1

{

3-4

l i

l l

An examination of the algorithm shows that the uncertainties in chordal LEFM/ flow l

measurement can be categorized as follows:

l l

1) Hydraulic uncertainties-the uncertainties associated with measuring, in a test facility, the profile factor for a specific installation and applying this factor to an LEFM/ measurement system in a plant;

2) Geometric uncertainties-the potential biases associated with imperfect knowledge of the dimensions of the measurement system-the angles, lengths, and placements of the acoustic paths and the intemal pipe diameter. It will be demonstrated that for most uprates covered by this report, most biases due to geometry are imbedded in the profile factor and do not require separate accounting;

3) Time measurement uncertainties -the potential biases and randomly varying errors associated with measuring the transit times of the acoustic pulses, including the measurement or calculation of non-fluid delays in the energy transmission path; and

4) Correlation uncertainties-the biases in the correlations relating fluid temperature to its sound velocity and pressure, and the density and enthalpy of the fluid to its temperature and pressure.

3.4 Accuracy of Calorimetric Power Determinations With the LEFM/

Table 3-1 summarizes the uncertainties for the determination of volumetric feedwater flow, properties and thermal power using the measurements of the LEFM/. The table shows how

)

these uncertainties would apply in the calorimetric power determination for a typical 2-loop PWR or BWR. It shows that, in a typical plant, a power determination with the LEFM/

would be accurate to within t 0.6%, on a 95% confidence level basis.

An overview to the approach to the calculation of the uncertainties summarized in Table 3-1 is given in Appendix D, Identifying and Bounding the Uncertainties in LEFM/ Flow and Temperature Measurements. Appendix E, Basisfor Uncertainties in Determining Thermal Power with the LEFM/provides detailed calculational bases and assumptions for each uncertainty listed in the Table (it should be noted that for simplicity of presentation, many individual uncertainty elements have been combined in Table 3-1). Appendix E also derives the sensitivity coefficients which relate the uncertainties in specific LEFM/ measurements (e.g., time of flight of an acoustic pulse) to the uncertainty in the power determination. These coefficients are applicable to all LEFM/ chordal feedwater installations. The LEFM/ and plant configuration assumed for the calculations of Appendix E and for Table 3-1 will, in general, be bounding for most plants applying for a thermal power uprate on the basis of this topic'al report. In particular, an internal pipe diameter of 14 inches was chosen for the calculations; this diameter is at the lower end of the range for main feedwater lines. LEFM/

measurement errors will be smaller for larger diameter feedwater pipes carrying flow at velocities comparable to those assumed in Appendix E.

3-5 L___________--

The LEFM/ is able to support a thermal power measurement accuracy of 0.6% because its operational performance, and particularly its flow measurements, agree very clor ely with the physical principles on which these measurements are based. This agreement between principles and performance is demonstrated by the survey of calibration data for LEFM's in Appendix F, A Survey ofHydraulic Performance ofChordal Leading Edge Flow Meters.

This appendix describes the results of tests that have been performed on chordt.1 LEFMs at certified hydraulic facilities over the past 22 years. These tests have been performed for hydraulic configurations ranging from straight pipes to locations downstream of single bends, compound bends, headers and more complex fluid system components. The tt:st data provide the bases for budgeted profile factor uncertainties due to the facility itself, the LEFM electronics used for the hydraulic test, and the uncertainties associated with e(trapolating test facility results to feedwater system operating conditions. They also provide insight into the effects of upstream hydraulic geometry on the profile factor and on the velocity profile measured by the individual chordal velocity measurements of the LEFM/.

Table 3-1 Thermal Power Determination Uncertainties With a Chordal LEFM/

in a Typical 2-loop PWR or BWR I

i F6

r s e e cl ub d a s c n

r ra t

te T

mt f

s m oi r

o o

r s

as J

s t

r n

n t

e to l

a c

nc Ur o

e t

sd r

a n fp r a a

l g

s t

t e m

r Un n

ow tn i

e nu a e e

s e

C t

t t r

t l

r s

U y

S E

/M sso F

r e

r Md.

E cu d

L

[

sn r

e a

t h

9 e

m t

, M a

f e

t o

r s

A j' n

s

... s w

o o

i {

D to

'..k h

.),,

re P

R cu d

d s

s a

e n

n r

s a

7 r

kD" lu 7

l t

p 4

m g

c 3

f a

ic ts i

e n

t o

a s

e" l

m a

p

's r

U t

,I

. q U

e QN -

hc i

S 4, '-

ie k'

t U

sop m

,. h',

S, o

• i C

p 1

\\

.\\ $ 3 \\ g; g N e 6 i" ru ig '~ F g

1. 's

} f c '1 ~ (.

i l'

, %. )m,. l i

n go ini t s a ,7. e.i s c f c.n -,1 oe f r v ws la.in p e t e. .ti l i n w. .ig n U q. DO c ~ m. i n ^ o r t ce lE / 5 M F e ) E ru d t y ta p L d r n s n m ir.i e n o p e e i f Dt o a a u e s s t o s T io e u s pe ter r l p t t P w - a a r a o D o s o d w cd eC e 5 l h s d i I c l P t a x r e o a ec ie r( t ms d u u r et n d a s h a en t ts r a c a r se m i s t or ' n tat a t at r e s o e y s s i ls b s M t te lu s e ' a m o". y u m c p e r

• g, r

m i i s u d c s v e s la e p n ic i 1 P ; a,, t o o c h a;, a y o e te" C r a c e w C n c d e w u tel d su' J, p S o "' f l t u s e o s r r A" ta& pi c F I n d, 2 r l d c 3 or U _ _;: e;l-te e w a t r no e en la p r P u e y g lal ig a c p n _ 5 a o s ~ l t t F u h ac .l

s d

or e V ~ ts e f t d. e [8 l a e 5 n u e P e ta I

4 ? . Trarrsducer A PATH f^'~- g c+ g ^h v,,, L,,rn gg.l, Pam Q. g to V ~ BA ~ C - ^* s g \\ ABh PATH = lPATH Transducer g AB BA L,rg C=1r p i' t + tg, As . C = sound velocity of fluid at rest i l l l k Figure 3-3. LEFM/ Path Geometry l 39 L_E_-__u____.____.____________..__ . _ _ _... _ _ _ _ _. _ _ _ _ = _ _ _ _ _. _ -. _ _. _ _ _. -____________.______m

ll l llllli

111Ill,

( T 3 Y m. y . 5 2p e. Y 1-X s e b sgn ( is u oH re c u d y - s e na ic s ett r tn# la ip d i T o l s ca c t o e w n i B lo n x u 't J le o g f p a n S a e l c 0 t e 1 i P 3 lo .f nI 1, p n o I 1 S i / iI I. M a ts FEL 4. le e 3 r r tsr e e ey c c e scl s ub ug u r e d n dl u ed m si s u l ia s a u a o ig ns e n s nd n t a s r o r St a Th T m F r 3 T 1 q 9 0' ' llll\\'l

!J l 08 4 0 6 \\ \\\\ is is 4 0 0 I 0 p p 44 e 0 0 0 5 0 v 2 r 1 I 4 u A% I 4 n 0 C 0 0 o 8 i I 3 t a 06 l I 3 re r 04 o I 3 C 02 e I 3 r I 3

• a 0

u ) t 0 F ( r I 2 p T 0 e e 1 I 8 r 2 u p 1 m 3 t 0 a r 6 e e m 0 m I 2 T 4 e /M 02 I 2 F 0 E I 2 L 0 = I 0 8 1 5 0 3 I 6 1 e i 0 r 4 ug 1 i 0 l 2 F 1 0 I 0 1 0 l 8 06 - *wt I e"' ll l lt l!

Section 4 CURRENT CALORIMETRIC POWER DETERMINATIONS l 4.1 The Mass Flow Algorithm for Nozzle-Based Instrumentation In most US nuclear plants, feedwater flow is determined from the differential pressure developed across one or more flow nozzles in the feedwater lines together with the temperature and pressure of the feedwater. As discussed below, such determinations are subject to uncertainties which do not exist for determinations based upon LEFM/ measurements, and are not capable of on-line verification as are LEFM/ measurements. In most PWRs there is one such nozzle for each steam generator; in most BWRs there is one nozzle in each of the two main feedwater lines supplying the reactor vessel. A few plants use a single nozzle measuring the total feed flow as a supplement to the measurements in the individual feedwater lines, or as the sole basis for the feedwater flow measurement. The inference of mass flow from the nozzle pressure differential rests on the conservation of momentum through the area change of the nozzle. Reference 6 describes the principles of nozzle-based flow measurements in detail and develops the following mass flow algorithm for a typical flow nozzle: Wr = K4..on [p * /1 - g

• Y
• F4
• d'
• S (4-1) where Wr mass flow rate through the nozzle, (lbs/hr)

= a dimensional constant to convert the units of measurement of Keimnsen = diameter, mass density, and differential pressure to flow units, (Vin /hr.) p feedwater density, a function of the temperature and pressure of = the water at the nozzle, (Ibm /cu. in.) discharge coefficient of the nozzle, dimensionless. Ce = ratio of the nozzle throat diameter to the intemal diameter of p = the pipe at the upstream location of the static pressure tap; (dimensionless) adiabatic expansion factor which, for compressed water, is Y = 1.000 (dimensionless) thermal expansion factor for the nozzle throat, accounting for F. = the difference in the diameter of the throat at operating temperature (when the flow is being measured) versus the diameter of the throat during calibration, when cold; (in/in). diameter of the nozzle throat at calibration,(in.) d = 41

1 l Ap difference between pressure measured at a static tap upstream = of the nozzle and a tap at the nozzle throat,(psi.) An examination of Equation 4-1 above shows that the uncenainty of a mass flow measurement based on a flow nozzle is detennined by:

1) Hydraulics-The nozzle discharge coefficient characterizes the actual vs. ideal change in fluid momentum from upstream pipe to nozzle. His change is affected by the fluid properties, particularly as they impact the thickness and character of the fluid boundary layer, and by the momentum content of the entering flow stream, particularly transverse momentum components that are determined by the configuration of the piping upstream of the flow nozzle;

2) Geometry-The flow calculated with the algorithm requires a knowledge of the diameter of the nozzle throat, d, and also the diameter of the pipe at the upstream pressure tap (for the determination of the diameter ratio p);

3) Property Measurements and Correlations-ne determinations of the thermal expansion factor, F., and the fluid density p, rest on measurements of fluid temperature and pressure (the effect of pressure on nozzle dimensions is generally neglected). The thermal expansion factor also requires a characterization of the effective expansion of the nozzle / pipe assembly with its temperature. The density determination requires a conelation relating density to fluid temperature and pressure (generally derived from the ASME tables for compressed water). The fluid temperature is typically measured by one or more temperature sensors in each feedwater line. Sensors used are generally resistance temperature detectors (thermocouple are used occasionally). Sensors of either type require the correlation of the measured electrical property (e.g., resistance for an RTD) with the inferred property, in this case temperature. Pressure is generally measured at the feed pump discharge or at the steam generator drum, using a Bourdon tube or other force measuring transmitter; and

4) Differential Pressure Measurement-The measurement of the pressure difference between the static tap and the nozzle throat employs one or more differential pressure transmitters. These devices measure the net force between the taps using a calibrated elastic displacement or strain sensing transducer.

4.2 Accuracy of Calorimetric Power Determinations With the Curret. Instrumentation Table 4-1 is a summary of the uncertainties in the thermal power determination for the same l typical 2-loop plant used in Appendix E to calculate LEFM/ thermal power uncertainties presented in Table 3-1. The uncertainties presented for each of the instruments used in the measurement-the temperature, pressure and differential pressure instruments -include typical allowances for the uncertainties in secondary standards used in their calibration, for biases introduced by differences between the calibration condition and the operating condition (e.g., the uncertainty in the biases caused by pressurization of the differential l l 4-2 ) ~

pressure transmitter casing), for random drift and for drift due to systematic, environmental factors (e.g., changes induced in analog electronics by ambient temperature, self heating errors in RTDs). The total uncertainty in the thermal power determination for the typical 2-loop plant of Table 4-1 is il.4 % (versus 0.6% with an LEFM/). The calculations on which Table 4-1 is based are included as Appendix A, Basisfor Uncertainties in Existing l l . Calorimetric Instrumentation. The instrument uncertainties used as inputs to Table 4-1 have been estimated to characterize realistic performance representative of currently installed calorimetric instrumentation. Generally speaking, historical transmitter calibration data demonstrate the ability to meet or exceed the performance quoted in Table 4-1. The realistic performance characterized in Table 4-1 does not account for common biases due to off-normal operation of nozzle-based systems, however. These include, for example, the tendency of flow nozzles to collect corrosion deposits on their throats, commonly referred to as nozzle fouling, which significantly alters their calibration. The tendency of nozzles to foul is widely discussed in the technical literature (See, for example, Reference 7). This bias is not typically accounted for by utilities since it results in a conservative (high) calorimetric power measurement. Other mechanisms result in non-conservative biases, as discussed below and in Sections 5 and 7 of this report. Actual performance data for nozzle-based flow indications have been trended at many utilities over the past five to ten years due to the increased awareness of nozzle fouling and its effect on feed flow indications. These data have not yet been compiled to develop a generally accepted industry-wide picture of field performance of nozzles. However, Reference 8 cites a compilation of data comparing the mass flow indications of a large number of nuclear plant flow nozzles with the indications of ultrasonic instruments (including a number of LEFMs similar in principle to the LEFM/ described in this report). Data extracted from this reference are plotted in Figures 4-1 and 4-2. The basic comparison of nozzle and LEFM flow indications is shown in Figure 4-1. Except for a few outliers at -3.0% difference, the distribution of flow differences is normal. These outliers represent cases where flow has bypassed the nozzle throat, introducing a systematic error. When these outliers are excluded from the nozzle / ultrasonic measurement comparison, results indicate nozzle-based flow measurement accuracies are generally consistent with the results of Table 4-1. Reference 8 also contains comparisons of feedwater RTD temperature indications with ultrasonic feedwater temperature measurements. Figure 4-2 summarizes those data, which indicate that in-service accuracies of the RTD temperature measurements are also generally consistent with the results of Table 4-1. The uncertainties presented in Table 4-1 are based on 95% confidence intervals, in accordance with the practice recommended in the ASME and ISA standards (References 2 and 3), and provide results that are directly comparable to Table 3-1. This implies that the 43

f probability of the power in the typical 2-loop plant of the table exceeding its licensed rating by more than the 1.4% calculated total uncertainty is about lin 44. (Appendix G). l The probability estimate above rests on the assumption that the data characterize the performance of the instruments on a continuous basis. The calibrations of the transmitters that transduce the pressure, differential pressure and thermometer resistance into electrical signals representative of the calorimetric process variables can be periodically checked. Normally such calibrations are performed once following each fuel cycle. But it is l impractical to check these calibrations continuously, nor is it possible to confirm the calibrations of the key primary elements for the thennal power determination--the flow nozzles and the resistance temperature detectors. Thus, while the conclusions from Table 4-1 are plausible, they cannot be confirmed on a continuous basis. Table 4-1. Thermal Power Determination Uncertainties with Current Instrumentation in a Typical 2-Loop PWR or BWR 1 Loop Uncertainty

• 2 Loop Uncertainty
• percent loop power, percent total power, 95 % confidence interval 95 % confidence interval

1. Hydraulles 0.5%

0.35 % Discharge Coefficient, including as built dimensions

2. Geometry 0.07 %

0.07 % Thermal expansion

3. Instrumentation 1.5%

1.3%

Transducers for differential pressure, temperature, and pressure used in the flow determination.

4. Feedwater Density

0.05%

=0.05% Correlation relating density to temperature and pressure.

5. Subtotal

1.58 %

1.35 % Mass Flow Uncertainty *

6. Feedwater Enthalpy

0.35 %

0.25 % Temperature and pressure instruments and l enthalpy correlation

7. Steam Enthalpy 0.23 %

0.23 % Pressure instruments and moisture uncertainty

8. Other Gains and Losses 0.07 %

0.07%

9. Total Power Determination Uncertainty

• 1.67 %

1.41 %

• The basis for all ux:rtainty elements in this table and for the combination of elements is contained in Basis for Uncertainties ' histing Calorimetric Instrumentation, Appendix A. It is noted that several elements in the table (e.g., itenc

,, and 7) contain both systematically related and uncorrelated components. 44 f. l w_-__-______________-._.

5 4 E0 M 4 F E L 5 n 3 a h 0 t e 3 ro m 5 2 le ) 0 le z z z o 2 z N on 5 /) 1 M F 0 E 1 L n a il il 5 0 le e z M zo n 0 ( ( e 5 c 0 n M e F r 0 e E i L f 1 f d n 5 a t h n 1 t ec s 0 r s 2 P e le e I 5 lz z 2 o N 0 3 5 3 6 4 2 0 8 6 4 2 0 8 6 4 0 2 2 2 2 1 1 1 1 1 j"-_.Rgj35 m8.$=

5 4 I 0 4 I 5 3 I 0 3 s 5 n 2 i o tac D 0 di T 2 n R I n er a u h 5 t t a 1 r e e ro p m m 0 )F eT M 1 F ( M E e L c F 5 n E 0 e r L e d f i n 6 f n d a 4 a 0 e e D r M u T ta R 5 r f e 0 p o m no D e s T 0 T i R r 1 a n p a m h o t 5 C ss 1 le 2 M 4 F 0 e E 2 r L u ig F 5 2 I 0 3 I 5 3 2 0 8 6 4 2 0 1 1 E$iS $$ B j5z

Section 5 BASIS FOR THE THERMAL POWER UPRATE WITH THE LEFM/ 5.1 Comparison of the Flow Algorithms for LEFM/ and the Current Instrumentation ~_ A comparison of the LEFM/ mass flow algorithm, Equation 3-1 defined in Section 3, with the nozzle mass flow algorithm, Equation 4-1 defined in Section 4, reveals some important similarities between the two instrument systems:

1) The algorithms of both systems rest on fundamental principles-the nozzle on conservation of momentum, the LEFM/ on distance traveled equaling the product of velocity and time. The calibration coefficients employed by each system-the discharge coefficient for the nozzle, the profile factor for the LEFM/ - are typically within 0.01 of unity, indicating that the actual performance of the instruments, when operating as designed, closely resembles the performance predicted by simple physical principles (this is not the case for other methods of flow measurement, such as those based on cross correlation of radioactive or turbulent disturbances). This attribute is extremely important. Because ofit, any significant error in the calibration process-for example, a problem with the calibration facility-will be detected,

~ ince it will result in a readily detected discrepancy between the calibration coefficient s determined by test, and that based on first principles.

2) Tne sensitivity of both systems to hydraulic profile is embedded in their calibration coefficients.

3) As-built dimensional biases of both systems are usually built into their calibration coefficients. This is inherent, because the calibration coefficient for each instrument

- the discharge coefficient for the nozzle and the profile factor for the LEFM/ -is normally back-calculated from the flow measurements ofthe calibration standard and the as-measured dimensions of the nozzle and spool piece respectively.

4) Both systems are dependent on their dimensions remaining in the as-calibrated state

~ (except for changes due to thermal expansion), though their sensitivities differ. The nozzle relies on throat diameter and intemal pipe diameter; the LEFM/ relies on face-to-face distances, intemal diameter, and path angles. ~

5) Both systems require a temperature-based dimensional correction for thermal expansion and hence rely on a knowledge of the coefficient of thermal expansion of the materials of construction of the primary element (i.e., the pipe nozzle assembly and the LEFM/ spool piece).

5-1

6) Both systems rely on correlations relating properties of the process fluid to measured state variables - density to temperature and pressure for both instruments, and temperature to sound velocity and pressure for the LEFM/.

Immediately after it is calibrated, a flow nozzle is capable of providing measurement accuracies in the 0.5% range, providing the differential pressure and fluid temperature measurements are made with laboratory grade, calibrated instruments (see for example the discussion of turbine heat rate testing in ASME-PTC-6, Reference 9). The less impressive flow measurement performance reflected in Table 4-1 and normally quoted for nuclear power plants-around 1.2% to il.5% for mass flow-comes about because of allowances for drift in service of the process variable measuring systems, and, in some instances, for allowances for nozzle calibration changes in service (e.g., due to corrosion deposits at the nozzle throat). The above comparison impi,ies that the chordal LEFM/ should be capable of qualitatively similar performance--accuracy in the 0.5% range. This report demonstrates that the conditions necessary to achieve such performance are met by the instrument described herein. Specifically, it is shown that:

1) The LEFM/ calibration complies closely with the physical principles on which it is based;

2) The LEFM/ provides assurance that critical dimensions of the spool piece do not change significantly in service; and

3) the accuracy with which the key variables of the LEFM/ -the times of flight of the acoustic pulses -are measured is significantly and demonstrably better than the accuracy with which the key variables of the nozzle-based measurement-differential pressure and temperature-are measured. The LEFM/ times are measured digitally with high precision.

5.2 Improvements in Calorimetric Accuracy: Basis for the Power Uprate As can be seen by comparing Tables 3-1 and 4-1, the LEFM/ provides measurements of feedwater mass flow and temperature leading to an uncertainty in thermal power of. 0.6%, substantially better than the 1.4% obtained with current instrumentation. To assess the increase in thermal power rating appropriate to the use of the LEFM/, it is first useful to interpret the meaning of the data of Tables 3-1 and 4-1 on a probabilistic basis. When they developed standards for the measurement of steam turbine heat rate in power plants, the ASME performed a series of Monte Carlo analyses which demonstrated that, if the uncertainty elements of a measurement system are calculated on a 2 standard deviation basis, the uncertainty in the overall measurement that results is characterized by a normal distribution with 2 standard deviations equal to the root sum square of appropriately weighted individual elements. 5-2

This result held even when the uncertainties ofindividual elements were not normally distributed. For example, a particular element might be characterized by a " roulette wheel" (flat) distribution between defined uncenainty bounds. It was subject only to one condition: that no single element dominate the calculation of the overall uncenainty. While it is not obvious, the tabulations of Tables 3-1 and 4-1 meet this condition. The profile factor uncertainty of the LEFM/ in Table 3-1 appears dominant, but is, in fact, made up of[ ]. Similarly, the instrumentation allowance for the current system in Table 4-1 appears dominant, but is in fact made up of numerous elements in several instruments - e.g., calibration and test equipment allowances, uncertainties in zero and span corrections for differential pressure transmitters operating under pressure, analog transmitter drift allowances, etc. Hence, it is concluded that the overall uncertainties described in Tables 3-1 and 4-1 are likely to be normally distributed. Furthermore, the sensitivity of the results to the nature of the elemental uncertainty distribution has been investigated as described in Appendix G. This investigation shows that the, distribution of the total uncertainty is likely to be normal whether the contributors are each normally distributed or distributed in roulette wheel fashion. Table 3-1 implies a distribution wherein one standard deviation of LEFM/ uncertainty is about i0.3% full power. As shown in Table 5-1 below, with this distribution there is essentially no chance (less than one in 3 million) that an operator using the LEFM/ to determine thermal power will exceed a power level 1.5% above that to which he is controlling (the odds have been computed on the basis of 5 standard deviations, Appendix G). Similarly, Table 4-1 implies a normal distribution of nozzle-based uncertainty with one standard deviation of 0.7%. As shown in Table 5-1, the odds of exceeding a power 3.5% above that indicated by the current instrumentation are similarly small. The foregoing is not intended to imply that the 2% overpower margin of Appendix K and Regulatory Guide 1.49 is not adequate. On the contrary, the implication of Table 4-1 is that the odds of exceeding the power for which core safety has been analyzed by as little as 1.5% are vanishingly small. It is clear from NRC discussions regarding adoption of revisions to Appendix K (Reference 10) that there is significantly more safety margin in loss-of-coolant analyses than 1.5%. Table 5-1. Probabilities and Odds Associated With Nozzle and LEFM/ Uncertainty Bounds Number of Venturi Nozzle LEFM/ Probability of Odds of Standard Bounds ( ) Bounds Operation Exceeding Deviations () Within Bounds Bounds on the High Side 1 0.7% 0.3% 68 % 1/6.3 2 1.4% 0.6% 95.4 % 1/44 3 2.1% 0.9% 99.7% 1/741 4 2.8% 1.2% 99.994 % 1/32,300 5 3.5% 1.5% 99.99994 % 1/3.3 million 5-3

To clarify the basis for a 1% power increase with use of the LEFM/ the results of Table 5-1 are shown graphically in Figures 5-land 5-2. Figure 5-1 illustrates the odds of exceeding uncertainty bounds on the high side on a logarithmic vertical scale. In Figure 5-1, the data for the LEFM/, shown on the top of the figure, are translated from a baseline of the current licensed power rating to a new baseline 1% above the current licensed power rating. The data for the current instrumentation are shown at the bottom of Figure 5-1, keeping the baseline at the current i licensed power level. Figure 5-1 illustrates that, even with a 1% power increase associated with use of the LEFM/, the probability of operation above 1.02 times the current licensed power level is still lower for the LEFM/ than for the current instrumentation and no power increase. Figure 5-2 is a curve which focuses on the power level at which the probability curves for the LEFM/ and current instrumentation intersect. The intersection occurs at a power level at which the probability of operation is the same. This power level occurs at about 101.8 percent of current licensed power. Thus, it is less likely that a plant would operate above 101.8 percent power using the LEFM/, with a one percentpower increase in place, than with the current instrumentation and no power increase. This is the basis for the one percent power increase with the LEFM/. There are two assumptions critical to the preceding discussion of thermal power margin. First, the necessity of an uncertainty distribution that is normal has been discussed and, based on the ASME studies and Appendix G, is satisfied. The second is that Tables 3-1 and 4-1 (or, in the case of current instrumentation, some alternative accounting) actually describe the performance of the instruments in service. Verification that the LEFM/ is operating within its design bounds is provided continuously, as mentioned above and discussed in detail in the next section. But there is no comparable on-line assurance that current instrumentation is operating within its design bounds. 5.3 Benefits of On-Line Verification To illustrate the benefits of on-line verification, Figure 5-3 shows the results of a survey of sustained overpower events reported in Licensee Event Reports from 1981 through 1997. The 51 identified events have been categorized by cause in order to examine whether they would have been preventable with the on-line verification capabilities of the LEFM/ system. Figure 5-3 illustrates that the LEFM/ with on-line verification would have prevented all significant sustained overpower events. Looking at the extremes, six cases have been reported in Licensee Event Reports where steady state overpower has occurred in an amount not consistent with the probability predictions implied by Table 4-1; i.e., operation at 1.4% or more beyond the licensed power level. The causes for these events are summarized below in Table 5-2. 5-4

Table 5-2. Sustained Overpower Events Above 1.4% and Their Causes LER Reported Reported Reported Cause of Event Number Power Duration Excursion 82-002 2.7% 46 days Differential pressure transmitter found out of tolerance. 86-025 1.5% 21 days Differential pressure transtnitter drift. 88-035 2%-3% 10 days Hole in venturi pressure tap. 91-012 2.09 % 5 years Core power calculation error; improper density compensation. 94-002 2.6% 8 months Perimeter bypass flow of venturi feed nozzles. 94-002 1.4% 10 months Perimeter bypass flow of venturi feed nozzles. In all cases but one, the sustained overpower event was the result of the instrumentation system (transmitters or nozzles) failing to operate as designed. The sixth case occurred because of improper density compensation. This case would also have been prevented bv use of the LEFM/,[ 1 It is the LEFM/'s ability to confirm on-line that it is performing within its design bases, as well as its high accuracy, that justifies a power uprate with its use. With an uprate of 1%, the LEFM/ will assure that the probability of exeeding the analyzed power level (i.e.,1.02 times the current licensed rating) by as little as 0.5% is nealigibly small. 5.4 Using the LEFM/ to Control Thermal Power With the existing instrumentation, for each feedwater flow measurement, the differential pressure transmitters provide an output proportional to the differential pressure across the flow nozzle. Resistance thermometers (or thermocouple) measure the feedwater temperature. Typically, these outputs are supplied to the plant computer where the density and enthalpy are calculated with the aid of synthesized ASME steam tables. The thermal power is then calculated, also by the plant computer. It is anticipated that a licensee will make use of the LEFM/ mass flow and temperature measurements by directly substituting the LEFM/ indications for the nozzle-based mass flow indication and the RTE temperature indications in the plant computer. The plant computer would then calculate en+halpy and thermal power as it does now. As an alternative, the calorimetric power can be manually calculated, using the LEFM/ indications and following a prescribed procedure, in either of these cases, the use of the LEFM/ will be limited to the calorimetric power determination. LEFM/ indications will not be used as a correction factor on venturi-based measurements that are in turn used to compute power in the plant computer, because the venturi nozzle drift allowance necessary to this process would lead to a thermal power calculation not in 5-5

compliance with the uncertainty estimates of this topical report. However, the nozzle-based feedwater flow will continue to be used for feedwater control and for any other function that it currently fulfills. Further, the nozzle indication may be corrected periodically on the basis of the LEFM/ indication, so that it serves as a backup in the event of an LEFM/ system failure. While this report is focused on operation at full power, it should be noted that the LEFM/ provides accurate flow and temperature indications from synchronization to full power. It may be used for thermal power determinations following synchronization at 10% to 15% power (when feedwater heating commences) and up to full power, with an accuracy better than the present instrumentation. 5-6

e n i t o ar 0 t p 0 a e1 u 0, t 0 n 0 e 3, m 1 3 u h / r it t w M s M in F E { tn L M 0 e F r h 3 t E r i

0..

u W L 1 C 0 d 0 n 3, a 1 2 n 3 o i t 0 a 0 tn 0, e l 0 m 0 u 3, r t 3 sn I 0 0 tn 3, 1 e 2 r 1 r 3 1 4 u 0 7 C 2 0 h t 1 1 i W l 47 s 7 l 5 eve L 14 4 re wo P g jU i n de ecx E 0 f 1 o

0..

y 1 t i il ba b 1U or P 1 5 e rug iF gn d i tn e t 011 a 0 e s r ,il I r n 1 I ; 0 ~ r r u e e c 1 w Cil op

_ M 0 3 0 1 r I e w o I P . M I de s ne I c 5 i i 2 L 0 1 2 0 1 f e o t a y t r in p )e u s i c n i d0 e V 1 0 ic 42 l n h t i t 1 i \\I e l w n s r e d r F v u e M \\ I c L \\g\\ f T o r E \\ e L \\ I t t n w g e o I c P I M I \\ g\\ 5 re g p 1 n I 0 i ( r d n 1 e e o I w e it o c a P x t I E ne f m I yo u r t t i s l n I i b i 0 a tn 1 b e 0 o r 1 r ru I P C I 2 5 er I u I ig F 5 0 0 5 0 5 0 5 01 2 2 1 i 0 0 0 0 0 0 0 0 0 0 0 0 0 0 _ M _ & gg,@s*f REa_y _ M T ll l ll

io r 0 d d o b d. 3 ta e e r ic t t o f a a r 1 i le le r r es e r r t v r n e e e e r r e n u u w m o l t t i a a pu b r r r n e e l t o p p a s h mr mr min b o o r t i e r e r r e w r e t r t h r e r e h le ot o t t n eo n b we wio r f 7 a o o lom lot t 9 f f la c 9 n e e u r u r v d r d e 1 t c u e e s el hl t i 1 r e ea n a P Fi F c Of 8 b 9 B 1 s, 'R 0 E b d. 2 L n 1 i de t ro peR s tn ev b E re b '" woprev O d b i en 0 a t 1 s d. u 1 S 3. 5 e ru ig F n n ii u b yr " " n a O i n a a n a O r t i r i, i _" " i r r i l t t t t d. i i 1

Section 6 ON-LINE VERIFICATION OF LEFM/ ACCURACY 6.1 Elements of the LEFM/ Measurement As described in Section 3.2 the uncertainties of a chordal LEFM/ used for the determination of thermal power can be grouped for analysis into four categories: (1) Hydraulic uncertainties (2) Geometric uncertainties (3) Time measurement uncertainties (4) Property determination correlation uncertainties The nature of the uncertainties in each of these major uncertainty categories--hydraulic, geometric, time measurement and property determination-and the means for ensuring that these uncertainties remain within design bounds are discussed in the paragraphs that follow. 6.1.1 Hydraulic Uncertainties Again recalling the LEFM/ algorithm, the profile factor relates the four weighted velocity. chord length products measured by the LEFM/ to volumetric flow. The uncertainties in profile factor include the following contributors: Calibration facility uncertainties: The facility's weigh tank scale, water thermometers, and fill-time-measuring hardware (all of which make up the flow standard for the profile factor determination) are traceable to NIST and have uncertainties typically certified to be less than [. ]. i i l l i I 6-1

) 1 6.1.2 Geometric Uncertainties An evaluation of the potential biases due to as-built dimensions of a spool piece is performed to ensure that the measured profile factor accords with its expected value, within dimensional and time measurement tolerances. There are two small geometric uncertainties that are carried for the LEFM/, as installed. One relates to the alignment of the spool piece relative to the upstream pipe. It is bounded by a calculational procedure, and, ifinstallation tolerances are met, is less than 0.1%. The second has to do with the uncertainties in the thermal expansion of the spool piece at the temperature at which it is used to measure feedwater flow versus its temperature at calibration. This uncertainty is also small and is primarily due to the uncertainty in the thermal expansion coefficient for the spool piece material. l . 6.1.3 Time Measurement Uncertainties L To provide an understanding of the uncertainties in the time measurements of an LEFM/, it 1 is necessary first'to provide a brief description of the signal processing employed by the instrument. The sequence of events in the measurement of the time-of-flight of a single electrical / acoustic pulse is diagrammed in Figure 6-1. i o2 j 1 -- _ j

4 1 I l l I l 4 i 1 ) 6-3

i 6.1.4 Uncertainties in Property Measurements The principal uncertainties in the determination of feedwater temperature, density, and enthalpy are: (1) The geometric uncertainties in the fluid acoustic path length. In the volumetric flow l. determination, biases in the face-to-face distances between transducer wells, due to [ fabrication tolerances and other effects, are en. bedded in the profile factor, however, a l knowledge of the face-to-face distance is required for the sound velocity determination (from which temperature is determined). Hence, the uncertainty in the face to face 7 dimension enters into the uncertainty in temperature and other properties. (2) The uncertainties in the time of flight measurement, including the uncenainties in the non-fluid delay, and (3) The uncertainties in the temperature / pressure / sound velocity correlation and the temperature / pressure / density correlation. (4) The uncertainty in fluid pressure, both of the feedwater at the location of the LEFM/ - spool piece and of the steam produced by the steam generator. The sensitivity of the LEFM/ indications to feed and steam pressures is very small. [ L 1 6.2 Bounding LEFM/ Elemental Errors . Table 6-1 lists the LEFM/ elemental errors and describes in some detail the processes by which they are bounded. The table items are taken from the detailed analysis of LEFM/ uncertainties in Appendix E. Table 6-1 shows that the bounds of every uncertainty that may contribute to the overall thermal power uncertainty are confirmed to be within their design values by one or more of the following methods: 6-4 a

1) they are inherently limited by design (e.g., uncertainties in temperature and density correlations);

2) they are confirmed to be within design bounds during pre-delivery testing (e.g., hydraulic tests to establish profile factor and its uncertainty, factory acceptance tests to establish reciprocity of time ineasurements of the LEFM/ electronics); and

3) they are confirmed to be within design bounds by tests during commissioning (e.g.,

F . l.

4) they are continuously monitored by the LEFM/ on-line verification system. Table 6-1 also indicates how the on-line verification system provides means whereby all uncertainties that might plausibly change in service are reconfirmed by at least one of the following methods:

. they are continuously checked by the LEFM/'s self testing feature; they are monitored and annunciated by LEFM/ diagnostics and alarms; and/or e they are confirmed to be within design bounds by periodic tests. The right-most column of Table 6-1 indicates whether conditions outside design bounds are alarmed. As shown in T:Jule 6-1, alarms are provided for all conditions requiring them. The on-line verification design is discussed funher in the section below. 6.3 On-Line Verification of Accuracy 6-5

a i l P l i i i 4 l i I 1 4 l J i i I i 4 6-6

r

---

o l

l ) \\ 6.4 Quality Measures in Design, Fabrication and Factory Acceptance Testing of the i LEFM/ l It is current utility practice to treat the measurement of feedwater flow, feedwater temperature, or other variables used in the determination of thermal power as not safety-related*, since they are functioning as calibration instruments. Use of the LEFM/ for thermal power detennination is classified in accordance with current practice, as a thermal power determination input which is not safety-related. However, mindful ofits importance to safe plant operation and consistent with its use as a precision calibration instrument, the LEFM/ system is designed and manufactured under Caldon's Quality Control (QC) Program (Ref.13), which provides for configuration control, deficiency reporting and correction, and maintenance. Specifically: 3

1) All engineering calculations relating to the LEFM/ design or the determination ofinputs for the LEFM/ are documented and checked and reviewed by a qualified engineer independent of the engineer who prepared the calculation.

2) All factory acceptance tests are conducted in accordance with written procedures, with appropriate pass / fait criteria. Conformance of the LEFM/ with the acceptance criteria of the test procedures is confirmed by the independent observations of at least two engineers.

3) All hydraulic calibration tests are likewise conducted in accordance with written test procedures. Again, conformance of the LEFM/ with the acceptance criteria of these procedures is confirmed by the independent observations cf at least two engineers. In addition, the calibration facility issues an engineering report summarizing and certifying the laboratory test results.

4) ne hydraulic test facility measurement apparatus, including the weigh tank, scale, time measurement equipment and temperature instruments, are traceable to calibration standards at the National Institute of Standards and Technology (NIST).

• In seme plants, feedwater flow and/or temperature instruments are used as direct inputs to the reactor protection system or for another automatic safety function. In these cases. the instruments are classified as safety-related. and would continue to be used for these functions. The LEFM/ would not be used for these functions. Its use would be limited to thermal power determination.

6-7 c_.

l

5) Factory acceptance test equipment, including oscilloscopes, precision time measurement standards, and tools used to measure the critical dimensions of the spool piece, are traceable f

to NIST calibration standards, and measurements are made and documented in accordance j l with controlled procedures. j l

6) LEFM/ installation and alignment (commissioning) tests are perfonned and documented i

in accordance with controlled procedures. Commissioning data are independently reviewed l by an engineer qualified to perform this task. This reviewer certifies that acceptance criteria are met. The oscilloscopes used in performance of these procedures are also calibrated to traceable (NIST) standards. ] I i Although its use for calorimetric input is not safety-related, the LEFM/ systen. sofhvare was developed and is maintained under a Verification and Validation (V&V90 gram car.-istent with IEEE 7-4.3.2 and NQA-2a1990, Subpart 2.7. The V&V Program follows a spiral develop.nent methodology, appropriate for an existing product with subsequent improvements. This methodology permits iterative review and improvement of design documents such as specifications, test procedures and design descriptions, while maintaining design documentation self-consistent by performing traceability analyses. He V&V Program was applied to all system i software and firmware, and includes detailed code reviews. In addition, the entire LEFM/ system was evaluated using an Integrated System Design Review. The V&V Program documents are maintained under revision control at Caldon, and have been audited by several utilities and by an independent quality assurance consultant. NOTE: Table 6-1 is proprietary in its entirety. h 6-8 L___-_-________________-___

l l t I l l l l 1 1 i 1 i ) 1 1 i i 1 l 1 i i h. i 1 I I I r l I i ._ _J

l l l o-i i i L __ __._

I l r i i i I l 1 i i 1 l r i I 1 l l l j l 1 1 6 l l M m I I i l l I l i i r h 1

f. r h Yc

l l l l l t 1 I Figure 6-1. LEFM/ Acoustic Pulse Transit Path t \\ l \\ t I 6-13

I I l Section 7 RELIABILITY 7.1 History of LEFM Technology 1 1 The original commercial design of the LEFM chordal transit time flow measurement system was developed by Westinghouse in the early 1970's as an outgrowth of technology used in the Polaris Missile System. Its earliest cornmercial applications were made at hydroelectric power facilities, to measure penstock flow. In 1973 and 1974, the technology was applied at Prairie Island Unit 2 to measure reactor coolant flow. Spool pieces were welded into the cold legs downstream of the elbows at the outlet of each steam generator. Hydraulic tests were conducted at Alden Research Laboratory to measure the l profile factor for the configuration, and the accuracy of the resulting volumetric flow measurement was determined. The temperature of the coolant at the spool piece location was also measured and the accuracy of this measurement was estimated. The flow measurements made by this LEFM were used as a basis for calculations of reactor coolant flow in subsequent j Westinghouse PWRs. The original Prairie Island Unit 2 installation has not been in continuous l service, but has been periodically tested and is still operable. 1 In the late 70's and early 80's, Westinghouse provided chordal LEFM systems to eleven U.S. nuclear plants for feedwater flow measurement: 1 i Caldon, Inc. purchased the LEFM technology from Westinghouse in 1989. Subsequently, design improvements to the chordal system signal processing and operating software, firmware and i electronic hardware have been made. These improvements have been incorporated into the LEFM design over the years since 1989. l

In addition to improving the chordal LEFM system, Caldon also developed an externally mounted version of the LEFM which uses the same fundamental transit time principles, similar software and identical electronics. However, the externally mounted LEFM fires acoustic pulses l through a mounting wedge, solid couplant and pipe wall, through the fluid, and then through the l pipe wall, couplant and wedge again before being received by the opposir.g transducer. Because of these intervening media, and because of physical limitations which require mounting on the l pipe diameter as opposed to chordal locations, the accuracy of these extemal meters is in the l range of 1 %, as opposed to the chordal meter accuracy of about l ]. The superior l accuracy of the chordal system is the basis ofits use for this power uprate program. Nevertheless, because of the extemal LEFM's good accuracy, its relative ease ofinstallation and the ability to install it on-line, and because the LEFM is not susceptible to the fouling biases l which cause venturi calibration errors, utilities have been installing and using the extemally l mounted LEFM for feedwater flow measurement. Caldon has permanently installed external systerns in fourteen plants for feedwater flow measurement in the U.S. and abroad. l 7.2 Nuclear Operating Experience With the Chordal LEFM; Component Reliability Table 7-1 summarizes hours of operating experience and failure data for Westinghouse and l Caldon chordal LEFM system components installed to measure feedwater flow in nuclear power l plants. There are a total ofnearly [ ] hours of operating experience using the chordal l LEFM in feedwater applications. Table 7-1 also shows calculated mean time between failure on a component level, and on a system level. [ l l 7.3 Nuclear Plant Operation With LEFM/ Out of Service i i l 7.4 LEFM Operating Experience in Other Applications l While there has been significant experience with LEFM's applied to nuclear plant feedwater flows, the majority of operating LEFM systems are in non-nuclear applications. The accuracy I-and reliability of these non-nuclear LEFM systems has been excellent, and they are discussed below in support of the reliability and accuracy claims made in this topical report. There are more than 112 operating LEFM systems measuring water flows in hydroelectric penstocks and water and petrochemical plants in the U.S. and abroad. He LEFM configuration 7-2

installed in large (4 to 30 foot diameter) penstocks has intemally mounted transducers, drill-mounted on the inner diameter of the penstock to form the four chordal flow measurement paths. (In penstocks, the chordal LEFM design is employed.) In all, there are an estimated 1,400 system-years, or over 12 million system-hours of operation of LEFM systems in these applications. Only 14 repairs or replacements of any kind have been made to these systems since 1990. In the mid 70's Westinghouse provided LEFM chordal systems for petroleum product flow measurement and leak detection in the Trans Alaska Pipeline. These systems employ 48 inch spool pieces and were the precursor to nuclear feedwater systems. The systems were installed at the suction and the discharge of each of 12 pumping stations along the 800 mile route. Caldon developed a prototype replacement system because of obsolescence in the existing system. Caldon's prototype was installed and tested in the 1991-95 time frame. During a recent 6 month proof test, in which one flow measurement was compared with the other, a repeatability of 0.07% ofindicated flow was demonstrated (repeatability is the key performance parameter for leak detection). In addition, a 3,000 to 1 turn-down ratio (maximum to minimum valid flow indication) was also demonstrated for both instruments. The reliability of the LEFM electronics in Alaska has also been excellent; the pipeline company is now commencing the procurement of replacement electronics for the remaining Westinghouse systems. 7.5 Survey of Overpower Events Reported in Licensee Event Reports (LER's) and Information Notices As described briefly in Section 5 of this report, a subject search of all Licensee Event Reports (LER's) was made to identify the root causes for past reported over-power incidents, and project the impact of operation with the LEFM/ on events of this type. This review is documented in Reference 14. The search identified 51 LER's which have been filed since 1981 for sustained operation in excess oflicensed power level. Figure 5-3 summarizes the results of the search. As shown in Figure 5-3, the LEFM/ with on-line verification would have prevented all significant overpower events. Use of the LEFM/ in accordance with this topical report would prevent transmitter and nozzle related over power incidents, as well as power calculation errors associated with its inputs to the power calculation. Thus, based on LER's reported to date, the number of overpower events would be expected to fall to about 35 % of the current experience rate with use of the LEFM/. In addition to overpower events, Licensee Event Reports and NRC Information Notices were also surveyed for events associated with ultrasonic instruments. Three such events were found. r The first was an overpower event reported in an LER. The cause was described as corrosion l l between transducers and pipe on a Controlotron externally mounted ultrasonic flow meter used for feedwater flow measurement (Reference 15). The buildup caused a reduction in signal strength, resulting in erroneous time measurements and operation at a reported 0.2% above licensed power level. This event would not likely occur with the chordal LEFM/, and ifit had, would have been detected and annunciated by one or more LEFM/ diagnostics before an l 7-3

i i unacceptable power error had resulted. Because chordal LEFM/ transducers are mounted inside { corrosion resistant wells, the interface between the transducer and the well is less likely to corrode or degrade than is the comparable interface on an extemally mounted ultrasonic instrument. Should a degradation in the interface occur, the reduction in coupling would produce an increase in the gain required to produce an acceptable signal. Should the gain increase approach an unacceptable level, the LEFM/ would alarm before the flow reading was affected. The second event was a miscalibration of venturi feedwater flow measurements using a Westinghouse model of the LEFM system that had been installed in the early 1980's. This event resulted in a 1% non-conservative feedwater flow indication, but did not cause overpower operation since the plant was operating at 98 % power for other reasons at the time. This event was caused by signal strength degradation which was not corrected by normal maintenance, and 3 was not reported by the older LEFM model in use. The event would have been prevented by use of the LEFM/ system, because the on-line verification features of the LEFM/ would not permit operation outside the conuhissioned set-up as did the older LEFM system in use. The third event was an application of a Controlotron extemally mounted (" clamp-on") ultrasonic flow meter during a high-head safety injection flow balance test. In this event, reported in NRC Information Notice 95-08 (Ref.16), the flow measurement was found to be in error by more than 1 % due to calibration in straight pipe and subsequent application in piping downstream of bends The piping configuration in the plant caused swirling flow at the measurement location, "hich caused an undetected error in the flow measurement. This type of event would not occur with the chordal LEFM/ for several reasons, enumerated below: 1 I l 1 ) l 7-4 { E_____________________________..__._______

,l1J1l li i l I-i' a a taD eru l ia F dna sruo I I gn i tare p O ra 5 e 7 lcu N la i cre m moC M FE t L 1 -7 e l baT l l l l l l ll llllL

Section 8 CONDITIONS FOR USE OF LEFM/ WITH 1% POWER INCREASE There are two conditions a licensee must satisfy, as a minimum, to make use of the LEFM/ System for a 1% thermal power increase in accordance with this topical report. The licensee must:

1. In."all an LEFM/ System certified in accordance with the Sample Certification shown in Figure 8-1. (An existing LEFM system can be upgraded and certified by Caldon for this purpose.)

'2. Appy for a license amendment to increase the referenced thermal power limit in the license and modify Technical Specifications to (a) ensure periodic confirmation of the absolute and relative accuacy of the time measurement, and (b) reflect the increased plant power level. Figure 8-1. Sample Certification CERTIFICATION THE LEFM/ SYSTEM MEETS REQUIREMENTS FOR INCREASED PLANT THERMAL POWER This certificate provides notice that the LEFM/ system, serial number installed in Unit at Nuclear Power Plant and commissioned on , meets the requirements of Topical Report Number The documentation supporting this certification is kept on file at Caldon, Incorporated, and includes the following: ^

1. Design Specification Checklist for accuracy, reliability and on-line verification capability.

2. Drawings showing verification of critical dimensions and materials.

3. Complete Hydraulic Test Report for site-specific piping model(s).

4. Peninent sections of Factory Acceptance Test Reports.

5. Commissioning Test Report, including alignment and benchmark testing.

6. Configuration Control Documentation.

Z This certification is made by , Caldon, Inc. date This certification and the supponing documentation has been reviewed by: , Caldon, Inc. date 8-1

Section 9 l REFERENCES 1 )

1. NRC Regulatory Guide 1.49," Power Levels of Nuclear Power Plants", Rev.1, December 1973.

2. ANSI /ASME Power Test Code 19.1 - 1985, Part 1 Measurement Uncertainty, Reaffirmed 1990.

3. ISA-RP 67.04, Part II," Methodologies for the Determination of Setpoints for Nuclear Safety.

Related Instrumentation", Approved September 1994. 4. [ l

5. ASME Steam Tables, Sixth Edition.

6. " Fluid Meters and Their Theory and Application (Report of ASME Research Committee on Fluid Meters)", Sixth Edition.

7. 8.

9. ASME Power Test Code 6," Turbine Heat Rate Testing"

10. Federal Register / Vol. 53 No.180/ Friday, September 16,1988/ Rules and Regulations 2

11. EPRI P EP Survey #95-003 of 4/28/95," Maximizing Reactor Thermal Power"

12. NRC SSINS #0208 of 8/22/80, Discussion of" Licensed Thermal Power"

13. Caldon Quality Control Manual.

14. [

l ~ 15. Licensee Event Report No. 92-018-00, Dated 09/24/92, " Licensed Power Exceeded Due to i Inaccurate Feedwater Flow Indication", Kewaunee Nuclear Power Plant.

16. NRC Information Notice No. 95-08," Inaccurate Data Obtained With Clamp-On Ultrasonic Flow Measurement Instruments", January 30,1995.

l 91

APPENDIX A BASIS FOR UNCERTAINTIES OF EXISTING CALORIMETRIC INSTRUMENTATION \\ The contents of this Appendix are proprietary to Caldon,Inc. ) i

APPENDIX B GENERAL PRINCIPLES OF THE CHORDAL LEADING EDGE FLOWMETER l The contents of this Appendix are proprietary to Caldon, Inc.

APPENDIX C VERIFICATION OF THE LEFM/ WATER TEMPERATURE ALGORITHM The contents of this Appendix are proprietary to Caldon,Inc. .I i l l 3 l f ( ) l I I l

l ): I I } I f APPENDIX D IN LEFM/ FLOW AND TESI T RE SI A \\ TS \\ 1 1 J The contents of this Appendix are proprietary to Caldon,Inc. I ) 1 l l l ) l

APPENDIX E BASIS FOR UNCERTAINTIES IN DETERAUNING THERMAL POWER WITH THE LEFM/ FLOW AND TEMPERATURE MEASUREMENTS The contents of this Appendix are proprietary to Caldon, Inc. l. u l-o l f l

APPENDIX F A SURVEY OF IIYDRAULIC PERFORMANCE OF CHORDAL LEADING EDGE FLOWMETERS The contents of this Appendix are proprietary to Caldon, Inc. 4

l 1 l ) APPENDIX G ESTIMATES OF THE PROBABILITY OF EXCEEDING POWER MARGINS OF UP TO FIVE STANDARD DEVIATIONS The contents of this Appendix are proprietary to Caldon,Inc.

APPENDIX H CALDON FLOW MEASUREMENT APPLICATIONS The contents of this Appendix are proprietary to Caldon, Inc. \\ /}}