ML20028E961

Forwards Assessment of Hydraulic Peformance of WNP-1 Containment Recirculation Sumps (Hydraulic Evaluation of Design of Containment Recirculation Sumps) & (Hydraulic Model Study of Containment Recirculation Sumps).

ML20028E961
Person / Time
Site:	Washington Public Power Supply System
Issue date:	01/11/1983
From:	Bouchey G WASHINGTON PUBLIC POWER SUPPLY SYSTEM
To:	Harold Denton Office of Nuclear Reactor Regulation
Shared Package
ML20028E962	List: ML20028E961 ML20028E997
References
GO1-83-0011, GO1-83-11, NUDOCS 8301280301
Download: ML20028E961 (5)

Text

4 -

Washington Public Power Supply System P.O. Box 968 3000 GeorgeWashingtonWay Richland, Washington 99352 (509)372-5000 Docket No. 50-460 January 11, 1983 G01-83-00ll Mr. Harold R. Denton, Director Office of Nuclear Reactor Regulation U.S. Nuclear Regulatory Commission Washington, D.C. 20555 Attention: Document Control Desk

Subject:

NUCLEAR PROJECT NO. 1 RECIRCULATION SUMP TESTING

Reference:

1) Letter, DL Renberger, Supply System to Olan D. Parr, NRC, same subject, dated 05/15/79

2) Letter, DL Renberger, Supply System to BJ Youngblood, NRC, same subject, dated 08/14/80

3) Letter, BJ Youngblood, NRC to DL Renberger, Supply System, same subject, dated 10/01/80

4) NRC Generic Letter No. 82-14 In References 1 and 2, the Supply System requested approval to perform model tests to verify recirculation sump design and available NPSH in lieu of the in-plant tests committed to in the WNP-1 PSAR.

In Reference 3, the NRC indicated we might be relieved of our commitment to perform in-plant testing but asked questions that could not be answered without selecting a contractor and developing a test program. After developing the test program we elected to continue with the testing.

Forty copies of the ' -5t program and results are enclosed as requested by Reference 4.

0 0\

8301280301 830111 I i f3%

PDR ADOCK 05000460

Y O H. R. Denton Page 2 January 11, 1983 G01-83-00ll The Supply System again requests NRC approval to not perform the in-plant sump testing committed to in the PSAR. As explained in Reference 1, we believe formal approval of this deviation from the PSAR is required because the commitment was made in response to a specific NRC request and only after lengthy discussion with the staff.

Attached to this letter is a summary of the test program and a discussion as to why we believe in-plant testing is not required.

G. D. Bouchey Nuclear Safety & Regulatory Programs GDB: AGH:caa Attachments cc: M. Thadani, NRC NS Reynolds, Debevoise & Liberman G. Valentenyi, UE&C, PA (8U6)

FDCC (899)

ORM (847) l l

l l

i m e - -

ATTACHMENT APPLICABILITY OF HYDRAULIC MODEL TESTING IN THE PREDICTION OF WNP-1 RECIRCULATION SUMP PERFORMANCE Regulatory Guide 1.79, Revision 1, Section C.1.b.2 requires that a preoperational recirculation test be performed to demonstrate the capability to realign valves and pumps to recirculate coolant from the containment sumps into the Reactor Coolant System. It specifies that the testing should include taking suction from the sump to verify vortex control and acceptable pressure drops across screening, suction lines, and valves. It also requires that the testing verify that the available net positive suction head (NPSH) is greater than that required at accident conditions.

Although the Regulatory Guide calls for an in-plant test, activity on other dockets indicates that the NRC recognizes that the adequacy of the system in reference to vortex control and pressure drop can be effectively demonstrated by a hydraulic model test. Model tests offer many other advantages as indicated below and, therefore, many plants have used them to prove the adequacy of their sump design.

The advantages of a scaled hydraulic model test are:

- A properly designed scale model, constructed partially of transparent materials, allows easier observation of vortex activity and the cresence of trapped air within the sump structure.

- Laboratory testing of a scale model facilitates accurate measurement of screen, sump and suction pipe entrance losses.

- Laboratory testing also allows a greater range of parameters than is feasible with an in-plant test. For example, various combinations of flow conditions (flow rate, screen blockage, etc.) can be tried and the most critical ones selected for further testing.

- If design improvements are needed, they can be tried and proven more easily and economically on a model than on the plant.

In light of the above considerations, a complete hydraulic investigation (including a model study) of the WNP-1 recirculation sumps was conducted, although these have been designed according to the cr'teria of Regula-tory Guide 1.82 and are not expected to show any design deficiency.

Alden Research Laboratory, who has considerable expertise in recircu-lation sump studies, was contracted for the study. The study was conducted systematically in two phases. The first phase consisted of a hydraulic evaluation of the sumps in light of theoretical and model i studies reported in literature, as well as ARL's own accumulated expe-i rience. This study showed that the sumps have all the desirable features of a good sump design and were not likely to experience any significant problems. However, to confirm these predictions, a physical l scale model test was conducted. The findings of both phases of study l .tre contained in the attached report: " Assessment of Hydraulic Performance of the WNP-1 Containment Recirculation Sumps".

I

The principal features of the test program and the findings are summarized below:

1. The model was c,nstructed to a geometric scale of 1:2.5; its boundaries (

• tending sufficiently to include all physical features likely to affect the hydraulic conditions.

The model was operated according to the Froude law of 4 similarity and at large Reynglds numbers (2.1-3.6 x 10 for the sumps and 1.62-2.43 x 10 for the outlet pipes). The geometric scale and the Reynolds numbers chosen are large enough t- eliminate scale effects and to yield flow patterns and loss soefficients directly applicable to the plant.

2. Tests were conducted with severely perturbed approach flow conditions, using various flow distribution and screen blockage schemes. Screen blockage ranged up to 75% although Regulatory Guide 1.82, Section C.7, requires consideration of only 50%

blockage. These, together with low sump water depth and high outflow rate yield more severe flow conditions that the plant is likely to experience.

3. The model was also operated under the equal velocity rule (i.e., with pipe velocities equal to prototype pipe velocities) for some of the critical tests. Since the latter velocities are much greater than t's Froude-scaled velocities, it is further assured that scale effects are eliminated and conser-vative measures of vortex activity and head losses are obtained.

4. The tests showed that even under the worst screen blockages and flow distributions, at equal velocity and Froude-scaled velocity, the vortices were no more severe than surface dimples. No air drawing vortices were encountered under any test conditions, nor was any accumulation of air observed under the cover plate, as it is to be vented.

5. The swirl angles measured in the suction pipes were small.

As a result of decay agong the suction pipes, these will be reduced to less than 2 at the pump inlets, a value too small to degrade the pump performance.

6. The head loss coefficients representing the combined effects of screens, trash racks, sump geometry and suction pipe entrance, were determined to be 0.06 for the DHR sump and 0.15 for the CSS sump. Since scale effects have been eliminated, these values are directly applicable to the plant and may be used in calculations to verify whether the pumps have sufficient net positive suction head available (corrected for temperature) under accident conditions.

The only remaining purpose of an in-plant recirculation test would be to physically confirm the pressure drop calculations and verify that adequate net positive suction head is available to the DHR and CSS pumps. As indicated in the sump data section of the test report, the NPSH available as computed, is at least 67% greater than the NPSH required. Due to this large margin, it is believed that the normal calculational methods are sufficient to show that the DHR and CSS pumps are always assured of the required NPSH.

The above description of the test program, the test results and their application clearly demonstrate the adequacy of the WNP-1 DHR and CSS sumps. We believe that the model test, as performed by the Alden Research Laboratory, meets the intent of the pre-operationsal recircu-lation test required by Regulatory Guide 1.79 and therefore an in-plant test is not necessary.

I}.

,1 . . .

ASSESSMENT-DF THE HYDRAULIC-PNRFORMdNC - J OF THE WNP-1LCONTAINMENT. a

^'

RECIRCULATION SUMPS -

. o.

g

~

HYDRAULIC EVALUATION OFTHE D5 SIGN OFL f CONTAINMENT RECIR'CULATION SUMPS ,

I .

MAHADEVAN PADMANABHAN HYDRAULIC MODEL STUDY OF CONTAINMENT RECIRCULATION SUMPS -

BY; ,

WILLIAM W. DURGIN '

, , '[

JOHN F. NOREIKA1 / -

9-c

~

WASHINGTON PUBLIC POWER SUPPLY SYSTEM -

p _ _  : .

v.

NUCLEAR PROJECT NUMBE,R 1 (WNP-1)' ,

r my PREPAR8D FOR .

lH M BDQlUBOIS.

& Constructors M i i

' ~

PREPARED BY ,

ALDEN~ RESEARCH LABORATORY: '

WORCESTER : -POLYTECHNIC INSTITUTE ,

- ~

DECEMBER 1982 ,

, . n . :..

r' g R I .,4 9