ML20211B325

Requests That Proprietary Amended Response to RAI Re Proposed Amend to Allow Use of Elbow Taps for Measuring RCS Flow,Be Withheld,Per 10CFR2.790(b)(4)

ML20211B325
Person / Time
Site:	South Texas
Issue date:	09/12/1997
From:	Liparulo N WESTINGHOUSE ELECTRIC COMPANY, DIV OF CBS CORP.
To:	Collins S NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM)
Shared Package
ML20046D806	List: ML20211B325 ST-HL-AE-5752, Forwards Proprietary Amended Response to Request for Addl Info on Proposed Elbow Tap Tech Spec Table 2.2-1 & Section 3/4.2.5.Proprietary Encl Withheld,Per 10CFR2.790
References
CAW-97-1168, NUDOCS 9709250165
Download: ML20211B325 (38)

Text

._.-

,p s

Mg g $ygtggg Nucleat Setvices Division Electric Corporation so,333 Pittsbutgh Pemsytvata 16?30 0355 CAW 97-1188 SAE/FSE M-0222 September 12,1997 Document Control Desk U.S. Nuclear Regulatory Commission Washington, D.C. 20555 Attention: Mr. Samuel J. Collins APPLICATION FOR WITHHOLDING PROPRIETARY INFORMATION FROM PUBLIC DISCLOSURE

Subject:

' Revised Responses to NRC Questions on South Texas Project Elbow Tap Submittal'

Dear Mr. Collins:

The proprietary information for which withholding is being requested in the above referenced report is further identified in Affidavit CAW 97-1168 signed by the owner of the proprietary information, Westinghouse Electric Corporation. The affidavit, which accompanies this letter, l

sets forth the basis on which the information may be withheld from public disclosure by the Commission and addret:,2s with specificity the considerations listed in paragraph (b)(4) of 10 i

CFR Section 2.700 of the Commission's regulations.

Accordingly, this letter authorizes the utilization of the accompanying Affidavit by Houston Lighting and Power Company.

Correspondence with respect to the proprietary aspects of the application for withholding or the Westinghouse affidevit should reference this letter, CAW-97-1168 and should be addressed to the undersigned.

i l

l Very truly yours, N. J. Liparulo, Manager Equipment Design and Regulatory Engineering cc: Kevin Bohrer/NRC (12H5) t "The mission ofNSD is to provide our custe ners with pople, equipment and servsces that set the standards ofescellence in the nuclear industry."

l 9709250165 970918 PDR ADOCK 05000498 p

PDR

CAW-97-1168 AFFIDAVIT t

COMMONWEALTH OF PENNSYLVANIA:

$s COUNTY OF ALLEGHENY:

Before me, the undersigned authority, personally appeared Henry A. Sepp, who, being by me duly sworn according to law, deposes and says thet he is authorized to execute this Affidavit on behalf of Westinghouse Electric Corporation (" Westinghouse") and that the averments of fact set forth in this Affidavit are true and correct to the best of his knowledge, information, and belief:

~1 i

r fi Henry A. Sepp, Manager Regulatory and Licensing Engineering Sworn to and subscribed i

before e this /d day of _

,1997 s

<-i

/

Notarw MA ec seat W comrms,t,on exp$,ymg,yg Notary Public hr "

,:.an4 A%0Cutdn ot Wgg l

i

_ _ ~__

o 3-CAW-97 Il68 (1)

I am Manager, Regulatory and 1.icensing Engineering, in the Nuclear Services Division, of the Westinghouse Electric Corporation and as such, I have been specifically delegated the furtion of reviewing the proprietary information sought to be withheld from pub!!c disclosure in connection with nuclear power plant licensing and rulemaking proceedings, and am authorized to apply for its withholding on behalf of the Westinghouse Energy Systems Business Unit.

(2)

I am making this Affidavit in conformance with the provisions of 10CFR Section 2.790 of the Commission's regulations and in conjunction with the Westinghouse application for withholdimt accompanying this Affidavit.

(3)

I have personal knowledge of the criteria and procedures utilized by the Westinghouse Energy Systems Business Unit in designating information as a trade secret, privileged or as confidential commercial or financial information.

(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 Comm' ' an in determining 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 confidence by Westinghouse.

(ii)

The information is of a type customarily held in conddence by Westinghouse and not customarily disclosed to the public. Westinghouse has a rational basis for determining the types of Information customarily held in confidence by it and, in that connection, utilizes a system to determine when and whether to hold certain types of information in confidence. The application of that system and the substance of that system constitutes Westinghouse policy and provides the rational basis required.

Under that system, information is held in confidence if it falls in one or more of several types, the release of which might result in the loss of ar existing or potential competitive advantage, as follows:

(a)

The information reveals the distinguishing aspects of a process (or component, str:: Cure, toci, metbod, etc.) where prevention of its use by any of mcm. awn.i

-3 CAW-97.II68 Westinghouse's competitors without license from Westinghouse constitutes a competitive economic advantage over other companies.

(b)

It consists of supporting data, inclu?mg test data, relative to a process (or component, structure, tool, me'. nod, 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 Westinghouse, its customers or suppliers.

(e)

It reveals aspects of past, present, or future Westinghouse or customer funded development plans and programs of potential commercial value to Westinghouse.

(f)

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

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

(a)

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

(b)

It is information which is marketable in many ways. The extent to which such information is available to competitors diminishes the Westinghouse ability to sell products and services involving the use of the information.

(c)

Use by our competitor would put Westinghouse at a competitive disadvantage by reducing his expenditure of resources at our expense.

m,cunen,,

.. = -.

0

-4 CAW-971168 l

(d)

Each component of proprietary information pertinent to a particular competitive advantage is potentially as valuable as the total competitive advantage, if competitors acquire components of proprietary information, any one component may be the key to the entire puzzle, thereby depriving Westinghouse of a competitive advantage.

(e)

Unrestricted disclosure would jeopardize the position of prominence of Westinghouse in the world market, and thereb give a market advantage to the competition of those countries.

(f)

The Westinghouse capacity to invest corporate assets in research and development depends upon the success in obtaining and maintaining a competitive advantage.

(ill)

The information is being transmitted to the Commission in con 0dence and, under the provisions of 10CFR Section 2.790, it is to be received in confidence by the Commission.

(iv)

The information sought to be protected is not available in public sources or available information has not been previously employed in the same original 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 ar ropriately marked in " Revised Resp (mses to NRC Questions on South Texas Project Elbow Tap Submittal," (Proprietary), September,1997 for South Texas Project, being transmitted by liouston Power and Light Company letter and Application for Withholding Proprietary Information from Public Disclosure, to the Document Control Desk, Attention Mr. Samuel J. Collins. The proprietary information as submitted for use by llouston Power and Light Company for South Texas Project Nuclear Power Plants is expected to be applicable in other licensee submittals in response to certain NRC requirements for justification of use of RCS Dow veri 0 cation using elbow taps.

His information is part of that which will enable Westinghouse to:

wwnwo.un

=.

o

• CAW 97.ll68 (a)

Frovide elbow tap methodology.

(b)

Establish appropriate instrument uncertainties associated with elbow tap measurements.

(c)

Assist the customer to obtain NRC approval.

Further this information has substantial commercial value as follows:

(a)

Westinghouse plans to sell the use of similar information to its customers for purposes of meeting NRC requirements for licensing documentation.

(b)

Westinghouse can sell support and defense of RCS flow verification methodology using elbow taps to its customers in the licensing process.

Public disclosure of this proprietary information is likely to cause substantial harm to the competitive position of Westinghouse because it would enhance the ability of l

competitors to provide similar licensing support documentation and licensing defense services for commercial power reactors without commensurate expenses. Also, public l

disclosure of the information would enable others to use the information to meet NRC requirements for licensing documentation without purchasing 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 Westinghouse effort and the expenditure of a considerable sum of money.

In order for competitors of Westinghouse to duplicate this information, 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 testing and analytical methods and performing tests.

Further the deponent sayeth not.

vnc m smun-

i o

1 A1TAClIMENT 4 Amended Response to Request for AdditionalInformation NON PHOPRIETARY l

l l

Att+.chment 4 ST ill.AE 5752 Page I of 29 NON - PROPRIETARY SOUTil TEXAS PROJECT AMENDED RESPONSE TO REOUEST FOR ADDITIONAL INFORMATION PROPOSED _ AMENDMENT TO ALLOW USE OF ELilOW TAP APs TO MEASURE IWACTOR COOLANT SYSTEM FLOW RATE

1. Section 2.0 states that surveillance requirement (SR) 4.2.5, which currently requires the performance of precision heat balance measurements every 18 months, will no longer specify the method to be used for RCS flow measurement. Explain the elimination of the reference to RCS flow measurement methodology from the SR (i.e., why not reference both the precision heat balan:e measurements and the specific elbow tap methodology described in the application). Additionally, the proposed SR does not provide a reference as to when the measurements are to be taken. Section 2 of the submittal references "beginning of cycle" as does the analysis. Include such a reference in the SR or justify its omission. Also include the power level at which the surveillance is to be performed at in the proposed SR (per analysis assumption) or justify its omission.

RESPONSE: In the proposed SR 4.2.5, the reference to a specific flow measurement method was removed in order to be consistent with an approved change to t'.js specification for another licensee. The SR will be revised to reference the elbow tap Ap methodology and precision heat balance method. The elbow tap Ap methodology assumptions to perform the RCS flow measurement at beginning of cycle and at or above 90% of rated thermal power are no difTerent than for the exi> ting precision heat balance methodology. A notation will be added to the SR to reflect that the flow measurement is performed at beginning of cycle and greater than or equal to 90% of rated thermal power, in addition, the associated bases will include a document reference for the elbow tap Ap methodology.

2. Based on the change in RCS flow measurement methodology, are any revisions required to the revised thermal design procedure? Normally the RTDP references the calorimetric. lias streaming uncertainty increased from the assumptions made in the RTDP? Are the uncertainties for the elbow tap transmitters zerced out in the South Texas RTDP?

RESPONSE: No revisions attributable to changing the RCS Flow ver;dcation are necessary to the RTDP input values for this plant. Tt.e uncertainty value used in the RT9P calculations bounds both the performance of a calorimetric at the beginning of each cycle and the use of nonnalized cold leg elbow taps only.

The hot leg streaming values used in the uncertainty calculations for the calorimetric measurements for the normalization of the cold leg elbow taps and the RCS flow calorimetrics performed at the beginning of each cycle are appropriate for the measurement. Since the plant condition has not changed, the same hot leg streaming values are used in the calculations.

The uncertainties for the cold leg elbow tap transmitters reflect the conditions of normalization, either the beginning of cycle (calorimetrics performed each cycle) or calibration

~

0 STIll AE 5752 Page 2 of 29 The uncertainties for the cold leg elbow tap transmitters reflect the conditions of normalization, either the beginning of cycle (calorimetrics performed each cycle) or calibration each refueling outage (utilization of normalized cold leg elbow taps only). Thus, there are different input values used for the two different uncertainty calculations.

Specifically, transmitter uncertainties that are considered to be normalized (zero error) when beginning-of cycle calorimetrics are performed have input values for the cold leg elbow tap uncertainty calc Ttion.

3. Describe how the calorimetric uncertainties will be accounted for in the proposed methodology given that the proposed cibow tap correlation method relies on previous calorimetrics.

Ri!SPONSil: Since the baseline calorimetric flow was determined by averaging all of the previous cycles' measurements, the bounding RCS flow calorimetric measurement uncertainty, based on the most limiting set of cycle specific conditions for either Unit 1 or Unit 2, was used to detennine the calorimetric input to the cold leg elbow tap uncertainty calculations. Thus, the nonnalize.tlon conse vatively reflects the assumption ' hat all of the cycle calorimetric uncertainties were as large as the bounding value. Potential modifications to the calorimetric uncertainty, due to cordition changes, are not necessary unless the baseline calorimetric flow is redefined.

4. Section 3.4.2 states that calorimetric flows from all fuel cycles are evaluated for use in defining baseline calorimetric flow. What does the term " evaluated" mean? Was the average for all cycles used? Ilow many cycles of data were utilized?

RIiSPONSil: The calorimetric flows for each fuel cycle were compared to the best estimate flow, if the measured flow was within 12.8% (calorimetric flow uncertainty) of best estimate, then the measurement was ce nsidered valid, in the interest of facilitating the review process, the South Texas Project decided to follow a more conservative procedure previously approved by the NRC, in which many calorimetric flow measurements were used to define the baseline flow. 13aseline calorimetric flow is based on the average of all cycles (Cycles 1-7 for Unit 1, Cycles 16 for Unit 2), with no corrections for hydraulles or hot leg streaming biases. This procedure is conservative because the baseline flow will be biased low since it includes the effects of any steam generator tube plugging and hot leg streaming.

l l

l i

t

=

~ -.

-=

STill AIL 5752 Page 3 of 29

5. Section 3.4.2,"llaseline Elbow Tap Ap" states that the average op from all elbow taps is used as the basis foi the baseline elbow tap coeflicient. Considering potential differences in installatlon and individual elbow tap anci hydraulic characteristics, explain why this approach is acceptable. The op dah supplied seems to indicate that loop 4 is consistently different from the other loops. Is this measurement an indication of different flow characteristics or a sensing element installation efTects? Would separate flow coeflicients for each elbow tap better reflect flows? If you choose to use the proposed averaging method, provide a justification of this method including (but not limited to) a comparison of this method to using individual elbow tap correlations and show that your proposed method is conservative.

Identify all places (analyses, calculation methods, etc.) that may be affected by this approach and specify whether this would result in a net benefit or penalty (i.e., why are you proposing this approach in lieu of correlating each elbow tap individually?).

l(liSPONSII: The elbow tap flow measurement procedure uses elbow tap Aps to determine only the relative change in flow from the baseline calorimetric flow measurement. The absolute value of flow is determined by nonnalization, so installation differences and individual cibow tap and hydraulic characteristics have no impact on the measurement of flow changes. Since the three elbow taps in a loop are measuring the same flow and are equally sensitive to flow changes (within repeatability), and since the elbow taps are used only for determining flow changes, the loop flow changes defined by the average op or by the individual Aps would be exactly the same. Therefore, flow coefficients for each elbow tap are not necessary. Comparisons of flows determined from loop average elbow tap Aps with flows determined from the average of all cibow tap Aps at the South Texas Project, shown on Tables 1 and 2, have identified insignificant differences, so flow coefficients for each loop are also not necessary. Considering this data and similar data from other plants, the use of an average Ap simplifies the process while imposing no penalty or loss of flow measurement accuracy.

The ability of cibow taps to accurately detennine changes in flow was confirmed by measurements of different flows at Prairic Island Unit 2. Data was collected with one and two pumps running, resulting in a loop flow difTerence of about 8%, a larger change than expected for effects such as tube plugging at any plant. The measurements showed very good agreement between cibow tap and Leading Edge Flow Meter measurements of the flow differences, as shown on Table 3.4-1 in reference 2.

Any set ofindividual cibow tap Ap measurewr.,s will show that the elbow tap Aps difTer in magnitude. These differences, which are most likely due to small difTerences in elbow dimensions, installation dimensions or actual flow, have no impact on the measurement of flow changes, as confirmed by the measurements at Prairie Island Unit ?. The Ap differences measured at the South Texat Project are not considered to be unuma!.

l e

o ST ill AE 5752 l' age 4 of 29

6. Section 3.4.2," Flow Verification for Future Cycles," states the average of all elbow tap APs measured at or near full power... What is considered at or near full power?

RESPONSE: At or near full power is defined as greater than or equal to 90% of rated thermal power (RTP). [

ya

7. Section 3.4.2," Elbow Tap Flow Measurement Procedure," states that calorimetric flows that fell well outside the allowance (either high or low) should not be used in defining baseline flow, llow was the allowance and screening criteria determined? Was this procedure used at South Texas and were any calorimetric measurements deleted from the baseline flow detennination.

RESPONSE: The calorimetric flows for each fuel cycle were compared to the best estimate flow. If the measured flow was within 12.8% (calorimetric flow uncertainty) of best estimate, then the measurement was considered valid. No measurements at STP were outside this criteria.

In the interest of facilitating the review process, the South Texas Project decided to follow a more conservative procedure previously approved by the NRC, in which many calorimetric flow measurements were used to define the baseline flow. Baseline calorimetric flow is based on the average of all cycles (Cycles 1-7 for Unit 1, Cycles 1-6 for Unit 2), with no corrections for hydraulics or hot leg streaming biases. This procedure is conservative because the baseline flow will be biased low since it includes the afTects of any steam generator tube plugging and hot leg streaming.

8. The RTDP calorimetric flow uncertainty is given as a 95/95 value. Do the elbow tap flow measurements and best estimate hydraulic analysis provide equivalent results with regard to the 95/95 value? In Section 3.4.2 of your submittal you proposed to compare R to 1.004 *R'.

In Section 3.5.2 of your submittal you stated that the best estimate flow analysis has an accuracy of *2%. Discuss how this uncertainty (associated with your best estimate analytical methodology) is accounted for. Provide a more detailed justification of your proposal to l

allow cibow taps to exceed the best estimate flow by as much as 0.4% and still be accepted as l

a valid measurement of flow.

Explain this from an RTDP uncertainty assumption l

perspective as well as from the perspective of uncertainties assumed in other analyses that i

include an RCS flow assumption. Provide all places (analyses, calculation methods, etc.) that are affected by this credit.

RESPONSE: The uncertainty for the cold leg cibow tap fiow measurement is calculated on y

the same basis as the uncertainty for the RCS indicated flow used in the RTDP analyses, i

Therefbre from an uncertainty point of view, the elbow tap measurement results are comparable to RCS flow calorimetric results, without the undue influence of hot leg streaming in later cycle core designs. The cold leg elbow tap measurement alone is sufficient to satisfy any assumptions of the safety analyses. Ilowever as an additional check to assure

. the utilization of a conservative result, Westinghouse includes a comparison of the cold leg elbow tap flow change with the best estimate hydraulic analysis projected flow change for

~

~

AnachmInt 4 STilL AE 5752 Page 5 of 29 that specific cycle. The purpose of this second comparison is to decrease the possibility of a non-conservative influence due to an unforeseen cause. In the event the cold leg elbow tap flow change is significantly higher than the projected change in best estimate flow, i.e., the ratio of the measured flow to baseline flow is more than 0.4% greater than the ratio of projected best estimate flow to baseline best estimate flow, then the process provides that the calculated RCS flow should be based upon the utilization of 1.004 times the ratio of the best estimate flows times the baseline calorimetric flow. Effectively the lower of the two flows (measured vs. best estimate) is used for the verification of operation within the safety analyses assumptions. The value 1.004 (0.4 %) is based on a conservative estimate (smaller magnitude) of the repeatability uncertainty of the cold leg elbow tap measurements. This estimate is based on the uncertainties that could change between cold leg cibow tap measurements from one cycle to the next (or baseline cycle to projected cycle). This acceptance criterion was chosen based on the expected variability of the measurements.

Since this comparison is a conservative, secondary check and can only result in a reduction of the RCS flow (when compared to the measured flow) for verification purposes, no additional uncertainties are necessary for inclusion.

The best estimate flows are not determined on the same basis as the calorimetric or cold leg cibow tap measured flows. The best estimate flows are, as the name implies, a best estimate.

I

]* since it is the ratio of the projected best estimate flow to the baseline best estimate flow that is utilized in the comparison. The significance of the accuracy is further reduced by the manner in which the ratio is used, i.e., only when the ratio of the measured flows is greater than the ratio of the best estimate flows by more than 0.4%.

Therefore, the uncertainty of the best estimate flows has not been included in any uncertainty calculations. As noted previously, this is a secondary check for the purposes of conservatism and additional uncertainty accounting is not necessary.

As noted above, the value of 1.004 corresponds to a conservative (small) estimate of the repeatability of the cold leg elbow tap flow measurement. As primary reliance is placed on the cold leg elbow tap measurement, which is considered to be a 95/95 value, comparison within a conservative estimate of the repeatability is believed to be reasonable and prudent.

On a more rigorous basis, a larger acceptance criterion could be justified for the comparison difTerence, e.g., a more thorough accounting of the repeatability uncertainties could be included based on plant specific hardware. The plant specific value for the South Texas Project is greater than the repeatability value of 0.4 % used as the acceptance criterion. Thus the use of the 0.4 % value causes a conservative evaluation.

From a pmely theoretical point of view, the utilization of the secondary check, comparison with the best estimate flow, is not necessary. The comparison was included in the process for conservatism. The fact that the cold leg elbow tap flow exceeds the best estimate flow is not by itself indicative of a flawed cold leg cibow tap measurement. [

]* Therefore, Westinghouse recommends the secondary

Anuhment 4 o

ST llL AE 5753 Page 6 of 29 check with the acceptance cr terion based on the repeatability of the cold leg elbow tap i

measurements. From an RTDP point of view, cold leg cibow tap flow exceeding the best estimate flow does not by definition result in failure to verify flow within the bounds and accuracy of the RTDP calculations. The cold leg elbow tap measurement is a consistent and reliable process for verification of RCS flow. The uncertainties for this measurement process have been determined and appropriately bounded or explicitly included in the RTDP calculations. As noted, the acceptance criterion is conservatively determined, based on the repeatability of two cold leg elbow tap measurements, baseline and current cycle, and thus provides a reasonable criterion for a prudent check. The magnitude of this criterion does not have an efTect on the RTDP calculations and thus is not included in the RTDP process.

9. Section 3.5 states that the component Ap accuracy for the Prairie Island measurements was established by calibrations to be within 1% of the measured Ap. The sum of the ops measured across the reactor and steam generator were within 1% of the pump Ap, confirming measurement accuracy. Explain how this confirms accuracy of the measured Ap.

RESPONSE: Reactor coolant pump head was measured acmss taps located on the pump suction and pump discharge pipes. Another tap was located on the hot leg pipe. Reactor vessel Ap was measured across the hot leg and pump discharge taps. Steam generator Ap was measured across the hot leg and pump suction taps. Since all measurements used common taps, the sum of reactor ver.sel and steam generator Ap measurements would be expected to be equal to the pump head measurement. The test data confirmed that the sum of the reactor vessel and steam generatot Aps were within 1% of the pump head.

10. In Section 3.6.3, " Evaluation of Calorimetric Flows," the second bullet references the

" baseline calorimetric flow defined above." Where is this reference in the submittal?

RESPONSE: The baseline calorimetric flow definition is contained in the first bullet in Section 3.6.3, " Listed at the bottom of the column is the average of the cycle flows, conservatively def1ned to be the baseline calorimetricflowfor the unit." That is, the baseline calorimetric flow is the average of previous cycles' measured calorimetric RCS flows, and the value is listed at the bottom of the column labeled " MEASURED CAL" in Table 3.6-3 of the submittal.

I1. Provide and explain plant specific data and or experience of the temperature profile in the hot leg as a result of streaming and its plant specific effect on temperature readings. Provide infomiation on this phenomenon for each cycle considered in your proposed methodology.

Explain the effect of power level on this phenomenon. In addition explain how the hot leg temperature was obtained for the calorimetrics used (i.e., was the high, average of each loop or average of all RTDs used?).

RESPONSE: llot' leg temperature streaming gradients on the pipe circumference have been measured at a few plants between 1968 and 1996, and hot leg gradients have been inferred from the three hot leg RTD measurements in each loop at several plants since 1988. Prior to implementation of reactor core low leakage loading patterns (LLLPs), the measured hot leg circumferential temperature gradients and hot leg RTD differences were no more than 10 F.

ST.ll!-AE 5752 Page 7 of 29 After LLLP was implemented, calorimetric flow measurements began to indicate apparent decreases in RCS flow, and measured hot leg RTD difTerences increased to as high as 16'F.

Since core exit temperature measurements showed that the core exit gradients had increased significantly (from 25*F to as high as 60 F), it was concluded that the hot leg temperature stremning gradient had increased, and that the 3 point hot leg temperature measurements were being biased by the streaming gradients. The large core exit temperature gradients were caused by low-powered fuel assemblies at the edge of the core, so it was theorized that colder water in these gradients was not completely mixing with the hotter water as it entered the hot leg pipe, and was flowing between the downstream hot leg RTDs or sample scoops of the RTD bypass system. Since the RTDs were not measuring the colder water, the measured hot leg temperature was being biased high.

Other analyses and measurements support the cenclusion that LLLP is causing apparent decreases in calorimetric flow measurements, including the correlation described in Section 3.3.4, but best shown by the elbow tap flow measurements. There are no measurements that define the entire temperature gradient within the hot leg pipe, but 3 dimensional thermal-hydraulle models of the reactor vessel upper plenum and hot leg pipe support the theory that cold water areas exist in the hot leg pipe well downstream from the reactor vessel. A hot leg temperature streaming test performed during a 1995 96 fuel cycle at a 3-loop plant (Vandellos 2) provided a clear demonstration of the cold water effect. Figure 1 shows the mea:ured temperature gradient on the pipe surface relative to the temperatures measured by the three RTDs located at the same axial position. Figure 1 shows a large cold region at the bottom of the hot leg pipe in one loop, and shows that none of the RTDs detected this area of cold water. The RTD measurement is clearly biased well above the actual average hot leg temperature.

1 The measurements and analyses summmized above clearly demonstrate that the core power distributions of the LLLP core pattern cause larger hot leg streaming gradients, resulting in flow and temperature measurement l,iases. Based on the available data, the biases appear to change linearly with power, decreasing to zero as AT decreases to zcro at zero power.

STP specific plots of hot leg RTD measurements at the beginning of each cycle are provided in Figure 2 and Figure 3. The STP specific configuration of the hot leg RTDs is shown in Figure 4. Figures 2 and 3 confinn the existence of hot leg streaming. The temperature streaming differences vary from loop to loop and cycle to cycle as core power distribution changes, but the differences are not unusual and are smaller when compared with measurements at other plants. The three RTDs in a. loop can indicate that the streaming gradient is as large as the difference that was measured, but do not necessarily indicate the maximum difTerence that may exist in the hot leg, as illustrated by the example shown in Figure 1. Since there can be differences between the actual and measured hot leg temperature gradients, the measured temperatures provide no delimite information that can be used to define the streaming phenomenon.

The hot leg temperature measurements used in the calorimetric flow calculations are based on the average of the three RTD measurements in a loop.

0 ST ilL AE 5752 Page 8 of 29

12. Provide plant specific configurations of hot leg RTDs and elbow taps. With regard to RTDs, Figure 3.3-2 seems to indicate that an average of the three RTDs (assuming that they are located at 0,120, and 240 degrees) would eliminate the difference. Explain the streaming efTect in this case and include a temperature profile plot for the hot leg cross section. Provide similar plots for South Texas.

RESPONSE: STP specific configurations of the hot leg RTDs and elbow taps are shown in Figure 4. The hot leg RTDs are located at 0,120', and 240' in all loops e South Texas Project Units 1 & 2, assuming O'is at the top of the pipe.

Figure 3.3 2 was intended to generally illustrate a streaming gradient, and was not based on South Texas Project data, if the gradient in Figure 3.3 2 were perfectly linear, the measurement from the three RTDs would have no streaming error, regardless of magnitude and orientation of the gradient, llowever, as seen on the attached Figure 1, gradients are not linear. Skews and bulges in the gradient cause the streaming measurement biases.

The measurements shown on Figure 1 can be used to estimate the temperature profile across

'he hot leg cross section, but additional measurements inside the pipe would be needed to produce an accurate temperature profile. Based on Figure 1, [

j+=

13. In your correlations you neglect any changes in specific volume due to cold leg temperature changes. Provide a justification of this approach with regard to potential future changes in parameters that may affect the specific volume of the cold leg water (e.g., temperature and pressure).

RESPONSE: The elbow tap flow measurement procedure described in Section 3.4.2 does consider changes in cold leg specific volume. The basic equation is described in the response to Question 30. Any future elbow tap Ap flow measurement will use cold leg specific volume to determine RCS flowrate.

For comparison purposes, previous cycles' elbow tap flows only included significant cold leg specific volume changes. In some cases, the differences in cold leg temperature were small and the effect on flow was negligible (sensitivity of about 0.07 % flow per F), and cold leg specific volume corrections to the flow coefficient were not necessary. For the South Texas Project, the cold leg temperature differences for the early cycles were small while the changes for the last two cycles at both plants were significant (as shown on Table 3.6-2) and the appropriate cold leg specific volumes were applied.

14. An implicit assumption in your correlation methodology is that the correlation coeflicient (K) in the following equation remains constant. Ilow will your methodology address changes to this coefficient should they occur given that you're proposing to freeze the current coeflicients.

W = K (p Ap)"

- ~ -

l ST llL-AE 5752 Page 9 of 29 RESPONSE: A review of phenomena that might affect the flow coeflicient (fouling deposits as experienced in feedwater venturi nozzles, dimensional chang,es, upstream flow conditions),

discussed in Section 3.4.1, identified no phenomenon that would impact an elbow tap flow measurement. Over the long term, comparison of cibow tap flow measurements with the best estimate flow trend would provide an indication of a change in flow coeflicient, if a change were to occur, As described in Section 3.4.2, the elbow tap flow measurement would not be used if the measured flow change exceeded the best estimate flow change by more than the repeatability allowance for the elbow tap measurement. The elbow tap flow rneasurement would be used for all other cases, even if the measured flow would be below the best estimate flow trend, which would be a conserwtive condition.

15. Section 3.4.1 states," tests have demonstrated that elbow tap flow measurements have a high degree of repeatability.. " Were such tests performed for configurations similar to plant installations - with short stretches of straight pipe upstream of the elbow taps? Provide results and conclusions. (i.e., justify your assumption of repeatability for the plant specific configuration in light of the lack of straight pipe upstream of the elbow taps.)

RESPONSE: The tests mentioned in Section 3.4.1 relate to statements from the reference (ASME Fluid Meters). In our judgment, supported by the evaluation of plant data and the results of the test described below, the statements in the reference are considered to be applicable regardless of the upstream piping configuration. The accuracy of a measurement of absolute flow is affected by the upstream piping, but the repeatability of a measurement of changes in flow is not afTected. Concerning testing of the specific piping configuration in the RCS, Section 3.4.1 and Table 3 4-1 present the comparison of Leading Edge Flow Mater and elbow tap measurements at Prairie Island Unit 2 over a period of 11 years. The comparison indicated that the ditTerences were weil within the allowance for repeatability defined for the elbow taps. The STP elbow tap pipe configuration and pipe layout are similar to the Prairie Island configuration, as shown in Fig tre 4.

16. Explain the statement " Repeatability and accuracy are improved when all elbow tap AP measurements are used" in Section 3.4.2 and discuss the benefits realized from this approach.

RESPONSE: There are three channels, each composed of an elbow tap, transmitter, and associated electronics, for each cold leg. [

]* which is a more conservative value. Westinghouse recommends the utilization of all available instrument channels to assure as accurate a measurement as reasonably possible.

Multiple independent measurements of the same parameter, e.g., three measurements of cold leg elbow tap Ap on the same loop, increase the confidence associated with the magnitude of the measurement.

ST HL AE 5752 Page 10 of 29

17. Section I-5-4 of " Fluid Meters, Their Theory and Applications," 6th Edition, Howard S.

Bean, ASME, New York,1971; reference I to your submittal, states, "Let the values of pi,...

be arithmetical mean values obtained by averaging over the whole section, A, and. if the fluid motion is not strictly steady (laminar) but turbulent. a time average over the section." How was this considered / applied in your proposed methodology?

RESPONSE: RCS flow is turbulent, and pressure and differential pressure measurements fluctuate randomly at a low frequency. As with all Ap measurements, the elbow tap Ap measurements fluctuate, so the Ap measurement used in the procedure is based on the average of several measurements obtained over a short time period (e.g.,15 to 30 minutes).

In addition, the elbow tap flow measurement uncertainty, defined in Appendix A, includes an allowance for elbow tap Ap process noise.

18. Sections 1-5-27 and I-5-55 of" Fluid Meters, Their Theory and Applications," 6th Edition, lloward S. Bean, ASME, New York,1971; reference 1 to your submittal, states, that "a uniform fluid velocity was assumed, thus neglecting any effect of normal stream turbulence."

In addition, " Application, Part 11 of Fluid Meters, Sixth Edition 1971, Interim Supplement 19.5 on Instrument and Apparatus," ASME, New York,1972; provides guidance on length of piping required upstream of the elbow taps to ensure uniform velocity profiles Considering j

that the recommended length of pipe does not exist upstream of the elbow taps in your plant, discuss how this was addressed in your proposed methodology.

RESPONSE: The piping configuration upstream of the elbow taps affects the accuracy of an absolute flow measurement, but the elbow tap flow measurement procedure uses the measured Ap to determine flow changes, not absolute flow. Therefore, the configuration considerations are not applicable. Also refer to the response to Question 5.

19. " Flow Measurement Engineering Handbook," R. W. Miller, McGraw-Hill Book Company, New York,1983; gives an accuracy for elbow taps of *4%. Additionally, this source states l

that the minimum length of piping as presented in Fluid Meters (see question above) is necessary to hold piping bias errors, due to piping influence, to less than 0.5%. It further states that an additional 0.5% should be added to the flow-coeflicient accuracy value for any decrease in length of pipe. Discuss how each of these factors was addressed in your proposed methodology. List ark ustify the accuracy and precision values used in your methodology and explain how they were applied.

RESPONSE: Refer to the responses to Questions 5 and 18. None of the uncertainties mentioned-in the question apply to the proposed cibow tap flow measurement procedure.

The elbow tap flow measurement uncertainty is quantified in detail in Appendix A to reference 2.

20. I)escribe how feedwater venturi fouling and its effect on feedwater measurement uncertainty was addressed in the calorimetrics that will be used to derive the elbow tap coefficients. List the uncertainties for feedwater measurement and explain how they were used to determine the overall uncertainty. In addition, discuss how feedwater venturi area expansion factor uncertainties were accounted for.

I

ST-llL AE 5752 Page 11 of 29 RESPONSE: Table A-3, Calorimetric RCS Flow Measurement Uncertainties, provides all of the component uncertainties utilized in the bounding RCS flow calorimetric measurement uncertainty calculation. Basic presumptions of this uncertainty calculation are that each of the cycle specific measurements was performed at full power, at the beginning of each cycle, and with a clean feedwater venturi in each loop. Thus, no allowance was included in the uncertainty calculation for the effects of feedwater venturi fouling.

The feedwater measurement uncertainties are noted under the categ @, "Feedwater Flow" and "Feedwater Enthalpy" on Table A 3. These uncertainties are: an allowance for the basic accuracy of the venturi, the effect of temperature uncertainty on the venturi thermal expansion coefficient, the potential effect of material differences on the venturi thermal expansion coefficient, the efTect of temperature uncertainty on feedwater density, the effect of pressure uncertainty on feedwater density, the Ap measurement uncertainty, the effect of temperature uncertainty on feedwater enthalpy, and the effect of pressure uncertainty on feedwater enthalpy. [

]* to the combination of all of the uncertainties. The uncertainties were combined in a manner equivalent to the equation noted on page 14 of WCAP-13441, " Westinghouse Revised Thermal Design Procedure Instrument Uncertanty Methodology South Texas Project Units 1 & 2."

As noted above, two allowances are made for uncertainties associated with the feedwater venturi thermal expansion coefficient. The first is a simple allowance for the temperature uncertainty. The second is a conservative allowance for material variation. These are noted on page 12 of WCAP-13441. These are [

]* to the uncertainty calculation for the RCS flow calorimetric.

21. Discuss the type of quality assurance review performed on the analytical model. Ilow does this meet the requirements of 10 CFR Part 50 Appendix B7 In addition, provide data relating to changes in the plant (steam generator tube plugging, fuel design changes, etc.) that have affected flow condition and confinn that the model was able to predict these changes accurately. What confirmatory checks were performed to ensure accuracy of the analytical model at South Texas? Since the RCS loops experience changes at different rates (e.g.,

steam generator tube plugging) provide this data on a loop specific basis.

RESPONSE: The analytical model is incorporated into a compt.ter code that is under

- configuration control within Westinghouse, inputs to the computer code are subjected to an independent verification process, identified in Westinghouse Quality Assurance procedures, t

that confirms inputs and results.

This procedure is the same procedure to which Westinghouse performs all their safety related calculations and evaluations.

The best estimate flow analysis, described in Section 3.5, predicts RCS flow and defines expected changes in flow for a new cycle. Best estimate flow is applied only as a check on the elbow tap flow measurement, but is not used alone to verify flow.

l l

Attchment 4 ST-IIL-AE 5752 Page 12 of 29 The best estimate flow analysis considers several factors, described in Section 3.5, in defining flow changes. RCS flow resistances for the South Texas Project are listed on Table 3 in response to Question 31. The analysis recalculates steam generator flow resistance for a specified number of tubes plugged, and flows for each loop with different amounts of tube plugging are defined. The relationship between the flow reduction and number of tubes plugged is [

]* Other factors considered in the best estimate flow analysis include changes in the reactor core flow resistance due to fuel design changes and the effect of removing thimble plugs en core bypass flow [

]*

Referring to Section 3.6, best estimate flow changes for impeller smoothing and tube plugging were calculated for the Soutt Texas Project. These changes were found to be in good agreement with the measured elbow tap flow changes, considering the repeatability allowance for the elbow taps. Comparisons for several other plants show similar results. A comparison of best estimate and elbow tap total flows for a 3-loop plant with average tube plugging that reached 16%, with a plugging imbalance of 7%, both before and after steam generator replacement, also showed good egreement (whhin 0.4%), even with the relatively large hydraulics changes and imbalanced loop tube plugging.

22. Periodic confirmation of cibow tap characteristics is important in ensuring reliability, llow will you do this? In the proposed methodology you are proposing to use the best estimate model to confimi your cibow tap readings. The best estimate methodology is used to confirm that your " measured' flow is in agreement with your expected (predicted) flow.

Discuss the actions that you intend to take (evaluation, recalibrations, etc.) should the elbow tans read higher / lower than your best estimate predictions to ensure that flow

" measurements" are obtained in an acceptable manner and that they are not replaced by the unconfirmed analytical method. The use of" unconfirmed analytical method" in this question refers to a prediction of flow without confirmation through actual measurement. The staff requests that the licensee also commit to 1) notify the NRC of any changes to the hydraulic flow model in a manner which affects the results of the model and 2) contact the NRC for i

further review of the methodology if the elbow tap-determined flow rate exceeds (becomes less conservative than) the analytically determined fiow rate.

RESPONSE: Measurements from several plants have shown that best estimate and elbow tap flow changes have been consistent with each other over several cycles. There have been no indications of a significant difference in these trends.

Over the long term, comparison of elbow tap flow measurements with the best estimate flow trend would provide an indication of a change in a flow coefticient, if a change were to occur. Given the high confidence level for predicting and measuring changes in flow, it is unlikely that a change in an elbow tap flow coefficient would be needed, and a change would not be considered unless i

a trend were defined by more than a single measurement. As described in Section 3.4.2, the elbow tap flow measurement would not be used if the measured flow change exceeded the best estimate flow change by more than the repeatability allowance for elbow taps. The elbow tap flow measurement would be used for all other cases, even if the measured flow change were below the best estimate flow change by more than the repeatability allowance.

thus providing a conservative indication of measured flow. This comparison of measured to predicted flow rate changes provides a conservative upper bound to the flow measurement. If

ST HL-AD5752 Page 13 of 29 the absolute measured flow rate exceeds the absolute best estimate flow rate, and the flow rate change is outside the repeatability allowance, then an evaluation will be performed to assess the measurement validity. To aid in these future potential evaluations, STP will

- continue to obtain 1/")% calorimetric data at beginning of cycle to be used in re-evaluating the elbow tap methodology if required.

It is assumed that changes to the hydraulic flow model refer to changes in methodology rather than changes such as the number of steam generator tubes plugged or the addition of intermediate flow mixing grids in the fuel. STP will: 1) notify the NRC of any known changes to the hydraulic flow model methodology, and 2) contact the NRC if the measured absolute elbow tap flow rate exceeds the absolute best estimate flow rate when the measured flow change exceeds the predicted flow change by more than the repeatability allowance.

23. Discuss the effect of vibration and turbulence induced noise on elbow tap measurement and how this was accounted for in your proposed methodology Address erosion and deposit formation with respect to the elbow tap instrument tube connection to the RCS legs. For example, if ti.e throughwall penetration initially terminates with a sharp edge at the leg inner wall surface, does this sharp edge change with time due to flow impingement? Ifit changes, what is the effect an indicated behavior? If it does not change, what is the basis for that conclusion?

RESPONSE: RCS flow is turbulent, and pressure and differential pressure measurements fluctuate randomly at a low frequency. As with all Ap measurements, the elbow tap Ap measurements fluctuate, so the Ap measurement used in the procedure is based on the average of several measurements obtained over a short time period (e.g.,15 to 30 minutes).

In addition, the elbow tap flow measurement uncertainty, defined in Appendix A, includes an allowance for elbow tap Ap process noise.

The instrument penetration through the elbow wall terminates inside with a sharp edge, ground free from burrs or other irregularities. The velocity in the elbow is low (44 feet per second or less at South Texas Project) relative to crosion of stainless steel piping, and large velocity changes and particle ionization associated with feedwater venturi fouling do not exist in the elbow. Neither of these phenomena would affect the instrument penetration. In support of this conclusion, Section 3.4.1 and Table 3.4-1 present the comparison of Leading Edge Flow Meter and elbow tap measurements at Prairie Island Unit 2 over a period of 11 years. The differences between these measurements were well within the repeatability allowance for the elbow taps, so the effect of erosion or deposit formation was not evident.

24. In Section 3.4.2 of your submittal you stated, "If a known hydraulic change (e.g., tube plugging) was made before a cycle, calorimetric flow for the cycle should be adjusted so all flows have a common hydraulic baseline." Was this approach used in your proposed methodology? If so, discuss how this was done and justify this approach.

RESPONSE

An alternate methodology recommended by Westinghouse to determine baseline flow would include corrections to calorimetric flow measurements used to define a baseline flow. The corrections are based on the best estimate of flow changes due to known hydraulics changes, so the calorimetric measurements would have a common hydraulics

Anachment 4 ST llL-AE-5752 Page 14 of 29 basis. Otherwise, the baseline calorimetric flow measurement would include some of the flow changes also detected by the elbow taps.

In the interest of facilitating the review process, the South Texas Project baseline flow was based on the average of all cycles (Cycles 1-7 for Unit 1, Cycles 1-6 for Unit 2), with no corrections for hydraulics changes or hot leg streaming biases.

25. Section 3.6.1 states, "Considering all of the above, the overall impact of the hydraulic changes was expected to be 0.3 to 0.9% flow over seven cycles of operation,..." However, Section 3.5.2 states that the best estimate flow analysis has an accuracy of *2%. Explain how you were able to predict the flow changes to a better accuracy than that of the analytical model. Similar statements were also made for Unit 2.

RESPONSE: The best estimate flow analysis accuracy of i2% applies to the ability to predict the actual total flow, but does not apply to the ability to predict changes in flow. The ability to predict the impact on flow due to hydraulics changes such as reactor core design changes (Ap tests are performed for new fuel assembly designs) or changes in tube plugging has been shown to be accurate by elbow tap flow measurements at many plants. The ability to predict total flow before or aRer a change is still 2%, but the abilhy to predict the change in flow is considered better than 10.2%.

26. Section 3.6.2 states that AP measurements were obtained at 70% power for some of the cycles. Additionally, this section states that a decrease of 1.2% flow from zero to 100%

power and 0.4% from 70% power to 100% power exists. Justify these statements. How was this data obtained? Are you assuming a linear relationship between flow and power? If so, why? Section 3.6.3 further states that in addition to the calorimetrics at 70% power, calorimetric data was also obtained shortly aller full power was attained. Why was this data not used instead of the 70% data? Also explain the statement, "Another adjustment was made in nonnalizing flows to the baseline flow to account for the decrease in cold leg temperature in Cycles 6 and 7..."

RESPONSE: RCS flow decreases slightly as reactor power is increased, since hot leg temperature, specific volume and velocity increase relative to the temperature, specific volume and velocity at zero power. Similar but smaller velocity increases occur in the reactor core and steam generator tubes as the temperature in these regions increases toward the full power Tavg. The increased velocity in these regions increases the RCS flow resistance, resulting in the flow reduction. The decrease in flow from zero to 100% power is plant specific, and depends on the difference between hot leg and cold leg specific volumes and the flow resistance of the affected regions. The relationship between flow and power is not linear, but the differences from a linear relationship are lost in roundoff to a significmit value.

For the South Texas Project, RCS flow decreases by 1.2% as reactor power increases from zero to 100%, and decreases by 0.4% as power increases from 70% to 100%, based on a best estimate flow analysis. As stated in Section 3.5.2, the predicted change in flow due to a reactor power change was measured and confirmed by Leading Edge Flow Meter and pump

ST-llL AD5752 Page 15 of 29 input power measurements at Prairie Island Unit 2. At plants where elbow tap flows were measured both at zero power and at 100% power, the change in flow could be detected.

Earlier STP Technical Specifications required an RCS flow measurement prior to 75% RTP.

Therefore, during these early cycles, precision calorimetric data and elbow tap Ap data were collected near 70% power.

As part of an instrument alignment procedure, precision calorimetric data were also obtained at 100%; however, the 100% instrument alignment procedure did not require collection of elbow tap data. Therefore, the calorimetric flows

!isted on Table 3.6 3 and used to define the baseline flow were obtained when the plants were at 100% power. The elbow tap Ap data for Unit 1 Cycles 1 through 5 and Unit 2 Cycles 1 through 3 were obtained at approximately 70% power and adjusted as described above.

STP instituted reduced Thot starting in Unit 1 Cycle 6 and Unit 2 Cycle 5. The statement in Section 3.6.2 is related to the elbow tap flow coefficient adjustment for cold leg specific volume, and is consistent with the elbow tap flow measurement procedure described in Section 3.4.2. The applicable equation is described in the response to Question 30. In some cases, the differences in cold leg temperature are small and the effect on flow is negligible (sensitivity of about 0.07 % flow per F), and cold leg specific volume corrections to the flow coeflicient are not necessary. For the South Texas Project, the cold leg temperature differences for the early cycles were small, while the changes for the last two cycles at both plants were significant (as shown on Table 3.6-2) and the appropriate cold leg specific volumes were applied.

27. Explain the statement in Section 3.6.4 that less precision was used when averaging elbow tap data during early cycles and how this was accounted for in uncertainty terms. Also explain the statement that the difference between elbow tap and best estimate flows would be about 1% if impeller smoothing actually occurred before Cycle 1 (i.e., What was assumed for impeller smoothing and when did it occur? Justify this assumption). Explain the statement for Unit 2 that if Cycle I had been used to define baseline calorimetric flow, the flow difference in Cycles 5 and 6 would be larger and would be a more representative indication of low leakage loading pattern impact.

With regard to the statement that based on comparisons of adjusted calorimetric flows in Table 3.6-3, the Cycle 6 flow is almost 2%

below the Cycle I flow, it appears that a large portion of that difference is attributable to impeller smoothing and tube plugging.

Present your understanding of what this 2%

reduction is attributable to.

RESPONSE: Elbow tap Ap data collected during the early cycles contained fewer measurement samples than during later cycles, so the repeatability uncertainty of the Ap measurements may have been larger than in later cycles, resulting in a larger deviation from the best estimate trend. Ilowever, the overall uncertainty for the cold leg cibow taps includes an allowance of [

]* flow for elbow tap Ap process noise which is sufficient to envelope the variation or imprecision of this measurement parameter.

Impeller smoothing results in a flow reduction of about 0.6% flow, based on Leading Edge Flow Meter and pump input power measurements at Prairie Island Unit 2.

The flow reduction appears to occur during (or by the end of) the first cycle of operation, based on elbow tap measurements at other plants. For the South Texas Project, an impeller smoothing

Attahment 4 ST-HleAE-5752 Page 16 of 29 flow reduction of 0.6% was used. If some impeller smoothing had taken place in Unit 1 before the elbow tap measurements were obtained, the best estimate and elbow tap flows would be in closer agreement.

A more representative calorimetric baseline flow without conservatisms (as discussed previously) would have used Cycle 1 to define the baseline flow rate. Table 3.6-3 shows that the Unit 2 Cycle 6 measured flow (394,116 gpm) is 96.8% of the Cycle 1 measured flow (406,944 gpm). Aller adjusting the Cycle 6 measured flow for the hydraulics changes of 1.4%, the Cycle 6 adjusted flow (399,712 gpm)is 98.2% of the Cycle 1 measured flow. The hydraulics changes resulted from impeller smoothing and tube plugging, as listed in Table 3.6-1. The 1.8% remaining difference (Cycle 6 adjusted flow of 98.2%) is considered to be caused by a LLLP-induced hot leg streaming bias.

28.The first plots of Figures 3.6-1 and 3.6-2 show that the elbow tap flows are higher than the best estimate flows whereas the second plots show the reverse is true.

Explain this difference. Provide a plot which includes all of the following (on the same plot) for comparison and/or the data for such a plot: 1) elbow tap flows (per your proposed methodology),2) calorimetric flows, and 3) best estimate flows.

RESPONSE: The first plots of Figures 3.6-1 and 3.6-2 are normalized to a baseline flow.

The calorimetric plot shows each cycle's calorimetric flow as a percent of baseline calorimetric flow (average of all cycles). The best estimate plot shows each cycle's best estimate flow as a percent of the baseline best estimate flow (Cycle 1). The elbow tap plot show, ich cycles' cibow tap flow as a percent of the baseline calorimetric flow.

The second plot of Figures 3.61 and 3.6-2 show actual flows and are not normalized. The raw elbow tap plot shows the calculated flow using the elbow tap methodology without applying the best estimate flow confirmation as described in Section 3.4.2. The elbow tap using procedure plot shows the actual flow determined using the elbow tap flow measurement with application of the best estimate flow confirmation procedure. Table 4 summarizes all data plotted on Figures 3.6-1 and 3.6 2 and Figure 5 provides plots which include all of the following (on the same plot) for comparison: 1) cibow tap flows (per the proposed methodology),2) calorimetric flows, and 3) best estimate flows.

29. Please address each of the items in Attachment 1, " Staff Guidelines for Use of Elbow Taps for RCS Flow Rate Measurement," item 1.0, "Use of Elbow Taps," and 2.0, " Assurance to Show that the Elbow Tap Correlation Remains Viable." Explain how your proposed method to measure RCS flow rate addresses each of the items of the above guidelines. Some specific questions pertaining to these items are given below.

Staff Guidelines for Use of Elbow Taps for RCS FRw Rate Measurement The RCS flow rate is one of the inputs for calculation of the Departure from Nucleate Boiling Ratio (DNBR). The transient and accident analyses include as inputs the initial condition of RCS thermal design flow. The minimum RCS flow rate requirement in the Technical Specifications (TS) is consistent with the assumed RCS thermal design flow.

l ST HL AE-5752 Page 17 of 29 The criteria established in 10 CFR 50, Appendix A, require a high degree of assurance that specified acceptable fuel design limits (SAFDL) are not exceeded. The SAFDLs for anticipated operational occurrences (AOO) are that neither DNB nor melting at the fuel centerline occurs.

De results of the safety analyses calculation are used to assure that the SAFDLs are met. The nuclear industry has developed Limiting Safety System Settings (LSSS) methodologies which combine uncertainties statistically, lhe validity of such methodologies requires that input uncertainties be statistically valid.

De staff believes that the most important potential safety need directly associated with RCS flow rate is maintenance of an adequate margin to prevent departure from nucleate boiling. The next safety need is providing a reactor trip due to a low RCS Gow rate, with the concerns being departure from nucleate boiling and an overtemperature condition. However, the importance of RCS flow rate to reactor trip is diminished by other trip parameters, such as loss of pump power, too large a temperature difference between the hot and cold legs, or high pressure; trip parameters that will often cause a trip prior to a flow rate trip.

The staff has reviewed the use of elbow taps for RCS flow measurement previously for ar nuclear power plant and has developed guidelines for the acceptability of the use of cotu sg cl%v taps to measure RCS flow rate. This guidance follows.

1.0 Us.* of Elbow Tg in using elbow taps for indication of RCS flow rate, one should assure that:

1.1 There is reasonable confirmation that the elbow tap correlation used to determine RCS flow rate is accurate to within a known uncertainty and bias or that the pe-ceived rate (correlation determined rate including uncertainty and bias) is less than the actual flow rate.

1.2 There is reasonable confirmation to assure that the proposed method of dete mining RCS flow rate remains within acceptable bounds of accuracy, Reasonable confirmation that the originally determined RCS flow rate is accurate to within a known uncertainty and/or that it is less than the actual flow rate should be supported by either:

(1) Applicable flow test data that correlates RCS flow instrumentation to flow rate, or (2) Some other method of correlating RCS flow rate to the RCS flow instrumentation.

2.0 Assurance to Show that the Elbow Tan Correligion Remains Viable 2.1 With an cibow tap conelation and an acceptance bound established, there is need to assure l

that the correlation remains viable. A reasonable approach is to provide an analysis program that correlates all physical changes in the RCS flow path to the RCS flow rate and use this for confirmation of the elbow tap measurements of flow rate.

(1) To confirm this, one should demonstrate that the elbow tap-based flow rate indication is within the uncertainty bound that was established for the elbow tap correlation when compared to an analysis prediction. Primary emphasis upon the analysis is acceptable since no change in cibow tap correlation is anticipated when physical changes are made in the plant.

(2) An acceptable analysis program is one that accurately calculates plant changes in pump performance, core bundle changes, SG tube plugging, SG tube sleeving, SG replacement, and any l

other physical changes in the RCS that affect the RCS flow rate (the same criteria as applied to the above confirmation process).

l

ST41L-AE-5752 Page 18 of 29 2.2 Acceptability is established by comparing analysis results with available plant data.

(1) If the proposed elbow tap correlation is "best-estimate," then the staff will expect a direct comparison of elbow tap determined flow rate based upon the proposed elbow tap correlation.

(2) If the elbow tap correlation is conservative, then the staff will expect two comparisons -

one with a best-estimate correlation coefficient that provides the best fit to the analysis and the other with the proposed co, :lation.

(3) If the elbow tap determined flow rate crosses over and becomes less conservative than the analytically determined flow rate, the NRC should be contacted for further review of acceptability.

RESPONSE

Item i.1 & l.2: The elbow tap flow measurement procedure uses elbow tap Aps to define the change in RCS flow from the baseline cycle tn the new cycle. The change in flow is then applied to correct the baseline calorimetric flow for the change measured by the elbow taps.

Calibration of the elbow tap flow elements is not required. The elbow tap flow measurement uncertainty, defined in Appendix A, includes the uncertainty for the baseline calorimetric flow measurement (s) and bounds the uncertainty for the repeatability of baseline elbow tap Ap and new cycle elbow tap Ap measurements used to define the flow change. Calorimetric flow measurements have been in use at almost all plants and the measurement uncertainty is well defined. Almost all plants also use elbow tap Aps for relative flow measurements, for daily and monthly flow verification throughout the cycle, and for the loss of flow reactor trip function, so the performance and uncertainty of cibow tap Ap measurements is also well defined. The comparison of elbow tap Aps to determine flow changes has been shown to be accurate at many plants and has been confirmed by comparison with the Leading Edge Flow Meter measurements at Prairie Island Unit 2. Therefore, the elbow tap flow measurement procedure will define RCS flows to within a known uncertainty.

Item 2.1: Westinghouse has been applying its best estimate flow analysis successfully in determining flow estimates and evaluating the effect of system hydraulic changes since the l

program was developed in connection with the Leading Edge Flow Meter tests that were perfomted at Prairie Island Unit 2 in 1974-75. The analysis, described in Section 3.5 and responses to Questions 21, 25, 26, and 31, considers the physical changes affecting RCS flow, including steam generator flow resistance changes due to plugging, sleeving or

- replacement, fuel assembly design changes or bypass flow changes, and pump performance.

The analysis estimates differences in loop flows due to imbalanced plugging or different pump performances, and calculates reverse flows with pumps shut down, flows at zero to full power, and from ambient to normal temperatures. It is concluded that the best-estimate analysis is equivalent to the analysis program recommended by the NRC.

Item 2.2: The elbow tap flow measurement procedure uses baseline and new cycle elbow tap Ap measurements to define the change in flow based on the standard relationship between flow and Ap. No calibration or data curve fit is required. The flow change is applied to the l

baseline calorimetric flow to determine flow for a new cycle.

The resulting flow is l

considered to be accurate to within the measurement uncertainty defined in Appendix A. No l

additional uncertainty is enmidered to be necessary. Refer to the response to Question 22 for l

l l

ST HL-AE 5752 Page 19 of 29 the South Texas Project's commitments to notify the NRC concerning the comparison of measured elbow tap and best estimate flow rates.

30. Provide the correlation used to measure RCS flow rate by the use of elbow taps. If pu use a correlation for the elbow taps 1 hat is in the form of:

m = K](AP)(p) where:

m

= the RCS flow rate K

= the elbow tap flow correlation coefficient AP

= the elbow tap pressure drop, and p

= the cold leg density please provide the values of correlation coefficient K for each of the 3 taps in each loop of the cold legs and infomiation on (1) how they were determined, including background of the data used, and (2) what conservatism has been applied for these values, it is noted that because the elbow taps are not calibrated in a laboratory environment, as the feedwater venturi meters are, but only normalized against the calorimetric heat balance, we expect that a conservative margin will be applied in this method.

RESPONSE: The equations used to define RCS flow for a new cycle are described in Section 3.4.2. Combining equations I through 4 from Section 3.4.2 results in the following equation:

FCF = BCF * ((Apr

• Vr)/(APb *Y))

b where:

FCF

= future cycle flow (total), gpm BCF = baseline calorime:ric flow (total), gpm Apr

= future cycle elbow tap Ap (average), inches vr

= future cycle cold.eg specific volume (average), cu ft/lb Ap3

= baseline cycle ellow tap Ap (aveiage), inches

= baseline cycle col i leg specific volume (average), cu ft/lb vn The (Ap

• v) terms define the total fic w coefficient for the baseline measurement and for each subsequent cycle. Since the three elbow taps in a loop are measuring the same flow and are equally sensitive to changes in flow (within repeatability), and since the elbow taps are only used for determining changes in flo.v, flow coefficients for each elbow tap measurement are not necessary. Comparisons of flo.vs determined from loop elbow tap Aps with flows determined from the average of all elbow tap Aps at several plants have identified insignificant differences, so flow coefficients for each loop are not necessary. The elbow tap measurement uncertainty includes allowances for elbow tap Ap process noise, sensor and instrument rack calibration accuracy, sensor and instrument rack measurement and test equipment, sensor and instrument rack temperature efTects, and sensor and instrument rack drift [

ST-HL AE 5752 Page 20 of 29 pac

^

31. Regarding the confirmation of the correlation uticient K, a hydraulic analytic prediction method is needed to confirm that the elbow tap-based flow rate indication is within the uncertainty bounds.

Please describe your hydraulic analysis program for estimating RCS flow rate including the inputs needed and the breakdown of items used in the calculations such as: RCS loop pressure drops, flow fractions used to adjust the pressure drops in the downcomer and the core regions, pressure drops for fuel, steam generator (SG) pressure diaps, reactor coolant pump head, RCS pump wear, and tube plugging. Provide the results of using this program end also provide its accuracy in predicting RCS flow rate. The results should show that the analysis program can predict past physical changes, such as from SG tube plugging and sleeving, core bundle changes, etc., within acceptable bounds of accuracy.

RESPONSE: Westinghouse has been applying its best estimate flow analysis successfully in estimating flow and evaluating the efTect of system hydraulic changes since the program was developed in 1974-75. The methodology is supported by measurements at Prairie Island Unit 2, the only plant where RCS flows have been measured very accurately with the Leading Edge Flow Meter. The analysis, described in Section 3.5.4 and 3.5.5 and responses to Questions 21,25,26 and 29, considers the physical changes affecting RCS flow, including steam generator flow resistance changes due to plugging, sleeving or replacement, fuel assembly design changes or bypass flow changes, and pump performance. The analysis estimates differences in loop flows due to imbalanced plugging or different pump performances, and calculates reverse flows with pumps shut down, flows at zero to full power and from ambient to normal temperatures. As stated in Section 3.5, best estimate flows are expected to be within *2% of the actual flow. It is concluded that the best estimate analysis is equivalent to the analysis program recommended by the NRC.

Elements considered in the analysis are listed in Table 3, along with the component flow d

2 resistances for South Texas Project Units 1 & 2, specified in units of E " feet /gpm at loop average flow. RCS loop pressure drops are based on the sum of steam generator, RCS piping and reactor vessel nozzle flow resistances and loop flow. Reactor core and internals pressure drops are based on reactor core and internals flow resistances and total flow. Tube plugging is an input to the analysis, and steam generator flow resistance is recalculated with flow through the reduced number of tubes and with the appropriate change in tubing friction factor. The relationship between the flow reduction and number of tubes plugged is nonlinear, increasing from 0.2% to 0.3% flow per percent plugged as tube plugging increases from zero to 20%.

ST-ilL AE-5752 -

Page 21 of 29.

TABLE 1 SOUTH TEXAS PROJECT UNIT 1 ELBOW TAP DIFFERENTIAL PRESSURES

. Q tle 1

.2 3

4 5

6 7

(Baseline)

Trans;nitter -

IT-417 492.67

.49230 492.01 487.97 493.84 48637 487.17 FT-418 508.69 505.0~:.

511.84 506.66 511.76 503.70 504.42 FT 419 486.92 486.21 489.27-482.00 486.99 478.85 478.95 Loop A avg 496.09 494.51 497.71 492.21 497.53 489.64 490.18 17-427 457.20 458.74 464.07 461.54-459.94 459.53 462.22 IT-428 502.08 496.02 503.45 497.54 483.69 495.65 503.99.

FT-429 492.59

- 493.09 496.25 495.38 494.24 492.21 494.80 Loop B avg 483.96 482.62 487.92 484.82 479.29 482.46 487.00

. FT-437 485.62 483.28 48138 483.50 488.10 478.21 479.96 IT 438 520.65 512.51 513.94 50730 520.57 50637 504.61 FT-439 4% 31 488.47 49631 488.25 499.13 485.42 486.91 Loop C avg 500.86 494.75 497.21 493.02 502.60 490.00 490.49 1T 447 45339 446.59 457.42 452.13 450.49 448.43 447.43 FT-448 474.08 470.76 470.00 469.04 468.62 467.49 467.49 FT 449 453.64 456.18 453.98 453.78 456.32 451.56 451.74 Loop D avg 460.37 457.84 460.47 45832 458.48 455.83 455.55 Plant avg 485.32 482.43 485.83 482.09 484.48 479.48 480.81 T cold deg. F 562.7 561.1 563.6 563.10 562.6 557.5 556.5 Sp. Vol. Cu lV#

0.02173 0.02168 0.02176 0.02174 0.02173 0.02156 0.02153 ELBOW TAP FLOWS IN PERCENT OF BASELINE FLOW Loop A avg 100.00 99.73 100.23 99.63-100.15 98.96 98.94 Loop 11 avg 100.00 99.75 -

100.48 100.11 99.52 99.45 99.85 Loop C avg 100.00 99.27 99.70 99.24 100.17 98.52 98.50 Loop D avg 100.00 99.61 100.08 99.80 99.79 99.12 99.02 COMPARISON OF FLOW CALCUl.A TIONS Avg of Loops 100.00 99.59 100.12 99.69 99.91 99.01 99.08 Plant avg (17)-

100.00 99.59 100.12 99.69 99.91 99.01 99.07 DitTerene 0.00 0.00 0.00 0.01 *

-0.01

• 0.00 0.00 *

• Indicated differences result from calculated.alues before roundoff.

r-ST HL-AE-5752 Page 22 of 29 TABLE 2 SOUTH TEXAS PROJECT UNIT 2 ELBOW TAP DIFFERENTIAL PRESSURES Qvie 1

2 3

4 5

6 (Baseline)

Transmitter FT-417 478.27 485.08 477.71 472.51 478.71 471.56 FT-4 I8 506.07 509.70 500.65 496.72 504.47 495.60 FT 419 478.44 483.22-474.79 ~

471.27 478.23 471.75 Loop A avg 487.59 492.67-484.38 480.17 487.14 479.67 IT 427 437.55 441.78 438.01 436.77 440.29 435.91 FT-428 469.07 468.52 464.81-453.08 467.86 458.62 FT 429 457.33 459.06 453.74 446.37 455.10 447.04 Loop Bavg 454.65 456AS 452.19 445.41

- 454.42

-447.19 FT-437 538.75 528.57 524.02 522.13 527.74 518.47 17-438 517.39 518.52 519.34 512.22 522.90 512.89 FT-439 483.97 486.43 490.85 481.23 491.31 484.06 Loop C avg 511.04 511.17 511.40 505.19 513.98 505.14 FT-447 461.54 469.43 463.74 457.58 461.34 459.30 FT-448 486.49 487.47 484.86 482.24 485.84 480.75 FT-449 456.30 460.10 456.57 455.44 459.27 454.93 Loop D avg 468.11 472.33 468.39 465.09 468.82 464.99 Plant avg 480.35 483.16 -

479.09 473.96 481.09 474.25 T cold deg. F

$62.7 562.8 563.2 563.6 556.7 556.8 Sp. Vol. Cu fV#

0.02173 0.02173 0.02175 0.02176 0.02154 0.02154 ELBOW TAP FLOWS IN PERCENT OF BASELINE FLOW Loop A avg 100.00 100.52 99.72 99.30 99.52 98,75 Loop Bavg 100.00 100.20 99.77 99.05 99.54 98.74 Loop C avg 100.00 100.01 100.08 99.49 99.85 98.99 Loop D avg 100.00 100.45 100.08 99.75 99.64 99.23 COMPARISON OF FLOW CALCULA TIONS -

Avg of Loops 100.00 100.30-99.91

'99.40 99.63 98.93 Plant avg (12) 100.00 100.29 99.91 99.40 99.64 98.93

- Difference 0.00 0.00

• 0.00 0.00 0.00

• 0.09

• Indicated differences result from calculated values before roundoff.

f

ST llL-AD5752 Page 23 of 29 TABLE 3 SOUTil TEXAS PROJECT RCS COMPONENT FLOW RESISTANCES i!alil l!alL2

+ ac o

k

ST ilL-AE 5752 Page 24 of 29 TABLE 4

SUMMARY

OF PLOTTED FLOW DATA Unit i Cycle Calorimetric Flow Best Estimate Elbow Tap Normalized Flow Normalized Flow Normalized Flow w/o BE Flow using

(%)

(gpm)

.(%)

(gpm)

(%)

Confirmation

' Procedure (gpm)

(gpm) i 100.2 404716 100.0 407472 100.0 404092 404092 2

100.5 406124 99.4 405028 99.7 402887 402887 3

101.9 411628 99.4 405028 100.1 404305 403274 4

100.3 405104 99.4 405028 99.7 402745 402745 5

99.1 400544 99A 405028 99.9 403743 403274 6

99.2 400880 99.2 404212 99.0 400126 400126 7

98.9 399656 99.1 403804 99.1 400393 400393 Unit 2 1

101.1 406944 100.0 405756-100.0 402456 402456 2

100.9 406188 99.4 403320 100.3 403632 401641 3

100.1 402988 99.4 403320 99.9 401927 401641 4

100.6 404852 99.4 403320 99.3 399771 399771 5

99.3 399644 99.4 403320 99.6 40101I 401011 6

97.9 394116 98.6 400076 98.9 398178 398178 l

L i

i

e e

ST IIL AE-5752 Page 25 of 29 Figure 1 4 3C l

l l

C 2,

ST-HL AE-5752 Page 26 of 29 Figure 2 Unit i Hot Leg RTD's LoopI Loop 2 630.0

- 630.0 628.0 628.0 m

626.0 e.A- - -

-/l*s

/

g 626.0

/

e, -

/

624.0 -

\\

624.0

% g-- _, A

-a s

c 1

.s cb 622.0

- ps9 L 622.0 <

a., !

,,,4 7 * *--- --,

N q

620.0 "

" 21244'*

620.0

---I

- l 618.0 f

! 618.0 -

e H

n

- -*- - 410X

t s 616.0 -- - -*- - 420X 616.0 - -

614.0 -

410Y 614.0 --

420Y

^

682.0

" " * *

• 410Z 612.0

" *- " 420Z

- 610.9 610.0 I

1.

2 3

4 5

6 7

1 2

3 4

5 6

7 Cycle Cycle Loop 3 Loop 4 630 630.0 q

628 E 628.0 s

/

g p'

d l 626

-a

/

-+% g 626.0 s

\\

e

/

a, 624 624.0

-/

~~*

\\

622 622.0

.g c.,.

--4,,,....,

,,.es, g

jp afl*

j 620.0 j 620

'o 1

-l

~

-1

[ 618 5,

! 618.0 s

s 616-

- -*- - 430X 616.0 -

- -*- - 440X 614 ---

430Y 614.0 --

440Y 4

• • • 430Z 612.0 a ' 440Z 612 610 610.0

~

l 2

3 4

5 6

7 1

2 3

4 5

6 1

Cycle Cycle

e,.

ST-HL-AE 5752 Page 27 of 29 Figure 3 Unit 2 Hot Leg RTD's impi Loop 2 630.0 630.0 628.0 628.0 sS 626.0 C-A 626.0 f

s s'e

,s N.

~*e 624.0 p--o.,,

624.0

/* *+.r{ \\

c g

.s' F 622.0 <

L 622.0 [

s

.....*d**....,,...*%.,

0 618.0 -

l 618.0 r

s w...... m 616.g

- -*- - 410X 616.0

- -*- - 420X 614.0 -

410Y 614.0 -

420Y 612.0 -

-- * * * *' 410Z 612.0 -- " * ** " 420Z 610.0

+

g q

610.0 l

2 3

4 5

6 1

2 3

4 5

.6 Cycle Cycle Loop 3 Loop 4 630.0 630.0 62R.0 628.0 g-9 s

626.0 p '--

v-~

626.0 n.*"

/

a 624.0

/-

e9 624.0

,e j w.,,

C C

y, A.

. s b 622.0 622.0

> ---. n p -'

s 620.0 o' * * * * ' A ;,, * * *

• 2*;

I 620.0

',-g._..-

L i

(

g

)L-p

! 618.0 8 618.0 s

616.0 ---

- -*- - 430X 616.0

- -*- - 440X 614.0 --

430Y 614.0 -

^

440Y 612.0 -

a"

• 430Z 612.0 - ' * *" 440Z 610.0 610.0 I

I I

I 2

3 4

5 6

1.

2 3

4 5

6 Cycle Cycle f.

a

, ~.

1 l

)

ST-IIL-AE-5752 Pege 28 0f 29 Figure 4 l

J Impi lL410X tmp2 TL420X Imp 3 T14430X loop 4 TIM 40X l

l l

RTD Thermowc!!

/ sN l20* '

i

'N, p/

s Impi TE-410Z Loopi TE410Y loop 2 Tik420Z RTD Loop 2 TE-420Y tmp3 TL430Z Thermowell Imp 3 TIM 30Y l

Loop 4 TD440Z Imop 4 TE 440Y ilot Leg Pipe Looking Toward the Steam Generator y-i s,

• ['

i a

_, n t

t s-it d

U-y n'

y SECT. " A -A" Cold Leg Elbow Tap Orientation 4

i

~.

O s

Attachm:nt 4

,e.

ST HL-AE 5752 -

Page 29 of 29 Figure 5 South Texas Unit 1 RCS Flow History 4:5000 a

,r -%

As0000 e

s

,/

's s 40$000 Q-a"----

_...... *...... - ;> 's_

1

...e.

.s 400000

-u f

C

- -*- - c.wetrsa E 395000 Bat talmate

• * * ** *

• Ltbew Tep (w/o DE Flew 390000 cennrmsenen) -

e Elben Tap using Precedere 3N5000 I

2 3

4 5

6 7

Cycle Number South Texas Unit 2 RCS Flow History 41$000 - ---

410000

+ - - - - -

~

Nm _ __,,

"~~Y%s m

g

......... r s s

.gg(Hy)g 4_%_.

{

- -+= - Calorimetrie

%g'

• 395000 Best Latimate N

• * * ** *

• Elbow Top (w/o BE Flow 390000 Ceanraatte )

Dbew Tap untag Preesdure 385000 1

1 2

3 4

5 6

Cycle Number

n O

o

ATTACIIh!ENT 5 Amended Response to Request to AdditionalInformation PROPRIETARY CLASS 2C