ML20054L448

Application for Amend to App a of License DPR-34,releasing Facility from 70% Power Restriction.Response to ORNL 820315 Questions Encl

ML20054L448
Person / Time
Site:	Fort Saint Vrain
Issue date:	07/06/1982
From:	Lee O PUBLIC SERVICE CO. OF COLORADO
To:
Shared Package
ML20054L444	List: ML20054L443 ML20054L448 ML20054L451 ML20054L453 ML20054L459
References
NUDOCS 8207080130
Download: ML20054L448 (9)

Text

BEFORE THE UNITED STATES NUCLEAR REGULATORY COMMISSION In the Matter of the Facility Operating License)

)

of ) Docket No. 50-267

)

PUBLIC SERVICE COMPANY OF COLORADO )

Application for Amendment to Appendix A of Facility Operating License License No. DPR-34 0F THE PUBLIC SERVICE COMPANY OF COLORADO FOR THE FORT ST. VRAIN NUCLEAR GENERATING SiATION This application for Amendment to Appendix A of Facility Operating License, License No. DPR-34, is submitted for NRC review and approval.

Respectfully submitted, PUBLIC SERVICE COMPANY OF COLORADO By S/0. R. Lee .

O. R. Lee, Vice President KELLV, STANSFIELD & 0'DONNELL Bryant O'Donnell Robert E. Thompson Public Service Company Building Denver, Colorado 80202 Attorneys for Applicant 8207080130 820706 PDR ADOCK 05000267 P PDR

STATE OF r0LORADO )

) ss.

CITY AND COUNTY OF DENVER)

0. R. Lee, being first duly sworn, deposes and says; That he is Vice President of Production of Public Service Company of Colorado, the Licensee herein; that he has read the foregoing Application for Amendment to Appendix A of Facility Operating License and knows the contents thereof, and that the statements and matters set forth therein are true and correct to the best of his knowledge, information and belief.

S/0. R. Lee

0. R. Lee Subscribed and sworn to before me this 6th day of July, 1982.

Witness my hand and official seal.

My commission expires: C/to d E3h 7 10h 3 .

O

_da -

)

hublic p*/

Notary

ATTACHMEF_ A P-82229 Response to ORNL Questions ORflL Letter Dated Naren 15, 1982 i

l I

i l

l l

., ,n .-

. Page 2 LCO 4.1.7 ORNL: "We would recommend, however, that the method of calculating ' core average outlet temperature' be defined in LCO i 4.1.7 since the primary concern is with the erroneous

! temperature indications. The most accurate calculation would use flow weighting (based on orifice positions), with measured temperatures for regions 1-19 and 21-31 and calculated i temperature for the other seven regions."

J 1 PSC Response: A definition of calculated core average outlet

, temperature is currently provided in Technical Specification

. 2.20. As a result of Type II flow-induced region outlet temperature measurement errors, it will be necessary to add new definitions. In order to maintain the current structure of the Technical Specifications to the greatest possible extent, the i Specification 2.20 will remain as is and new definitions 2.21

! and 2.22 will be added. The new definition has been based, as

! recommended by ORNL, on a combination of measured and calculated j region outlet temperatures, as appropriate.

While the flow weighting method mentioned by ORNL is the most I accurate method for calculation of core average outlet l temperature, it is not as easy for the operator to implement as 4 other methods. The current Technical Specification 2.20 does not employ flow weighting to define core average cutlet temperature nor is the revised Technical Specification 2.21 based on flow weighting. Based on ORNL's comments we did go back and compare average outlet temperature calculations for Cycle III operation utilizing data at 60?s, 70?;, and 90!4 power on

the basis of flow weighted averages and arithmetic averages. We found that the two calculations were within 5 F. Given this comparison, and the fact that flow weighted averages are extremely cumbersome for the operator, we believe the proposed definition is most appropriate.

ORNL: "In choosing ' comparison' regions....., we recommend that

! the LC0 should specify that paired regions should be of the same

' type', (e.g., only 5-column regions should be compared with other 5-column regions)."

PSC Response: It is preferable, but not necessary, that paired j regions be of the same type. During RT-500K testing, provisions

were made for comparing 5-column regions with 7-column regions, and vice versa. In one instance, such a comparison was successfully used. Orifice valve control tables will be i generated for any of the four possible combinations of regions 4

(5-5, 7-7, 7-5, 5-7). When possible, paired regions will be of the same type. However, given that appropriate calculations are made, the operator should not be restricted from choosing other regions should it become desirable to do so.

I Page 3

! ORNL: "As in the original LCO, the revised version should

! include a ' Basis' section that has the original explanation plus a discusssion of the Type II flow errors."

]

PSC Response: The revised Technical Specifications include bases. For LC0 4.1.7 the original basis has been retained to

the extent that it remains applicable to the revised LCO.

i GA Report on FSV Tests up to 100% Power

! ORNl.: "The report noted (page 2-3) that tests showed successful operation up to 100% power and 5.0 psi core pressure drop. Does this mean higher core pressure drops will not be tried or

}

allowed (vs. FSAR design value of 8.4 psi)?"'  !

PSC Response: No such implication was intended. As the reactor approaches an equilibrium core configuration, the' core pressure drop at 100% power will approach the design value of 8.4 psid.

ORNL: "In the discussion of measured region outlet temperature redistributions (page 4-4), it is noted that during a power

? increase one expects all regions (outlet temperatures) to 1 increase. The extent of a change and its polarity, however,

depend on the changes in power-to-flow (P/F) ratio and core inlet temperature (Tci). Based on the data given for the ,

l '

l November 5, 1982, redistribution event, the mean outlet

temperature change due to the change in P/F alone would only be l about 5 F, while the GA calculations (Figures 4-13 and 14) show l generally larger increases. The effect of the T ci change should be noted in such an analysis, since a large T ci increase would ,

tend to reduce or even reverse the direction of outlet

temperature changes from the errant regions."

i i

PSC Response: The expected (calculated) region outlet '

i temperature increases shown in Figures 4-13 and 14 reflect both j the power-to-flow ratio and the measured core inlet temperature

before and after the redistribution event. In the final report, the discussion on Pages 4-4 and 4-16 has been slightly expanded to clarify this point.

ORNL: "We have no problems with the discussion and analysis of j nuclear channel deviations, core reactivity perturbations, and core resistance changes, but would suggest keeping an eye on the drift in nuclear channel No. 5 as seen in Figure 4-5. The drift  !

i rate appears to be approximately 1% per 3 hours3.472222e-5 days <br />8.333333e-4 hours <br />4.960317e-6 weeks <br />1.1415e-6 months <br />." (

I

Page 4 PSC Response: The observed drift rate of approximately 1% per 3 hours3.472222e-5 days <br />8.333333e-4 hours <br />4.960317e-6 weeks <br />1.1415e-6 months <br /> does not appear to be related to the redistribution phenomenon. A possible cause for the " drift" could be a change in the rod shadowing effect as the shim group or regulating rod positions changed. We will continue to monitor the drift in future operations.

ORNL: "In the analysis of Type 11 flow effects (Section 4.2.5),

it should have been noted that the derivation of calculated errors (T g - T g in Figure 4-16) requires specifying a region outlet gas temperature gT as well as the Type II gas flow and temperature. An earlier GA analysis of this effect used the TAC 3D code and assumed g T = 1,200 F, Tyy (Type II gas) = 900 F, and a heat transfer coefficient to the outside of the thermocouple sleeve based on 50% power operation (R. H. Boonstra, February 6, 1978). In order to predict as much as a 400 F Type II error (Figure 4-16) it would be necessary for T

g

-T yy to be at least 500 F. Assuming the current analysis is based on an extrapolation from Boonstra's work, the differences in assumptions should be considered when interpreting the data.

For example, estimates of some of the Tyy values can be obtained from gap thermocouple readings: for the November 5, 1981, 82%

power event, TC No. 5. approximately 800 - 900 F for RR No. 37; ,

TC No. 14 approximately 900 F for the RR No. 35; and TC No. 9 approximately 1,200 F for RR No. 36. From the high value for RR No. 36's Tyy, one would expect a minimal impact of Type II flow on the RR No. 36 response during a redistribution, and this is borne out by the data (page 4-3)."

PSC Resro g : ORNL's observations regarding the calculations in Section 4.2.5 are essentially correct. In the final report, the discussion in Section 4.2.5 has been clarified with regard to i

the assumed temperatures of the Type 11 flow and of the actual l

region outlet helium.

ORNL: "The GA analysis concludes that ef fect of T>oe II flow errors would show up only in regions at the fM end of the thermocouple strings. ..We agree that., .it appears that Tyne II flow errors are minimal elsewhere. However, there may be some cases where it could be sionificant, such as instances where

adjacent regions have significantly different sutlet l temperatures or where an unusually cool gap flow occurs."

PSC Response: As noted in Section 4.2.6.1 of the report, both calculations and thermocouple traverses have confirmed that only the gap flow between regions and the side reflector can be of sufficiently low temperature to cause significant discrepancies.

l Differences in outlet temperatures between adjacent regions i sufficient to cause significant Type II flow-induced outlet temperature measurement errors have not been experienced during normal operation. In 1976 and 1977, tests were performed

Page 5 (RT-417) in which the outlet temperatures of adjacent regions were intentionally adjusted to maximize their difference. The purpose of these tests was to determine whether the outlet temperature of one region could influence the indicated outlet temperature of the adjacent region. Although adjacent region outlet temperature swings as high as 320 F were achieved, no significant effects were detected.

ORNL: "The most suspect area (other than the NW border) for Type II flow errors is the SE border, at the far ends of the thermocouple strings. The report argues, however, the Type II flow from the cooler reflector (in the SE) is inhibited by the 8 inch long ' bullet nose' on the end of the thermocouple string.

Also, the jaws flow and the Type II flow discrepancies differ in the way they respond to pressure drop and power changes, and in their time response characteristics (page 4-31 and Figure 4-18).

While data from the latest tests tend to confirm these points, some earlier results are not as convincing. For example, the responses of region 36 (NW) and region 25 (SE) to the December 13, 1980, redistribution at 60*4 power are nearly identical."

PSC Response: ORNL's observation that the response of Region 36 and Region 25 to the December 13, 1980, redistribution are nearly identical is correct. However, the response of these regions at that time was characteristic of a change in jaws flow, not a change in Type II flow. As discussed in Section 4.3.2 of the report, the redistribution phenomena observed in the seven NW boundary regions are explained by a combination of changes in jaw flow, increased bypass cooling, and Type II flow effects. The relative extent to which each of these contribute to the observed behavior of any one of the seven regions during a given redistribution varies from event to event. In the case cited by ORNL, the contribution of Type II flow to the observed response of Region 36 was minimal, and jaws flow effects dominated.

As is also discussed in Section 4.3.2 of the report, the measured outlet temperatures of some of the boundary Regions 21-31 are cooler than expected after a redistribution due to a combination of changes in jaw flow and increased bypass cooling.

Phenomena (.e.g, transient response) characteristic of Type II flow have not been observed in these regions.

Page 6 ORNL: "Since this particular conclusion (that Type II flow errors now effect only the NW border regions) is vital to the setting of the LCO 4.1.7 limits, we recommend that evidence of Type II flow errors in other regions be looked for whenever redistribution events occur. Such evidence should include consistency checks for region outlet versus steam generator inlet temperatures. In any case, if a non-NW border region behavior is somewhat suspect, we would recommend that it not be used as a comparison region."

PSC Response: ORNL's recommendation that if a non-NW boundary

-region behavior is suspect, it should not be used as a comparison region is reflected in the planned method of operation. As noted in the final report and in these responses, there is no evidence of Type II flow-induced errors in regions other than the seven NW boundary regions. However, appropriate technical specification surveillance requirements have been

. proposed in order to assure that suitable comparison regions are

,used.

I l

I l

l

I l

ATTACHMENT B P-82229 Testing and Operation of Fort St. Vrain Up To 100% Power I

i l

l

.. . . . . ~ . . - _ . - . - _ _ _ - - - _ _ _ _ _ _ - . _ - - - - - . - _ _ _ - - . . _ _ _