ML20072J680

Forwards Responses to Questions in Re Proposed Tech Spec Change to Reduce Measurement Uncertainty for RCS Flow Rate

ML20072J680
Person / Time
Site:	McGuire
Issue date:	03/28/1983
From:	Tucker H DUKE POWER CO.
To:	Adensam E, Harold Denton Office of Nuclear Reactor Regulation
References
TAC-49164, NUDOCS 8303300119
Download: ML20072J680 (8)

Text

,

e e

DUKE POWER GOMPANY P.O. HOX 33180 CHARLOTTE, N.C. 28242 HALH. TUCKER TELEPHONE (704) 373-4531 www eneminewr war m eemoocciion March 28, 1983 Mr. Harold R. Denton, Director Office of Nuclear Reactor Regulation U. S. Nuclear Regulatory Commission Washington, D. C.

20555 Attention:

Ms. E. G. Adensam, Chief Licensing Branch No. 4 Re: McGuire Nuclear Station Docket No. 50-369

Dear Mr. Denton:

Attached are responses to the questions included in the March 17, 1983 letter by Elinor G. Adensam. The questions concerned the proposed Technical Specifi-cation change for McGuire Unit 1 to reduce the measurement uncertainty for RCS flow rate. Please note that these responses were discussed with the NRC staff on March 23, 1983. We will be available to meet with the NRC staff to discuss this further, if necessary.

Very truly yours, hb.

f,/h/

N Hal B. Tucker REH:jfw Attachment cc:

Mr. James P. O'Reilly, Regional Administrator Mr. W. T. Orders U. S. Nuclear Regulatory Commission Senior Resident Inspector Region II McGuire Nuclear Station 101 Marietta Street, NW, Suite 2900 Atlanta, Georgia 30303 oZl SJSEo@SSSShi P

/

1. Question r

Table 1 provides the equation for calculating flou. Table 2 provides the

?

uncertainties associated with calculation of loop flou. Since Table 2 does not include any uncertainty associated with primary systm net heat losses, please confim that this tem is neglected and results in a conservative determination of toop flou.

Response

Increasing primary side heat losses results in a less conservative determina-tion of RCS flow. Primary side heat loss uncertainties were assessed to be

,j zero due to the negligible effect of this parameter on the total RCS flow un-certainty.

2. Question r

Does the assessment of uncertainty in the measurement of feeduater flow agree i

vith any published industry standard or publication or was it developed based i

on physical equations for flou? It is expected that the uncertainty in feed-water flow measurement has been the subject of previous investigations. The staff would like to knou if such data was used and if not, what expertise vas used for this evaluation.

Response

The assessment of uncertainty in the measurement of feedwater flow was developed

[

from the physical equations. Feedwater flow measurement has not been the sub-ject of any Duke Power Company studies. We.have no knowledge of any existing industry studies. The expertise used in developing the feedwater flow uncertainty was Duke Power Company staff using standard uncertainty techniques. The assess-p ment of feedwater flow uncertainty has been discussed with Westinghouse per-J.

sonnel and they are in aggreement with the techniques used in determining it.

\\

3. Question The component error in differential pressure measurement made to determine l

feed:0ater flou is a very too number. Please provide the basis for the value used and how it is meas.wed.

P Re.sponse l

The differential pressure measurement made to determine feedwater flow is indeed a low number. The instrument specified was designed to yield a very low uncer-tainty number. The differential pressure cell uses a fused quartz bourdon tube.

The fused quartz crystal exhibits the lowest hysteresis creep and is one of the most perfectly elastic materials known. The feedwater differential pressure is measured by a Ruska DDR-6000 direct reading differential pressure gauge which I

uses a quartz tube cell. Manufacturer's specifications are as follows:

A) Standard Accuracy =10.008% RDG B) 90 Day Stability = 0.004% FS C) DVM Repeatability 10.001% RDG For full scale range of 50 PSID and a differential pressure of 8.9 PSID

.,x-_a u

_...w Pg. 2 of 4 j

t

[i Response (con't) c at 75% power level, the differential pressure uncertainty is 10.033%. contains the specifications of the Ruska DDR-6000.

i 4.

Question Please further clarify how the feeduater temperature is measured and the basis for its component error.

Response

Each feedwater temperature is measured by precision instrumentation consist-l ing of a continuous-lead type J thermocouple and an icebath reference junction.

The feedwater thermocouple output is measured by a Leeds and Northrup 914 Numa-i tron DVM with a 0-40 millivolt range. Refer to Figure 1 for a comparison between F

process feedwater temperature measurement and precision feedwater temperature measurement. Component uncertainties are as follows:

f Thermocouple Calibration 1 0.25'F Readout Calibration 1 0.03*F Standards Lab Calibration. Uncertainty - (USL)

I USL = / IU z.

_. / (!.25)2 + (1.03)2 0.25'F i

=

Additional conservatism is added to this measurement uncertainty:

2 X USL = 2 X 0.25'F = 0.5*F S.

Question The dominant consideration in the total steam enthalpy error is moisture carry-over, which is an estimated value. Also, the net pump heat addition uncertainty is an estimated value. Since these factors are onesided, i.e., not negative values, what is the basis fbr using the RSS method fbr their consideration in flow measurement uncertainty?

Response

The dominant consideration in the total steam enthalpy error is moisture As the estimated moisture carryover increases, it results in a carryover.

more conservative determination of RCS flow. The estimated moisture carryover used in the RCS flow uncertainty analysis was the Westinghouse guaranteed value l

at'the steam generator outlet, i.e., 0.25%. The expected moisture carryover per Westinghouse is 0.10% at full power, with less moisture carryover at lower power levels. Previous industry experience with this type steam generator indi-cate the moisture carryover may exceed the guarantee value at initial startup.

In view of this, the 1.25% assessed for the moisture carryover is a two sided figure and can be combined using the RSS method. Pump, power uncertainty has been conservatively estimated to be 2%.

In the actual test, pump power is a measured parameter with uncertainty contributed from the process instrumentation i

used to measure pump motor current and voltage. The uncertainty of these measure-ments are two-sided, thus the RCS pump power can be combined in the precision calorimetric RCS flow measurement uncertainty using the RSS method.

.. <..--.. e a.-, u

a... :. -

2,.--...

-.... -. =.

e Pg. 3 of 4 F

J L&N Numatron Plant Computer and Process Control 5

j oo 0

O Type J Thermocouple i

E i

sv-

+ Ambient Reference a

}

Icebath %

Junction V

V f

e 1

i Feedwater -+ l l

F-a.

L Thermowe'TTs

?

L A

i

!T r

1 l

l-1 l,

1 I

Figure 1 I

Final Feedwater Temnerature ficasurement Schematic

/

Process Versus Precision Comnarison I

a

Pg. 4 of 4 I

1 6.

Q'testion 1

Please provide further clarification of the measurement errors associated with hot and cold leg temperature measurements. H00 is DVM used to measure resistance and hou did you get the measurement span in *F and RTD' calibration in

%7 Be explicit how this measurement is made.

Response

The reactor coolant hot and cold legs temperature measurements are performed utilizing a digital resistance bridge and four lead resistance temperature detectors. Two of the leads are used to pass a current through the RTD while the other two leads measure the emf across the RTD. A three point calibration is performed at the factory on each RTD to determine the coefficients of the third-order polynomial curve which characterizes each specific RTD. The RTD resistances are then utilized v. the characteristic equations for the different RTD's to iteratively calculate RCS temperature. The RTD's used for the precision calorimetric are narrow range RTD's with calibrations spans as follows: hot legs =

?

530*-650*F, cold legs = 510*-630*F.

The accuracy of this calibration guaranteed by Westinghouse is 10.2% of the measured span, i.e., 10.2% X 120*F = 0.24*F.

The DVM accuracy is 0.15% of range, i.e., 10.15% X 120*F = 0.18'F.

7.

Question Please provide a copy of the interface requirements established by Westinghouse

^

that would be used by Duke Power Co., to assure that measuremente vould be made in a manner as assumed in the analysis and with the required accuracy.

Response

i In this matter, no interface requirements established by Westinghouse were utilized by Duke Power Company. Duke Power Company staff engineers have devel-oped the techniques utilized in this test during the previous twenty-five years in similar testing applications. The uncertainty analysis reflects actual test measurement instrumentation uncertainties. The test procedure defines the test techniques and instrumentation utilized in this test and provides assurance that future measurements will be made in the manner presented in the subject uncer-tainty analysis. All instrument and measurement uncertainties are conservative and consistent with the Improved Thermal Design Procedure (ITDP) study performed by Westinghouse.

b 9

e

~. -......

ATTAOBENT 1 Ruska DDR-6000 Specifications f

i h

l l

I I

e-Dir:ct RI. ding Prcmura G:gea 4 #

RUSWh

~

(

SPECIFICATIONS

• For full scale ranges to 2500 PSI REPEATABILITY:

0.002% F.S.

SHOCK TOLERANCE:

5 g/10 m.s. with No Resets, or STABILITY, 0.004% F.S.

15 g/8 m.s. Resetting Offsets UNEARITY:

0.001% F.S. (included above)

WARM-UP TIME:

2 Hours. Instrument can be "ON" RESOLUTION.

0.001% F.S.

indefinitely.

HYSTERESIS.

Not Detectable BOURDON TUBE 150% Full Scale to 1000 PSI SLEW RATE:

5 Sec. Maximum Full Scale PROOF PRESSURE:

125% Full Sca!e over 1000 PSI ANALOG OUTPUT:

DifferentiaI 0-11.5 Volts DC Max.

CASE WORKING Aluminum 1000 PSI Impedance,less than 5000 ohms.

PRESSURE:

Steel 5000 PSI TEMPERATURE RANGE:

Operate: 10 to 36*C CASE PROOF PRESSURE: Aluminum 2000 PSI Storage: 0 to 60*C Steel 20000 PSI BOURDON TUBE VOLUME: 1 cc Plus Fittings CASE VACUUM EFFECT:

Negligible CASE VOLUME:

180 cc Plus Fittings FITTINGS:

' %" NPTF Rear Bulkhead TILT SENSITIVITY:

0.002% FS perDegreeTilt STANDARD Bourdon Tube Relief Valve MATERIALS IN CONTACT Ouartz, Teflon,416 Stainless OVERPRESSURE set at 110% F.S. Case Relief Se. at WITH PRESSURE MEDIA: Steel, Polyethylene, Aluminum, PROTECTION:

10 PSI Brass, Buna N STANDARD FILTERS:

On Tube and Case MATERIALS IN CONTACT Aluminum, Brass, Copper, MOUNTING:

Standard 19" Relay Rack.

WITH REFERENCE MEDIA: Buna N, Polyethylene, Teflon, Cabinets Optional Quartz, Sapphire, Epoxy,416 SIZE:

19" Wide x 7"High Panel,16" Deep Stainless Steel, Alnico and Carbon WEIGHT:

27 Pounds Net,45 Pounds Gross PRESSURE MEDIA:

Clean, Dry Gas Compatible with Materia!s of Construction.

(Specifications based on 100,000 count fullscale

{

(Liquid - Special Order) instruments. Other models differ slightly.)

For full scale ranges above 2500 PSI, if different from above REPEATABILITY:

0.015% F.S-PRESSURE MEDIA:

Clean Dry Nitrogen or Gas STABILITY:

0.030% F.S.

Compatible witn Materials of LINEARITY:

0 020% F.S.

Construction. For Liquid Sers. ce, Pressure Medium is Silicone Oil.

SLEW RATE:

Consult Factory MATERIALS IN CONTACT Copper, Buna N, Polyethylene, WARM-UP TIME:

5 Hours (Can be left on WITH PRESSURE MEDIA: Teflon, Quartz, Sapphire, Epoxy, indefinitely) 416 SS, Alnico, Carbon Steel PROOF PRESSURE:

15,000 PSI Case,125% F.S.

Quartz MATERIALS IN CONTACT Quartz, Teflon,416 SS, WITH REFERENCE MEDIA. Polyethylene, Buna N FITTINGS:

NBS Femate for %" Tubing SIZE:

14"x12%"x16" For Cabinet WEIGHT:

55 Pounds Net For Cabinet

~

19"x12%"x16" For Rack Type 65 Pounds Net For Rack Type MODEL NUMBERS For Full Scales to 2500 PSI (Indicate Range and Units) 6000-801 - Aluminum Body 6000-802 -Steel Body For Full Scafes over 2500 PSI (Indicate Range and Units) 6000-803 -Gas Service-Cabinet Type s

6000 804-Gas Service-Rack Mount 6000-805 -Gas Service-Rack Mount with Manual Controller Manifold 6000-806 - Liquid Filled-Pressure Side 6001-21 Panel Blank-For Cabinet Mounting when DVM is not ordered.

Note: For 230 V, use -861, -862, etc. See page 5 for DVM Readout Model Numbers.

' Note: See " Definitions and Accuracy" page 23, for explanation of terms and discussion of accuracy.

4 dor 78

p.u.

_._m._..___..r..<_.___.m..-._,...._

NUS

. i

~

1 l

l DEFINITIONS AND ACCURACY f

DIRECT READING REPEATABILITY The ability of the instrument to reproduce outputs when conditions:

the same known pressure is applied to it under the same

a. Normal warm-up of instrument

~

conditions from either direction. This term includes

b. Zero setting daily I'

linearity on instruments up to 2500 psi and is valid for a

c. Based on 100,000 counts Full Scale with Full Scale

[i 24-hour period.

pressure not exceeding 2500 psi

d. Applied within the context of all other specifications W

DIRECT READING STABILITY n

}

The same as Direct Reading Repeatability, but extended TOTAL ACCURACY to a 90-day period.

To determine the total 24-hour accuracy of a mea-LINEARITY sure mode reading, the following terms must be added The maximum deviation of any output reading from the algebraically:

corresponding point on a straight line drawn through the calibrated end points and mid-point based on a 90-day Term:

Value, as Supplied by RUSKA

('.'.seriod.

a. Direct Reading i

p I

. SLEW RATE Repeatability 0.002% of Full Scale

'[

The nominal time to traverse from zero to full scale, or

b. Digital Voltmeter l

from full scale to zero.

Repeatability 0.001% of Reading 1

System volume, vacuum pumping time, and tempera-

c. Calibration 20.0055.% of Reading to 50 psi l'

ture will affect slew time in either measure or control Standard Accuracy 20.008%of Readingabove50 psi i

modes.

2' Stated values are average estimates.

After 90 days, the values of terms b. and c.will depend on the means of recertification employed by the user.

i p

NOTE: The foregoing definitions and the following For 90-day accuracy, substitute stability specifications

]

stated values are valid in accordance with the following for repeatability.

f i

0 H

1

?

li u

a i

I:

Na 78 l-23

.