ML15240A154

Staff Observations of Sump Strainer Head Loss Testing at Alden Laboratory for Generic Safety Issue 191

ML15240A154
Person / Time
Site:	Saint Lucie
Issue date:	09/08/2015
From:	Robinson L Safety Issues Resolution Branch
To:	Victor Cusumano Safety Issues Resolution Branch
Robinson L
References
Download: ML15240A154 (9)

Text

September 8, 2015 MEMORANDUM TO: Victor G. Cusumano, Chief Safety Issues Resolution Branch Division of Safety Systems Office of Nuclear Reactor Regulation FROM: Lindsay R. Robinson, Reactor Systems Engineer

/RA by ARussell for/

Safety Issues Resolution Branch Division of Safety Systems Office of Nuclear Reactor Regulation

SUBJECT:

STAFF OBSERVATIONS OF SAINT LUCIE UNIT 1 SUMP STRAINER HEAD LOSS TESTING AT ALDEN LABORATORY FOR GENERIC SAFETY ISSUE 191 On August 4 and 5, 2015, U.S. Nuclear Regulatory Commission (NRC) staff traveled to the Alden Research Laboratory in Holden, Massachusetts to observe testing associated with the resolution of Generic Safety Issue 191 (GSI-191). The objective of this trip was to observe sump strainer head loss tests being conducted for St. Lucie Unit 1. The participating NRC staff members were Ashley Smith and Lindsay Robinson of the Safety Issues Resolution Branch in the Division of Safety Systems and Marioly Diaz-Colon and Paul Klein of the Steam Generator Tube Integrity and Chemical Engineering Branch from the Division of Engineering. The staff interacted with personnel from ALDEN Research Laboratory, Inc. (Alden), Enercon Services, Inc. (Enercon), Calvert Cliffs Power Station (Calvert Cliffs), and Wolf Creek Generating Station (Wolf Creek).

The enclosure summarizes the staffs visit on August 4-5, 2015.

CONTACT: Lindsay R. Robinson, NRR/DSS/SSIB 301-415-4115

September 8, 2015 MEMORANDUM TO: Victor G. Cusumano, Chief Safety Issues Resolution Branch Division of Safety Systems Office of Nuclear Reactor Regulation FROM: Lindsay R. Robinson, Reactor Systems Engineer

/RA by ARussell for/

Safety Issues Resolution Branch Division of Safety Systems Office of Nuclear Reactor Regulation

SUBJECT:

STAFF OBSERVATIONS OF SAINT LUCIE UNIT 1 SUMP STRAINER HEAD LOSS TESTING AT ALDEN LABORATORY FOR GENERIC SAFETY ISSUE 191 On August 4 and 5, 2015, U.S. Nuclear Regulatory Commission (NRC) staff traveled to the Alden Research Laboratory in Holden, Massachusetts to observe testing associated with the resolution of Generic Safety Issue 191 (GSI-191). The objective of this trip was to observe sump strainer head loss tests being conducted for St. Lucie Unit 1. The participating NRC staff members were Ashley Smith and Lindsay Robinson of the Safety Issues Resolution Branch in the Division of Safety Systems and Marioly Diaz-Colon and Paul Klein of the Steam Generator Tube Integrity and Chemical Engineering Branch from the Division of Engineering. The staff interacted with personnel from ALDEN Research Laboratory, Inc. (Alden), Enercon Services, Inc. (Enercon), Calvert Cliffs Power Station (Calvert Cliffs), and Wolf Creek Generating Station (Wolf Creek).

The enclosure summarizes the staffs visit on August 4-5, 2015.

Docket No.: 50-335

Enclosure:

Trip Report CONTACT: Lindsay R. Robinson, DSS/SSIB 301-415-4115 DISTRIBUTION:

Public ASmith LRobinson JStang ARussell GKulesa RidsNrrPMStLucie PKlein MDiaz SSmith MYoder VCusumano RidsRgn2MailCenter Accession No.: ML15240A154 NRR-106 OFFICE NRR/DSS/SSIB/RSE NRR/DSS/SSIB/BC NRR/DSS/SSIB/RSE NAME LRobinson/ARussell for VCusumano/JStang for LRobinson/ARussell for DATE 08/31/2015 09/02/2015 09/08/2015 OFFICIAL RECORD COPY

OBSERVATIONS OF TESTING AT ALDEN TEST FACILITY AUGUST 4 AND 5, 2015 DOCKET NO.: 50-335 OBJECTIVE On August 4 and 5, 2015, U.S. Nuclear Regulatory Commission (NRC) staff observed sump strainer head loss tests associated with the resolution of Generic Safety Issue 191 (GSI-191).

The objective of this trip was to observe the experimental testing to better understand the St. Lucie Unit 1 approach to GSI-191 resolution and to inform staff review of similar methods that may be used by other licensees.

This particular test (identified by Alden as 1152PSLTST-455-00) was conducted by Alden to determine, by incremental debris addition representing increasing Loss-Of-Coolant Accident (LOCA) pipe break sizes, the maximum pipe break size that would still maintain acceptable Emergency Core Cooling System (ECCS) performance at St. Lucie Unit 1. The configuration of the test strainer assembly and the flume geometry remained constant. The strainer approach velocity was constant except for flow sweeps performed to characterize the debris bed prior to chemical precipitate addition. The quantities of particulate and fiber, along with the particulate-to-fiber ratio, varied during the test to be representative of each incremental pipe break size.

OVERVIEW OF FACILITY The test tank occupies a 12.5 ft x 5 ft x 5.5 ft area and contains a strainer, simulated containment piping, and the geometry to simulate the strainer keyway configuration inside containment (see Figure 1). The test strainer area is 360.6 ft2 and matches the configuration of one keyway module in the plant. For the head loss tests, the test fluid is borated and pH adjusted with sodium hydroxide to be more representative of the plant compared to using potable water. Total volume of water in the tank is about 165 ft³ or 1,232 gallons. The tank is elevated above the ground to create space for piping, a heat exchanger, sensory equipment, test leads, and a main recirculating pump. There are two debris hoppers located next to the tank to provide for mixing and entry to the test tank for fine debris additions. The fine and small fibrous debris solutions were created in mixing tanks located above and next to the test tank.

TEST SETUP To set up the head loss testing, the fibrous debris is prepared to meet either fiber fines (Class 2) or fiber smalls (sizes varying from multiple attached or interwoven strands, also known as Class 3 fiber fines, up to 6 in length) requirements. The fiber debris solution is prepared by placing a predetermined amount (based on procedure) of 2 inch x 2 inch squares of baked Nukon into a mixing container (see Figure 2). Using high powered water jets on a timed interval, the Nukon is broken down into fines until it reaches the desired fine characteristic as dictated by the procedure. Once the fiber debris solution is prepared as either fines or smalls, additions of particulates, Calcium-Silicate (Cal-Sil), and dirt and dust are added to the mixtures and manually combined with a large stirring rod.

ENCLOSURE

The amount of Nukon, particulate, Cal-Sil, dirt and dust used is determined by the break size being evaluated (see Table 1).

In the test setup (see Figure 3), the fine fiber debris solution, as prepared above, is poured into each debris hopper in equal portions. The debris hopper then further breaks down the fibers before adding the debris solution into each end of the test tank. Once the fiber enters into the test tank, the suction strainer then draws suction on the solution using the main recirculating pump. A small portion of the solution is then heated by the heat exchanger before returning to the debris hopper where it gets returned to the test tank or the mixing flow lines. The small fibrous debris solution has a similar transport flow path in the test setup except that it is added by pouring it through a cross-sectional pipe segment at the front of the test tank only instead of debris hoppers at each end of the test tank. The test loop is held at a constant 120°F by directing a portion of the overall recirculation flow over an in-line heat exchanger. The test temperature helps lower the air content in the test tank fluid. The flow rate is set to match a target approach velocity of 0.0026 ft/sec. The initial debris additions to the test represented break size G (Table 1), which simulated an 8.5-inch break.

Figure 1: Test Tank Design Concept, Isometric View Figure 2: Nukon Fiber Before and After Preparation

Figure 3: Piping and Instrumentation Diagram (P&ID)

Table 1: Debris Content in Breaks of Interest Pipe ID Cal-Sil Particulates Nukon (lbs) Temp-Mat (lbs)

Designation (in) (lbs) (ft3) Fines Smalls Fines Smalls A 42 1170.16 11.71 513.9 570.24 6.08 6.41 B 31 341.34 7.94 351.88 385.1 3.41 1.94 C 31 556.95 9.29 282.21 285 3.78 3.14 D 30 258.69 7.76 332.51 369.16 48.58 68.61 E 30 531.59 8.96 278.67 284.2 4.26 5.06 F 10.126 88.44 7.33 88.7 77.34 0.5 0.22 G 8.5 61.33 7.39 53.09 38.21 1.33 1.24 TEST PERFORMANCE The head loss testing began by adding the fine fiber plus particulate debris solution (see Figure 4). Once all the fine debris solution for a given break size was added, then the small fiber plus particulate debris solution was added. The testing alternated between adding fine debris solution and then adding the small debris solution until all required debris was added to reach a specified break size per Table 1. Each addition took at least 90 minutes before the next fiber debris solution was added to the tank. Over time, the sump strainer became covered with fibrous debris. The small fibers and some fines settled to the bottom of the tank and along the suction screen (see figure 5).

Figure 4: Test Tank 5 Minutes into Designation G and 40 Minutes into Designation G, Respectively Figure 5: Clean Test Tank and Test Tank 20 hours0.833 days <br />0.119 weeks <br />0.0274 months <br /> After Initial Fiber Addition, Respectively

While batches of fiber and particulate were being added to the maximum debris head loss test, Alden personnel began preparing chemical precipitates for later additions to the test. Alden had the chemicals and necessary equipment for generating large quantities of both Aluminum Oxyhydroxide and Sodium Aluminum Silicate (SAS) precipitates using the methodology outlined in WCAP-16530-NP-A, Evaluation of Post-Accident Chemical Effects in Containment Sump Fluids to support GSI-191. The NRC staff had the opportunity to observe preparation of a 500 gallon batch of SAS precipitate.

For the preparation of SAS precipitate, potable (tap) water was mixed with aluminum nitrate (Al (NO3)3) and sodium ortho-silicate (Na4SiO4) solution. The ingredients were added slowly and mixing was allowed to continue for 1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br /> to allow the chemical reaction to complete.

Approximately one hour after the precipitates were prepared, the Alden staff began a settlement test in a graduated cylinder. The one-hour precipitate settlement was approximately 9.3 milliliters (mL) out of 10 mL. In other words, the cloudy portion of precipitate covered up to the 9.3 mL mark with 0.7 mL of clear fluid from the 9.3 mL mark to the 10 mL mark. This measurement easily met the one-hour settlement criteria specified in WCAP-16530-NP-A and was acceptable to the NRC staff. Figure 6 shows various points in the preparation of SAS chemical precipitate.

Figure 6: Preparation of SAS chemical precipitate

The lead test engineer for Alden also walked down the piping from the chemical precipitate tanks to the test tank and described how precipitate addition would be performed. Although the NRC staff was not able to observe chemical precipitate addition to the St. Lucie Unit 1 head loss test due to testing delays and schedule constraints, the NRC staff has observed chemical precipitate addition to head loss tests during previous visits to the facility. As an example, see Staff Observations of Testing for Generic Safety Issue 191 during July 12-14, 2010 trip to Alden Test Facility for PCI Strainer Tests, (Agencywide Documents Access and Management System (ADAMS) Accession Number ML102160226).

TEST RESULTS The NRC staff observed multiple additions of debris but was unable to observe the entire test due to the length of the test and schedule constraints. At the time of the staffs departure on August 5, all debris representative of a 30-inch break and approximately one half of the incremental debris to go from the 30-inch break to a 42-inch break had been introduced to the test loop. Head loss at that point was approximately 1 ft H20. The NRC staff and the St. Lucie representatives had a closeout discussion and agreed to have a follow-up phone call after the current test and subsequent thin bed test had been performed.

SUMMARY

The NRC staff became more familiar with the St. Lucie Unit 1 approach to addressing GSI-191 and the Alden Test Procedure, 1152PSLTST-455-00, being used to evaluate its maximum head loss requirements that must be met for ECCS.

Overall, the trip was effective and will assist the review of the licensees submittal to closeout GSI-191.

FOLLOW-UP PHONE CALL The NRC staff, St. Lucie representatives, and Alden representatives participated in a phone call to discuss the results from the St. Lucie Unit 1 maximum load and thin bed strainer tests and the plans for the third confirmatory strainer head loss test. The licensee representatives indicated that the chemical precipitate addition to the initial head loss test (representing a maximum calcium silicate, near maximum fiber load) resulted in approximately a 4 times increase in head loss from the conventional fiber and particulate debris head loss. The total head loss was still well within the acceptance value.

The total head loss from the thin bed test was approximately 25 percent of the first maximum load test demonstrating that the thin bed condition is not more challenging with respect to strainer head loss for St. Lucie Unit 1. The licensee indicated that the plans for the confirmatory test are to add approximately 10 percent more fiber and 30 percent less calcium silicate relative to the first test. After some discussion on debris sequencing, the licensee indicated that they will be adding all fine fibrous debris first followed by all small fibrous debris. The NRC staff requested and the licensee agreed to have an additional phone call after the confirmatory test results are available.