Text

Responses to Requests for Additional Information Regarding Generic Letter 2003-01, Control Room Habitability
	ML073110132
Person / Time
Site:	Callaway
Issue date:	10/30/2007
From:	Graessle L; AmerenUE
To:	; Document Control Desk, Office of Nuclear Reactor Regulation
References
	GL-03-001, ULNRC-05448
	Download: ML073110132 (34)
	v • d • e

AmerenUE PO Box 620 Callaway Plant Fulton, MO 65251 October 30, 2007 U.S. Nuclear Regulatory Commission Attn: Document Control Desk Mail Stop P1-137 Washington, DC 20555-0001 ULNRC-05448 WAmereU Ladies and Gentlemen:

DOCKET NUMBER 50-483 CALLAWAY PLANT UNIT 1 UNION ELECTRIC CO.

FACILITY OPERATING LICENSE NPF-30 RESPONSES TO REQUESTS FOR ADDITIONAL INFORMATION REGARDING GENERIC LETTER 2003-01, "CONTROL ROOM HABITABILITY"

References:

1. AmerenUE Letter ULNRC-04885, "Response to Request for Information on Generic Letter 2003-01, 'Control Room Habitability,"' dated August 11, 2003.

2. AmerenUE Letter ULNRC-05104, "Response to Generic Letter 2003-01, 'Control Room Habitability," dated December 17, 2004.

3. AmerenUE Letter ULNRC-05298, "Amended Response to Generic Letter 2003-01, 'Control Room Habitability," dated June 6, 2006.

4. NRC Letter dated November 7, 2006, from Jack Donohew, NRC to Charles Naslund, AmerenUE.

5. AmerenUE Letter ULNRC-05427, "Response to Request for Additional Information Concerning Generic Letter 2003-01, 'Control Room Habitability,"' dated July 16, 2007.

By letter dated August 11, 2003 (Reference 1), Union Electric Company (AmerenUE) transmitted its preliminary response to NRC Generic Letter 2003-01, "Control Room Habitability." In the response AmerenUE identified that it would need to develop an alternate method of integrated in-leakage testing because of the Callaway Plant Control Room/Control Building design. Ameren UE contracted with Brookhaven National Laboratory and performed this alternate integrated in-leakage testing for Callaway Plant on September 17-19, 2004 using the Atmospheric Tracer Depletion (ATD) method. The results of the testing were transmitted by AmerenUE letters dated December 17, 2004 and June 6, 2006 (References 2 and 3, respectively).

a subsidiary ofAmeran Corporation K4 a.

ULNRC-05448 October 30, 2007 Page 2 During its review of the responses, the NRC developed a request for additional information (RAI) that was transmitted by NRC letter dated November 7, 2006 (Reference 4). AmerenUE responded to that RAI in its letter dated July 16, 2007 (Reference 5).

From its ongoing review of AmerenUE's responses concerning control room habitability, the NRC staff has transmitted further requests for additional information (RAIs) for which responses are needed in order to support completion of the NRC's review. Accordingly, this letter provides AmerenUE's responses to the NRC's RAIs in Attachment 1. Within the attachment each of the individual questions/requests contained in the associated RAI is stated and immediately followed with AmerenUE's response. Text from the NRC's RAIs is shown in italics. to this letter provides the updated report from Brookhaven National Laboratory, titled "Information Required for Alternative Test Methods to determine in-leakage of air in the Control Room Envelope," which incorporates the response information for the questions in Attachment 1.

Please contact Tom Elwood, Supervising Engineer, Regulatory Affairs and Licensing at 573-676-6479 for any questions you may have regarding this issue.

This letter does not contain any new commitments.

I declare under penalty of perjury that the foregoing is true and correct.

Sincerely, Luke H. Graessle Manager, Regulatory Affairs KRG/nls Attachments: 1) Responses to NRC RAI Questions Regarding Generic Letter 2003-01, Control Room Habitability

2) Information Required for Alternative Test Methods to determine in-leakage of air in the Control Room Envelope. Revision 2

ULNRC-05448 October 30, 2007 Page 3 cc: Mr. Elmo E. Collins, Jr.

Regional Administrator U.S. Nuclear Regulatory Commission Region IV 611 Ryan Plaza Drive, Suite 400 Arlington, TX 76011-4005 Senior Resident Inspector Callaway Resident Office U.S. Nuclear Regulatory Commission 8201 NRC Road Steedman, MO 65077 Mr. Jack N. Donohew (2 copies)

Licensing Project Manager, Callaway Plant Office of Nuclear Reactor Regulation U. S. Nuclear Regulatory Commission Mail Stop O-8G14 Washington, DC 20555-2738

ULNRC-05448 October 30, 2007 Page 4 Index and send hardcopy to QA File A160.0761 Hardcopy:

Certrec Corporation 4200 South Hulen, Suite 422 Fort Worth, TX 76109 (Certrec receives ALL attachments as long as they are non-safeguards and public disclosed).

Electronic distribution for the following can be made via Other Situations ULNRC Distribution:

C. D. Naslund A. C. Heflin L. H. Graessle G. A. Hughes S. A. Maglio S. L. Gallagher L. M. Belsky (NSRB)

T. B. Elwood B. F. Holderness J. B. Little R. A. Pohlman Ms. Diane M. Hooper w/o Mr. Dennis Buschbaum w/o Supervisor, Licensing TXU Power WCNOC Comanche Peak SES P.O. Box 411 P.O. Box 1002 Burlington, KS 66839 Glen Rose, TX 76043 Mr. Scott Bauer w/o Mr. Stan Ketelsen w/o Regulatory Affairs Manager, Regulatory Services Palo Verde NGS Pacific Gas & Electric P.O. Box 52034, Mail Stop 104/5/536 Mail Station 7636 P.O. Box 56 Phoenix, AZ 85072-2034 Avila Beach, CA 93424 Mr. Scott Head w/o Mr. John O'Neill w/o Supervisor, Licensing Pillsbury Winthrop Shaw Pittman LLP South Texas Project NOC 2300 N. Street N.W.

Mail Code N5014 Washington, DC 20037 P.O. Box 289 Wadsworth, TX 77483 Missouri Public Service Commission w/o Governor Office Building 200 Madison Street PO Box 360 Jefferson City, MO 65102-0360

Attachment I to ULNRC-05448 Responses to NRC RAI Questions Regarding Generic Letter 2003-01, Control Room Habitability In its letter dated August 11, 2003 Union Electric Company (AmerenUE), transmitted its preliminary response to NRC Generic Letter 2003-01, "Control Room Habitability." In the response AmerenUE identified that it would need to develop an alternate method of integrated in-leakage testing because of the Callaway Plant Control Room/Control Building design. In its submittals dated December 17, 2004 and June 6, 2006, AmerenUE reported the results of its alternate integrated in-leakage testing for Callaway Plant which was performed September 17-19, 2004 using the Atmospheric Tracer Depletion (ATD) method. AmerenUE's responses to NRC Generic Letter 2003-01 and the RAIs dated November 7, 2006 are currently under review by the NRC staff.

To continue the review of AmerenUE's responses concerning control room habitability,the NRC staff requests the following additionalinformation.

Question L.a.

Which Dominion NuclearPower Plant was the testing performed at or for?

Response

Unfiltered in-leakage tent testing of safety-related equipment external to the Control Room Envelope (CRE) was performed at the Surry Power Station (Dominion Generation).

Question 1.b.

What was the test procedure or associatedtest report identifier and what were the date(s) of the testing? Was this test or test report included in the listed references?

Response

The primary boundary test procedure was a multi-tracer version of the single tracer constant injection technique of the American Society for Testing and Materials (ASTM) consensus standard E-741 method - referred to as the Brookhaven National Laboratory (BNL) Air Infiltration Measurement System (AIMS) test.

Testing was performed from January 15-18, 2004. Details of the results of the primary envelope boundary AIMS testing and the ancillary tracer component testing are in the report, "Multi-Tracer Testing at Dominion's Surry Power Station for Air In-Leakage Determination, 19 April 2004, TTC-1011." This reference was not previously listed.

Page 1 of 7 to ULNRC-05448 Question 1.c.

What was the time duration between the ASTM E741 method and the ATD method tests?

Response

Tracer component testing of three of the four emergency charcoal filter fan systems (about four hours duration for each test) was performed immediately prior to the CRE boundary testing (about a 36-hour period) and the fourth tracer component test immediately afterward.

The tent testing using two perfluorocarbon tracers (PFTs) [IPTCH and 2PTCH] following an ASTM E-741 approach and the Atmospheric Tracer Depletion (ATD) testing of the same component using two different PFTs [PECH and mPDCH] were performed simultaneously, using the same samplers and sample analyses for determination of the four PFTs. Two emergency fan systems supplied filtered air at the Main Control Room elevation. The zones were designated as Main Control Room Unit 1 and Unit 2. The other two emergency fan systems supplied the two Switch Gear rooms on the elevation below.

"rent Testing" of the negative pressure portion of the four systems was performed by installing a tent around the filter housing just downstream of the charcoal filter housing and up to the fan housing, including the fan shaft seal.

The plastic tenting was pressurized with a known concentration of PFTs. Their presence and magnitude in the fan discharge air confirmed unfiltered in-leakage was occurring and quantified the extent. Each test took 2.5 to 3.5 hours5.787037e-5 days 0.00139 hours 8.267196e-6 weeks 1.9025e-6 months .

Question 1.d.

What was the tent test volume? How well does this model the multi-zone situation described for the plant control room envelope?

Response

The tent test volume was estimated at 10 to 15 ft 3. Note 1: Exact knowledge of the volume was unnecessary as the volume had to sufficiently enclose a filtration system component section as described in the Question 1.c response. Note 2: The component tent testing had no direct dependency on the 265,000 ft3 CRE.

At the Surry plant, the unfiltered in-leakage into the CRE was determined using the multi-zoned AIMS technique, an E-741 technique, with the component unfiltered in-leakage (UI) being used as the basis to compare the ASTM E-741 and the BNL ATD approaches.

The component UI's were determined by the tent testing in two ways:

1) Measuring an added PFT tent concentration and the same PFT in the fan Supply Air -

an ASTM E-741 approach.

Page 2 of 7

Attachment I to ULNRC-05448

2) Measuring a Turbine Building background PFT2 and the same PFT2 in the fan Supply Air.

For the ASTM E-741 test the flow rate of the unfiltered in-leakage Qu, is related to the flow rate through the Supply Air QfSA by the following relationship (QfsA

CfsA/Ctent) where C is the concentration in the filter Supply Air (fSA) or the tent. For the ATD approach, the equation changes to reflect that the background concentration is that of the Turbine Building (TB) (QuI = QfSA

CfsA/CTB) provided that the charcoal exit concentration was zero in PFr2.

The equations for the ASTM E-741 and the ATD component testing approaches are the same when the charcoal bed exit concentration is zero; the only difference being that in the former, the reference PFT is that added to the tent and in the latter, the reference PFT is that measured in the background air of the Turbine Building which housed the filtration systems.

The full equation for the ATD approach is:

QuI = QfsA * (CfsA - CCh)/(CTB - Cch) Equation I such that if the concentration in the charcoal, Cch, is not zero, the unfiltered in-leakage, Qui, would be smaller.

Question i.e. Part 1 How should the comparison of the benchmark test result be understood, consideringcalculated values are being comparedfor correlationpurposes with calculatedupper bound values?

-Response The ATD values for the unfiltered in-leakage represent an upper bound, worst case scenario because the charcoal filters were assumed to be 100% efficient (Eff = 1.0), that is, Cch = 0 in Equation 1 above. Although the charcoal bed exit concentration was not measured, Equation 1 can be solved for the concentration after passing through the charcoal:

CCh = (QfsA*CfsA - QuI *CTB)/ (QfsA - Qui) Equation 2 Measured values for the unfiltered in-leakage in the control room envelope zones were obtained from tent testing results using the ASTM E-741 approach (AIMS procedure).

Equation 2 can be solved for the charcoal efficiency Eff = I - Cch/CTB = (1 - CfsA/CTB)*QfsA/(QfsA - Qui) Equation 3 The ATD unfiltered in-leakage results summarized in Table 1 (page 5) were computed using Equation 1, assuming the charcoal concentration was zero (Eff = 1.0). The measured tent test unfiltered in-leakage rates based upon the ASTM E-741 approach were used in Equation 2 to calculate the Surry charcoal bed efficiencies, which are also presented in Table 1.

Page 3 of 7

Attachment I to ULNRC-05448 Prior to testing at the Surry plant, the efficiency of the charcoal filters in removing PFI's was experimentally determined (Dietz, 2003, Dietz, 2003a Dietz, R, "Determination of Unfiltered In-Leakage by ATD and AIMS E 741 Techniques", National HVAC User Group (NHUG) 7/29/2003. Power Point Presentation). The BNL efficiency determinations for the same supplier's charcoal as used at the Surry plant, tested in the same bed thickness and space velocity, are presented in Table 2 (page 5) for the same two PFTs used in determining the Surry efficiencies.

Comparison of the calculated charcoal efficiency from the tent tests obtained using the AIMS procedure and BNL data collected independently shows excellent agreement, supporting the equivalence of the two methods for measuring unfiltered in-leakage, the exception being the Switch Gear Room 2 data.

In Switch Gear Room 2, the unfiltered in-leakage rates determined using both the tent test using the AIMS procedure and the ATD test procedure, significantly exceeded the 23-cfm squirrel-cage fan rate. This would only impact the tent test results because there was no guarantee that the tent concentration of the added PFT would be uniform; thus, the magnitude of the added tracer in the in-leakage air would be uncertain. For the ATD determination, the in-leakage, tracer concentration was that in the Turbine Building air, regardless of whether it came via the squirrel-cage fan or leakage past the tent. (Leakage past the tent had to occur because the fan only supplied 23 cfm, but the unfiltered in-leakage value was greater.)

Assuming the BNL determined charcoal efficiency of 0.9969 was applicable to the Switch Gear Room 2 case, the tent test result should have been 35 cfm, significantly exceeding the 23 cfm of the squirrel-cage fan.

Question i.e. Part2 If the charcoalexhaust concentrationshad been factored in, could the values shown in Table 5 for the ATD test be less than those shown for 'the Tent Testing? Could the values presented in Table 5 be inferred to show that the ATD test results are conservative relative to the ASTM E741 method in determining a value of unfiltered in-leakage when that may not actually be the case ?

Response

.BNL determined that the efficiencies for removing the PFT mcPDCH and the PFT PECH were 0.9981 +/- 0.0017 and 0.9969 +/- 0.026, respectively. The estimated in-leakage using the ATD tests was calculated using these efficiencies. These results are reported in Table 2. For the first 3 tests the corrected ATD in-leakage rates are within experimental error with the ASTM E-741 procedure. In the last test, the in-leakage rate exceeded the fan supply rate (23 cfm),

causing the difference between the tests.

In summary, the AIMS tent test and the ATD results demonstrate that the ATD results, in the absence of measurements at the charcoal exhaust, are, indeed, a conservative upper limit. In practice, the ATD methodology calls for the measurement at the charcoal exhaust, as was done for the determination of the unfiltered in-leakage at the Standard Nuclear Unit Power Plant Page 4 of 7

Attachment I to ULNRC-05448 System (SNUPPS) plants. This was not done at Surry, due to the Surry test not being an "official" ATD test of record.

Table 1, Surry CREVS Unfiltered In-Leakage and Charcoal Efficiencies CRE Region Tent Testing ATD (assumed ATD Charcoal (AIMS) 100% Filter (measured Efficiencies Efficiency) Filter (Per PFT)

CFM CFM efficiency)

CFM mcPDCH PECH Main Control 15.6 +/- 1.8 <17.8 +/- 2.1 15.4 +/- 1.8 0.9975 0.9979 Room Unit I Main Control 13.3 +/- 1.5 <15.8 +/- 1.9 13.3 +/- 1.6 0.9987 0.9963 Room Unit 2 Switch Gear 8.5 _ 1.0 <11.5 +/- 1.3 9.0 + 1.1 0.9958 0.9981 Room Unit 1 Switch Gear 26.8 + 3.1 <38 35.0 + 4.2 N/A* N/A*

Room Unit 2 1 1 1 1 1

Unfiltered In-Leakage exceeded fan supply rate and therefore estimates of filter efficiencies are unreliable.

Table 2, Brookhaven National Laboratory and Surry Charcoal Efficiency Comparisons by PFT BNL Surry mcPDCH PECH mcPDCH PECH 0.9963 0.9935 0.9975 0.9979 0.9996 0.9999 0.9987 0.9963 0.9962 0.9981 0.9958 0.9981 0.9990 0.9981 N/A* N/A*

Unfiltered In-0.9993 0.9948 N/A* N/A* Leakage exceeded Average 0.9981 0.9969 0.9974 0.9974 fan supply rate and Standard 0.0017 0.0026 0.0014 0.0010 therefore estimates Deviation of filter efficiencies are unreliable.

Page 5 of 7 to ULNRC-05448 Question 1.f.

Would the difference in values of unfiltered in-leakage obtained by the different test methods (ASTM E741 versus ATD) be expected to be more or less or about the same for a plant control room envelope test compared to that of the tent testing described? Does comparing upper bound values for the benchmark ATD tests impairthis extrapolationand if so how?

Response

The response to the question on section L.e shows that the differences in the reported unfiltered in-leakage results between the ASTM E-741 test and the ATD test was due to the assumption of 100% efficiency of the charcoal filter.

Using the ASTM E-741 unfiltered in-leakage, the estimate for the filter efficiency in Equation 2 produced estimates that were identical to the measured efficiencies within experimental error.

Alternatively, using the measured efficiencies and the ATD data provided estimates of unfiltered in-leakage consistent with the E 741 results within experimental error.

Question 2 In the Brookhaven document, section 1. Summary of the test method, Figure I Simplified schematic offlows into the Control room Envelope; a parameterCb is identified as Background Concentration. In section 3. Parameterspecifications;it refers to Figure1 and the unfiltered PFTconcentration(Cu) and shows an equation using Cu. Are "Cb" and "Cu" the same thing?

Response

Yes, the text has been clarified by changing Cb to Cu. in Figure 1 and the adjacent text in the revised Brookhaven National Laboratory report provided as Attachment 2.

Question 3 In the Brookhaven document section 10. Uncertainty(e.g., precision, accuracy) of results obtained with the test method; in the paragraphfollowing a table titled "Table 4..." it reads "Table 3 shows..." Was this paragraphintended to read "Table 4 shows..."?

Response

Yes, this has been changed in the text in the revised Brookhaven National Laboratory report provided as Attachment 2.

Page 6 of 7 to ULNRC-05448 Question 4 In the Brookhaven document section 12. Assessment that determines the acceptabilityof the alternative test; in the third paragraphit reads "Forexample, the NEI report (1) states..." What is the "(1)"

intended to refer to or convey?

Response

That is a typo that should refer to the reference NEI 2001. This has been fixed in the text in the revised Brookhaven National Laboratory report provided as Attachment 2.

Question 5 In the Brookhaven document section 12. Assessment that determines the acceptability of the alternativetest; in the fourth paragraphit reads "Table 5 summarizes the comparison..." Was this intended to read "Table 6 summarizes the comparison..."?

Response

Yes, in section 12 the table number and corresponding reference in the text have been changed to "Table 6" in the revised Brookhaven National Laboratory report provided as Attachment 2.

Page 7 of 7 to ULNRC-05448 Brookhaven National Laboratory EnergyEnvironment, and National Security Directorate Environmental Sciences Department Information Required for Alternative Test Methods to determine in-leakage of air in the Control Room Envelope.

Revision 2 August 30, 2007 Preparedby Date: 8/30/07 Terry Sullivan

/1 Technical Review by. Date: 8/30/07 John Heiser to ULNRC-05448 InformationRequiredfor Alternative Test Methods to determine in-leakageof air in the ControlRoom Envelope.

Regulatory guides describe methods acceptable to the NRC staff for demonstrating compliance with regulations. Regulatory Guide 1.197 (NRC, 2003) provides guidance on the type of information the staff needs to assess the capability of an alternative test method to demonstrate (Control Room Envelope) CRE integrity. Any alternative test method should incorporate characteristics for test attributes detailed in Section 4 of Appendix I of NEI 99-03 (NEI, 2001) with the clarification noted under regulatory Position 1, "Testing." Regulatory Guide 1.197 recommends the CRE integrity be tested using the approach specified in the American Society for Testing and Materials (ASTM) consensus standard E741, "Standard Test Method for Determining Air Change in a Single Zone by Means of a Tracer Gas Dilution," (ASTM, 2000). ASTM E741 was not designed for pressurized systems such as those found in some control rooms and alternative testing techniques are often used in these cases. Regulatory Guide 1.197 requires a response to the following 12 questions in order to permit NRC to judge the acceptability of testing techniques that are alternatives to those specified.

1. Summary of the test method During an emergency, air entering the control room envelope is filtered through a charcoal system using the Control Room Emergency Ventilation System (CREVS). The atmospheric tracer depletion (ATD) method uses perfluorocarbon tracers (PFTs) that are part of the normal background as surrogates for infiltration of unfiltered air into the Control Room Envelope (CRE). The four PFTs of 400 molecular weight are retained by the charcoal filters. Thus, if there were no unfiltered in-leakage (UI), at steady state, the concentration of these PFTs in the CRE would be the same low level as in the outflow of the charcoal filters; a slightly higher level in the CRE would mean a slight amount of UI. Figure 1 provides a simplified representation of the system. Through measuring of these PFTs in air outside the CRE (e.g. background (unfiltered) concentration CJ)immediately after passing through the charcoal filters (filtered concentration Cf), and within the CRE (CcRu), and using the known flow rate through the filter system, an accurate measure of in-leakage (Q,) flow rate can be obtained.

Control Room Envelope

,Cj.= 6harco.al Filtered Air Conrcentration Qfc CORE(Q . .....

= 1flbw rate filter-ed air Qu*Cb Cb= Background Concenitration 1

Q= flow rate': unfiltered air Figure 1 Simplified schematic of flows into the Control room Envelope to ULNRC-05448 To perform appropriate ATD determinations, it is essential that measurements be made under steady state (SS) conditions - that is, the output depleted concentrations of the charcoal systems being used must be constant and the concentrations in the locations being sampled must no longer be changing. In principle, the stable performance of charcoal cells should be ascertained ahead of the actual in-leakage testing to be sure they are appropriately characterized and any portable charcoal systems used to accelerate the approach to steady state (SS) should also be evaluated; the systems should also be checked for potential interferences from other components already on the charcoals. In practice, these measurements are usually made at the time of actual in-leakage testing.

The schematic in Figure 1 is simplified and can be adjusted to address more complicated situations such as the potential for in-leakage into equipment rooms or other areas that are part of the CRE. In many cases, the CRE is not a single well-mixed zone - a requirement for successful implementation of any tracer testing. This was the case for SNUPPS (Standard Nuclear Unit Power Plant System) designed plants - the CRE in these systems consists of 3 distinct zones. The modeling of such situations is specified in the test design (TTC-TD-01, 2003) and developed specifically within the Wolf Creek and Callaway plants' testing procedures. For successful implementation, the mixing within such distinct zones must be demonstrated as well mixed.

It is important to note that the integrated ATD testing is a comprehensive approach to determining UI in a CRE - regardless of the pathway by which such UI arrives at the CRE.

This is especially relevant to CREs which are contained within other zones or buildings of the plant such as at the Wolf Creek and Callaway plants.

2. Description of the test apparatus and tolerances; The test apparatus brought to the Nuclear Power Plant consists of air sampling equipment to measure the PFT concentrations. These are described in detail below. In addition, calibrated ancillary measurement tools (flow meters, voltmeter, temperature recorders, and delta pressure meter) need to be performance-verified and miscellaneous materials (polyurethane tubing - 1/8" OD by 1/16 in ID, shipping containers) as well as appropriate data sheets need to be prepared prior to testing. The flow meters are used to measure flow into the sampling equipment.

In order to improve mixing in certain zones and to reduce the time required to reach equilibrium, the utility may be asked to supply additional equipment - floor fans to enhance mixing and portable charcoal filter systems to accelerate the removal of background level of tracers down to the steady state level needed for accurate test results.

The test apparatus used back at Brookhaven consists of the sampler desorbers and the gas chromatograph (GG) system used to analyze the samples collected at the plant. The capability of the GC analysis system is matched to that of the field sampling equipment, to provide sufficient sample volume (i.e., sampling duration) to meet the measurement to ULNRC-05448 needs with sufficient precision.- An example of this matching of capabilities is provided next, for the case of SNUPPS designed plants.

Depending on the ventilation systems' charcoal-filtered rates and volumes of the zones, the predicted ATD performance suggests sampled air will have fractional depletions running from 1.0 (outside air into the Control Building (CB) - thus, no depletion), to

-0.25 (CB steady state levels with charcoal filtration operating), to 0.001 to 0.01 (at steady state levels within the control room), and to <0.001 (Emergency Ventilation System-EVS charcoal filtered discharge air). Thus, collecting adequate sample volumes to quantify each of these levels and to automate that collection as much as possible are important goals.

Four types of sampling systems can be used - two automatic collection systems and two types that require manual change of sampling tubes. The Sequential Air Sampler (SAS) is a 20-tube automated sampler that collects air at high sampling rates (up to 550 mL/min) and is used where fractional depleted concentrations will be low - in the CR. The Air Pro (AP) is a 2-channel pump that is also capable of high flow-rate (up to 600 mL/min) sampling on two sampling tubes simultaneously and from ductwork where there is several inches of vacuum or pressure; the sampling tubes must be manually changed. The Brookhaven Atmospheric Tracer Sampler (BATS), an automated system with 23 tubes, is a low flow-rate (50 mL/min) sampler for where concentrations are higher and samples are needed from many different elevations. Lastly, many Personal Air Samplers (PASs) with moderate flow rates (150 mL/min) can be used at different locations on the CB elevations to verify good mixing.

The tolerances of the GC system and the flow rate determinations of the samplers are considered in the development of the testing plan and procedures. The GC system has a limit of detection (LOD) for the multiple PFTs of about 40 counts (GC peak area counts);

thus, the minimum counts desired in a sample is about 4000 counts. Normal background air has an average surrogate ambient PFT (mcPDCH) concentration of -7700 counts/L.

With the SNUPPS design used at the Wolf Creek and Callaway plants, the following arrangement of samplers was used, Figure 2, to allow appropriate sample volume collection for the control room envelope (CRE) which includes the sample locations near RA (return air grill), CREVS (Control Room Emergency Ventilation System), Equipment Room A, ERa and Equipment Room B, ERb. The type of sampler and sample duration are specified in Table Ia. Samples were also collected from the Control Building, Figure 3, which includes sample locations either near the return air or supply air grills or the CBEVS, Control Building Emergency Ventilation System) inlets and outlets, Table lb to ULNRC-05448 aS SAS A AP N Figure 2 Control room Envelope Sample Locations used at the Wolf Creek plant.

to ULNRC-05448 Table la. Control Room Envelope Sampling Systems and Sample Quantity, Duration, Rates, and Locations Enve- Sampling Chan Tubes lope Zone Locations Sampler(5) #/Qty /Site Comments CRE CR RA grill #1 SAS"' #1 1 20 1-h samples (16) from -1600

2 SAS #2 1 " Friday to 2-h samples by 0900 Saturday (4) - 24 h total Near CR operator AP(2) #1 1,2 3,3 h samples on the Saturday only ERa Near CREVS SAS #3 1 20 1-h samples (16) from --1600

" @ 300 cfm RA SAS #4 1 if Friday to 2-h samples by 0900 Saturday (4) - 24 h total CREVS CREVS inlet AP #2 1 5 h samples at -300 mL/min; 2 on Friday (1500-1700) & 3 on Saturday CREVS outlet " 2 3 h samples at -600 mL/min; 1 on Friday (1500-1800) & 2 on Saturday; 2 in series ERi Near CREVS AP #3 1 4 h samples at 600 mL/min on

" Near RA grill " 2 4 Saturday only

-Each SAS (Sequential Air Sampler) automatically changes tubes (up to 20 times), sampling at -550 mL/min (-33L/h)

(2)All Air Pro (AP) channels at 550 to 600 mL/min unless otherwise indicated; tubes are manually changed (3)Brookhaven Atmospheric Tracer Sampler (BATS) automatically changes tubes (23), sampling at -50 mUL/min (-3L/h)

(4)All Personal Air Samplers (PASs) at 150 mL/min; tubes are manually changed; sampling only on 2nd day (5)A total of 4 SASs, 4 APs, 6 BATS, and 14 PASs will be used

- the BATS 23 sampling tubes are permanently-installed stainless steel tubes containing adsorbent

- the other 3 samplers use glass sampling tubes with adsorbent; these are called CATS (capillary adsorption tube samplers) to ULNRC-05448 Table lb. Control Building Sampling Systems and Sample Quantity, Duration, Rates, & Locations Enve- Sampling Chan Tubes lope Zone Locations Sampler #/Qty /Site Comments CB El 2000 Active RA grill BATS'3-#1 1 11 2-h samples (1700 Friday to 1500 Saturday) 4 areas PASs(4) 4 4 h samples; 2 each SWGR1/2 on Saturday El 2016 Active RA grill BATS #2 1 11 2-h samples (1700 Friday to 1500 Saturday) 4 areas PASs 4 4 h samples, 3 in Corrl/2, 1 in ERa-Saturday CBEVS inlet BATS #3 1 10 5 1-h samples @ 1700 Fri; 5 on Saturday CBEVS outlet AP #4 1 3 h sam; 1st tube; 1 1500-1800 Friday; 2 Sat.

2 3 h sam; 2ndtube; 1 1500-1800 Friday; 2 Sat.

El 2032 Active RA grill BATS #4 1 11 2-h samples (1700 Friday to 1500 Saturday)

" 3 areas PASs 3 4 h samples from 3 locations on Saturday E120731/2V2 Active RA grill BATS #5 1 11 2-h samples (1700 Friday to 1500 Saturday)

" 3 areas PASs 3 4 h samples from 3 locations on Saturday Any of 4 Active SA grill BATS #6 1 11 2-h samples (1700 Friday to 1500 Saturday)

Total primary sampling tubes: -230 to ULNRC-05448 ROM (SEPARATIONJ GROUPS 2.4,f6) ii EL. 2073' - G' EL. 2476 L.ONER CA&LE SPI'WAN~G RM.,UNIT M'rR.(EGKO35)

.COMT. F64, PFISM~RIZA7)ON SYS FILXER UN IT CFGI(OZAý

-ACCESSCC4 L F.kWCI L UbWT (GO-3)

EL.2OI6'- 0

-MISSILE SH-IELDING FOR INTAKE LOUVRES I I gSfTf!5 EL. 1981-0-"

Figure 3 Wolf Creek Control Building Envelope with floor elevations.

to ULNRC-05448

3. Parameter specifications; The major parameters that need to be measured to implement ATD are the flow rates through the charcoal-filtered EVS(s), the PFT concentration in the normal outside air (pre-charcoal filter), the PFT concentration in the air after passing through the charcoal filters in the EVS(s) (PFT depleted) and inside the CRE [and the CB in the case of SNUPPS plants] (PFT depleted plus PFTs from UI). If there were no UI, the air concentration in the CRE would equal the concentration after passing through the charcoal filters. The charcoal filters on the CREVS are typically effective at removing over 99% of the PFTs from the air; their performance is determined as part of the testing.

Thus, any increase in the concentration in the CRE above that in the exhaust of the charcoal filters indicates UI. The magnitude of the UI is calculated using the flow rates of the filtered supply air; those rates and their uncertainties are either measured by the plant using their standard procedures or by a tracer determination using a separate PFT.

Using the simplified schematic in Figure 1, the ATD approach needs to use the measured filtered flow and PFT concentration (Qf and Cf, respectively), the unfiltered PFT concentration (Cu), and the PFT concentration in the control room envelope (CcRE).

Performing a mass balance for steady-state conditions, the unfiltered flow rate can be determined from:

Q. = Qf*Fdep/(1 - Fdep) (1)'

where Fdep is (CCRE - Cf)/Cu.

Note that for the SNUPPS-designed plants, the actual equations for Uls into the control room and equipment rooms of the CRE and into the CB are somewhat more complex due to the possibility of leakage from the CRE into the CB, and the presence of equipment rooms in the CRE that receive some of the supply air from the CREVS.

SNUPPS Control Building Balance Equations.

At steady state (SS), the rate of UI into the CB is given by:

QTI-c0 = Qf-CB - (CcB - Cf)/(Camb - CCB) + 8QCR (CcB - CcR)/(Camb - CCB) (2) where the first term on the right accounts for the tracer depletion by the CB filtered supply-air (SA) rate and the second term accounts for the fraction, F, of the exfiltrating CR pressurization air that enters the CB. That fraction, 6, could range from 0 to 1; thus, the calculated rate of UI will be a range rather than a discrete value. The "C" parameters are the measured concentrations in the Control Building (CB), filtered supply air (f), background (amb), and the Control Room (CR). The Q parameters are flow rates from the areas defined by the subscripts, e.g. QfCB is the filtered air from the control building, and QCR is the control room air flow When the rate of exfiltration from the CR is not considered (F,= 0; the 2nd term disappears) and the depleted concentration from the CB charcoal filter (Cf) was close to zero, then Eq. 2 reduces to:

QUI-CB = Qf-CB " CCB/(Camb - CCB) = Qf-CB FCB/(l - FCB) (3) which is Eq 1, where Fcb is the depletion ratio CcB /Camb.

to ULNRC-05448 SNUPPS Control Buildin2 Balance Equations At steady state, the rate of UI into the CR is given by:

QUI-CR = Qf-CR (CCR - Cf)/(Camb - CCR) - I/9 QEqRm CCR/(Camb - CCR) (4) where the first term accounts for the filtered air entering the CRE and the second term accounts for exchange between the CRE and Equipment Room. The constant, 1/9, is the nominal filtered SA rate (2000 cfm) divided by the AC fan rate (18000 cfm). The filtered SA concentration (Cf) is close to zero and given that the rate of the filtered SA is about 2,200 cfm and the flow rate into the active equipment room (QEqRm) is about 350 cfm (such that 1/9 of the 350 cfm is about 2% of the Qf-cR), then Eq. 4 reduces to:

QUt-CR = 0.98

Qf-cR - CcR/(Camb - CCR) = 0.98

Qf-cR ° FCR/( - FCR) (5)

For both the CB and the CR, the explicit solution forms (Eq. 2 and 4, respectively) were used to calculate the final unfiltered in leakage (UI) results.

SNUPPS Equipment Room A Balance Equations A material balance around this zone, which includes the CR Filtration System, was performed with the assumption that a portion (EI) of the total out-leakage (-350 cfm) from the CR enters Equipment Room A in addition to the 350 cfm directly from the CR AC System plus any UI directly into that zone. The assumption is that the higher pressure in the CR will allow some fraction of its total out-leakage to enter the Equipment Room.

The resulting steady state solution for UI into the ERa is givenby:

QUI-ERa = SI- °QuIC R+ 350)1]dpR JQER +a" .. Q~+/-I.jU-R+301Fp 6 1 - FdcpERa where QERa is the 350 cfm flow rate from the CR AC system directly into equipment room A (equivalent to the 300 cfm return from this zone back to the filtration system plus the 50 cfm of pressurization air in this zone), the 350 cfm is the CR pressurization rate, the Fdep are for the respective depleted concentration ratios, and El is defined above (8C might range from 0.1 to certainly no more than 0.6 of the total CR out-leakage entering the ERa). The factor 8/9 Of QERa is because the 18,000 cfm of the CR AC system only contains 16,000 cfm of CR recycle air.

SNUPPS Equipment Room B Material Balance A material balance around this zone was done making the assumption that a fraction (82) of the CR out-leakage and a fraction (F3) of that from Equipment room A enter Equipment Room B, along with its UI. The steady state solution is:

Qui- .350 + OUI.ERa)+/- EIQUI.CR+ 350)1 + E*(QuI.CR +3 5)Fde E*.R=I_.* _F.Ria._(50 + OuI-ErJlEdcCR (7)

ERb= I - FdepERb where the terms have been previously defined. The solution depends on the previously determined UI rates into the CR (precisely determined) and the ERa (reasonably bounded) and on estimates for 6 2 and F3; note that El would be chosen earlier to bound the ERa UI rate.

to ULNRC-05448

4. Material requirements; The ATD method requires sampling equipment to collect the PFTs from the air and portable fans and portable charcoal filter systems as discussed in response to item 2.

There are no tracer gases released and no analyses performed at the plant and thus, no GC instruments or operating gases at the plant.

5. Safety implications of the test (e.g., personnel safety, impact on plant operations and on plant equipment);

Safety and operational concerns with this testing are minimal. Sampling equipment is battery operated and low voltage. The tracer is already present in the atmosphere and additional injection of new material is not required; thus, the safety of the PFTs is not an issue. However, the tracers are all perfluorocarbons and their Material Safety Data Sheets (MSDSs) show them to be chemically inert and biologically inactive. PFTs have been used extensively in atmospheric dispersion testing and in ventilation testing in homes and buildings.

The ATD method has minimal impact on plant operations. In the usual nuclear plant, to use this method, the Control Room Emergency Ventilations System (CREVS) must be operated long enough to reach a steady state depleted concentration in the CRE; This time is a function of the filtered flow rate through the CREVS and the volume of the CRE. In addition to the sampling equipment, the only other change to plant operations is the use of additional fans to improve mixing (only in those few zones where good mixing is inherently not present) and the short-term use (a few hours) of portable charcoal filtration systems to more quickly reach the steady state conditions necessary to quantify unfiltered in-leakage. After the portable charcoal systems are turned off, the CRE depleted concentrations are sampled over a remaining time sufficient for steady state to be achieved with just the CREVS charcoal filters running. Plant operations are not changed.

The ATD method does not influence the performance or safety of plant equipment. All systems are operated in accordance with standard operating procedures. The only intrusion into plant equipment involves the use of probes placed in some ductwork to collect air samples for determination of depleted PFT levels. Full details of needs are given in the specific testing plans and procedures.

6. Preparations before initiation of the test; Building plans are reviewed to determine air flow pathways and the magnitudes of re-circulating (mixing) and filtered flow rates. Prior to testing, data collection is performed of all room volumes and nominal flow rates within the CRE (and CB for SNUPPS plants). This information is needed to determine the time to reach steady state conditions, assure that mixing is sufficient, and to perform a preliminary sampling design; other than an accurate knowledge of the filtered supply-air rates, zonal volumes and other flow rates are not relevant to quantification of UI. If the time to reach steady state is unacceptably to ULNRC-05448 long or mixing is not sufficient, provisions are made to bring in charcoal filter systems and fans. Complete details on the generic approach are specified in Test Design Procedure, (TTC-TD-01, 2007).

Using the provided flow rates and room volumes, a detailed sampling plan is developed.

The plan defines sample locations, sample volumes, total number of samples, and test duration. Based on the sample plan, sampling equipment is emplaced or installed at ductwork as required. Plant-specific details are provided in a testing plan and procedures document, provided to each plant prior to testing.

7. Calibration of the test equipment; The Tracer Technology Center Quality Assurance Plan addresses calibration of test equipment. The overarching standard for calibration is the Brookhaven National Laboratory Calibration Subject Area which addresses the identification of equipment to be calibrated and the related calibration requirements. The frequency and rigor of calibration of the equipment items is tailored according to the potential impact on the environment, safety, health, and quality of the analysis. Measuring or test equipment used to monitor processes or generate data, important to the project or activity, is calibrated and maintained. Automatic sampling equipment is tested to confirm it is operating within specifications immediately prior to shipping to the plant for the tests.

The gas chromatographs undergo a standard checkout procedure prior to sample analysis.

Pressure flow meters used in this test are-primary standards calibrated at the factory and do not require further calibration for one year. All pressure flow meters were used within one year of purchase. Voltmeters and micrometers are calibrated annually.

The implementing procedures within the Tracer Technology Center (TTC) are the following:

TTC-TP-03, Calibration of Active CATS Pumps (Flow rate, Battery Voltage and inlet pressure)

TTC-SA-01, QA for Calibration Standard TTC-SA-02, Carrier Gas QA Procedure TTC-SA-03, Chromatograph Maintenance and Performance Checks TTC-SA-04A, Chromatograph GC-1 Pre-Sample Analysis Checkout Procedure TTC-SA-04B, Chromatograph GC-2 Pre-Sample Analysis Checkout Procedure TTC-SA-07, CATS/BATS Calibration Standard Loading Procedure

8. Description of the test procedure; The objective of the ATD testing is to demonstrate that unfiltered in-leakage into the Control Room Envelope (CRE) are below the design basis used in the dose assessment calculations. The ATD tests are to be performed in the least intrusive fashion possible while meeting this objective. For SNUPPS-designed plants, this means allowing for normal Control Building (CB) and CRE ingress/egress. The test is done with a secondary objective to obtain a worst-case CB and CRE configuration and adjacent-zone HVAC operation to provide an upper bound on unfiltered in-leakage. This latter objective, to ULNRC-05448 however, is controlled by the plant operators - such that test conditions are consistent with the limiting conditions in their licensing basis. The multi-tracer ATD method meets these objectives.

The first step in the test procedure is to review the dimensional plans of the facility and the CB/CR ventilation diagrams to determine important inputs needed to meet the measurement objectives. For SNUPPS-designed plants, with the Control Building and Control Room Emergency Ventilation Systems (CB/CREVS) operating, the CB unfiltered in-leakage (UI) should be less than 300 cfm and the Control Room (CR) UI should be less than 10 cfm. The 10 cfm for the CR was an assumed UI due to ingress and egress - the CR boundary UI is assumed to be zero.

The presence of the CR inside a mostly-filtered CB environment is what adds complexity to the determination of CR- and CB-UI and confounds the typical single tracer gas techniques.

Using the atmospheric tracer depletion (ATD) of the normal background concentration of four (4) perfluorocarbon tracers (PFTs) of 400 molecular weight by the charcoal filters, in-leakage can be measured. The PFTs act as a direct surrogate for contamination in the outside air during an incident - once the original levels in the CR and CB have been displaced, that is, reached steady state with the EVS running. The CB-determined flow rate of UI represents the flow rate of unfiltered outside air leaking directly into the CB. The CR determination represents the equivalent rate; a value of 10 cfm could be 100% from the outside air or could be the equivalent of 40 cfm of CB air that is leaking in but contains only Y4 of the normal outside air concentration of contamination. The ATD testing will not provide the pathway - only the equivalent magnitude.

To perform appropriate ATD determinations, it is essential that measurements be made under steady state conditions - that is, the output depleted concentrations of the charcoal systems being used must be constant and the concentrations in the locations being sampled must no longer be changing. The stable performance of the CR and CB EVS charcoal cells should be ascertained ahead of the actual in-leakage testing to be sure they are appropriately characterized. Any portable charcoal systems to accelerate the approach to steady state should also be evaluated. Their times to steady state are strictly a function of the condition of the charcoal. The systems should also be checked for potential interferences from other components already on the charcoals.

The attainment of steady state concentrations in the CB and CR are dependent on the volumes and flow rates through these regions. Detailed calculations were performed for the SNUPPS plants to determine the time to reach steady state. The calculations indicated that it would take several days in some rooms of the plant and suggestions were made to add fans and charcoal filtration systems to improve mixing and accelerate the depletion process.

Based on the calculated time to reach steady state, a sampling sequence is defined. The sequence is selected to collect a few air samples as the PFT levels approach steady state and several samples after steady state is predicted to occur. The PFT levels are measured using a gas chromatograph at Brookhaven National Laboratory. The values are reviewed to demonstrate that steady state has been reached and the steady state values are used to calculate in-leakage.

to ULNRC-05448

9. Manner of calculating in-leakage and associated error from test results; Using the measured flow rates and steady state PFT concentrations, a mass balance is performed to determine the unfiltered in-leakage. The basic equation for the CB, CR, and Equipment Rooms A and B are provided in Section 3.

The Design Basis Accident assumes unfiltered in-leakage into the CB at a rate of 300 cfm, Measuring the steady state value of PFT depletion, Fdep, the value of unfiltered in-leakage, QuI, can be determined. The Control Room Envelope is within the Control Building. Therefore, the depletion will be even greater within the CRE.

The results of the tests are provide in Table 2 (Wolf Creek) and Table 3 (Callaway).

These plants have two independent air treatment systems (Train A and Train B) and both were tested. Using the equations in section 3, the following UI rates and uncertainties were computed concentration results:

Table 2. Unfiltered In-Leakage (UI) Rates, cfm at the Wolf Creek Plant Train A Test Train B Test CR 6.9 + 0.4 10.5 + 2.6 CB <63 14.2 + 3.0 Equipment Room 5.6 +/- 2.2 2.1 +/- 0.7 B (ER1501)

Equipment Room 23.0 +/- 4.9 32.3 +/- 13.5 A (ER1512)

The CR UI rates are just about at the values assumed in the DBA originally submitted, although that during the Train-A testing is statistically lower than 10 cfm. The tightness of the CR is essentially independent of which train was operating.

Details of the calculation are found in (Dietz, 2004a, 2004b, 2004c).

Table 3 Unfiltered In-Leakage (UI) Rates, cfm at the Callaway Plant Train A Test Train B Test CR 10.1 +/- 0.9 4.7 +/- 1.1 CB 69 to 97 109 to 166 Equipment Room 0.9 +1.0 3.4 + 0.3 B (ER1501)

Equipment Room 21.2 + 2.4 1.5 +/- 0.7 A (ER1512)

During the Train-A testing, the CR UI rate was just about at the value assumed in the design basis accident originally submitted (10 cfm); for Train-B testing, the rate was half that value. The two rates are statistically different; one would expect the tightness of the CR to be independent of which train was operating - unless there were differences in the pressurization air rates (nominally 400 cfm) from the CB. Those rates had not been measured.

to ULNRC-05448 The CB was reasonably tight - at about 1/3 to 11/2 of the DBA value of 300 cfm. The UI was less during the Train-A testing when the outside air (OA) rate was 833 cfm versus during the Train-B testing - consistent with the lower OA rate of 674 cfm. Thus, pressurization air rates have a strong inverse impact on UI rates.

The UI rates in Table 3 above into the Equipment Rooms are different from those into the CR and CB. The latter are the total UI rates into those locations whereas for the Equipment Rooms, they are additionalUI rates - calculated using assumptions for flow communications between the CR and the Equipment Rooms. . For 3 of the 4 cases, other than CR air (with its proportionate amount of UI) at 350 cfm that is deliberately discharged into the active Equipment Room, there is little additional UI. However, Equipment Room A in the "A" train testing did have 21 cfm of additional UI.

As originally assumed, the CRE is not a single zone - there are statistically different UI rates into the CR, and the two equipment rooms. Multizone systems are known to be a problem for the standard ASTM E 741 methods.

10. Uncertainty (e.g., precision, accuracy) of results obtained with the test method; A procedure for calculating the accuracy of the results using the test method is described in TTC-DP-02, Data Processing for ATD Testing. A major advantage of the ATD testing procedure is that the error is primarily a function of the error in the measured concentration of PFT. This error can be minimized through designing the test to collect larger volumes of air for sample analysis when concentrations are expected to be low. In contrast, for pressurized CREs, traditional tracer tests are limited by the absolute error in the measured filtered supply air rate into the CRE. For typical systems this can be 5 to 10% of the total air flow. Thus, for a 1000 CFM system, the accuracy of traditional tracer tests is limited to + 50 to 100 CFM. For systems with expected low unfiltered in-leakage rates, this is not accurate enough to quantify that rate.

The general equation for calculating uncertainty is:

AQui, fraction = [(AFdep (1+Fdep/(1-Fdep)) 2 + (AQf) 2]1/ 2 (3)

Where, AQui is the fractional uncertainty of the unfiltered in-leakage rate, AFdep is the fractional uncertainty in the fractional depletion Fdep, the ratio of the.PFT concentration inside the CRE versus that outside the CRE, and AQf is the fractional uncertainty in measured flow rate into the CRE through the CREVS. For low values of Fdep (< 5%) the second term, Fdep/(l-Fdep) , is small compared to 1 and Eqn 3 is approximated as AQui, fraction = [(AFdep) 2+ (AQf) 2]1 /2 (4)

The uncertainty estimate depends on the number of samples collected at steady state. For a single sample the error in AFdep has been determined to be bounded by the relationship (TTC-DP-02, 2003):

(AFdep) 2

= 0.01 + (50/Adep) 2 (5)

Attachment 2 to ULNRC-05448 where, .01 is the squared value for the analytical uncertainty in the measured PFT due to uncertainties in sample volume and area under the peak, 50/Adep is the counting statistics error where Adep is the number of counts for the PFT surrogates. Using Eqns 4 and 5, the fractional uncertainty is estimated from the following equation.

AQui, fraction = [0.01 + (50/Adep) 2 + (AQf)2]1/ 2 (6)

Table 4 provides examples of total and percentage error under different conditions and shows the total error of this method is between for a 12 and 15% of the estimated value for a wide range of conditions including unfiltered infiltration rates as low as 30 CFM. In contrast, the error estimate of the standard tracer techniques is directly proportional to the error in measured flow rate. For example, in case 2 with a 7% uncertainty in flow into the system, AQf, using the standard tracer techniques would provide an error estimate of

+/-140 CFM. The uncertainty in this estimate is almost a factor of 5 greater than the estimate of the unfiltered infiltration rate obtained using the ATD method with the same conditions.

Table 4 Calculated error under different in-leakage conditions 1 2 3 Qf, cfm 1200 2000 4700 AQf, fraction 0.06 0.07 0.07 Fdep(PFT)ss 0.140 0.0150 0.0101 Adep, cts 7000 750 530 Qui, cfm 195 30 48 AQui, cfm (%) 23 (12%) 4.2 (14%) 7.4 (15%)

Table 4 shows that the fractional error in the estimate of unfiltered in leakage is 12 - 15%

for a wide range of conditions. This level of accuracy is adequate for demonstrating compliance with design basis in-leakage rates of a few cfm.

In the case where 3 or more measurements are taken at steady state, there are enough separate determinations for Fdep (PFT at steady state) to use their average in Eqn. 3 to get the average Qui and the standard deviation of that average, SDF, to get the uncertainty in Qui as:

AQui, fraction = [(SDF) 2 + (AQf) 2 ] 1/2 (7)

Where SDF is the standard deviation in Fdep, which is the ratio of CRE and unfiltered PFT concentrations.

11. Correlation of the results of the alternative test method with a test performed in accordance with regulatory Position 1.1 to ULNRC-05448 The ATD method was compared to a traditional injection method using a tent testing procedure at Dominion's Surry Nuclear Power Plant (Dietz, 2004e). The tent testing was conducted January 15 - 18, 2004. The Surry CRE (Control Room Envelope) consisted of 5 zones on 2 elevations - four of which had separate ventilation systems and the 5 h, a stairwell between the 2 elevations with no ventilation. The primary boundary test procedure was a multi-tracer version of the single tracer constant injection technique of the ASTME741 method - referred to as the Brookhaven AIMS test. With the exception of the four emergency fan systems, all safety-related ventilation equipment was inside the CRE; thus, separate component testing of these 4 emergency systems was required.

A tracer component injection test of three of the four emergency charcoal filter fan systems was done just before the CRE boundary testing and the 4 th, afterwards -

following the standard ASTM E 741 approach. "Tent Testing" of the negative pressure portion of the 4 systems was performed by installing a tent around the filter housing just downstream of the charcoal filter housing and up to the fan housing, including the fan shaft seal. The plastic tenting was pressurized with a 25 cfm fan and two tracer permeation sources (IPTCH and 2 PTCH) were placed at the fan in-take. This procedure is consistent with ASTM E741. Each test took 2.5 to 3.5 hours5.787037e-5 days 0.00139 hours 8.267196e-6 weeks 1.9025e-6 months , not counting tent and sampling setup.

Simultaneously, an Atmospheric Tracer Depletion (ATD) determination was performed for the same component using 2 other PFTs (PECH and mPDCH). using the same samplers and samples. The 4 PFTs are simulataneously quantified during sample analysis. Two emergency fan systems supply filtered air at the Main Control Room elevation. The zones were designated as Main Control Room Unit I and Unit 2, and the other two -supplied the two Switch Gear rooms on the elevation below.

Sampling was comprised of three consectutive min samples collected at each of 2 discharge-air locations while periodically confirming the tent-air concentration by extracting mL aliquots onto Capillary Absorption Tube Samplers (CATS) from each of 4 locations in the tent. The actual fan flow rate was measured using a rotary-vane anemometer and found to be 23 cfm. The ratio of the PFT concentration in the discharge air to that in the tent air times the 1000 cfm charcoal-bed fan rate (provided by Dominion) gave the unfiltered in-leakage rate in that portion of each system.

The 2 ATD PFTs were measured in the Turbine Building air (entering the EVS inlet) and at the fan discharge; the ratio of the discharge to the inlet concentration times the nominal fan rating (1000 cfm), gave the reported ATD results. The ATD result is presented as an upper bound because the concentrations of the two tracers used for ATD were not measured in the charcoal exhaust. Regardless, the agreement, shown in table 5, demonstrates that the ATD test results are comparable to the ASTM E741 method ("tent testing") results with the exception of Switch Gear Room 2.

to ULNRC-05448 Table 5 CREVS Unfiltered In-Leakage, cfm CRE Region Tent Testing ATD (assuming ATD 100% Filter (measured Efficiency) Filter efficiency)

Main Control Room 15.6 +/- 1.8 <17.8 + 2.1 15.4 + 1.8 Unit 1 Main Control Room 13.3 +/- 1.5 <15.8 + 1.9 13.3 + 1.6 Unit 2 Switch Gear Room 8.5 1.0 <11.5 + 1.3 9.0 +1.1 Unit 1 Switch Gear Room 26.8 + 3.1 <38 35.0 + 4.2 Unit 2 To estimate the actual in-leakage from the ATD tests requires knowledge of the charcoal filter efficiency. Recalling, Eqn (1)

Qu = Qf*Fdep/(1 - Fdep) (1) where Fdep is (CCRE - Cf)/Cu.

where the Q's are flow rates and the C's are concentrations. The subscripts, u, f, CRE, refer to the unfiltered, filtered, and tent test envelope. In the Surry tests, Cf was not directly measured. This is why the ATD tests provide an upper bound. The unfiltered and filtered concentrations can be related by the efficiency of the charcoal filter (Eff).

Cf = (1-Eff)*Cu (8)

In the tests at the Surry Plant, Cf was not measured. In the calculations in Table 5 labeled 100% filter efficiency, it was assumed that the charcoal efficiency was 100% and therefore, Cf would be zero. Therefore, values in this column in Table 5 are an upper bound for the unfiltered in-leakage.

Prior to testing at the Surry plant, the efficiency of the charcoal filters in removing PFTs was experimentally determined (Dietz, 2003a). The efficiency for removing mPPDCH was 0.9981 + 0.0081 and for PECH was 0.9969 +/- 0.026. Substituting Eqn 8 into Eqn 1 and using these efficiencies, the estimated in-leakage using the ATD tests. These results are reported in the last column of Table 5. For the first 3 tests the corrected ATD in-leakage rates are within experimental error with the ASTM E-741 procedure.

In the Switch Gear Room 2 tests, the unfiltered in-leakage rates determined by both procedures, tent test and ATD, exceeded the 23-cfm squirrel-cage fan rate. This would only impact the accuracy of the tent-test results because there was no guarantee that the tent concentration of the added PFT would be uniform; thus, the magnitude of the added

Attachment 2 to ULNRC-05448 tracer in the in-leakage air would be uncertain. For the ATD determination, the in-leakage tracer concentration was that in the Turbine Building air - regardless of whether it came via the squirrel-cage fan or leakage past the tent (leakage past the tent had to occur because the fan only supplied 23 cfm but unfiltered in-leakage was higher).

In summary, this treatment of the tent-test and ATD results shows that the ATD results, in the absence of measurements at the charcoal exhaust, are, indeed, a conservative upper limit. In practice, the ATD methodology calls for the measurement at the charcoal exhaust, as was done at the SNUPPS plants. This was not done at Surry, because we were not conducting an "official" ATD test.

12. Assessment that determines the acceptability of the alternative test in lieu of a test performed in accordance with regulatory Position 1.1.

The ASTM E 741 standard was not devised with consideration of multizones within buildings; in particular, that standard declares that "single zones [CRE] within multizone buildings are difficult to isolate such that they exchange air only with the outside and not to other zones..."(section 3.1.7.1) (ASTM, 2000). Further, most CREs are, themselves, not single zones as was shown to be the case at the SNUPPS designed plants. The Atmospheric Tracer Depletion (ATD) was developed as a new industry standard consistent with ASTM E 741 specifically for pressurized CREs with low design-basis unfiltered in-leakages (Uls). A unique feature is the comprehensive nature of the test; it directly determines total UI, regardless of the in-leakage pathway and the zones contiguous to the CRE.

ATD avoids the multizone issues Mentioned in ASTM E 741 The normal outside air has a known concentration of ambient PFTs and, at steady state with the control room emergency ventilation system (CREVS) running, the CRE has a measurable depleted concentration. That ratio is a direct determination of unfiltered in-leakage - regardless of the pathway and the connection to surrounding zones. Although its strength is in the stated application, it can be used in neutral-balanced envelopes as well.

The ATD tests rely on using charcoal filters present in the Control Room Emergency Ventilation System to remove background concentrations of PFTs in the air. The PFT levels can be measured accurately down to 1% or less of background using readily achievable sample volumes. Therefore, the method is capable of accurately measuring in-leakage to less than a few cfm. For pressurized CREs, other approaches that tag the filtered supply air and measure the difference between the concentration inside the CRE versus that in the supply air, are not accurate for small in-leakage rates such as those that exist at Wolf Creek or Callaway. For example, the NEI report (NEI, 2001) states that traditional "tracer gas testing uses flow measurements for positive pressure control rooms, which increases the overall uncertainty of the test result. If the actual unfiltered in-leakage is small (< 100 cfrn) and the pressurizing air flow is relatively large (>1000 cfm), the uncertainty in the air flow measurement causes the accuracy of the tracer gas test to become very poor." For example, if the pressurizing flow is 2000 cfm and the uncertainty in this measure is 5%, this leads to an error band of at least +/- 100 cfm.

to ULNRC-05448 When this error is compared to the measured in-leakage, the overall test uncertainty can exceed 100 percent of the measured value. In contrast, the ATD method was shown to determine unfiltered in-leakage of less than 10 cfm (+/-25%) in systems with total flow rates of 2000 cfm.

A detailed comparison of the ATD method and the ASTM Standard E-741 was performed as part of a larger review of infiltration detection techniques (Dietz, 2003b).'

The review included analysis of each of the 18 major elements or sections of the standard with respect to the four tracer techniques used to measure unfiltered in-leakage. The techniques are SF6 (sulfur hexafluoride) decay, SF6 injection, AIMS (Air Infiltration Measurement System) based on injection of PFTS and the focus of this report, the ATD method. Of these 18 major elements, 14 were specifically applicable. Within the 14 applicable elements there were 108 sub-elements. Table 6 summarizes the comparison between four tracer techniques used to examine unfiltered in-leakage and the ASTM Standard.

Table 6 Comparison of T'Iacer Test Methods with ASTM Standard E741 sub-elements.

SF 6 Decay SF 6 Inject AIMS ATD Test Meets Sub-element 80 83 84 69 (percentage of subtotal) (89%) (89%) (91%) (92%)

Test does not meet Sub-element 9 9 8 5 Uncertain 1 1 - 1 Subtotal 90 93 92 75 Not applicable 18 15 16 33 TOTAL 108 108 108 108 When this review was conducted in the end of May 2003, the reviewer was not aware of the exceptions to the standard as noted in the Control Room Habitability Guidance Report (NEI, 2003 Appendix EE, p. EE-1). It will be seen that many of the exceptions noted in the NEI document were also noted in the Dietz, 2003b review as "not applicable" (na).

The comparison of the ATD technique and the elements in the ASTM Standard E-741 showed that there was agreement on 92% of the applicable elements (Dietz, 2003b). In comparison, the standard approaches for measuring in-leakage met only 89% of the ASTM Standard elements. Thus, the ATD technique is consistent with the ASTM Standard E-741.

In addition to improved accuracy compared to other in-leakage tests, the ATD offers less disruption of plant operations, minimal NPP staff support, and minimal impact on plant operations.

to ULNRC-05448 References ASTM, 2000 American Society for Testing and Materials (ASTM) Standard E741, "Standard Test Method for Determining Air Change in a Single Zone by Means of a Tracer Gas Dilution." 2000.

Dietz, 2003a Dietz, R, "Determination of Unfiltered In-Leakage by ATD and AIMS E 741 Techniques", NHUGS 7/29/2003. PowerPoint Presentation.

Dietz, 2003b Dietz, R., "Applicability of ASTM E741 to four techniques for measuring unfiltered in-leakage," August, 2003.

Dietz, 2004a Dietz, R., "Atmospheric Tracer Depletion (ATD) Testing for Unfiltered In-Leakage at Wolf Creek Nuclear Plant", TTC-2004-06, Brookhaven National Laboratory, June 24, 2004.

Dietz, 2004b Dietz, R., "Atmospheric Tracer Depletion (ATD) Testing for Unfiltered In-Leakage at Callaway Nuclear Plant", TTC-2004-04, Brookhaven National Laboratory, June 26, 2004 Dietz, 2004c Dietz, R., "Procedure for Atmospheric Tracer Depletion (ATD) Testing for Unfiltered In-Leakage at Wolf Creek Nuclear Plant", TTC-2004-03, Brookhaven National Laboratory, August 10, 2004.

Dietz, 2004d Dietz, R., "Wolf Creek ATD In-Leakage Final Results," TTC-2004-05, Brookhaven National Laboratory, December 22, 2004.

Dietz, 2004e Dietz, R., Wilke, R., Weiser, R., Vignato, G., "Multi-Tracer Testing at Dominion's Surry Power Station for Air In-leakage Determination, TTC-101 1, Brookhaven National Laboratory, April 19, 2004 Dietz, 2006 Dietz, R., "Callaway ATD In-Leakage Final Results - Amended Report" TTC-2006-02, Brookhaven National Laboratory, May 26, 2006.

NEI, 2001 Nuclear Energy Institute, "Control Room Habitability Guidance", NEI 99-03, rev 0, June, 2001.

to ULNRC-05448 NEI, 2003 Nuclear Energy Institute, "Control Room Habitability Guidance", NEI 99-03, rev 1, March, 2003.

NRC, 2003 Nuclear regulatory Commission, Regulatory Guide 1.197, "Demonstrating Control room Envelop Integrity at Nuclear Power Stations," May 2003.

TTC-DP-02, 2003 Dietz, R., "Data Processing for ATD Testing Procedure Number: TTC-DP-02" revision 0 August 1, 2003.

TTC-TD-01, 2007 Tracer Technology Center Procedure, "Nuclear Control Room Air In-Leakage Test Design", TTC-TD-01, Revision 1, May 10, 2007.

TTC-SA-02, 2003 Tracer Technology Center Procedure, "Chromatograph Carrier Gas Acceptance", TTC-SA-02, Revision 0, August 1, 2003.

TTC-SA-04A, 2003 Tracer Technology Center Procedure, "GC 1 Chromatograph Pre-Analysis Checkout Procedure", TTC-SA-04A, Revision 0, July 02, 2003.

TTC-SA-04B, 2003 Tracer Technology Center Procedure, "GC2 Chromatograph Pre-Sample Analysis Checkout Procedure", TTC-SA-04B, Revision 0, July 24, 2003.

TTC-SA-04, 2003 Tracer Technology Center Procedure, "Sample Analysis Procedure", TTC-SA-05, Revision 0, July 26, 2003.

TTC-SA-06, 2003 Tracer Technology Center Procedure, "CATS Rack Maintenance Procedure", TTC-SA-06, Revision 0, July 24, 2003.

TTC-SA-07, 2003 Tracer Technology Center Procedure, "CATS/BATS Calibration Standard Loading Procedure", TTC-SA-07, Revision 0, July 23, 2003.

TTC-TP-02, 2003 Tracer Technology Center Procedure, "Preparation of Permeation Sources", TTC-TP-02, Revision 0, July 10, 2003.

TTC-TP-03, 2003 to ULNRC-05448 Tracer Technology Center Procedure, "Calibration of Active CATS Pumps", TTC-TP-03, Revision 0, August 1, 2003.

TTC-TP-04, 2003 Tracer Technology Center Procedure, "Preparation of Capillary Adsorption Tube Samplers", TTC-TP-04, Revision 0, August 1, 2003.