B17444, Forwards Two Mnps,Unit 2 Independent Corrective Action Verification Program (ICAVP) Drs That Have Been Placed in Unresolved Status as Part of ICAVP Insp by Parsons Power for Review

Forwards Two Mnps,Unit 2 Independent Corrective Action Verification Program (ICAVP) Drs That Have Been Placed in Unresolved Status as Part of ICAVP Insp by Parsons Power for Review

ML20154A882
Person / Time
Site:	Millstone
Issue date:	09/23/1998
From:	Bowling M NORTHEAST NUCLEAR ENERGY CO.
To:	NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM)
References
B17444, LCR-H97-16, NUDOCS 9810050031
Download: ML20154A882 (131)

Text

_ _ .. , . .. . . - -

l '"% Northeast Roge Feny Rd. (Route 1% Wuedord, Cr 06385 L Nuclear Energy waiston Noaea, rower station l Northeast Nuclear En rgy Company P.O. Box 128 Waterford, Cr 06385-0128

+

(860) 447-1791 Fax (860) 444-4277

'ne Northeast Utilitics System SEP 2 31998 Docket No. 50-336 B17444 U.S. Nuclear Regulatory Commission Attention: Document Control Desk Washington, D.C. 20555 Millstone Nuclear Power Station, Unit No. 2 Independent Corrective Action Verification Program Unresolved Discrepancy Reports The purpose of this letter is to submit for NRC review two Millstone Unit No. 2 ICAVP discrepancy reports that have been placed in unresolved status as part of the ICAVP inspection by Parsons Power. The designation of these two discrepancy reports as unresolved has occurred after extensive reviews and discussions held under the ICAVP protocol by Northeast Nuclear Energy Company (NNECO) and Parsons Power.

As part of the ICAVP protocol for Millstone Unit No. 2, NNECO requests that the NRC evaluate NNECO's position on discrepancy reports DR-0027 (Item 2) and DR-0312 (Item 1), and provide a final determination of whether or not Millstone Unit No. 2 is within its licensing bases for the specific topics addressed in this submittal.

Attachment 1 provides a summary of the NNECO position on the subject discrepancy reports and why NNECO believes that Millstone Unit No. 2 currently meets its licensing bases for each issue. NNECO concludes therein that the discrepancy reports should be classified as non-discrepant as part of the final record for this program.

Attachment 2 provides a copy of the handout used at the August 17,1998, ICAVP meeting to discuss DR-0027. Attachment 3 provides a copy of DR-0027, and Attachment 4 provides a copy of DR-0312. These attachments are provided as supplemental information and do not expand the licensing bases of the unit.

l There are no commitments contained within this letter. A001, i 9810050031 980923 PDR

- fl l ADOCK 0500o336 P PDR ,

083422-8 REY.13-95

. _ _ ._ _ ._ _.m.. _ _ . , _ _ _ _ . _ . _ _ _ _ _ _ _ _ _ _ _ _

U.S. Nuc! car Regulatory Commission B17444\ Page 2 i

Should there be any further questions on the information provided in this letter, please contact Mr. R. T. Laudenat at (860) 444-5248.

Very truly yours, NORTHEAST NUCLEAR ENERGY COMPANY Martin L. Bowling, Jr. V Recovery Officer - Technical Services cc: L. J. Callan, Executive Director of Operations H. J. Miller, Region 1 Administrator D. G. Mcdonald, Jr., NRC Senior Project Manager, Millstone Unit No. 2 D. P. Beaulieu, Senior Resident inspector, Millstone Unit No. 2 E. V. Imbro, Director, Millstone ICAVP inspections W. D. Lanning, Director Millstone inspections J. P. Durr, Chief, inspections Branch, Millstone Inspections D. L. Curry, Parsons Power Group l

l 1

I

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

l l Docket No. 50-336 l- 817444 i

(

t Attachment 1 Millstone Nuclear Power Station, Unit No. 2 Specific Information On Discrepancy Report DR-0027, " Enclosure Building Filtration & Containment / Enclosure Building Purge System Design," Rev. 5, dated October 8,1997, and Discrepancy Report DR-0312, " Water in Diesel Oil Storage Tank-(T47A) Could Enter Diesel Oil Supply Tanks," Rev. 3, dated January 27,1998.

1 l

1 l

1 1

l l

l l

1 1

l.

l l

l September 1998

U.S. Nuclser Regulatory Commission B17444\Attechmznt 1\Page 1 1

Millstone Nuclear Power Station, Unit No. 2 i

Specific Information On Discrepancy Report DR-0027, " Enclosure Building Filtration & Containment / Enclosure Building Purge System Design," Rev. 5, dated October 8,1997, and Discrepancy Report DR-0312, " Water in Diesel Oil Storage Tank (T47A) Could ,

Enter Diesel Oil Supply Tanks," Revision 3, dated January 27,1998 1

DR-0027 l l Parsons Power Group Discrepancy Report DR-0027, " Enclosure Building Filtration and Containment / Enclosure Building Purge System Design," Rev. 5, dated October 8,1997

! Item 2i Breach of Enclosure Buildina Intearity Due to Excessive Neaative Pressure

-(Main Exhaust Fans Ooeratina and Exhaust Damper 2-AC-11 Opened)

In item 2, Parsons Power has postulated the failure of a non-safety damper (2-AC-11) to close, which could challenge the enclosure building integrity. On this basis, Parsons l has determined this item to be a discrepant significance Level 1 finding.

l l NNECO has concluded that the issue reported in item 2 of DR-0027 does not represent l a discrepant condition because the issue postulated by Parsons Power is outside the i licensing bases of Millstone Unit No. 2. The licensing bases (including the design bases) for Millstone Unit No. 2 do not require the pathway that includes damper 2-AC-11 to meet single failure criteria for its isolation function during purging of the Enclosure l Building. This conclusion was provided in DR-0027 responses dated October 28, 1997, and May 1,1998, and was reviewed with the NRC and Parsons Power in a public j meeting held on August 17,1998.

Specifically, the Millstone Unit No. 2 configuration is described by an evaluation documented in Licensee Event Report (LER) 94-040-02, " Ventilation Design Deficiency l Affecting Enclosure Building Integrity," dated September 11,1995. LER 94-040-01, dated March 10,1995, reported, in part, a single failure vulnerability associated with damper 2-AC-11. The single failure would create a release path from the Enclosure Building to the atmosphere without charcoal filtration following a Loss of Coolant Accident (LOCA) during an Enclosure Building purge operation. A Probabilistic Risk l Assessment (PRA) was completed to evaluate the safety significance of the single i failure deficiency (Attachment 1 to LER 94-040-02). The PRA concluded that the plant j is designed to adequately and safely mitigate consequences of a LOCA, and is at no l

further risk due to radiological releases now than when previously evaluated at the time

of initial licensing.

l As part of LER 94-040-02, NNECO provided results of a licensing and design bases

( review of the Containment and Enclosure Building Purge System and Enclosure l

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

U.S. Nuclur Regulatory Commission B17444\ Attachment 1\Page 2 Building (Attachment 1 to LER 94-040-02). This review included reconsideration of NRC Qunstions and Answers applicable to these systems during the plant's initial licensing process, and the NRC's Safety Evaluation for Millstone Unit No. 2 dated May 10,1974.

As part of their eva'.uation of the LER, the NRC prepared a Safety Evaluation dated March 28,1996.* This NRC review concluded that in the unlikely event of a failure of damper 2-AC-11 to close during purging operation, the condition would be detected l and operator actions taken to mitigate the condition in sufficient time as to not be of '

sufficient consequence to support the requirement for a backfit modification.

NNECO recognizes that the NRC's decision not to invoke a backfit requirement in 1996, while based on the same failure as postulated in DR-0027, would not necessarily encompass the DR-0027 scenario if different consequences resulted. However, l NNECO has reviewed the DR-0027 postulated scenario for consequences, even '

though it is outside of the current licensing and design bases, and concluded that there would not be an increase in the consequences of the event previously evaluated by the NRC in the 1996 Safety Evaluation.

_DR-0312 Parsons Power Group Discrepancy Report DR-0312, " Water in Diesel Oil Storage Tank (T47A) Could Enter Diesel Oil Supply Tanks," Rev. 3, dated January 27,1998 Item 1: Water in Diesel Oil Storace Tank T47A Could Enter Diesel Oil Suoolv Tanks in item 1, Parsons Power has postulated the ingress of water into the underground Diesel Oil Storage Tanks. When the water transfers to the Diesel Oil Supply Tanks during several design basis scenarios the diesel generators would shutdown a short time later, causing a condition that would be outside Millstone Unit No. 2's design basis.

NNECO has concluded that the condition described in Discrepancy Report DR-0312 does not represent a discrepant condition. NNECO has evaluated each of the design basis scenarios that could be at issue relative to the diesel generators and the potential for failure of the units due to water intrusion into the underground Diesel Oil Storage Tank and concluded that these conditions are within the licensing bases of Millstone Unit 2. The scenarios evaluated include:

. Extended Full Power Operation Followed By A Loss Of Normal Power, With Or Without A LOCA; M USNRC Memorandum from C. H. Berlinger to P. F. McKee, " Millstone 2, Safety Evaluation -

f Enclosure Building Single Failure Vulnerabilities (TAC M93652)," dated March 28,1996.

- U.S. Nuclur R:gulatory Commission  ;

B17444\ Attachment 1\Page 3 l

. Extended Full Power Operation Followed By Unit Shutdown in Accordance i With The Technical Specifications Prior To Arrival Of A Probable Maximum  !

Hurricane (PMH) And Associated Design Basis Flood Levels; e Extended Full Power Operation Followed by a Unit Manual or Automatic Trip Due a Seismic Event Without a LOCA.

NNECO notes that in each instance, the licensing bases are met.

The Diesel Oil Storage Tank was designed to maintain leak tightness for the maximum water levels associated with the PMH, which supports our licensing basis. This is evident by the placement of the single open-to-atmosphere connection to the tank (i.e., ,

the vent pipe with flame arrestor) which is at approximate elevation (+)25'-0", a point .  !

which is higher than the maximum design basis wave runup level.  !

J NNECO notes that the Diesel Oil Storage Tank (T47A), and its associated transfer  ;

pumps (P47A&B) and piping, are designated in the plant's original design basis as non-Category 1 systems. Diesel oil has been identified in the FSAR as being automatically l transferred from the underground diesel oil storage tank to each of the diesel oil supply tanks. The diesel oil supply tanks fulfill the licensing basis requirements for Category 1 fuel supply.-

l 1

l

. _ . _ _ . . _ __ ._. .. .......__._..._._.._m . _ . _ . . - . _ __ . . _ _ . . _ _ _ . . _ . . _

Docket No. 50-336 B17444 i

I i

Attachment 2 J

Millstone Nuclear Power Station, Unit No. 2 DR-0027 Handout Used at ICAVP Meeting l

i l

l l

September 1998 l i

1

M '

k L

h r

l NRC/ Parsons DR 0027 .

i Enclosure Building Design I

Over. view, Mike Ahern Five Issue DR Rated Significance Level 1 Issues Discussed in Detai: in Previous Public Meetings with Parsons One Remaining Level 1 Issue; Enclosure Building Integrity following a single Failure of 2-AC-11 NU Concludes Millstone 2 Complies with .

License Bases, Public Health and Safety Assured j

~# &

Issue Description, Clark Maxson Millstone 2 Enclosure Building Ventilation Layout Scenario

- Infrequent Purge Operation, research of a sample year concluded Enclosure Building purge occurs less than 7% of the time

- LOCA Coincident with Single Failure of 1-AC-11 to close

- Results in Main Exhaust Fans and EBFS Fans Pulling on Enclosure Building -

i i

FROM OUTSIDE AIR ,

1. ~

FROM , -

~

SECONDARY CONTAINMENT

• RADWASTE VENT ENCLOSURE SYSTEM AC-8 )k i

TO UNIT 2 FROM Z~II STACK g uEL HANDLING *C-8

-* (P I 2-HV-299 SYSTEM ,

j l (A.B.C.)

9 PLENUM F-34A 2-AC-22 AC-fZ- _

(F-348 ) CLOSES OPENS -

(F-34 C ) ON CIAS _, ON CIAS TO UNIT I AC-57 M STACK MAIN #-

t EXHAUST CONTAINMENT 2-EB-54 FANS H Y OU SIDE U IT ,

F-25A l H 2-EB-43 3 (Z-1) F-23 AC-130 (EBFS FANS) A i I

' ' 9 CLOSES (Z-2 )

, T AC-7 8

AC-6

~

ON CIAS STOPS ON CIAS F-258 h '

AC-4 6

AC-5 EBFS FANS START ON EBFAS , LC LC EBFS AND CONTAINMENT & ENCLOSURE BUILDING PURGE SYSTEM SKM-MCR-ESFS.cgn i

License Bases Lack of single failure isolation for 2-AC-11 was addressed at the time of Plant License and later reaffirmed Post Startup in 1995 Both times concluded 2-AC-11 does not need to be Single Failure Proof based on acceptable consequences

$ b t

s License Bases, Original 1973-1974 XRC/SU Question and Answers XRC Question "... demonstrate flow in purge lines will be inward following a LOCA including failure of AC-1 or AC-11"

License Bases, Original (cont.) .

i XKECO Response- Evaluation of bounding AC-1 failure determined that with 2 EBFS fans running, that flow would still be into the Enclosure Building i

NRC SER for Millstone 2 dated May 10,1974-

" Based on our review of the proposed design and predicted performance of the EBFS, we have concluded that the system meets the intent of GDC .

41,42,43, and 64

e License Bases, Post Startup -

1995 Engineering Analysis in support of LER 94-040-02:

- Postulating a single failure of 2-AC-11 during purging is beyond 1:he License Bases

- Negligible Public Safety Impact based on the

~

probability of occurrence and consequences .

I

License Bases, Post Startup (cont.)

- Instrumentation would a ert operators to the potential release and terminate it by salutting off the main exhaust fans

- Thus, the plant is adequately and safely designed to mitigate the consequences of a LOCA

- No modifications are required to eliminate the

~

2-AC-11 single failure

License Bases, Post Startup (cont.)

NRC reviewed LER 94-040-02 and the NRC Safety Evaluation concluded:

The fact that the vulnerability exists only during purging of the Enclosure Building reduces the probability by at least an orc.er of magnitude", and .

License Bases, Post Startup (cont.)  !

The fact that the potential release pati is

monitored for radioactiviy provides a high degree of confidence that manual action would be quickly taken to terminate the release by shutting off the main ventilation fans"

License Bases, Post Startup (cont.)

Based on these considerations, the staff accepts the licensee's position that the correction of the AC-11 single failure

! vulnerability is unnecessary. It is also noted that the licensee has performed an Integrated Safety Assessment Program cost benefit analysis...and determined that the l potential safety benefit is insignificant." .

4  %

i Beyond License Bases, "What If" .  ;

Instrumentation would a..ert operators to the potential release and terminate it by shutting off the main exhaust fans j Enclosure Builc ing will not 3e significantly impacted

- 2nclosure Building is c.esignec. anc. model tested to greater than 8 in. w.g.

-Maximum possiie Enc osure Bui;.c.ing negative pressure is 6 in. w.g. -

< . I Beyond License Bases, "What If" (cont.)

Low probability event, purge operation is infrequent '

Conclusion, Public Health and Safety is Assured

6 Conclusions, Mike Ahern Millstone 2 Complies with License Bases Significant Design Margin Covers "What If" Public Health and Safety Assured

~

Docket No. 50-336 B17444

! Attachment 3 Millstone Nuclear Power Station, Unit No. 2 Discrepancy Report DR-0027 I

l l

l l

[

r l

l i-I i

September 1998

- , --' ~- -

SEP-01-98 11:28 FROM:PERCGT MILLSTnr ts10-655-25tr9 TO rrmFRr* ' P'Y'F : 17 l

i 1 PARSONS POWER GROUP INC. "

l 2075 Morgantown Road, Reading PA lCAVP MILLSTONE UNIT 2 19607 l

( DISCREPANCY REPORT (610) 855-2000 . FAX: (610) 855-2509 i

I DR NUMBER

DR-0027 DR TITLE:  !

Enclosure Bldg Filtration & Containment / Enclosure Bldg Purge System Design REVISION: S ISSUE DATE: 10/8/97 ORIGINATING GROUP: 2

! SIGNIFICANCE LEVEL: UNRESOLVED  !

DISCREPANCY Backcround i

Operation of the Enclosure Building Filtration System (EBFS) is credited in the Loss of Coolant j

Accident (LOCA) analyses for the calculation of otTsite dose. Maintaining the Enclosure Building i

Filtration Region (EBFR) at a negative pressure ensures any containment penetration leakage into the  !

EBFR remains in the Enclosure Building for controlled release via the installed filtration system. Thus, the offsite dose is reduced due to filtration and elevated release via the Unit 1 stack.

' {

} System Desian Per FSAR Section 6.7.4.1, the EBFS can maintain the EBFR under a minimum negative pressure of 0.25 in.wg with one fan ' operating. De capacity of one fan is 9,000 CFM and the design in leakage rate

. injo the EBFR is 2560 CFM Assuming the failure of the purge supply damper 2-AC-1 as the single failure, both EBFS fans are relied upon to operate in order to maintain the minimum pressure of-0.25 in.  !

wg. in the EBFR. In order to handle the additional 8400 CFM ofin. leakage through the open damper both fans are required to operate.

The Enclosure Building is designed to a maximum negative pressure of 2.0 in.wg.

Item 1: Breach of Enclosure Buildine Intenriev due to Excess Nerative Pressure (EBFS Fans Or,eratinal in the event of an emergency condition, an Enclosure Building Filtration Actuation System (EBFAS) signal will start the two EBFS fans F-25A & F 25B. The fans will run until shutdown by the plant operators. Damper 2-AC-1 (EB air supply isolation damper) will close upon receipt of the emergency signal, if not in the normally closed position. FSAR Section 6.7.2.I states that 2.0 in.wg. is the maximum differential pressure that the enclosure metal siding can sustain and still maintain ~ it's leak-tight characteristics. Since the EBFS does not have pressure control provisions to prevent exceeding the building ~ maximum pressure limit, a potential exists for breaching the integrity of the enclosure building when two fans operate with damper 2.AC-1 closed.

l Page i of 16

! DR-0027. DOC i 4

SEP-Ot-98 11:29 FTut: PEF.CGI t11Li s10NE A i O-D'h'-E*309 TU: FORRE WuE 18 Item t B,each of Enewure B..iidine integrity DumFsessive Neative prenure(Main Exhau DEntituumd Exhaust Damper 7.AC-ll Onenedt if a CI AS occurs while purging the Enclosure Building, the purge supply fan F 21 and damp automatically stopped and closed. respectively. The Enclosure Building purge exhaust damper AC-8 remains open. The main exhaust fans will continue to operate and, if damper 2-AC-Il fails open, air from the Enclosure umiding until the fans are turned of f manually following a Unit 2 Sta radiation alarm. The f "SS is also activated automatically and both fans operate. -

Damper AC-l is a pneumatic damper. The sudden closure of this damper while the main exhaust fans are exhausting air from the Enclosure Building,could cause a sudden increase in negative pressu building. The design exhaust rate from the building is 32.000 CFM and the operating pressure in th main exhaust plenum is 5.5 in.wg. (Dwg 25203-26057). This pressure is significantly higher than th 2.0 in.wg maximum pressure limit for the building.

When the main exhaust fans are operating, the air exhausted from the Enclosure Building (via exhaust damper AC-11) is mixed with exhaust air from the other buildings prior to discharge to the Unit 2 stack.

It is possible that the main exhaust fans will continue to operate together with the EBFS fans. The negative pressure induged by the main exhaust fans in the building is a back pressure to the EBFS fans and will cause the EBFS fans to operate to the left of the their combined fan curve, thus, increasing the building negative pressure.

An analysis of this potential breaching of Enclosure Building leak-tightness integrity does not exist.

NRC Safety Evaluation Attached to MP2-DE 96-0485 (Reference I.1) addressed the failure of non-sa damper AC-Il from the perspective ofreleases via the main exhaust path. NNECo committed to perform certain operator actions to shutdown this release path following receipt of a high radiation signal.

However, the Safety Evaluation did not consider the potential for excessive negative pressure in the

' Enclosure Building due to damper AC Il remaining in the open position. The closure of damper AC-Il may be necessary to ensure Enclosure Building integrity.

i

[ tem 3: Two Fan Oncratine Capacity below Design in leakare with Damner 2-AC-l in Onen Position _:

~ Cons.ider the case for which a LOCA occurs with damper 2-AC-1 in the open position. In addition.

consider the single failure ofdamper 2-AC-1 as failing in the open position. For this scenario, the design building in leakage is 2560 + 8400 = 10,960 CFM. Since the in leakage exceeds the design capacity of one fan (9000 cfm) it is concluded that both fans must operate to achieve the design 0.25 in.wg negative pressure. Using FSAR Figure 6.7 3 to estimate the two fan operating capacity indicates that the system may not be capable of handling the in-leakage. Thus, the design minimum -0.25 in.wg building pressure may not be achieved.

Item 4: inability to Maintain Minimum Negative Pressure with One EBFS Fan Operating and AC-1 Ooened According to the NRC Safety Evaluation Attached to MP2 DE 96-0485 (Reference 1.1):

In the event of a LOCA or MSLB during purging, with failure of actuation signal CHl-CIAS, damper AC-l would fail to close and fan F-25A would fail to start. This combination oflack of -

isolation (AC-1) and reduced filtered exhaust capability (loss of one train of exhaust / cleanup) ,

would prevent the secondary contain, ment from functioning properly as a fission product cleanup system for primary containment leakage as the single operating F-25 fan would not have

• sufficient capacity to establish and maintain the necessary negative pressure in the unisolated Enclosure Building...

. . Although the licensee claims that corrective action is not required by the original licensing basis, a modincation has been proposed to eliminate the AC 1 vulnerability. A gravity damper Page 2 of 16

( DR-0027. DOC ,

SEP-01-93 11:29 FRLMPEECGI titLLSTOtE 6te-EGS-2509 TO:rOUGERE PM 19 would be installed as shown in the drawing. It would be weighted such that operation of purg fan F-23 opens it, but a -0 25 wg. vacuum due to operation of an EBFS fan would not cause it to open. This action would eliminate the AC.) single failure condition "

(

According to the NOTE 2 under 4.1.13 of SP 2609E, "Approximately 5 pounds of force applied to suction damper, AC-130. counter weight lever is sutti ient to open the damper." The damper size is 47" x 47" For conservatism, use half of the damper asca as the effective area that is subjected to a diflerential pressure. Thus, a pressure differential of about 0.13 in. wg. will open the gravity da According to drawing 25203-26057, the static pressure in the vicinity of AC-130 ( pressure point 13)i 0.11 in.wg. Test data (attached to Reference B.13) shows that one EDFS fan operating can cre negative pressures than 0.25 in, wg (-0.35 to 0.75 in.wg).

In addition, during containment purge using the EBFS containment cleanup flow path, AC-1 is ma opened, but the purge supply fan must not be started (refer to 4.1.12 of SP 2314B). Thus AC-130 is relied upon to open by the differential pressure created by the EDFS fan for makeup air during containment cleanup.

The above contradicts the AC-130 performance requirement as stated in the AC-1 resolution. AC-130 will open at less than th.e design negative pressure of 0.25 in wg, with the purge supply fan F-23 shutdown.

Calculation Review To resolve items 1 and 3 above, the calculations noted under Reference B were reviewed. The review found a generic problem in that the calculations are out of date and have several analytical problems. A preliminary review of the UIRs for the system (Reference C) initially conilrmed the same finding.

However, the U1Rs were not specific enough to indicated that items I and 3 above will be corrected.

Reauest for Additionallnformation RAI-468 was issued on 09/17/97 requesting the following:

1 1.

Copies of the latest completed test procedures, performance data and operational data used to

, measure building pressure, fan flow rate, and building in !cakage rate for the following conditions:

a.

One fan operating with purge supply damper 2-AC-1 in the open and close positions.

b.

Two fans operating with purge supply damper 2- AC-1 in the open and close pos[tions.

- I

2. The basis for in leakage rate of 2560 CFM referenced in FSAR Section 6.7.4.1.

3.

Documentation which identifies the design features used to prevent the EBFR from exceeding the  !

maximum allowed 2.0 in. wg. negative pressure.

l

4. Up. dated calculation or other documentation that determines the system capacity with one fan operating and two fans operating.

The purpose of the request was to determine: ,

=  !

The as-built leak tightness of the building and the air in-leakage rate.

The capacity of the system when two fans are operating and damper 2 AC-1 is opened,

. i If the pressure in the building does not exceed the maximum allowed of 2.0 in.wg. when two l fans are operating and damper 2 AC-1 is closed. '

If the CMP had recognized the potential problem of breaching the integrity of the building l when two fans are operating with damper 2.AC-l closed.

Page 3 of 16 DR-0027. DOC

)

SEP-01-99 11 30 nW1:FEP.CGI tilLLCHOE 610- trh_"E3 TO:FO AERE P%L:20 l

i NU responded to the request on 09/26/97 by prosiihng copies of the following:

• Niemo MP2-DE-96 0485 (Refcrence G.1)

UIRs 3129,3171,956,2224,984 and 971 SP 2609A. II, C, D. E (Reference E )

i l

Evaluation of H Al-468 information The review of the above Surveillance Procedures (SPs) indicated the system is tested wi operating. The parameters tested andbmlding recorded. negative pressure and the time to achieve the design pressure are The SPs do not test the system w'th two fans operating. Thus, the capability of the system t

maintain the building minimum design pressure, when damper AC-1 is in the open position, ha verified. Assurance that the building integrity pressure limit is not exceeded when the damper closed position has not been demonstrated.

UIRs 3171,956,2224,984, and 971 indicated that the CMP has recognized the need to update the

! existing calculations. The UIRs, however, are not sufficiently detailed to indicate that the specifle discrepancies are recognized.

Item 5: UIR 3129 Conclusions and Corrective Action The CMP via UIR 3129 recognized the need for a new analysis / calculation to provide the system operating curve and operating procedure to test both fans operating simultaneously.ne UIR Recommended Disposition Details are repeated below:

"1.

AR 97019618-01 is written for CMP to evaluate need for additional surveillance or test t

based on AR 97019618-03 analysis esults.

2.

AR 97019618 02 is written for CMP to verify that EBFR siding to sustain and still maintain d ,

leak-tight characteristics at upper limit of 2 inch of W.G. negative pressure.

3.

' ' AR 97019618-03 is written for CMP to generate new calculation showing system operating curve with one fan operating and also, when two (2) fans operating in parallel. Upon completion evaluate for AR 97019618-01."

. The UIR Final Disposition is repeated below:

' " Expert Panel: AGREES with Recommended Disposition Details.

AR 97019618-01, CMP to evaluate need for additional surveillance or test based on AR 97019618-03 analysis results. AR 9701%18-02, CMP to verify that EBFR siding to sustain and maintain leak-tight characteristics at 2 inch of W.G. AR 97019618 03, CMP to generate new

• calculation showing system operating curve with one fan and also, when two (2) fans operating in parallel. Upon completion evaluate for AR 97019618-01."

It is clear from the above that damper 2-AC-1 concerns were not recognized by the UIR. The need for a * -

test is subject to the analysis results and not mandated. Mandated testing is required sineetat bestcan -

analysis is subjective for the EBFS. Testing for in leakage rate was not addressed, therefore, degradation of the building lenk tightness characteristics can not be monitored.

The UIR (section 1, item 1) states "...two (2) fans in operation must be capable of maintaining a negative l pressure in the EBFR less than the upper limit of 2 inches W.G." The UIR(section 2, item 1) states

" System performance calculation for the fans are inadequate." UIR section 2, item 4 states "There is no l

Page 4 of 16 DR-0027. DOC

-- . . -- .- - .- ~ __ . - .. - . . . ~ . _ . , . . _ ~ . - ~ - . - ~ . -

SEP-01-98 11:30 FROM:PE&CGI r11LLSTOtE 610-855 2509 TO:FOlEERE P45E:21 procedure developed toEOFR."

in the test these two fans operating simultaneously and record a maximum negative

( Maintaining the structural integrity of the EB is essential in order to take credit for the filtered, ele release path used to meet 10CFRl00 release limits. llowever, the system design calculatio program are inadequate to demonstrate that the system meets its design requirement. However, the UIR states as conclusion 1 (Section 2),"This UIR has been determined not to require a CR and has not iden' tified a potential salety significant condition." The Final Disposition Section of the UIR did contradict this statement.

This conclusion is inconsistent with the information presented in the UIR.

NMECo UIR 3129 did not recognize the potential safety significance of the consequence of Enclosure Building integrity.

Basis for Significance Level 1:

One of the safety functions of EBFS is to collect and process potentially radioactive airborne particles and gases in the EBFR following a LOCA and limit the site boundary radiation doses to the 10CFR100 requirements. Due to the lack of supporting calculations and/or test procedures /results the ability of th EBFS to perform its primary safety function cannot be assured.

Discrepancies identified may:

1.

Breach Enclosure Building integrity due to excess negative pressure (items I & 2), and 2.

Fail to maintain the minimum required negative pressure with design building in-leakage (items 3 &

4).

NNECo UIR-3129 did not recognize the potential safety significance of the consequence of breaching the Enclosure Building integrity.

(

4

< S. ; 9

.e u

• r .

Page 5 of 16 1l DR-0027. DOC .

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

~

SEP-01-99 11:30 FROr1:PERCGI 11ILLSTONE 610-GSS-2btD TO:FCUGERE PAGE:Ed ItEFEllENCES:

i A. FSAR l

1. Section 5,3 Enclosure Building '

2. Chapter 6 Engineered Safety Features Systems

3. Section 6.7 Enclosure Huilding Filtration System

4. Section 9.9.2 Containment and Enclosure Building Purge System .

5; Figure 6.71 EBFR Negative Pressure vs. Time Afler DDI

6. Figure 6.7-3 Enclosure Building Filtration System Fan Performance Curve B. Calculations

1. I K21 01,10/3 tn4, Enclosurc Building Filtration System

2. IK21-03,8/14/73, EBFS Nw & Delta P Calculation

3. IK2104,8/23/73, EBFS P Drop Calculation

4. 1K121-05,6/11n3, EBFS In-leakage

5. 1 K21-06,6/11/73, In leakage Calculations for Various Rooms

6. IK21-08,4/.6/71, Enclosure Building Filtration System

7. I K21-I I, 3/24/71, Summary of Calculation of Air Flow through the EDFS Filter Units with Given Crack Areas

8. I K21-14,12/02/69, EBFS Fan Selection

9. 1K21-15,8/21/69 Enclosure Building Filtration System

10. IK21-17,8/18/69, Enclosure Building Filtration System

11. IK21-18,12/04/69, Pressure in EBFR Transient Conditions

12. NUSCo Calc. XX XXX 10RA "EBFS Initiation Time Effect on LOCA Dose" Rev,01,12/05n8 13, 2 ENG-174,10/4/91, Air Flow through a 4" Hole from the Enclosure Building at Design Pressure C. UIRs 959 976 981 987 956 977 982 988 971 978 984 3129

~ ~

972 979 985 3171 974 980 986 D. Operating Procedures

1. OP 2314B Containment and Enclosure Purge, Rev 16

2. OP 2314G Enclosure Building Filtration System,Rev i1 E. Surveillance Procedures I. SP 2609 A, EBFS and Control Room Vent. Operability Test, Fac.1, Rev 12

2. SP 2609 B, EBFS and Control Room Vent. Operability Test, Fac. 2, Rev 14

3. SP 2609 C, Enclosure Building Integrity Verification, Rev 5

4. SP 2609 D, Enclosure Building Filtration System Filter Testing Refuel, Rev 10

5. SP 2609 E, Enclosure Building Filtration System Testing, Rev 6 <

F. RAls

1. RAI-0221,08/07/97

2. RAI-0415.09/0997

! 3. RAI 0468,09/17/97

(' Page 6 of 16

,; OR-0027. DOC -

SEP-Ol-98 11:31 FROr1:PEPI,GI t11LLSTC14- tite-855-2509 TO:FOL6ERE PEE: 2.5 G. Drawings:

25203-26028, shl, Rev 30 25203 26028, sh5, Rev 15 25203 26028, sh4, Rev 7

( 25203 26028, sh2, Rev 35 25203 26059, Rev i 25203-26057, Rev 0 25203-26028, sh3, Rev 10 25203-29640, Rev 1

11. Technical SpeciGcation

1. LCO and SR,3/4.6.5, Secondary Containment '

2. Bases,3/4.6.5, Secondary Containment

1. Miscellaneous 1

Memo MP2-DE-96-0485, AC-l and 1I "MP2 Enclosure Building Secondary Containment Integrity Single Failure Deficiencies" and NRC Memorandum (Carl Berlinger to Philip Mckee, Dated 3/28/96) covering disposition of this apparent deficiency along with the safety evaluation.

2.

Millstone inspection Report 97-02, June 24,1997, page 55 of 91: E8.2 (Closed) Unresolved item 50 336/95-25-03; Enclosure Building Filtration System Single Failure Vulnerability D. R. Ramos Tier-2 10/8/97 Originator Group Date

(

1 i

Page 7 of 16 DR-0027. DOC _,

i l

l l

SEP-01-93 11:31 FRO 1:PE&CGI NILLSTOE 610-835-2' A B TO: F WhERE f%E! P4 l --

EVAL,UATIDN f, @ BASIS VALID 0 BASIS INVALID - CLOSED O PREVIOUS 1,Y IDENTIFIED j

, IlY NNECo - C1.OSED

( Basis Valid l

=

R. T. Glaviano 10/8/97 Croup Lead Date REVIEW AND APPROVAL Reviewed: E.A. Blocher 10/8/97 Deputy Project Director Date t

k'# ' Approved: D.L. Curry 10/8/97 Project Director Date Forwarded to NNECo, NEAC, and NRC:10/8/97 Posted to WWW:10/13/97 Date Date i

Page 8 of 16 i DR-0027. DOC

sm a n 90 I t : 31 mnH:PEtcGt mt I snDE NG-509 iu:m mu WE: 25

~

t .

5USIStAlW OF NNECu PRUFD5ED CORitEC flVE ACllON NU has concluded that item I has been previously discovered and is considered to be a significance level 3 discrepancy. and postulated tan / damper scenarios in DR Items 2. 3. 4, and 5 cue non discrepant A summary of the conclusion for each item is listed in the Conclusion Continuation. CR M2-97-2294 has been i provide any follow-on activities associated with this DR.

Item 1; NU has concluded that the issue reported in item 1 of Discrepancy Report DR-0027 is a discrepan condition previously identified by NU in UIR No. 3129 and is a Significance level 3. UIR No. 3129 identified that a calculation or procedure does not exist to verify the enclosure building upper limit negative pressure of 2 in. w.g.. Corrective Actions consisting of creating a new calculation, evaluating the enclosure building verses the new calculation results, and review the need for a new surveillance procedure to test both fans were initiated; reference AR 97019618, assignments 01,02, and 03.

Item 2: NU has concluded that the issue reported in Item 2 of Discrepancy Report DR-0027 does not represen a discrepant condition. NU believes that the original plant licensing basis does not require damper AC-11 to be subject to single failure criteria. Ulis is based on the lack of redundancy in the system design and research of the licensing basis documentation. In addition, probability analysis has been performed which indicates that based on the relatively short amount of time that the enclosure building is being purged, the probability of occurrence of the single failure for damper AC-ll is low and the risk to public safety was determined to be negligible. The postulated single failure question for damper AC-1I was previously reported via LER 94-040.

Item 3: NU has concluded that the issue reported in Item 3 of Discrepancy Report DR-0027 does not represent a discrepant condition. A new counterbalanced gravity damper, AC-130, has been installed upstream of damper AC-1 which has been designed and is tested to remain closed at a negative pressure of 0.25 in. w.g.. This new damper will provide the necessary isolation of the supply air flow path so that the Enclosure Building Filtration Fans will be able to draw the required negative pressure of 0 25 in. w.g..

Ltem 4: NU has concluded that the issues reported in item 4 of Discrepancy Report DR-0027 do not represent a j

discrepant condition. First, the five pound applied force was provided in surveillance procedure SP2609E to assist the operator in determining what method to use to manually open the damper for testing and it is not a design requirement. Testing ofdamper AC-130 shows that it opens at a negative pressure of approximately 0.40 rn. w.g . Second, the normal flow path for makeup air for containment purge using the Enclosure Building Filtration System flow path is not through damper AC-130 but through AC-3. AC-3 is manually positioned during the test procedure to allow enclosure building air to be used for makeup.

Item 5: NU has concluded that the issue reported in item 5 of Discrepancy Report DR-0027 does not represent a discrepant condition. The Enclosure Building integrity was not considered an issue because a preliminary review indicated that the fan performance was not considered sufficient to reach a negative pressure which would challenge the enclosure building design. Single failure vulnerabilities were also addressed in past assessments and documentation, the results of which were deemed applicable to this issue. Based on this engineering documentation, the engineering staff and the UIR expert panel, who approves the UIR resolution, l concurred that the issue was not safety significant and a CR was not deemed necessary.

Second response received from NNECo on 05/07/98.

Dispositiont This res nse provides additional information to the initial DR-0027 re nse, M2-IRF-00481.

Item 1: Breach of En losure Building Integrity due to Excess Negative PressurekBFS Fans Operating) '

l Calculation 97-EBF-02000-M2, Rev. O," Enclosure Building inleakage and Negative Pressure," dated  !

12/18/97 calculated the negative pressure in the EB with both the EBFS fans operating in parallel. The

)

Page 9 of 16 D R-0027. DOC l

-- . - - . , - . . - - , , -. -- . . - - -- ~. -- . ~ - - -

Sr.rNt! 40 1t:30 rI&l:PESCGt MILLSTONE 6106 MJ TO:FOWERE fY G:26 i

' calculatmn tesuhs estabhsh that the negative pressure is 0 5 in, w.g. which is below the negative p in, w.g. described in the FSAR. Calculation 97-EDF 02000-M2. Rev. O is currently being revised. All required I FS AR (section 6.7) changes and procedure changes associated with the calculation results will be made following approval of Revision 1. The response to DR-0426 will address the calculation revision and the associated regtured FSAR and procedure changes.

l As previously stated in M2-1RF-00481, NU has concluded that the issue reported in Item 1 of l Discrepancy Report DR-0027 is a discrepant conditinn previously identified by NU. as documented in UIR 3129 and is a Significance Level 3 discrepancy. Significance Level 3 was chosen because the design basis was not

[ fully verified as the formal pressure calculations were not performed.

Item 2: Breach of Enclosure Building Integrity Due to Excessive Negative Pressure (Main Exhaust Fans j Operating and Exhaust Damper 2-AC-Il Opened):

l The previous response, M2-IRF-00481, to DR-0027. Item 2, stated that a new calculation will be created to determine the ED perfonnance assuming the failure modes identified in DR-0027, item 2. Due to the variables associated with the single failure scenario and the lack of test data a calculation could not be performed. Instead, Technical Evaluation, M2-EV-98-0095 was prepared to describe the single failure scenarios associated with the CEBPS Isolation dampers 2-AC-1 and 2 AC-11 and providejustification that the Enclosure Building Filtration System (EBFS), CEBPS and the Enclosure Building (EB) meet their design and licensing basis.

The Technical Evaluation concludes that the 2-AC-11 single failure scenario condition ofhaving both the MES fans and the EBFS fans drawing down the EB would not impact the EB leak-tightness integrity based on the original qualification testing of the EB. As previously stated in M2 IRF-00481, NU has concluded that the issue reported in Item 2 ofDiscrepancy Report DR-0027 does not represent a discrepant condition. NU considers the single failure scenarios associated with 2-AC-1 and 2-AC-11 beyond the original licensing and design basis of the CEBPS and no further corrective actions are required.

Note: He required FSAR changes associated with the single failure scenarios are addressed in UIR 2224, UIR 3367, and ACR M2-96-0788. In addition, all required FSAR (section 6.7) changes and procedure changes associated with the calculation change results will be made following approval of the calculations. The response to DR-0426 will address the revision to Calculation 97-EDF-02000-M2, Rev. O, and the associated required FSAR and procedure changes.

Item 3: Two Fan Operating Capacity below Design In-leakage with Damper 2-AC-1 In Open Position:

The previous response, M2-IRF-00481, to DR-0027, Item 3, stated that the 2-AC-1 single failure scenario was eliminated by implementati6n of PDCR MP2-041-95. Surveillance testing per SP2609E verifies that the required negative pressure will be maintained with one EBFS fan operating and 2-AC-1 open.

Calculation 97 EBF-02000 M2, Rev. O, calculated the inleakage into the EB to be 8,700 cfm for one EBFS fan operating. The calculated inteakage value does not match the inleakage value provided in the FSAR (section 6.7). Calculation 97-EBF-02000-M2, Rev. O is currently being revised. All required FSAR (section l 6.7) changes and procedure changes associated with the calculation results will be made following approval of Revision 1. "Ihe response to DR-0426 will address the calculation revision and the associated required FSAR and procedure changes.

l

( NU has concluded that the issue reported in item 3 of Discrepancy Report DR-0027 is a discrepant l condition previously identified by NU, as documented in UIR 3129 (calculation) and UIR 3367 (FS AR section Page to of 16

DR-o027. DOC

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

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

.SEP 01 P 11:33 FROt1: PEF.CGI tilLLG1Ut4- 610' n N ' TO:FOUGERE ' PM!. : $_'(

'6.7 and 5.3.5). NU considers item 3 a Significance Level 3' discrepancy based on the FSAR EDFS is capable of perfomiing its intended function as verified by surveillance testing per SP2609E.

i Items 4 and 5 No additional response. Response provided in M2-IRF-00481 concluded that Items 4 and 5 do not represent discrepant conditions

Conclusion:

This response provides additional information to the initial DR-0027 response [ M2-NU has concluded that DR-0027, has identifted a condition previously discovered b'y NU which requires correction. NU considers the issues identified in DR-0027 to be a Significance Level 3.

lltall As previously stated in M2-IRF-00481, NU has concluded that the issue reported in Item 1 of Discrepancy Report DR-0027 is a discrepant condition previously identified by NU, as documented in UIR 3129 and is a Significance level 3 discrepancy. The response to DR-0426 will address the revision to Calculation 97-EBF-02000-M2, Rev. O and the associated required FSAR and procedure changes.

Item 1 Technical Evaluation M2-EV-98-0095 concludes that the 2-AC-1I single failure scenario would not impact the EB leak-tightness integrity. As previously stated in M2 IRF-00481, NU has concluded that the issue reported in item 2 of Discrepancy Report DR-0027 does not represent a discrepant condition. NU considers the single failure scenarios associated with 2 AC-1 and 2-AC-11 beyond the original licensing and design basis of the CEBPS and no further corrective actions are required.

Item 3 NU has concluded that the issue reported in Item 3 of Discrepancy Report DR-0027 is a discrepant condition previously identified by NU, as documented in UIR 3129 and UIR 3367. NU considers Item 3 a Significance Level 3 discrepancy based on the required FSAR changes. The EBFS is capable ofperforming its intended function as verified by surveillance testing per SP2609E.

Items 4 and 5 No additional response. Response provided in M2-IRF-00481 concluded that Items 4 and 5 do not represent discrepant conditions l

1r

• Page 11 of 16 DR-0027. DOC

~ ,w . - s , - , . , , - -,n - ~ yy

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

SEP-01-90 11:34 FROt1:PE&CGI 11ILLSTOrn 610-855-2509 TO:FGmERF PAGE:28 COMMENT ON NNECo PROPOSED CORRECTIVE ACTION f

-( REVISED

~

General DR-0027 of the responsecontains

/ commentsseveral on this DR.complex technical issues. The below chronology is presented t DR Chronology:

l 10/08/97 - Preliminary DR-0027 irsued to NNECo.

10/28/97- NNECO issued respor.se to DR-0027 I

12/22/97 - NNECo issued Calcu'.ation 97-EBF-02000-M2 Rev. 0 01/19/98 - Working meeting conducted to discuss the DR-0027 issues. NNECO committed to revise the i response.

03/24/98 - Preliminary DR-042[> issued to NNECO. This DR iisted additionalissues related to Enclosure )

Building Ventilation System.

05/07/98 - NNECo issued revised response to DR-0027. This response tied resolution of DR-0027 to reso of DR-0426.

06/30/98 - NNECo issued response to DR-0426 and revised Calculation 97-EBF-02000-M2 Rev 1.

08/17/98 - Meeting at NNECo to discuss DR-0027. )

)

i l

The comments provided herein represent the initial Parsons comments on the NNECo responses.

Item 1: Breach of Enclosure Building Inteerity due to Excess Nenative Pressure (EBFS Fans Oneratine) t

-(

UIR-3129 identified that the existing supporting calculations and testing of the 2 EBFS Fan operating con did not support the -2.0 in.wg. building negative pressure limit (See Item 5). NU performed calculation 97-EBF-02000 M2 Rev 1, using the average 1993 as-tested inleakage condition of 8700 cfm at -0.35 in.wg. to calculate the expected maximum negative pressure for a two-fan operating configuration. Calculation 97-EBF-02000412-Rev 1 reports an expected negative pressure of-0.5 in.wg at 10600 cfm inleakage for the two fan operating condition.

The PS AR states the Enclosure Building design inleakage rate is 2560 CFM at -0.25 in wg. This DR identified that operation of 2 EBFS fans at this building integdty condition will cause the Enclosure Building negative pressure to exceed the FSAR-stated limit of-2 in.wg. NU did not perform an analysis to verify this discrepancy il or include this configuration in calculation 97-EBF-02000-M2 Rev 1.

Proposed Corrective Actions:

NU acknowledged the need to change the FSAR to incorporate the calculation results and proposes two corrective actions:

1) Change the FSAR-stated building design negative pressure limit to match the value (-9.75 in.wg.) det in the 1972 qualification testing, and

2) Change the building design inlaaNe value to 8700 cfm. 4 L

i Page 12 of 16

-( - DR-0027. DOC ,

1 w,.p,.- ., y m - - .

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

SEP-01 "O t t 135 FROM:Pti&CGT t1TILSTOtr h10-055-2509 To:m IGFRF PE 29

~~ -

Parsons Cor55I5bts -

j 1) The qualification test chamber configuration differs from the actual plant configuratio .

Scaling up the qualification test contiguration to match the plant contiguration gives a p inleakage of 46 cfm at -9.75 in.wg. The average actual building inleakage is 8700 cfm a .

rationale has been presented to document the applicability of the 1972 qualification test Millstone Unit 2 contiguratinn. Thus, the qualification test is not a valid basis for changing the En Building design negative pressure limit.

2) Changing the building design inleakage rate to 8700 cfm should maintain the enclosure bui

-2.0 in.wg limit for the two ran operating condition. It would also be necessary to specify aminimum required inteakage to ensure building pressure remains below the -2.0 in.wg limit for the tw configuration. Note. however, that changing the building design inleakage value will aggravate the iss identified in item 2 below.

This item remains DISCREPANT and is classified as Significance Level 3. This item remains implementation of the proposed corrective action to raise the building design inleakage value. It is that raising the inleakage value will aggravate the issue identified in item 2 below and will also invalida inleakage value used in LER 94-040 02.

Item 2: Breach of Enclosure Building Integrity Due to Excessive Negative Pressure (Main Exhaust Fans Operating and Exhaust Damper 2-AC-Il Opened):

NNECo Response:

The damper failure vulnerabilities described in this DR have been previously addressed and found by NU of sufficiently low probability to be below regulatory significance. Per NU, the fan combinations and system alignments in conjunction with the damper single failure are not part of the MP2 original licensing basis.

Parsons Comment:

In response to NNECo LER 94 040 02, the NRC issued an SER (attached to NU memo MP2-DE-96-0458, dated 9/9/96) which evaluated the AC-11 single failure scenario. The NRC evaluation of the radiological conseguence of the AC-11 failure considered:

1. Infrequency of the subject operating mode

2. Timely mitigation of the unfiltered release via c,perator action to secure the main exhaust fans on a high radiation alarm

3. With AC-11 open, the five fan operating condition will not cause excessive negative pressure in the buildin because the three main exhaust fans have suction demands that can be satisfied from other sou EBFS fans with a combined operating capacity of 13,900 CFM will create a vacuum in the building.

Enclosure Building inleakage will not exceed the 2560 cfm licensing basis value.

The NRC concluded that a backfit was not required based on the above items.

Regarding the above factors:

1.

Parsons concurs that the NRC evaluated an operating frequency of approximately 600 hours0.00694 days <br />0.167 hours <br />9.920635e-4 weeks <br />2.283e-4 months <br /> per year, which is less than 10% of the plant operating time.

l 2.

Parsons concurs that operator action to shutdown the Main Exhaust Fans on a high radiation alarm is l appropriate.

' Page 13 of 16 DR-0027. DOC l

l

SEft Ut-9G 11:35 FROt1:PESCGI tilLLSTOtE 610 M W sO') TO:FOUM RE PME : 30 m

3.

The FSAR & LER stated inleakage value of 2560 cfm, combined with the 13,900 cfm fan ca an excess fan capacity of about i1,000 cfm. However, these are not the actual plant conditions. The actual

{ inteakage of 8700 cfm, combined with a calculated fan capacity of10,600 cfm. gives an exc of approximately 2.000 cfm. Thus, this implied basis for the NRC's SER is negated by the actu conditions not matching the values stated in the LER.

According to NU's latest evaluation, a five fan operating mode, at worst, could create a :6.0 in, wg.

pressure. This is excessive when compared with the building licensing basis -2.0 in.wg. maximum le integrity limit. Thus, the evaluation result negates the basis for the NRC's SER that the five fan combination will not cause excessive negative pressure in the building.

Parsons concludes that the SER basis, i.e., the plant configuration and performance reported in the L does not match the installed configuration.

The Parsons concems regarding the five fan operating scenario are as follows:

1. An excessive negative building pressure (-6.0 in, wg.) results from a 5 fan operating configuration.

2. Leak tight integrity further, degrades (seam cracking) due to the excessive negative pressure causing permanent increase in building leakage. SP 23148 states that to prevent the potential for seam crackin building must be maintained between +0.4 and -0.4 in, wg. This limitation was the basis for modification PDCR 2-91-77 and PDCR 2-32-84.

3. Operators shutdown the Main Exhaust Fans.

4. EBFS fans actual excess capacity of 2000 cfm is insufficient to handle the increased leakage due to the degraded building and AC-11 open. Thus the EBFS cannot maintain the minimum -0.25 in.wg. pressure.

Proposed Corrective Action:

NU proposes to:

l.

! Change the FSAR-stated building design negative pressure limit to match the value (-9.75 in.wg.)

determined in the 1972 qualification testing, and

2. Increase the building design inleakage value to 8700 cfm.

Parsons Comment:

1.

As stated in issue 1 above, changing the building pressure to -9.75 in.wg. is not supported by the 1972 qualification test, and

2. With an increased Enclosure Building design inleakage value (to 8700 cfm), degraded leakage integrity, and AC-11 open, the EBFS fans will not be able to maintain the minimum required negative pressure.

In accordance with the August 17,1998 meeting at NNECo, this issue is categorized as UNRESOLVED.

Item 3: Two Fan Operating Capacity below Design in leakage with Damper 2-AC-1 In Open Position Item a:

Parsons agrees that the installation ofdamper AC-130 by PDCR MP2-041-95 resolves the single failure i vulnerability of 2-AC-1. SP 2609E appropriately tests the damper for its proper function. I i

l FS AR section 6.7.4.1 discusses the two-fan tiow capability as sufficient to mitigate an AC-1 failure. FSAR section 5.3.5 specifies the installation of AC-130 as the means for mitigating an AC-1 failure. NU identified,

{

via UIR-3367, the need to update FSAR Section 6.7.4.1 to specify AC-130 as the means for mitigating an AC-1 j

failure. Parsons concurs that this issue is "PREVIOUSLY IDENTIFIED" byNNECo.

l l

j Page L4 of 16 DR-o027. DOC t

l

SEP-01 -98 11 : I3 FRG71:PE&CGI tilLLSTotE C'10 'W'5 -M&2 TO:FOUGERE PAGE:31 .

llcath:

The calculation 97 EBF-02000-M2, Rev i resuhs also indicated that when two fans are operating,

(

4 is 10,600 CFM and not 13,900 CFM as stated in the FSAR. Parsons considers this issue to be "PREV' IDENTIFIED" by NNECo NU acknowledged the need to change the FSAR to incorporate the calculation results, item 4: Innbility to Maintain Minimum Negutive Pressure with One EUFS Fun Operating und AC-1 Opened l

Parsons agre, s that the installation of damper AC-130 by PDCR MP2-041-95 resolves the single failure :

vulnerability of 2-AC-1. NU's initial response to DR-0027, M2-IRF-00481, provided additional information I that shows that the setpoint for AC-130 is 0.40 in. wg. The setpoint is correctly selected and will ensure that AC-130 will not prematurely open in the event that 2-AC-1 fails in the open position. SP 2609E approp tests the damper for its proper function. Parsons concurs that this item is NON DISCREPANT.

Item 5: UIR-3129 Conclusions and Corrective Action \

Several AR's were generated to address the issues raised in UIR 3129. The assigned actions included additionl calculations and testing. Lice'nsing basis issues, and the safety sigrtificance of UIR-3129, were not identified at the discovery complete date. They were subsequently identified in the calculation 97-EDF-02000-M2, Rev 1 results. Thus, this issue remains DISCREPANT as Significance Level 4.

Prepared:

h.$. l huko D.R. Ramos / R. Glaviano 8[2G,(93 Group Lead Date

/fno p f j VS*f W \

Reviewed: E.A. Blocher Deputy Project Director Date Approved:

W D.L. Curry e/m/9(

Project Director

)

Date 1

I Forwarded to NNECo, NEAC, and NRC: / [Q f" Posted to WWW:

Date Date Page 15 of 16 DR 0027. DOC

SEP-01-53 11:44 FROM:PERCGI MILLSTOtE 610-855-2509 TO FOUGERE- PAGE:01 l

l FINAL RESOLUTION '

l i Item 1: OPEN (Significance Level 3)

'(' ltem 2: UNRESOLVED Item 3: PPEVIOUSLY IDENTIFIED i

! Item 4: . NON-DISCREPANT.

l Item 5: OPEN (Significance Level 4)

E.A. U13 cher 3/4Y70 Depury Project Director Date i

r l

.(

l t

l a

Page 16 of 16

'. .(

DR-0027. DOC

~ . .

I lARSONS POWER GROUP INC.

ICAVP MILLSTONE UNIT 2 2675 Morgantswn Rs:d, Re: ding, PA DISCREPANCY REPORT 19607 (610) 855-2000

• FAX: (610) 855-2509 DR NUMHER: DR-0027 DR TITLE:  !

REVISION: .4Enclosure Bldg Filtration & Containment / Enclosure Bldg Purge Sys ISSUE DATE: 10/8/97 ORIGINATING CROUP: 2 SIGNIFICANCE I2V_EL:!'  !

-l l 1 DISCREPANCY I

Backcround Operation of the Enclosure Building Filtration System (EBFS)is credited in the Loss of Coo (LOCA) analyses for the calculation of offsite dose. Maintalmng the Enclosure Building (EBFR) at a negative pressure ensures any containment penetration leakage into the EBFR

- Enclosure Building for controlled release via the installed filtration system. Sus, the offsite dose reduced due to filtration and elevated release via the Unic-1 stack. '

System Desian Per FSAR Section 6.7.4.1, the EBFS can maintain the EBFR under a nunimum negative pre in.wg with one fan operating. The capacity of one fan is 9,000 CFM and the design in-leak the EBFR is 2560 CFM. Assuming the failure of the purge supply damper 2-AC 1 as the sin ,

both EBFS fans are relied upon to operate in onier to maintain the minimum pressure of-0.2 the EBFR. In order to handle the additional 8400 CFM ofin-leakage through the open damper b are required to operate.

The Enclosure Building is designed to a maximum negative pressure of 2.0 in.wg.

Item 1: Br h of Enclosure BuilAa-lee avdue to h== N==:ve P.ww (EBFS Fans Oi,ci4inal In the event of an emergency condition, an Enclosure Building Fihration Actuation System (E will start the two ERFS fans F-25A & F-25B. 'Ihe fans will run until shutdown by the plant oper Damper 2-AC-1 (Es n supply isolation damper) will close upon receipt of the wgf signal, ifno; in the normally closed position. FSAR Section 6,7.2.1 states that 2.0 in.w3. is the maximum differential pressure that the enclosure metal sidingscan susta' and still mamtain its leak-tight characteristics. Since the EBFS docs not have pressure control provisions to prevent exceeding the building maxim limit, a2-AC-1 damper potential closed.exists for breaching the integrity of the enclosure building when two fans oper Page 1 of 15 DR-0027. DOC l

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

Jwm 2 BreA1LQf_lln9.]93.teluiding imetrirv Due tp_ Excessive _ N_Sga, tite Ooeratina.20ilhhEd.LQ_a.mur 2-AC-!l_Omne_IR n;;

If a CIAS occurs while purging the Enclosure Building, the purge supply fan F

[ automatically stopped and closed. respectively, nc Enclosure Bm!- are ding purge edau remains open. The main cdaust fans will contmue to operate and ifr damper 2-AC .

from the Enclosure Building until the fans are tumed off manuallyn following a U alarm. %e EDFS is also activated automatically and both fans operate.

exhausting air from the Enclosurc Building could cause re building. He design exhaust rate from the building is 32,000 CFM and the opera exhaust plenum is - 5.5 in.wg. (Dwg 25203-26057).

in.wg maximum pressure limit for the buildmg. This pressure is significantly higher than the -2 0 When the main exhaust fans are operating, the air exhausted from the Enclosure Build damper AC-1 i)is mixed with exhaust air from the other buildmgs prior to discharge to t It is possible that the main exhaust fans will continue to operate together with the .

pressure induced by the main exhaust fans in the building is a back pressure to the EBFS fans and causepressure.

negative the EBFS faris to operate to the left of the their combined fan curve, thus, incre An analysis of this potential breaching of Enclosure Building leak-tightness integrity does not NRC Safety Evaluation Attached to MP2-DE-96-0485 (Reference I.1) addressed th damper AC-11 from the perspective ofreleases via the main exhaust path. NNECo com certain operator actions to shutdown this release path following receipt of a high radiation si However, the Safety Evaluation did not consider the potential for excessive negative pressure Enclosure Building due to damper AC-11 rmMa% in the open position. He closure of damper may be necessary to ensure Enclosure Building integrity.

(

Item 3 Two Fan Operatina Camr#v below Desien In-lc*a>>e with D== 2-AC-1 in Onen P

'4 Consider the case for which a LOCA occurs with damper 2 AC-1 in the open position. In addi consiar the single building in-leakage is failure of damper 2 AC-1 as failing in the open position. For this scenario 2560 + 8400 = 10,960 CFM. Since the in-leakage exceeds the design capacity of one fan (9000 cfm) it is concluded that both fans must operate to achieve the design 0.25 in pressure. Using FSAR Figure 6.7-3 to estunate the two fan operating capacity indicates that the maynotnot may be capable of handling the in-lcakage. Aus, the design minirnum -0.25 in, wg buildi be achieved.

Item 4: In=hility to Maia% Mi= =em Nana'ive Ibsure with one EBFS Fan Or,c.. :-e and According to the NRC Safety Evaluation Arrachad to MP2-DE-96-0485 (Reference I.I):

" in the event ofa LOCA or MSLB during purging, with failure ofactuation signal CH1 CIA damper AC-1 would fail to close and fan F-25A would fail to start. This combination oflack of isolation (AC-1) and reduced filtered exhaust capability (loss ofone train of exhaust / clean

. would prevent the secondary containment from functioning properly as a fission product clean system for primary containment leakage as the s*mgle operating F-25 fan would not have sufficient capacity to establish and maintain the necessary negative pressum in the unisolated Enclosum Building...

Page 2 or 15 DR-0027. DOC

~ __ . ___ _ _ __ - _ _ _ _ __ . . .- __ - . _ __ . _ _ .

PW4 4 M

. Although the hcensec claims that corrective action is not requited by the original lice a modification has been proposed to climinate the AC-1 vulnerability. A gravity damper would be installed as shown m the drawing. It would be weighted such that operation of purge fan F-23 opens it. but a -0.25 wg. vacuum due te operation of an EDFS fan would not cause it to open.

This action would ehminate the AC-1 single failure condition."

According to the NOTE 2 under 41.13 of SP 2609E, "Approximately 5 pounds of force applied to F suction dampct, AC-130, counter weight !cycr is sufficient to open the damper." The damper size is 47" For conservatism. use half of the damper area as the effective area that is subjected to a ditTeren pressure. Thus, a pressure differential of about 0.13 in, wg. will open the gravity damper. Accord drawing 25203-26057, the static pressure in the vicimty of AC-130 ( piessure point l3)is -0.111 Test data (attached to Referencc D.13) shows that one EBFS fan operating can create higher pressures than 0.25 in. wg. (-0.35 to -0.75 in.wg).

In addition, during containment purge using the EBFS containment cleanup flow path. AC-l opened, but the purge supply fan mus6 not be started (refer to 4.1.12 of SP 2314B). Hus, AC-130 is relied upon to open by the differential pressure created by the EBFS fan for makeup air during contai cleanup.

The above contradicts the AC 130 performance requirement as stated in the AC-1 resolution. AC-130 will open at less than the design negative pressure of 0.25 in. wg, with the purge supply fan F-23 shutd Calculation Review To resolve items I and 3 above, the calculations noted under Reference B were reviewed. The review found a generic problem in that the calculations are out of date and have several analytical problems. A preliminary review of the UIRs for the system (Reference C) initially confirmed the same finding.

However, the UTRs were not specific enough to indicated that items I and 3 above will be corrected.

Reauest for Addidaa=1Information RAI-468 was issued on 09/17/97 requesting the following:

1.

Copics of th latest completed test procedures, performance data and operational data used to measure building pressure, fan flow rate, and building in4eakage rate for the following conditions:

a.

One fan operating with purge supply damper 2-AC-1 in the open and close positions.

b. Two fans operating with purge supply damper 2- AC-1 in the open and close positions.

2.

The basis for in-leakage rate of 2560 CFM referenced in FSAR Section 6.7.4.1.

3.

Documentation which identifies the design features used to prevent the EBFR from exceeding the maximum allowed 2.0 in, wg. negative pressure.

4.

Up-dated calculation or other documentation that determines the system capacity with onc fan operating and two fans operating.

He purpose of the acquest was to determme:

De as-built leak-tightness of the building and the air in-leakage rate.

e The capacity of the system when two fans are operating and damper 2 AC-1 is opened.

If the pressure in the building docs not exceed the maximum allowed of-2.0 in. wg. when two l fans are operating and damper 2-AC-1 is closed.

t Page 3 of 15 DR-0027. DOC

If the CMP had recognized the potential problem of breaching the integrity of the bui two fans arc operating with damper 2-AC-1 closed.

NU responded to the request on 09/26/97 by providmg copies of the following:

I

• Memo MP2-DE-96-0485 (Reference G 1)

UIRs 3129,3171,956. 2224. 984, and 971 SP 2609A, B, C, D, E (Reference E )

Evaluation of R AI-468 Information The review of the above Surveillance Procedures (SPs) indicated the system is tested with only operating. The building negative pressure and the time to achieve the design pressure are the o parameters tested and recorded.

' The SPs do not test the system with two fans orerating. Thus, the capability of the system to achieve an maintain the building minimum design pressure, when damper AC-1 is in the open position, has not been verified. Assurance that the building integrity pressure limit is not exceeded when the damper is in the closed position has not been demonstrated.

UIRs 3171, 956,2224,984, and 971 indicated that the CMP has recognized the need to update the existing calculations. He UIRs, however, are not sufficiently detailed to indicate that the specific discrepancies are recognized.

Item 5: UTR-3129 thb% and Cm.w.tive Action The CMP via UIR 3129 recognized the need for a new analysis / calculation to provide the system op curve and operating procedure to test both fans operating simultaneously. The UIR Rw.eucaded Disposition Details are rcpeated below:

"1.

AR 9701%18-01 is written for CMP to evaluate need for additional surveillance or test base on AR 97019618-03 analysis n:sults.

f 2.

AR 9701961842 is written for CMP to verify that EBFR siding to sustam and still maintain leak-tight characteristics at upper limit of 2 inch of W.G. negative pressure.

3.

AR 97019618 03 is written for CMP to generate new calculation showing system operating curve with one fan operating and also, when two (2) fans operating in parallel. Upon completion evaluate for AR 97019618-01,"

He UIR Final Disposition is repeated below:

" Expert Pancl: AGREES with P-, = - DispositionDetails.

AR 97019618-01, CMP to evaluate need for additional surveillance or test based on AR 97019618 03 analysis results. AR 9701961842, CMP to verify that EBPR siding to sustain and maintain leak-tight characteristics at 2 inch ofW.G. AR 97019618 03, CMP togenerate new calculation showing system ep .G g curve with one fan and also, when two (2) fans operating in parallel. Upon completion evaluate for AR 9701 % 18-01."

It is clent from the above that damper 2-AC-1 concerns were not recognized by the UIR. De need for a test is subject to the analysis results and not mandated. Mandated testing is required since, at best, an

. analysis is subjective for the EBFS, Testing for in-leakage rate was not addressed, therefore, degradation of the building leak-tightness characteristics can not be monitored.

Page 4 of 13 DR-0027. DOC

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

- _ - - ~ . . - . . -

i The UIR (section 1, item i) states " . two (2) fans in operation must be capabic of pressure in the EBFR less than the upper limit of 2 inches W G."

The UIR (section 2, item 1) states

" System performance calculation for the fans are inadequate." UIR section 2, item 4 st; f procedure developed to EBFR."

in the test these two fans operating simultaneously and record a maximum negative pr!

Maintaining the structural integrity of the EB is essential in order to take credit for the filter release path used to mcct 10CFR100 release limits.1-lowever, the system design c; program are inadequate to demonstrate that the system meets its design requirement. However,l states as conclusion ! (Section 2), "His UlR has been deterntined not to rcquire a CR and has not identified a potential safety significant condition." The Final Disposition Section of the UIR! '

contradict this statement.

This conclusion is inconsistent with the information presented in the UIR.

NNECo UIR integrity.

Enclosure Building 3129 did not recognize the potential safety significance of the conseq Basis for Significance Level 1:

One of the safety functions of EBFS is to collect and process potentially radioactive airbome gascs in the EBFR following a LOCA and limit the site boundary radiation doses to the 10CFR100 requirements. Due to the lack of supporting calculations and/or test procedures /results the abilityj '

EBFS to perform its prinuuy safety function mannt be assured.

I Discrepancies identified may:

1.

Breach Enclosure Building integrity due to excess negative pressure (items 1 & 2), and 2.

Fail to maintain the nunimum required negative pressure with design building in-leakage (item NNECo UIR-3129 did not recognize the potential safety significance of tne consequence of brea Enclosure Building integrity.

I i

i Page 5 of 15 DR 0027. DOC i

REFERENCES:

A. FSAR

1. Section 5.3 Enclosure Building

{ 2. Chapter 6 Engineered Safety Features Systems

3. Section 6.7 Enclosure Building Filtration System 4 Section 9.9.2

5. Figurc 6.7-1 Containment and Enclosure Building Purge System 6.

EBFR Negative Pressure vs. Time Afler DB2 Figure 6.7-3 Enclosure Building Filtration System Fan Performance Curve B. Calculations 1.

IK21-01,10/31n4, Enclosure Building filtration System

2. IK21-03, 8/14n3, EBFS Flow & Delta P Calculation

3. I K21-04, 8/23n3, EBFS P Drop Calculation 4 1K121-05,6/11/73, EBFS In-leakage 5.

I K21-06, 6/1In3, in-leakage Calculations for Various Rooms 6.

1 K2108, 4/6n1, Enclosure Building Filtration System 7.

IK21 11, 3/24/71, Summary of Calculation of Air Flow through the EBFS Filter Units w Crack Areas

8. IK21-14,12/02/69, EBFS Fan Selecuon 9.

IK21-15, 8/21/69 - Enclosure Buddag Filtration System 10.1K21-17, 8/18/69, Enclosure Building Filtration System 11.1K21-18,12/04/69, Pressure in EBFR Transient Conditions

12. NUSCo Calc. XX-XXX-10RA "EBFS Initiation Time Effect on LOCA Dose" Rey, 01,12

13. 2-ENO-174,10/4/91, Air Flow through a 4" Hole from the Enclosure Building at Desig C. UIRs 959- 976 981

( 987 956 977 982 988 971 978 984 i 3129 972 979 985 3171 974 980 986 D. Operating Procedures 1.

OP 2314B Containment and Enclosure Purge, Rev 16 2.

OP 23140 Enclosure Building Filtration System, Rev 11 E. Surveillance Procedums 1

2 SP 2609 A, EBFS and Control Room Vent. Operability Test, Fac.1, Rev 12 SP 2609 B, EBFS and C3ntrol Room Vent. Operability Test, Fac. 2, Rev 14

3. SP 2609 C, Enclosure Building Integrity Verification, Rev 5 4.

5. SP 2609 D, Enclosun: Building Filtration System Filter Testing-Refbel, Rev 10 SP 2609 E, Enclosure Building Filtration System Testing, Rev 6 F. RAls

1. RAl-0221,08/07/97

2. RAl-0415,09/0997

3. RAT-0468,09/I7/97 Page 6 of 15 DR-0027. DOC i

k~

G. Drawings:

25203-26028, sh 1, Rev 30 25203-26028, sh5, Rev 15 25203-26028, sh2, Rev 35 25203-26028, sh4, Rev 7 25203-26059, Rev i 25203-26057, Rev 0 25203-26028, sh3, Rev 10 25203-29640, Rev 1 11 Technical Specification 1.

LCO and SR,3/4.6.5, Secondary Containment

2. Bases,3/4.6.5, Secondary Containment I. Miscellaneous 1.

Memo MP2-DE-96 0485, AC-1 and 11 "MP2 Enclosure Building Secondary Containm Single Failure Deficiencies" and NRC Memorandum (Carl Berlinger to Philip Mckee covering disposition of this apparent deficiency along with the safety evaluation.

2.

Millstone inspection Report 97-02, June 24,1997, page 55 of 91: ES.2 (Closed) Unresolved item 336/95-25-03; Enclosure Building Filtration System Single Failure Vulnerability I

D. R. Ramos Tier-2 10/8/97  :

Originator Group Date Page 7 of 15 DR-0027. DOC I

\

= __ _ .

l

@ EtASIS VALID EVALUATION ~-*

l 0 DASIS INVALID - CLOSED

, 0 PREVIOUSLY IDENTIFIED BY NNECo - CLOSED l- Basis Valid l

1 1

I i

=--

R. T. Glaviano jo/s/97 l Group Lead i Date REVIEW AND APPROVAL Reviewed: E.A. Blocher 10/8/97 Deputy Project Director Date i

. Approved

D.L. Curry

. 10/8/97

(, Project Director Date Forwarded to NNECo, NEAC, and NRC:10/8/97 Posted to WWW:10/13/97 Date Date

?

l Page 8 of 15 DR 0027. DOC k.

l

u.i.e.r

^

~

SUMMARY

OF NNECo PROPOSED CORRECTIVE ACTION NU has concluded that item I has been previously discovered and is considered to be a s I discrepancy, and postulated fan / damper scenarios in DR Items 2,3,4, and 5 are

. A summary of non-dis the conclusion for each item is listed in the ConclusionCR Continuation.

M2 97-2294 has been issued to provide any fullow-on activities associated with this DR.

Item 1. NU has concluded that the issue reported in Item 1 of Discrepancy Report D condition previously identified by NU in UIR No 3129 and is a Significance level 3. UIR N that a calculation or procedure does not exist to verify the enclosure building o upper lim in. w.g.. Corrective Actions consisting ofcreating a new calculation evaluating the enclosure b verses the new calculation results, and review the need for a new surveillance procedure initiated; reference AR 97019618, assignments 01,02, and 03.

Item 7. NU has concluded that the issue reported in Item 2 of Discrepancy Report DR a discrepant condition. NU believes that the original plant licensing basis does not requir subject to single failure criteria. This is based on the lack of redundancy in the system des licensing basis documentation. In addition, probability analysis has been performed which the relatively short amount of time that the enclosure building is being purged, the probabil the single failure for damper AC-11 is low and the risk to public safety was determined to be postulated single failure question for damper AC-11 was previously reported via LER 94-040.

Item 3. NU has concluded that the issue reported in Item 3 ofDiscrepancy Report DR-0 a discrepant condition. A new counterbalanced gravity damper, AC-130, has been insta AC-1 which has been designed and is tested to remain closed at a negative pressure of 0.25 damper will provide the necessary isolation of the supply air tiow path so that the Enclosure Fans will be able to draw the required negative pressure of 0.25 in. w.g..

Item 4:

NU has concluded that the issues reported in Item 4 ofDiscrepancy Report DR-0027 do discrepant condition. First, the five pound applied force was provided in surveillance procedure S assist the operator in determining what method to use to manually open the damper for testing a

.I design requirement. Testing of damper AC-130 shows that it opens at a negative pressure in. w.g., Second, the normal flow path for makeup air for containment purge using the Enc Filtration System flow path is not through damper AC-130 but through AC-3. AC-3 is manual during the test procedure to allow enclosure building air to be used for makeup.

Item 5:

NU has concluded that the issue reported in Item 5 ofDiscrepancy Report DR-0027 does a discrepant condition. The Enclosure Buildingintegrity was not considered an issue because review indicated that the fan performance was not considered sufficient to reach a negative press challenge the enclosure building design. Single failure vulnerabilities were also addressed in past documentation, the results of which were deemed applicable to this issue. Based on this en documentation, the engineering staff and the UIR expert panel, who approves the UIR resolution, the issue was not safety significant and a CR was not deemed necessary.

Second response received from NNECo on 05/07/98.

Disposition: This response provides additional information to the initial DR-0027 response, M2-IRF Item 1: Breach of Encinsure Buildinn Inteerity due to Excess Negative Pressure (EBFS Fans Operatind Calculation 97-EBF-02000-M2, Rev. O, " Enclosure Building Inteakage and Negative Pressure," dat 12/18/97 calculated the negative pressure in the EB with both the EBFS fans operating in parallel. T calculation results establish that the negative pressure is 0.5 in. w.g. which is below the negative Page 9 or 15 DR4o27. DOC

, -- .. .- .- .- - - - . - -.- - ~ - - . - - . . - .

in. w g. described in the FSAR Calculation 97-ERF 02000412 Rev 0 is c, FS AR (section 6.7) changes and procedure changes associated with the calcula{  !

acquired FSAR and procedure changesapprova) nofandRevision the associated 1 The res Discrepancy Report DR-0027 is a discrepant condition pt ,

umented in UlR 3129 and is a Signi6cance Level 3 discrepancy.

verified as the formal pressure calculations were not performed.

Significance Level 3as was not fully chosen Item 2: Breach of Enclosure Buildine Inteerity Due to Excessive ans Operatine and Exhaust Damper 2-AC-Il Onenedh Neentive P to determine the EB performance assuming the failure mode -

em 2. Due to the variables Technical Evaluation, M2-EV-98-0095 was prepared t .

e CEBPS Isolation dampers 2-AC-1 and 2-AC-11 and providejustification n that th i

System (EBFS), CEBPS and the Enclosure Building (EB) meet their design and

)

MES fans and the EBFS fans drawing down the EB wo(

i thi original qualification testing of the EB. As previously stated in M2-IRF-00481 ,

e issue reported in Item 2 ofDiscrepancy Report DR-0027 does not represent a discr design basis of the CEBPS and no further corrective actions ar i

Note:

UIR 3367, and The required FSAR ACR M2-96-0788. changes associated with the single failure scenarios are add ,

In addition, all required FSAR (section 6.7) changes and procedure chang associated with the calculation change results will be made following approval of th FSAR and procedure changes.to DR-0426 will address the revision to Calcula ,

Item 3: Two Fan Operatine Canacity below Deslen In-leakane with Damner 2-AC-1 In On The previous response, M2 IRF-00481, to DR-0027, Item 3, stated that the 2-AC-1 s scenario was eliminated by implementation ofPDCR MP2-041-95.

Surveillance testing per SP2609E verifies the the required negative pressure will be maintained with one EBFS fan operating and Calculation 97-EBF-02000-M2, Rev. O, calculated the inledage into the EB to be 8,700 (EBFS fan operating. The calculated inleakage value does not match the inleakage value 6section 6.7). Calculation 97-EBF-02000-M2, Rev. O is currently being revised. All re Revision a nd 1. The response to DR-0426 will address the ca procedure changes.

c NU has concluded that the issue reported in Item 3 of Discrepancy Report DR-0027 ondition previously identified by NU, as documented in UIR 3129 (calculation) and UIR 3 is capable ofperforming its intended function as verified by su .

Page 10 of 15 DR 0027. DOC

Items 4 and 5 j f No additional response. Response provided in M2-IRF-00481 concluded that Items 4 and 5 represent discrepsnt conditions

\

Conclusion:

This response provides additionalinformation to the initial DR-0027 response. M24RF-0 NU has concluded that DR 0027, has identified a condition previously discovered by NU which req correction. NU considers the issues identified in DR-0027 to be a Significance Level 3.

Ite m 1 i

As previously stated in M2 FRF-00481, NU has concluded that the issue reported in Item 1 of '

Discrepancy Report DR-0027 is a discrepant condition previously identified by NU, as documente and is a Significance Level 3 discrepancy. The response to DR-0426 will address the revision to Calc EBF-02000-M2, Rev. O and the associated required FSAR and procedure changes.

Item 2 Technical Evaluation M2-EV-98-0095 concludes that the 2-AC-Il single failure scenario would not impact the EB leak-tightness integrity. As previously stated in M2-IRF-00481, NU has concluded that the reported in Item 2 of Discrepancy Report DR-0027 does not represent a discrepant condition. NU considers the single failure scenarios associated with 2-AC-1.and 2-AC-11 beyond the original licensing and desig CEBPS and no further conective actions are required.

Item 3 NU has concluded that the issue reported in item 3 of Discrepancy Report DR-0027 is a discrepant condition previously identifico by NU, as documented in UIR 3129 and UIR 3367. NU considers Item 3 a Significance Level 3 discrepancy based on the required FSAR changes. The EBFS is capable of

intended function as verified by surveillance testing per SP2609E.

Items 4 and 5 No additional response. Response provided in M2-IRF-00481 concluded that Items 4 and 5 do not represent discrepant conditions Page it of 15 DR-0027. DOC

l 70MMk.lT ON NNECo PROPOSED CORRECrlVE ACUON ]

Gen eral -

(

of the response / comments on this DRDR-0027 contains several complex techni DR Chronology; 10/08/97 - Preliminary DR-0027 issued to NNECo.

10/28/97 - NNECO issued response to DR-0027.

12/22/97 - NNECo issued Calculation 97-EBF-02000-M2 Rev. 0 01/19/98response.

- Working meeting conducted to discuss the DR-0027 issues. NNECO committed to rev 03/24/98 - Preliminary DR-0426 issued to NNECO. This DR listed additional issues related to Enclosure Building Ventilation System.

05/07/98 - NNECo issued revised response to DR-0027. This response tied resolution ofDR-0027 to re ofDR-0426.

06/30/98 - NNECo issued response to DR-0426 and revised Calculation 97-EBF-02000-M2 Rev 1.

The comments provided herein represent the initial Parsons comments on the NNECo responses.

Item 1: Breach of Enclosure Buildine Inteerity due to Excess Neentive Pressure (EBFS Fans Ooeratine)

UIR-3129 identified that the existing supporting calculations and testing of the 2 EBFS Fan operating co did not support the -2.C in.wg. building negative pressure limit (See Item 5). NU performed calculation 97-EBF-02000-M2 Rev 1, using the average 1993 as-tested inleakage condition of 8700 cfm at -0.35 in.wg calculate the expected maximum negative pressure for a two-fan operating configuration. Calculation 97-EBF-02000-M2 Rev 1 reports an expected negative pressure of-0.5 in.wg at 10600 cfm inleakage for the two fan operating condition.

The FSAR states the Enclosure Building design inleakage rate is 2560 CFM at -0.25 in wg. This DR identified that operation of 2 EBFS fans at this building integrity condition will cause the Enclosure Building negativ pressure to exceed the FSAR-stated limit ef-2 in.wg. NU did not perform an analysis to verify this discrepancy or include this configuration in calculatior. 97-EBF-02000-M2 Rev 1.

4

, Proposed Corrective Actions:

NU acknowledged the need to change the FS AR to incorporate the calculation results and proposes two corrective actions:

1) Change the FSAR-stated building design negative pressure limit to inatch the value (-9.75 in.wg.) dete in the 1972 qualification testing, and

2) Change the building design inleakage value to 8700 cfm.

Parsons Comments:

1) The qualification test chamber configuration differs from the actual plant configuration (See DR-0594).

3caling up the qualification test configuration to match the plant configuration gives a predicted building inleakage of 46 cfm at -9.75 in.wg. The average actual building inleakage is 8700 cfm at -0.35 in.wg. No rationale has been presented to document the applicability of the 1972 qualification test to the existing Millstone Unit 2 configuration. Thus, the qualification test is not a valid basis for changing the Enclosure Building design negative pressure limit.

Page t2 of15 DR o027. DOC

1 h 2) Changing the building design inleakage rate to 8700 cfm should u mairtain the en ng below the p -2 0 in.wg limit for the two-ran operating condition. It would also nimum be necessary l

g configuration. Note, however, that changing the buil identified in item 2 below. ue ng This item remains DISCREPANT and is classified as Significance Level 3 implementation of the proposed corrective action to raise the building design inleak .

l

! Item Operatine 2: and Breach Exhaustof Enclosure Damner 2-AC-11 Buildine Onenedh Inteerity Due to Excessive Neestive Press NNECo Response:

1 sufficiently low probability to be below regulatory signi I

alignments in conjunction with the damper single failure are not pan of the MP2 orig .

Parsons Comment:

In response to NNECo LER 94-040-02, the NRC issued an SER (attached to NU memo M dated 9/9/96) which evaluated the AC-11 single failure scenario. The NRC evaluation of ,

consequence of the AC-11 failure considered:

1. Infrequency of the subject operating mode, i

2. Timely mitigation of the unfiltered release via operator action,

3. Based on the design inleakage value of 2560 cfm, two EBFS fans (with a capacity o maintain fans, the required minimum negative pressure following operator action to shutdown the l, 4. negative Enclosure Building will perform as designed (implies that the building will not experiencc an exc pressure).

l The NRC concluded that a backfit was not required based on the above items.

Regarding the above factors:

L Parsons concurs that the NRC evaluated an operating frequency of approximately 600 is less than 10% of the plant operating time.

2.

Parsons concurs that operator action to shutdown the Main Exhaust Fans on a high radiation alarm appropriate.

3. The FSAR & LER stated inleakage value of 2560 cfm, combined with the 13,900 cfm fa l an excess fan capacity of about 11,000 cfm, which could make it possible to maintain the req pressure. However, these are not the actual plant conditions. The actualinleakage of 8700 can, com with a calculated fan capacity of 10,600 cfm, gives an excess fan capacity of approximat would not be sufficient to maintain the required negative pressure with AC-Il open. Thus, this basi

NRC's SER is negated by the actual plant conditions not matching the values stated in the LER condition would be further aggravated by degrading the building leakage integrity resulting from negative pressure experienced during a Sve-fan operating condition.

4.

In a five fan operating mode, the Enclosure Building will exhibit an excessive negative pressure i licensirg basis design inleakage rate. Thus, this basis for the NRC's SER is negated by condP. ions not matching the conditions stated in the LER.

I Page 13 of 15 DR 0027. DOC

\

i f

i - -.. . . - - -

i e NU responses to follow-up NRC questions,n thedoes .

e LER and notthe l ma The

1. Parsons concerns regarding the five fan operating scenario are as follows.

lI 2. An excessive negative building pressure results from a 5 fan operating configuration.

building leakage. Leak tight integrity further degrades due to the excess negat ncrease in 3

Operators shutdown the Main Exhaust Fans.

4.

l the EBFS cannot maintain the minimum -0.25 in.wg.. Thus .

pressureE Proposed Corrective Action:

NU proposes to:

1.

in the 1972 qualification testing, andChange the FSAR-stated . . .

building ermined 2.

Increase the building design inleakage value to 8700 cfm.

Parsons _ .nment:

1.

As statedtest, qualification in and issue I above, changing the building pressure to -9.75 in.wg. is not su 2.

With an increased Enclosure Building design inleakage value (to 8700 cfm), and will not be able to maintain the minimum required negative pressure.

This issue remains DISCREPANT at Significance Level 1.

Item 3: Two Fan Operatine Canacity below Desien in-leakane with Damner 2-AC-1 In One 11snu:

Parsons agrees that the installation ofdamper AC-130 by PDCR MP2-041-95 resolves vulnerability of 2-AC-1. SP 2609E appropriately tests the damper for its proper function.

3 FSAR section 6.7.4.1 discusses the two-fan flow capability as sufficient to mitigate an AC .

UIR 3367, the need to update FSAR Section 6.7.4.1 to p failure. Parsons concurs that this issue is "PREVIOUSLY IDENTIFIED" by NNECo.

Item b-iThe calculation 97-EBF-02000-M2, Rev 1 results also indicated that when two fans a Is '0,600 CFM and not 13,900 CFM as stated in the FSAR. Parsons considers this issue t IANTIFIED" results. by NNECo. NU acknowledged the need to change the FSAR to incorporate I

tem 4:

Onened Inability to Maintain Minimum Neentive hsure with One EBFS Fan Operatine and AC-1 P

v arsons agrees that the installation of damper AC-130 by PDCR MP2-041-95 resolves the single f t ulnerability of 2-AC-1. NU's initial response to DR-0027, M2-IRF-00481, provided addition hat shows that the setpoint for AC-130 is 0.40 in wg. The setpoint is correctly selected an AC-130 will not prematurely open in the event that 2-AC-1 fails in the open position tests the damper for its proper function. Parsons concurs that this item is NON-DISCREPANT

' Page t4 of15 DR-0027. DOC i

e

. . ~ . .

l Item 5: tilR-3129 Conclusions and C+rrective Action Several AR's were generated to address the issues raised in UIR 3129. t onal The ass

, calculations and testing. Licensing basis issues, and the safety significance of U

( the discovery results. complete date. They were subsequently identified -

in the ev i calculation Thus, this issue remains DISCREPANT as Significance Level 4.

b $. n kf akA Prenared: ER. Ram . Glaviano _

/ Y[-

Group Lead Date Reviewed:

ca3sk f'. A. Blocher Deputy Project Director Date Approved: 9/d D.L. Curry

// [

Project Dirbor Date Forwarded to NNECo, NEAC, and NRC:

Posted to WWW:

Date Date

(

FINAL RESOLtrnON Item 1: OPEN(Significance Level 3)

Item 2: OPEN(Significance Level 1)

Item 3:

PREVIOUSLYIDENTIFIED ltem 4: NON-DISCREPANT Item 5: OPEN (Significance Level 4)

A Bloc /2#

/

Deputy Project Director Date Page 15 of 15 DR 0027. DOC

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

. vo 2 , ~ m.,a. c au s. c m.,u es - -, i . e . mu wa ;u e v. or.Leocccuriv.elber.N s tw x:tJd.

PARSONS POWER GROUP INC. .

ICAVP MILLSTONE UNIT 2 2675 Morgantown Rsad, Reading, PA 1 9607 i

DISCREPANCY REPORT

\ -

(610) 855-2000

• FAX: (610) 855-2509 DR NUMBER: DR-0027 DR TTTLE:

Enclosure Bldg Filtration & Containment / Enclosure Bldg Purge System Design REVISION: 0-T ISSUE DATE: 10/8/97 ORIGINATING GROUP: 2 SIGNIFICANCE LEVEL: 1 l DISCREPANCY l

Backarr==d Operation of the Enclosure Building Filtration System (EBFS) is credited in the Loss of Coolant Accident

' . (LOCA) analyscs for the calculation of offsite dose. Mmnummg the Enclosure Building Filtration Region (EBFR) at a negative pressure ensures any containment penetration leakagc into the EBFR remains in the Enclosure Building for controlled release via the installed filtration system. Thus, the offsite dose is reduced due to filtration and elevated relcase via the Unit-1 stack.

4 System Desian Per FSAR Section 6.7.4.1, the EBFS can maintain the EBFR under a mimmum negative pressure of 0.25 in.wg with one fan operating. The capacity of onc fan is 9,000 CFM and the design in-leakage rate into the EBFR is 2560 CFM. Assuming the failure of the purge supply damper 2-AC-1 as the single failure, both EBFS fans are relied upon to operate in order to maintain the minimum pressure of-0.25 in. wg. in the EBFR. In order to handle the additional 8400 CFM ofin leakage through the open damper both fans are required to operate.

- The Enclosure Building is designed to a maximum negative pressure of 2.0 in.wg.

Item 1 Breach of N!asure Buildina Intenrity due to Excess Neantive P.g.ure (EBFS Fans Oceratino)

In the event of an emcrgency condition, an Enclosure Building Filtration Actuation System (EBFAS) signal will start the two EBFS fans F-25A & F-25B. The fans will run until shutdown by the plant operators.

Damper 2-AC 1 (EB air supply isolation damper) will close upon receipt of the emergency signal, if not in the normally closed position. FSAR Section 6.7.2.1 states that 2.0 in.wg. is the maximum differential pressure that the enclosure metal siding can sustain and still maintain its leak-tight characteristics. Since the EBFS does not have pressure control provisions to prevent exceeding the building maximum pressurc

' limit, a potential exists for breaching the integrity of the enclosure building when two fans operate with

damper 2 AC-1 closed.

/'

Page1ofg

-. h-0027. doc 4

l l

Attachment 2 ICAVP Response Form Response ID: M2-IRF-01767 RFI/RAI Number: NA' NA AR Number: NA DR Number: DR-0027 (Parsons) AR Number: 97024996 CR Number: M2-97-2294 M2-DRT-00027(NU)

System Number /Name or Program ID/Name:

2314G/EBFS

Subject:

Request for replacement copy of document (s) previously transmitted by Northeast Utilities to Parsons.

This discrepancy is rated as Significance Level I by Parsons.

Background:

DR-0027 response is for replacement of Response ID: M2-lRF-01635 O continuation Disposition:

Replacement copy required.

] Continuation

Conclusion:

Northeast Utilities is transmitting a replacement copy to Parsons of M2-IRF-01635 and any applicable documents for response to DR-0027. Transmitting M2-EV-98-0095 Rev. O Attachment 3 Page 6 - 21, and Attachment 4,5, and 6.

C Continuation Preparer: M, d,c;((#Adpugg Date: sf/J ff D. L. Goodman Independent Reviewer: Date:

NA Unit Lead Review: hM[ hi N Mattioli Date: f//2-/98 Technical Review Team Concurrence: / ///b Date:

I G. Pitman NU ICAVP Project Manager Approval: j Date:

/ ff

p. Fougere Page 1 of 1 i

w, vo :. .o.e: mma re:ws exd.Cc- A u- S -h a: iveo:xce:cuouumeeo mw : m item 2* Breach of Enclosure Building Intestnty Due to Excessive Nerative Pressure (Main Enhaust Fans OrwMina and Exhaust Damner 2-AC-11 Onenedi If a CIAS occurs while purging the Enclosure Building, the purge supply fan F-23 and damper AC-1 are automatically stopped and closed, respecuvely. The Enclosure Building purge exhaust damper AC-8 remains open. He main exhaust fans will continue to operate and, if damper 2-AC 11 fails open, draw air from the Enclosure Building until the fans are turned offmanually following a Unit 2 Stack high radiation.

alarm. He EBFS is also activated automatically and both fans operate.

Damper AC 1 is a pneumatic damper, he sudden closure of this damper while the main exhaust fans are exhausting air from the Enclosure Building, could cause a sudden increase in negauve pressure in the building. The design exhaust rate from the building is 32,000 CFM and the opersong pressure in the main exhaust plenum is - 3.5 in.wg (Dwg 25203-26057). His pressure is significantly higher than the -2.0 in.wg maxmnun pressure limit for the building. i I

When the main exhaust fans are operating, the air exhausted from the Enclosure Building (via exhaust damper AC-l1) is nuxed with exhaust air from the other buildings prior to discharge to the Unit 2 stack.

j It is possible that the main exhaust fans will continue to operate together with the EBFS Fans. The negative '

pmesure induced by the main exhaust fans in the building is a back pressure to the EBFS fans and will cause the EBFS fans to operate to the left of the their combined fan curve, thus, increasing the building negative pressure.

An analysis of this potential breaching of Enclosure Budding leak-tightness integrity does not exist.

NRC Safety Evaluation Attached to MP2-DE-96-0485 (Reference I.1) addressed the failure of non-safbty ,

damper AC-l 1 from the perspective ofreleases via the main exhaust path. NNECo committed to perform '

cenain operator actions to shutdown this relcase path following receiot of a high radiation signal.

However, the Safety Evaluation did not consider the potential for excessive negative pressure in the Enclosum Building due to damper AC-11 remaining in the open position. Ec closure ofdamper AC-11 may be necessary to ensure Enclosure Building integrity.

Item 3 Two Fan Ooereno Canacity below Desinn in-leakane with Damoer 2 AC-1 In Ooen Position-Consider the case for wi.ich a LOCA occurs with damper 2-AC-1 in the open position. In addition, consider the single failure of damper 2-AC-1 as failing in the open position. For this scenario, the design building in-leakage is 2560 + 8400 = 10,960 CFM. Since the in-Icakage exceeds the design capacity of one fan (9000 cfm) it is concluded that both fans must operate to achieve the design 0.25 in.wg negative pressure. Using FSAR Figurc 6.7-3 to estimate the two fan operating capacity indicates that the system may not be capable ofhandling the in-leakage. Thus, the design minimum -0.25 in. wg building pressure may not be achieved.

Item 4: Igt . Maintain Minimum Nenative Pressure with One EBFS Fan Ooerating and AC-1 Onened According to the NRC Safety Evaluation Attached to MP2-DE-96-0485 (Reference 1,1):

"...In the event of a LOCA or MSLB during purging, with failure of actuation signal CH 1-CIAS, damper AC-1 would fail to close and fan F-25A would fail to start. This combination oflack of isolation (AC-1) and reduced filtered exhaust capability (loss of one train of exhaust / cleanup) would prevent the secondary containment from functioning properly as a fission pmduct cleanup system for primary containment leakage as the single operating F-25 fan would not have sufficient capacity to establish and maintain the necessary negative pressure in the unisolated Enclosure Building..

Page 2 of 3

. .-0027. doc

w ;; 4.,o - 5 5 ; W

-nu A w av me aw ou ::u[ ~ ' Choo[c==mooo% i ebe i

. Although the licensee claims that corrective action is not required by the original licensing basis, a modification has been proposed to eliminate the AC 1 vulnerability. A gravity damper would be installed as shown in the drawing. It would be weighted such that operation of purge fan F 23

[ opens it, but a -0.25 wg. vacuum due to operation of an EBFS fan would not cause it to open.

This action would elimmate the AC-1 single fadure condition."

i l

According to the NOTE 2 under 4.1.13 of SP 2609E. "Approumately 5 pounds of force applied to F-23 suction damper, AC-130, counter6 wei ht lever is sufficient to open the damper." The damper size is 47" x 47" For conservatism, use half of the damper area as the effective area that is subjected to a differential pressure. Thus, a pressure differential of about 0.13 in. wg. will open the gravity damper. According to drawing 25203-26057, the static pressure in the vicinity of AC 130 ( pressure point (3) is -0.1 I in.wg.

Test data (attached to Reference B.13) shows that one EBFS fan operaung can create higher negative pressures than 0.25 in, wg. (-0.35 to -0.75 in.wg).

in addition, during contamment purge using the EBFS containment cleanup flow path, AC 1 is manually I opened, but the purge supply fan must not be started (refer to 4.1.12 of SP 2314B). nus, AC-130 is  ;

I relied upon to open by the differential pressure created by the EBFS fan for makeup air durmg containment  !

cleanup.

The above contradicts the AC-130 performance requirement as stated in the AC-1 resolution. AC-130 will open at less than the design negative pressure of 0.25 in. wg, with the purge supply fan F-23 shutdown.

Calculation Review To resolve items I and 3 above, the calculations noted under Reference B were reviewed. The review found a generic problem in that the calculations arc out of date and have several analytical probicms. A preliminary review of the UTRs for the system (Reference C) initially confumed the same finding.

However, the UIRs were not specific enough to indicated that items 1 and 3 above will be corrected.

t

( Reonest for Additional InArmation

RAI-468 was issued on 09/17/97 requesting the following

1. Copies of the latest completed test procedures, performance data and operational data used to measure building pressure, fan flow rate, and building in-lcaknge rate for the following conditions:

a. One fan operating with purge supply damper 2-AC-1 in the open and close positions.

b. Two fans operating with purge supply damper 2- AC-1 in the open and close positions.

2. De basis for in-leakage rate of 2560 CFM referenced in FSAR Section 6.7.4.1.

3. Documentation which identifies the design features used to prevent the EBFR from exceeding the maximum allowed 2.0 in. wg. negative pressure.

4. Up-dated calculation or other documentation that determines the system capacity with one fan operating and two fans operating.

De purpose of the request was to determine:

The as-built leak-tightness of the building and the air in-leakage rate.

ne capacity of the system when two fans are operating and damper 2 AC-1 is opened.

4 If the pressure in the building does not excced the maximum allowed of-2.0 in, wg. when two fans are operating and damper 2-AC-1 is closed.

Page 3 of 8

. . A)027. doc

_ _ _ _ .- m __ . _. _ _ _ _ , . , _ _ , _ ,_ _, ._ ,

St.,i-ee- W ,1e:41 e r<Lrt:M4<SCNS eGtWICM e10-d55-2569 IC:W5268520fA0018&3427 PAGE:W t

o' if the CMP had r4~4 the potennal problem of breachin6 the mtegnty of die butiding when two fans are operat og with damper 2 AC-1 closed.

l g NU responded to the request on 09/26/97 by providing copies of the following:

Memo MP2 DE-96-0485 (Reference G.1)

=

UIRs 3129,3171,956,2224,984, and 971

=

SP 2609A, B, C, D E (Reference E )

Evaluation of R AI-468 Inennation '

l l

The review of the above Survallance Procedures (SPs) indicated the system is tested with only one fan operating. The building negative pressure and the time to achieve the design prcasure are the only parameters tested and recorded.

The SPs do not test the system with two fans operating, hs, the capability of the system to achieve and l maintain the building minimum design pressure, when damper AC-1 is in the open position, has not been verified. Assurance that the building integrity pressure limit is not exceeded when the damper is in the closed position has not been demonstrated U1Rs 3171,956,2224,984, and 971 indicated that the CMP has recognized the need to update the existing calculations. The UIRs, however, are not sufficiently detailed to indicate that the specific discrepancies are recogruzed.

Item 5: UIR-3 l29 Conclusions and Corrective Action l

The CMP via UIR 3129 recognized the need for a new analysis / calculation to provide the system operating curve and operating procedure to test both fans operating simultaneously. The UIR Recommended l . Disposition Details are repeated below:

i "1.

AR 9701%18-01 is written for CMP to evaluate need for additional surveillance or test based on AR 97019618-03 analysis results.

l 2. AR 97019618-02 is written for CMP to verify that EBFR siding to sustain and still maintain leak-tight charactaistics at upper limit of 2 inch of W.G. negative pressure.

3 AR 97019618-03 is written for CMP to generate new calculation showing system operating curve with one fan operating and also, when two (2) fans operating in parallel. Upon completion evaluate for AR 970196I8-01."

1 The UIR Final Disposition is repeated below:

" Expert Panel: AGREES with Recomrnended Disposition Details.

AR 97019618 01, CMP to evaluate need for additional surveillance or test based on AR 97019618-03 analysis results. AR 97019618-02, CMP to verify that EBFR siding to sustain and maintain leak-tight characteristics at 2 inch of W.G. AR 97019618-03, CMP to generate new calculation showmg system operating curve with one fan and also, when two (2) fans operating in parallel. Upon cornplesion evaluate for AR 97019618-01."

L lt is clear from the above that damper 2-AC-1 concerns were not recognized by the UIR. The need for a j test is subject to the analysis results and not mandated. Mandated testing is required since, at best, an analysis is subjective for the EBFS. Testing for in-leakage rate was not addressed, therefore, degradation

of the building leak-tightness characteristics can not bc monitored.

4 Page 4 of 8 l .,4 0037. doc l I

_,i

_ _ _ _ .. . --~~ ~ ^ ~ " ^^

. L ..cm m w c .

e m es ,i io,,4 . ^i ivwe m ess csun iuse m ee:cueuuu mea 437 Pc.GE:06 l The UIR (scction 1, item 1) states "...two (2) fans in operation must be capable of mamr l pressure in the EBFR Icss than the upper linut of 2 frches W.G/* nc UIR (section 2, item 1) states

[ " System performance calculation for the fans are inaaequate." UIR section 2, item 4 states "Th!

procedure to test these two fans operating simultancously and record a maximum negative pressure i

developed in the EBFR."

j Maintauung the structural integrity of the EB is essential in orrier to take credit for the filtered, eleva release path used to meet 10CFR100 release limits. However, the system design calculat program are inadequate to demonstrate that the system meets its design requirement. However, the UIR .

states as conclusion 1 (Section 2), "nis UIR has been determined not to require a CR and has not identified a potential safety significant condition." The Final Disposition Section of the UIR did not contradict this statement.

This conclusion is inconsistent with the information presented in the UTR.

NNECo UlR 3129 did not recognize the potential safety significance of the consequence of brea Enclosure Building integrity.

Basis for Significance Level 1:

One of the safety functions of EBFS is to collect and process potentially radioactive airborne particles and gasca in the EBFR following a LOCA and limit the site boundary radiation doses ts the 10CFR100 requirements. Due to the lack of supporting calculations and/or test procedures /results the ability of the EBFS to perform its primary safety function cannot bc assured.

Discrepancies identified may:

1 1.

Breach Enclosure Building integrity due to excess negative pressure (items 1 & 2), and 2.

Fail to maintain the minimum required negative pressure with design building in-lenkage (items 3 & 4).

( NNECo UlR-3129 did not recognize the potential safety significance of the consequence of breaching the Enclosure Building integrity.  !

4 4

i

• 0027. doc l

4

- -,. . - , . . - . -. . ~ -. . - - . . - - . .- ..

_~...,,.--.i.

vu-uo- to - . . enu.ur,-r=ure e x m w . om :5:

c:=:n:-

t0:m535a5t_0000019s0437 rcsI:e7 i

l

REFERENCES:

A - FSAR i

(- 1 L Section 5.3

~ Enclosure Building

2. Chapter 6 Engineered Safety Features Systems

3. Secuan 6.7 Enclosure Building Filtration System

4. Section 9.9.2 Canemmment and Enclosure Building Purge System -

5. ~ Figure 6.7-1 EBFR Negative Pressure vs. Time After DB1

6. Figure 6.7-3 Enclosure Building Filtration System Fan Performance Curve B; Calculations 1.-

IK2101,10/31n4, Enclosure Building Filtration System

2. IK21-03, R/14#3, EBFS Flow & Delta P Calculation

' 3. IK2104, R/23n3, EBFS P Drop Calculation

4. !KI2105,6/11/73, EBFS In-leakage .

5 IK21-06,6/11/73, In-leakage Calculations for Various Rooms

6. I K21-08, 4/6/71, Enclosure Building Filtraten System 7.

IK21-11, 3/24#1, Summary of Calculation of Air Flow through the EBFS Filter Units with Given Crack Areas

8. 1K21-14,12/02/69, EBFS Fan Selection

9. IK21-15, 8/21/69 - Enclosure Building Filtration System

10. I K21-17, 8/18/69, Enclosun: Building Filtration System

11. IK21-18,12/04/69, Pressure in EBFR Transient Conditions

12. N1;SCo Calc. XX XXX-10RA "EBFS Initiation Time E&ct on LOCA Dose" Rev. 01,12/05/78

13. 2-ENG-174,10/4/91, Air Flow through a 4" Hole from the Enclosure Building at Design Pressure C. .UIRs

-(

959 976 981 987 956 977 982 988 971 978 984' 3129 972 979 985 317I 974 980 986 D. ' Operating Procedures

1. OP 2314B Contamment and Enclosure Purge, Rev 16

2. OP 2314G Enclosure Building Filtration System, Rev i1 E. Surveillance Procedures 1, SP 2609 A, EBFS and Control Room Vent. Operability Test, Fac.1, Rev 12

2. SP 2609 B. EBFS and Control Room Vent. Operability Test, Fac. 2. Rev 14

3. ' SP 2609 C, Enclosure Building Integrity Verification, Rev 5 4.' SP 2609 D, Enclosure Building Filtration System Filter Testing Refuel, Rev 10 5; ~ SP 2609 E, Enclosure Building Filtration System Testing, Rev 6 F. RAls

1. RAl-0221,08/07/97

2. RAl-0415,09/0997

3. RAI-0468,09/17/97 Pagc6 ofR i ,.(-0027 doc

^ ^ - - - - - -- - - - ' ~ ' - ~ ~ ^ ~ ~^'^~

-C n, - =C u c m m

,,, wo ;, 2 e a c' ._v_,.

ve- etu 2:: c:uw. i u s erseeo:cweeeleme<w PNE:as G. Drawings:

25203 26d28, shl, Rev 30 25203-26028, sh5, Rev 15 25203-26028, sh2, Rev 35 25203-26028, sh4, Rev 7

{

25203-26028, sh3, Rev 10 25203-26059, Rev 1 25203 26057, Rev 0 25203-29640, Rev i H. Technical Specification

1. LCO and SR, 3/4.6.5, Secondary Containment l 2. Bases. 3/4.6.5, Secondary Containment I. Miscellaneous 1.

Memo MP2-DE-96-0485, AC-1 and 11 "MP2 Enclosure Building Secondary Containmen Single Failure Deficiencies" and NRC Memorandum (Carl Berlinger to Philip Mckee, Dated covering disposition of this apparent deficiency along with the safety evaluation.

2.

Millstone Inspection Report 97-02, June 24,1997, page 55 of 91: E8.2 (Closed) Unrcsolved item 50-336/95-25-03; Enclosurc Building Filtranon System Single Failure Vulnerability D. R. Ramos . - -

Tier-2 10/8/97 Orieineor .Qroug D,at,

(

Page 7 of 8

[ 0027. doc

(

. , ,so_;, t o . -,c -, u u r-w

vn w' .ce vr- eiU-cc-Et tfd TC:89526852OOOOO186045 P94GE:OS EVALUATION ~

M BASIS VALLD - O BASIS INVAT m . CLOSED

, 0 PREVIOUSLYIDENTIFIED

( BY NNECo - CLOSED oasis Valid l

l d 4,<

R. i. C vinni 10/8/97 Group Lead Date REVTEW AND APPROVAL Rsviewed: E.A. Blocher 10/8/97 De uty Proj'ect Director Date a/rb

( Approved

D.L. Curry 10/8/97 Project Director Date l

l i

i l-l I .1037. doc

--. _ - = - ._

/s/ P. D. Hinnekamp 9/30/97 97-535 Form Approved by Effectne Date SORC Mtg. No. j

, -. . ~ . CR Form ca so. .

l I W.

Dg

&r Initiation ~ %f-97" A;lW '

Sectioal:.iTo beicompie t'ed. by9 nitihtor (ple.as.e; type:or print): 4 We.%tnR g u Organization identifying condition: Discovery date: 10/8/97 Affected Unit (s): - System #: 2390C _. - . _

M2 ICAVP Response Discovery time: 1700 10 22 30 CO

.__ .l. Condition description (including how

~~

condition was discovered, organization creating condition, what activity was in progress

~'

- when event was discovered): ~_

DR-0027. A discrepancy report concerning the Enclosure Building Filtration System has been issued by the ICAVP contractor.

11a specific condition description is contained in the attached Parsons Discrepancy Report DR-0027. __

- According to the IUAVP contractor, this is a Significance Level I (highest) discrepancy report. A Level 1 discrepancy is assigned to ._

a discrepancy by Parsons when the system does not meet its licensing and design bases and cannot perform its intended function.

- m _ , . _ _ _ _,_ . _ _ , --~ ,

~~ ~ ~ ~~

~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~

T.~ Component identi?ficaIion tiu~rnbel:~~~ ~ ~ ~ ~ ~ ~ ~ ~ ~

~ -' - . _;.

Method of Discovery: Ext. Oversight-- -- -- .

~~ C Continuation Sheet O

2. Immediate corrective action taken g

The system engineer (Phil Bauman) has been notified by Audix.

CMP is currently reviewing whether this issue had been discovered during the P!-7 review of the EBFS system.

TR# ' AWO# - -Eng. Disp.# Continuation Sheet O .

3. Recommended corrective action This CR should be assigned to Engineering for evaluation and disposition.

(

~

- uf/- Continuation Sheet O 4 Initiator Name: Craig B,Swannerf/ // Time: fL. / p.35 Phone No.: 3432 Initiator's Signature: 6. Jh_~ -- . Date: Ihd9 Cost Control Center: 82B -.

v Initiator Re ollow p: YES

~ ~

Sup'e' Isor Na'm'e:' Joe Fougere

" ~

''''''~~~~" Tim'e[~"/N[~~~~'~~~~'~~~

Supervisor Signature: Date: /.f1/f1 Phone No: 5526 Section 2rcTo be cotupleh by Operability /Reportability Screening DesigneenimAaE/wA4N Does CR have an actual or potential effect on plant or personnel safety, operability, Notes:

1.

reportability, (e.g., NGP 2.25, EPIP 4400) or plant cperation?

O Yes er Don't Know (Section 3 required to be completed.)

O No Designee Date Time ifcontinuation sheets (RP 4-1, Page 7) are required identify the section being continued by section number.

j{

' ^

Form RP4-1 Rev.5 Page 1 of 7 Sheet I k

1 p we.. - _'

Attachment 2 ICAVP Response Form Response ID: M2-IRF-00481 RFI/RAI Number: N/A AR Number: N/A l DR Number: DR-0027 (Parsons) AR Number: 97024996 CR Number: M2-97-2294 l M2-DRT-00027 (NU)

System Number /Name or Program ID/Name: Tier 1 i

Subject:

Enclosure Building Filtration & Containment / Enclosure Building Purge System Design.

This discrepancy is rated as Significance Level I by Parsons.

Btckground: Operation of the Enclosure Building Filtration System (EBFS)is credited in the Loss of Coolant l Accident (LOCA) analyses for the calculation of offsite dose. Maintaining the Enclosure Building Filtration Region (EBFR) at a negative pressure ensures any containment penetration leakage into the EBFR remains in the Enclosure Building for controlled release via the installed filtration system. Thus, the offsite dose is reduced l due to filtration and elevated release via the Unit-1 stack. (continued on page 2 of 22)

Continuation l

l l Disposition: NU has concluded that this DR identifies issues involving a composite of design bases, equipment capability, operator intervention and startup and operational testing issues. As such, it is not appropriate to

{, comment on NU's characterization of the individual issues as being valid, invalid, pre-discovered or of a

particular Significance Level and the disposition of each of these without first providing a lead in discussion of l the EBFS, Enclosure Building isolation features and NU's committed design of the system, as follows

(continued on page 10 of 22)

Continuation

Conclusion:

NU has concluded that Item I has been previously discovered and is considered to be a l significance level 3 discrepancy, and postulated fan / damper scenarios in DR Items 2,3,4, and 5 are non-

! discrepant. A summary of the conclusion for each item is listed in the Conclusion Continuation. CR M2 !

2294 has been issued to provide any follow-on activities associated with this DR. (continued on page 22 of 22)

Continuation l Preparer: /

Ont# Date: /# '

7!f7 G. Komosk

)

l Independent Reviewer: Y. / _ m / te:w/nh7 Unit Lead Review: Date: 7/97 '

F. Mattioli / 17

. Technical Review Team Concurrence: A // //d n Date: IC/cU / 92 NU ICAVP Project Manager Approval Date: /d

\

pougerV ,

f NU CMP Director Approval: / CA Date: /0 'AT-f 7 l R. Necci Page 1 of 22

Attachment 3 ICAVP RFI/RAI Response Form Continuation Sheet

)

Response ID: M2-IRF-00481 RFI/RAINumber: N/A AR Number: N/A DR Number: DR-0027 (Parsons) AR Number: 97024996 CR Number: M2-97-2294 M2-DRT-00027 (NU)

Bickground Continuation-System Desien Per FSAR Section 6.7.4.1, the EBFS can maintain the EBFR under a minimum negttive pressure of i 0.25 in, w.g. with one fan operating. The capacity of one fan is 9,000 CFM and the design in-leakage rate into the EBFR is 2560 CFM. Assuming the failure of the purge supply damper 2-AC-1 as the single failure, both EBFS fans are relied upon to operate in order to maintain the minimum pressure of

-0.25 in. w.g. in the EBFR. In order to handle the additional 8400 CFM of in-leakage through the open damper both fans are required to operate.

The Enclosure Building is designed to a maximum negative pressure of 2.0 in.w.g.. l Item 1: Breach of Enclosure Building Integrity due to Excess Necative Pressure (EBFS Fans Operatina)

\

In the event of an emergency condition, an Enclosure Building Filtration Actuation System (EBFAS)

signal will start the two EBFS fans F-25A & F-25B. The fans will run until shutdown by the plant ,

operators. Damper 2-AC-1 (EB air supply isolation damper) will close upon receipt of the emergency signal,if not in the nomially closed position. FSAR Section 6.7.2.1 states that 2.0 in. w.g. is the maximum differential pressure that the enclosure metal siding can sustain and still maintain its ?eak-tight characteristics. Since the EBFS does not have pressure control provisions to prevent exceeding the ,

building maximum pressure limit, a potential exists for breaching the integrity of the enclosure building when two fans operate with damper 2-AC-1 closed.

Item 2: Breach of Enclosure Building Intecrity Due to Excessive Nezative Pressure (Main Exhaust Fans Operatine and Exhaust Damper 2-AC-11 Opened):

If a CIAS occurs while purging the Enclosure Building, the purge supply fan F-23 and damper AC-1 are automatically stopped and closed, respectively. The Enclosure Building purge exhaust damper AC-8 remains open. The main exhaust fans will continue to operate and,if damper 2-AC-11 fails open, draw air from the Enclosure Building until the fans are turned off manually following a Unit 2 Stack high radiation alarm. The EBFS is also activated automatically and both fans operate.

Damper AC-1 is a pneumatic damper. The sudden closure of this damper while the main exhaust fans are exhausting air from the Enclosure Building, could cause a sudden increase in negative pressure in the building. The design exhaust rate from the building is 32,000 CFM and the operating pressure in the main exhaust plenum is - 5.5 in. w.g. (Dwg 25203-26057). This pre sure is significantly higher than the

-2.0 in.wg maximum pressure limit for the building.

Page 2 of 22

Attachment 3 ICAVP

! RFI/RAI Response Form Continuation Sheet i(

Response ID: M2-IRF-00481 RFI/RAI Number: N/A AR Number: N/A DR Number: DR-0027 (Parsons) AR Number: 97024996 CR Number: M2-97-2294 M2-DRT-00027 (NU)

B ckground Continuation: Item 2: (continued)

When the main exhaust fans are operating, the air exhausted from the Enclosure Building (via exhaust damper AC-11)is mixed with exhaust air from the other buildings prior to discharge to the Unit 2 stack It is possible that the main exhaust fans will continue to operate together with the EBFS fans. The negative pressure induced by the main exhaust fans in the building is a back pressure to the EBFS fans and will cause the EBFS fans to operate to the left of the their combined fan curve, thus, increasing the building negative pressure.

An analysis of this potential breaching of Enclosure Building leak-tightness integrity does not exist.

NRC Safety Evalcion Attached to MP2-DE-96-0485 (Reference I.1) addressed the failure of non-safety damper AC-11 from the perspective of releases via the main exhaust path. NNECo comtr"ted to perform certain operator actions to shutdown this release path following receipt of a high radiation

signal. However, the Safety Evaluation did not consider the potential for excessive negative pressure in the Enclosure Building due to damper AC-11 remaining in the open position. The closure of damper AC-11 may be necessary to ensure Enclosure Building integrity.

Item 3: Two Fan Operatine Capacity below Desien In-leakage with Damner 2-AC-1 In Open Position:

Consider the case for which a LOCA occurs with damper 2-AC-1 in the open position. In addition, consider the single failure of damper 2-AC-1 as failing in the open position. For this scenario, the design building in-leakage is 2560 + 8400 = 10,960 CFM. Since the in-leakage exceeds the design capacity of one fan (9000 cfm) it is concluded that both fans must operate to achieve the design 0.25 in.wg negative pressure. Using FSAR Figure 6.7-3 to estimate the two fan operating capacity indicates that the system may not be capable of handling the in-leakage. Thus, the design minimum -0.25 in. wg building pressure may not be achieved.

Item 4: Inability to Maintain Minimum Negative Pressure with One EBFS Fan Operatine and AC-1 Opened:

According to the NRC Safety Evaluation Attached to MP2-DE-96-0485 (Reference I.1):

"...In the event of a LOCA or MSLD during purging, with failure of actuation signal CH1-CIAS, damper AC-1 would fail to close and fan F-25A would fail to start. This combination of lack of isolation (AC-1) l and reduced filtered exhaust capability (loss of one train cf exhaust / cleanup) would prevent the j secondary containment from functioning properly as a fission product cleanup system for primary l containment leakage as the single operating F-25 fan would not have sufficient capacity to establish and l maintain the necessary negative pressure in the unisolated Enclosure Building..

Page 3 of 22

l Attachment 3 ICAVP RFI/RAI Response Form Continuation Sheet

!I Response ID: M2 IRF-00481 1 RFI/RAI Number: N/A AR Number: N/A DR Number: DR-0027 (Parsons) AR Number: 97024996 CR Number: M2-97-2294 M2-DRT-00027 (NU)

B ckground Continuation: Item 4: (continued)

... Although the licensee claims that corrective action is not required by the original licensing basis, a modification has been proposed to eliminate the AC-1 vulnerability. A gravity damper would be installed as shown in the drawing. It would be weighted such that operation of purge fan F-23 opens it, but a -0.25 wg. vacuum due to operation of an EBFS fan would not cause it to open. This action would eliminate the AC-1 single failure condition."

According to the NOTE 2 under 4.1.13 of SP 2609E, "Approximately 5 pounds of force applied to F-23 suction damper, AC-130, counter weight lever is sufficient to opu the damper." The damper size is 47" x 47". For conservatism, use half of the damper area as the effective area that is subjected to a differential pressure. Thus, a pressure differential of about 0.13 in wg. will open the gravity damper.

According to drawing 25203-26057, the static pressure in the vicinity of AC-130 ( pressure point 13) is -

0.11 m.wg. Test data (attached to Reference B.13) shows that one EBFS fan operating can create higher l negative pressures than 0.25 in wg. (-0.35 to -0.75 in.wg).

In addition, during containment purge using the EBFS containment cleanup flow path, AC-1 is manually opened, but the purge supply fan must not be started (refer to 4.1.12 of SP 2314B). Thus, AC-130 is relied upon to open by the differential pressure created by the EBFS fan for makeup air during containment cleanup.

The above contradicts the AC-130 performance requirement as stated in the AC-1 resolution. AC-130 will open at less than the design negative pressure of 0.25 in. wg, with the purge supply fan F-23 shutdown.

Calculation Review To resolve items I and 3 above, the calculations noted under Reference B were reviewed. The review found a generic problem in that the calculations are out of date and have several analytical problems. A preliminary review of the UIRs for the system (Reference C) initially confirmed the same finding.

However, the UIRs were not specific enough to indicated that items 1 and 3 above will be corrected.

i l

l Page 4 of 22

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

l l Att chm:nt 3 ICAVP RFI/RAI Response Form Continuation Sheet Response ID: M2 IRF-00481 RFI/RAI Number: N/A AR Number: N/A DR Number: DR-0027 (Parsons) AR Number: 97024996 CR Number: M2-97-2294 M2-DRT-00027 (NU)

B:ckground Continuation: Item 4: (continued)

Reauest for AdditionalInformation RAI-468 was issued on 09/17/97 requesting the following:

1. Copies of the latest completed test procedures, performance data and operational data used to measure building pressure, fan flow rate, and building in-leakage rate for the following conditions:

a. One fan operating with purge supply damper 2-AC-1 in the open and close positions.

b. Two fans operating with purge supply damper 2- AC-1 in the open and close positions.

2. The basis for in-leakage rate of 2560 CFM referenced in FSAR Section 6.7.4.1.

3. Documentation which identifies the design features used to prevent the EBFR from exceeding

( the maximum allowed 2.0 in, wg. negative pressure.  ;

4. Up-dated calculation or other documentation that determines the system capacity with one fan operating and two fans operating.

The purpose of the request was to determine:

The as-built leak-tightness of the building and the air in leakage rate.

The capacity of the system when two fans are operating and damper 2-AC-1 is opened.

If the pressure in the building does not exceed the maximum allowed of-2.0 in. wg. when two fans are operating and damper 2-AC-1 is closed.

If the CMP had recognized the potential problem of breaching the integrity of the building when two fans are operating with damper 2-AC-1 closed.

NU responded to the request on 09/26/97 by providing copies of the following:

Memo MP2-DE-96-0485 (Reference G.1)

UIRs 3129,3171,956,2224,984, and 971 SP 2609A, B, C, D, E (Reference E )

l t l

Page 5 of 22

- - = - - k *y n --\ 5, -e-J s

. =.= =

l Attachment 3 ICAVP

' RFI/RAI Response Form Continuation Sheet Response ID: M2-IRF-00481 RFI/RAI Number: N/A AR Number: N/A DR Number: DR-0027 (Parsons) AR Number: 97024996 CR Number: M2-97-2294 M2-DRT-00027 (NU)

Bickground Continuation: Item 4: (continued)

Evaluation of RAI-468 Information The review of the above Surveillance Procedures (SPs) indicated the system is tested with only one fan operating. The building negative pressure a.M the time to achieve the design pressure are the only parameters tested and recorded.

The SPs do not test the system with two fans operating. Thus, the capability of the system to achieve and maintain the building minimum design pressure, when damper AC-1 is in the open position, has not been verified. Assurance that the building integrity pressure limit is not exceeded when the damper is in the closed position has not been demonstrated.

UIRs 3171,950,2221,984, and 971 indicated that the CMP has recognized the need to update the existing calculations. The UIRs, however, are not sufficiently detailed to indicate that the specific discrepancies are recognized.

Item 5: UIR-3129 Conclusions and Corrective Action The CMP via UIR 3129 recognized the need for a new analysis / calculation to provide the system operating cur and operating procedure to test both fans operating simultaneously. The UIR Recommended t,.sposition Details are repeated below:

"1. AR 97019618-01 is written for CMP to evaluate need for additional surveillance or test based on AR 97019618-03 analysis results.

2. AR 97019618-02 is written for CMP to verify that EBFR siding to sustain and still maintain leak tight characteristics at upper limit of 2 inch of w.g. negative pressure.

3. AR 97019618-03 is written for CMP to generate new calculation showing system operating curve with one fan operating and also, when two (2) fans operating in parallel.

Upon completion evaluate for AR 97019618-01."

l l

l Page 6 of 22

- - . ~ ~ . -

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

i  !

\

1 l

J Attachment 3 {

ICAVP i RFI/RAI Response Form Continuation Sheet

(-

Response ID: M2 IRF-00481 RFI/RAINumber: N/A AR Number: N/A DR Number: DR-0027 (Parsons) AR Number: 97024996 CR Number: M2-97-2294 M2-DRT-00027 (NU)

BIckground Continuation: Item 5: (continued)

The UIR Final Disposition is repeated below:

" Expert Panel: AGREES with Recommended Disposition Details.

AR 97019618-01, CMP to evaluate need for additional surveillance or test based on AR 97019618-03 analysis results. AR 97019618-02, CMP to verify that EBFR siding to sustain and maintain leak-tight characteristics at 2 inch of w.g. AR 97019618-03, CMP to generate new calculation showing system operating curve with one fan and also, when two (2) fans operating in parallel. Upon completion evaluate for AR 97019618-01."

It is clear from the above that damper 2-AC-1 concerns were not recognized by the UIR. The need for a

{

test is subject to the analysis results and not mandated. Mandated testing is required since, at best, an '

analysis is subjective for the EBFS. Testing for in-leakage rate was not addressed, therefore, degradation j of the building leak-tightness characteristics can not be monitored.

The UIR (section 1, item 1) states "...two (2) fans in operation must be capable of maintaining a negative pressure in the EBFR less than the upper limit of 2 inches w.g." The UIR (section 2, item 1) states

" System performance calculation for the fans are inadequate." UIR section 2, item 4 states "There is no  ;

. procedure to test these two fans operating simultaneously and record a maximum negative pressure I developed in the EBFR."

Maintaining the structural integrity of the EB is essential in order to take credit for the filtered, elevated release path used to meet 10CFR100 release limits. However, the system design calculations and testing i program are inadequate to demonstrate that the system meets its design requirement. However, the UIR  !

states as conclusion 1 (Section 2), "This UIR has been determined not to require a CR and has not identified a potential safety significant condition." The Final Disposition Section of the UIR did not contradict this statement. This conclusion is inconsistent with the information presented in the UIR.

NNECo UIR-3129 did not recognize the potential safety significance of the consequence of breaching the Enclosure Building integrity.

Basis for Significance Level 1:

One of the safety functions of EBFS is to collect and process potentially radioactive airborne particles and gases in the EBFR following a LOCA and limit the site boundary radiation doses to the 10CFR100 requirements. Due to the lack of suppot.ing calculations and/or test procedures /results the ability of the L EBFS to perform its primary safety function cannot be assured.

Page 7 of 22

K Att:chment 3 ICAVP

[

RFI/RAI Response Form Continuation Sheet {!

( Response ID: M2 IRF-00481 RFURAI Number: N/A AR Number: N/A DR Number: DR-0027 (Parsons) AR Number: 97024996 CR Number: M2-97-2294 M2-DRT-00027 (NU)

Bickground Continuation: Item 5: (continued)

%0 Discrepancies identified may:

1.

Breach Enclosure Building integrity due to excess negative pressure (items 1 & 2), and 2.

Fail to maintain the minimum required negative pressure with design building in-leakage '

(items 3 & 4).

NNECo UIR-3129 did not recognize the potential safety significance of the consequence of breaching the Enclosure Building integrity.

REFERENrES:

A. FSAR

1. Section 5.3 Enclosure Building

( 2. Chapter 6 Engineered Safety Features Systems

3. Section 6.7 Enclosure Building Filtration System

4. Section 9.9.2 Containment and Enclosure Building Purge System

5. Figure 6.7-1 EBFR Negative Pressure vs. Time After DBI

6. Figure 6.7-3 Enclosure Building Filtration System Fan Performance Curve B. Calculations

1. IK21-01,10/31n4, Enclosure Building Filtration System

2. IK21-03,8/1493, EBFS Flow & Delta P Calculation 3.1K21-04,8/23n3, EBFS P Drop Calculation

4. IKI21-05,6/11D3, EBFS In-leakage

5. IK21-06,6/1In3,In-leakage Calculations for Various Rooms

6. IK21-08,4/691, Enclosure Building Filtration System

7. IK21-11, 3/2401, Summary of Calculation of Air Flow through the EBFS Filter Units with Given Crack Areas

8. IK21-14,12/02/69, EBFS Fan Selection 9.1K21-15,8/21/69 - Enclosure Building Filtration System

10. IK21-17,8/18/69, Enclosure Building Filtration System

, 11. IK21-18,12/04/69, Pressure in EBFR Transient Conditions

12. NUSCo Calc. XX-XXX-10RA "EBFS Initiation Time Effect on LOCA Dose" Rev. 01, 12/05n8

,i

13. 2-ENG-174,10/4/91, Air Flow through a 4" Hole from the Enclosure Building at Design Pressure Page 8 of 22

Attachment 3 ICAVP '

RFI/RAI Response Form Continuation Sheet Response ID: M2-IRF-00481 RFI/RAINumber: N/A AR Number: N/A DR Number: DR-0027 (Parsons) AR Number: 97024996 CR Number: M2-97-2294 M2-DRT-00027 (NU)

Background Continuation:

REFERENCES:

(continued)

C. UIRs 959 976 981 987 956 977 982 988 971 978 984 3129 972 979 985 3171 974 980 986 l

D. Operating Procedures

1. OP 2314B Containment and Enclosure Purge, Rev 16

2. OP 2314G Enclosure Building Filtration System, Rev 11

( E. Surveillance Procedures

1. SP 2609 A, EBFS and Control Room Vent. Operability Test, Fac.1, Rev 12 l

2. SP 2609 B, EBFS and Control Room Vent. Operability Test, Fac. 2, Rev 14

3. SP 2609 C, Enclosure Building Integrity Verification, Rev 5

4. SP 2609 D, Enclosure Building Filtration System Filter Testing-Refuel, Rev 10

{

5. SP 2609 E, Enclosure Building Filtration Sysum Testing, Rev 6 F. RAls

1. RAI-0221,08/07/97

2. RAI-0415,09/0997

3. RAI-0468,09/17/97 G. Drawings 25203-26028, sh1, Rev 30 25203-26028, sh5, Rev 15 25203-26028, sh4, Rev 7 25203-26028, sh2, Rev 35 25203-26059, Rev 1 25203-26057, Rev 0 25203-26028, sh3, Rev 10 25203-29640, Rev i 4 H. Technical Specification

. l. LCO and SR,3/4.6.5, Secondary Containment

2. Bases,3/4.6.5, Secondary Containmeat Page 9 of 22

=.

Att chment 3 l

ICAVP l RFI/RAI Response Form Continuation Sheet

('

Response ID: M2-IRF-00481 RFI/RAI Number: N/A AR Number: N/A DR Number: DR-0027 (Parsons) AR Number: 97024996 CR Number: M2-97-2294 M2-DRT-00027 (NU)

B ckground Continuation:

REFERENCES:

(continued)

I. Miscellaneous

1. Memo MP2-DE-96-0485, AC-1 and 11 "MP2 Enclosure Building Secondary Containment Integrity Single Failure Deficiencies" and NRC Memorandum (Carl Berlinger to Philip Mckee, Dated 3/28/96) covering disposition of this apparent deficiency along with the safety evaluation.

2. Millstone Inspection Report 97-02, June 24,1997, page 55 of 91: E8.2 (Closed) Unresolved Item 50-336/95-25-03; Enclosure Building Filtration System Single Failure Vulnerability Disposition Continuation:

Throughout the Nuclear Industry, the general topic of Secondary Containment Drawdown, Filtration and Release systems (EBFS in Millstone 2's case) is one which is tempered by a degree of practical implication which stands in the way of achievement of full redundancy and single active failure

( compliance. While the fans, filters, and, to a degree, the dampers involved, lend themselves to conventional redundancy via a two train duplication of equipment, the buildings which are drawn down by these systems do not. This is due, in part, to the practical matter of personnel passageways (doors) which are not of an airlock design and which are operated by non-licensed personnel. The design of the secondary containment is a hybrid in many respects, possessing minimal resistance to the effects of extemal phenomena which conventional safety related structures must be able to withstand. It is rationalized that the secondary containment affords an additional degree of protection; but not one of such significant impact that it warrants the normal full pedigree of design features.

Since plant startup in 1975, there have been inconsistent interpretations of the design basis for the Containment and Enclosure Building Purge System (CEBPS), the Enclosure Building Filtration System (EBFS) and the Enclosure Building (EB). A series of NRC (AEC) questions and NU responses reflecting the position of the EBFS, CEBPS, and EB were provided during the plant operating license process in 1973-1974. Communications continued in 1977-1981 on the subject of the seismic qualification of the EB and the associated penetrations. In 1994, an LER 94-040," Ventilation Enclosure Building Integrity," was initiated and a final position was provided relating to the license basis for the secondary containment and associated isolation and filtration systems.

This confusion is documented in the LER wherein NU describes two conditions in which the design does not satisfy single failure criteria. These are:

l - The failure of damper AC-1 to close upon receipt of a CIAS signal and the resultant inability to

'q achieve the required level of building drawdown, and,

- The failure of damper AC-11 to close upon receipt of a CIAS signal and the resultant release of radioactive gases via an unfiltered release path.

Page 10 of 22 1

Attachment 3 ICAVP RFI/RAI Response Form Continuation Sheet i

Response ID: M2 IRF-00481 RFI/RAI Number: N/A AR Number: N/A DR Number: DR-0027 (Parsons) AR Number: 97024996 CR Number: M2-97-2294 M2-DRT-00027 (NU)

Disposition Continuation:

The CEBPS functions to maintain a suitable environment in the EB during all non-design basis accident modes of operation. The Main Exhaust System (MES) provides the exhaust pathway, including the exhaust fans, for the CEBPS. The CEBPS is normally not in use during plant operation except when it is necessary to improve the environment in the EB and does not provide accident mitigation functions.

The 1973-1974 correspondence provided the license basis for the CEBPS and MES as non-QA and non-Seismic. In 1977, the CEBPS isolation dampers AC-1 (supply) and AC-11 (exhaust), including control circuits were upgraded to QA status.

The EBFS functions to collect and process potential containment leakage, to minimize radioactivity levels resulting from all sources of containment leakage in the event of a Loss Of Coolant Accident (LOCA). The EBFS is designed to maintain the Enclosure Building Filtration Region (EBFR) under a minimum negative pressure of 0.25 in w.g. with one fan operating. Both EBFS subsystems operating in parallel will not exceed the enclosure building design pressure of-2.0 in w.g.. NRC Safety Evaluation dated 5/10/74 concluded that the proposed design of the EBFS meets the intent of the GDCs 41,42,43, and 64.

The EBFR integrity is maintained during a LOCA by isolation of nonsafety-related ventilation systems that communicate with the EBFR and the design of the EB. The 1973-1974 correspondence provided a single failure evaluation of CEBPS isolation dampers AC-1 and AC-11. Dampers AC-1 and AC-11 receive a Containment Isolation Actuation Signal (CIAS) to close in order to isolate the EBFR. The correspondence concluded that the EBFR integrity will be maintained in the event of a failure of AC-1 or AC-11 following a LOCA. The 1979-1981 correspondence provided the license basis of the EB and the associated penetrations. The committed license basis does not require the EB to be designed to be functional subsequent to a Safe Shutdown Earthquake (SSE).

In 1994, potential single failure deficiencies with respect to AC-1 and AC-11 were identified in LER 2-94-040. Upon further evaluation, NU determined that single failure capability in all respects was not a committed design feature for the composite EBFS/ Enclosure Building system. The result of this round of submittals was NU's voluntary upgrade of the AC-1 design feature to include a weighted damper in series with AC-1 and the NRC's acceptance of operator action upon receipt of a stack monitor high rad alarm signal to manually terminate the unfiltered release in the case of AC-1l's failure. Low risk due mainly to low probability of occurrence coupled with decisive operator actions were factors leading to the NRC's concurrence with NU's position. The AC-1 single failure event was resolved with the implementation of Plant Design Change (PDCR) 2-041-95," Containment and Enclosure Building Purge System Damper Modifications". The PDCR installed a counterbalanced gravity damper in the CEBPS supply duct to provide redundancy for damper AC-1. The modification was a system upgrade to mitigate the postulated single failure of the Facility 1 CIAS signal to AC-1.

Page 11 of 22

n Attachment 3 ICAVP

! RFI/RAI Response Form Continuation Sheet l

(

Response ID: M2 IRF-00481 RFI/RAI Number: N/A AR Number: N/A DR Number: DR-0027 (Parsons) AR Number: 97024996 CR Number: M2-97-2294 M2-DRT-00027 (NU)

Disposition Continuation:

The evaluations by NU and the NRC in response to LER 94 ^40 for the single failure vulnerability of AC-11 determined that correction of the AC-11 vulnerability is not required. The determination was based on the low probability of the event and operator action to isolate the radiological release path. The DR-identified issues relating to the failure of damper AC-11 to isolate and the resultant impact on EB integrity are discussed in detail with the appropriate DR Items below.

ITEM 1: The Breach of Enclosure Building Integrity due to Excess Negative Pressure with Damper AC-1 Closed and Two EBFS Fans Operating.

When an Enclosure Building Filtration Actuation System (EBFAS) actuation signalis received, the two Enclosure Building Filtration System (EBFS) fans, F-25A/B, will automatically start and run. The enclosure building air supply isolation damper, AC-1, and the enclosure building main exhaust isolation

( damper, AC-11, will both close on receipt of the emergency signal. No other flow paths would be open 3 to the enclosure building during this type of event, hence, the two EBFS fans would draw the pressure down in the enclosure building. This is the expected response for the system.

The DR postula:es that because there is no analysis or surveillance testing to support operation of two EBFS fans in parallel, that the maximum design pressure for the enclosure building,-2.0 in, w.g., could be exceeded.

During the PI-7 Graded System Review for the EBFS, UIR No 3129 was generated and identified that the supporting calculations and the testing of the operating conditions with the two EBFS fans operating in parallel were inadequate. AR 97019618, with assignments 01,02, and 03 was initiated to track the corrective action. These AR assignments will track calculations and work activities to document the negative pressure within the enclosure building with one EBFS fan operating and with two EBFS fans operating in parallel, verify EB integrity and determine the need for additional surveillance or system testing. Calculation 97EBF-02000-M2 (in final preparation) was generated to determine the negative pressure in the EB with 2 EBFS fans operating in parallel in response to UIR 3129. In addition, Millstone Unit 2 Pre-Operational Test T2314GP,Rev.1,7/19/75, resulted in a pressure of approximately

-0.7 in, w.g. in the enclosure building with 2 EBFS fans operating. The calculation currently supports the test data with a calculated pressure of -0.6 in. w.g..

The pre-operational testing and results of the "in-preparation" calculation, that resulted from the disposition of UIR 3129, demonstrate the acceptability of the existing design. The calculation, developed as part of the corrective action to UIR 3129, will provide the formal documentation to i demonstrate that the building integrity is not jeopardized.

Page 12 of 22

l Attachment 3 l ICAVp l RFI/RAI Response Form Continuation Sheet l (

Response ID: M2 IRF-00481 RFI/RAINumber: N/A AR Number: N/A DR Number: DR-0027 (Parsons) AR Number: 97024996 CR Number: M2-97-2294 M2-DRT-00027 (NU)

Disposition Continuation: Item 1: (continued)

CONCLUSION:

Based on the above discussion, NU has concluded that the issue reported in Item 1 of Discrepancy Report DR-0027 is a discrepant condition previously identified by NU, as documented in UIR No. 3129 and is a Significance Level 3 discrepancy. Significance Level 3 was chosen because the design basis was not fully verified as the formal pressure calculations were not performed. The EBFS will p:rform ,

its intended design function as demonstrated by pre-operational testing and preliminary final calculation results.

I ITEM 2: The Breach of Enclosure Building Integrity Due to Excessive Negative Pressure with g Damper AC-1 Closed, a Single Failure of AC-11 such that it is in the Open Position, Main Exhaust Fans Operating, and EBFS Fans Operating.

If a Containment Isolation Actuation Signal (CIAS) occurs while purging the enclosure building, supply fan F-23 and damper AC-1 are automatically stopped and closed. EBFS fans F25A/B start and run drawing air from the enclosure building. Normally, main exhaust fans F-34A/B/C would continue to operate, damper AC-8 would go to its open position, and AC-11 would close which would isolate the main exhaust fan suction from the enclosure building. This is the designed response for the system (s).

The DR contends that if a single failure of damper AC-11 is postulated causing it to remain open, the main exhaust fans, in addition to the two EBFS fans, would, as a result, all be aligned and able to draw air from the enclosure building. The DR also postulates that the negative pressure created by the EBFS fans in conjunction with the main exhaust fans could exceed the maximum design pressure for the enclosure building, -2.0 in, w.g..

The issues raised in this item do not jeopardize the enclosure building design as listed below, based on the discussion that follows:

l

- The damper failure vulnerabilities described in this DR have been previously addressed by NU

and have been found to be of sufficiently low probability to be below regulatory significance.

l

- The fan combinations and system alignments in conjunction with the damper single failure are not

( part of the MP2 original licensing basis.

An assessment prepared to support the response to this DR indicates that the -2.0 in, w.g. pressure in the enclosure building with five fans operating can not be achieved.

Page 13 of 22

i Attachment 3 ICAVP

.. RFURAI Response Form Continuation Sheet i

Response ID: M2 IRF-00481 i RFI/RAI Number: N/A AR Number: N/A DR Number: DR-0027 (Parsons) AR Number: 97024996 CR Number: M2-97-2294 M2-DRT-00027 (NU)  !

i Disposit , Continuation: Item 2: (continued) i With regard to the scenario postulated in this DR item, Millstone 2's licensing basis does not require j AC-11 to meet single failure criteria. This is supported by the design of the system which does not l provide redundancy, which is further outlined in memo DE2-95-472. Memo DE2-95-472,"AC-1 and 11, MP2 Enclosure Building Secondary Containment Integrity Single Failure DeSciencies - Engineering Analysis", addresses the potential single failure of AC-11. This memo dennes the licensing bases for the EBFS and main exhaust system. The licensing basis identified does not require damper AC-11 to meet l

single failure criteria. It also refers to a probability analysis that concluded that, based on relatively little amount of time the enclosure building is being purged, that the probability of occurrence of the single failure scenario for AC-11 is low and the risk to public safety was determined to be negligible.

-Memo MP2-DE-96-458 also supports this single failure position and provides an NRC evaluation in

( which the single failure position was evaluated as acceptable. This memo addresses the results of the evaluation relative to the unidentified radiological release path reported in MP2 LER 94-040-01 and 02.

The memo identified that the licensing bases for AC-11 did not require the damper be subject to single 1

failure criteria. The analysis presented in this memo addressed a specinc condition in which the exhaust i system could be in service purging the enclosure building and damper AC-11 could fail in the open position thereby providing an unfiltered radiological release path to the environment. Subsequent to this analysis, this position was evaluated as acceptable by the NRC as indicated in the above noted memo.

i l

If AC-11 was postulated to fail open and both the main exhaust fans and the EBFS fans were operatmg, the potential for drawing a pressure greater than -2.0 in w.g. is not considered credible. A review performed in support of the response to this DR indicates that the pressure drops in the ducting and the cross-connected systems result in a pressure that would be substantially less negative than the design pressure of -6 in. w.g. at the main exhaust fans. Based on the five fan alignment presented in the DR, the flow paths from the auxiliary building, the fuel building, and the enclosure building through the main exhaust fans would be such that the main exhaust system would not significantly contribute to the negative pressure from the EBFS system and thus would not create a condition that would challenge the enclosure building's design limit. It should be noted that while the enclosure building metal siding pressure limit for maintaining leaktight characteristics is -2.0 in, w.g., the metal siding is designed and factory tested to perform without failure with a infiltration rate of one air change per 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> at a pressure of-8 in, w.g..

Page 14 of 22 f

w. ~e,n -

___ . a Attachment 3 ICAVP RFI/RAI Response Form Continuation Sheet l (

Response ID: M2 IRF-00481 RFI/RAI Number: N/A AR Number: N/A DR Number: DR-0027 (Parsons) AR Number: 97024996 CR Number: M2-97-2294 M2-DRT-00027 (NU)

Disposition Continuation: Item 2: (continued)

The prelinA.a review postulated both EBFS fans op rating, all three main exhaust fans operating, supply fan F-M shut off with AC-1 damper closed, and damper AC-11 open, and assumed the enclosure building is at -2.0 inch wg. Given this scenario with all five fans operating, the pressure at the common plenum is the same for all three flow paths feeding the plenum. These flow paths are from the auxiliary building, the fuel building, and the enclosure building. The maximum negative pressure in the plenum is 6 in. w.g. before opening the outside makeup air path into the plenum. Considering only three paths, the negative pressure in the auxiliary building and fuel building cannot be obtained given the infiltration in those buildings and the differential pressure across the filter housings. The auxiliary building and the fuel building would provide additional paths of air infiltration so that a -6 in. w.g. pressure would not develop at the plenum. As such, a -2 in, w.g. pressure in the enclosure building with five fans operating can not be developed.

I. Based on the discussion provided above, NU considers that the scenario identified in the DR can not occur based on the design of the system. Additional mitigating actions are provided as the operators will stop the main exhaust fans with an Unit 2 Stack Gaseous Alarm per procedure ARP 2590H (Corrective Action 3a). This procedural requirement was the NRC approved corrective action resulting from LER 94-040 to minimize an unfiltered radiological release out the Unit 2 stack.

The DR states that damper AC-11 is Non-QA. It should be noted that damper AC-11 was originally purchased Non-QA but was upgraded to QA status because it receives a CIAS signal to close post LOCA. Consistent with the licensing and design basis as described above however, it is not qualified to seismic criteria nor does it meet EEQ requirements.

CONCLUSION:

Based on the above discussion NU has concluded that the issue reported in Item 2 of Discrepancy Report DR-0027 does not represent a discrepant condition. NU does not consider the scenario presented in the DR to be consistent with the MP2 licensing basis. However, NU considers it important to formally document the design reviews performed to assess the issues presented in DR-0027. Therefore, NU has initiated additional corrective actions to those created for UIR 3129 and is performing a new calculation (AR 97019618-04) to document Enclosure Building performance assuming the failure modes identified l in DR-0027, Item 2.

Page 15 of 22

Attachment 3 (CAVP RFI/RAI Response Form Continuation Sheet Response ID: M2 IRF-00481 RFI/RAI Number: N/A AR Number: N/A DR Number: DR-0027 (Parsons) AR Number: 97024996 CR Number: M2-97-2294 M2-DRT-00027 (NU)

Disposition Continuation:

ITEM 3: The Operating Capacity of Two EBFS Fans may be below the Enclosure Building l

Design In leakage with a Single Failure of Damper AC-1 such that it is in the Open  !

Position.

If a LOCA occurs while purging the enclosure building, supply fan F-23 and damper AC-1 are automatically stopped and closed. EBFS fans F25A/B start and run drawing air from the enclosure i building. Normally, main exhaust fans F-34A/B/C would continue to operate, damper AC-8 goes to its open position, and AC-11 will close which would isolate the main exhaust fan suction. This is the expected design response for the system (s).

The DR contends that if one postulates a single failure of damper AC-1 causing it to remain open, the expected flow from the supply system,8400 cfm, in addition to the enclosure building infiltration,2560 cfm, equates to a flow of 10,960 cfm. With this flow rate, one EBFS fan would not be able to create the

( required pressure, -0.25 in. w.g.. Two EBFS fans would be necessary and their effectiveness to create I

the required negative pressure is questionable.

LER-94-040 identified the single failure vulnerability associated with AC-1. Memos M2-DE-96-458 and DE2-95-472 regarding MP2 Enclosure Building Secondary Containment Integrity Single Failure Deficiencies for AC-1 and AC-11, address the potential single failure of AC-1 and found that the assumption of single failure for these dampers was not consistent with the MP2 licensing basis. Despite this conclusion, NU considered the benefit of design improvements and the AC-1 single failure vulnerability was eliminated by implementation of Plant Design Change Record (PDCR) MP2-041-95, Containment and Enclosure Building Purge System Damper Modifications.

The PDCR installed a counterbalanced gravity damper, AC-130, on the inlet to purge supply fan F-23. The gravity damperis designed to close when the purge supply fan is isolated and will remain closed with a negative pressure of up to approximately 0.40 in. w.g. in the enclosure building. The damper will remain closed at enclosure building pressures less than 0.40 in. w.g. and open at negative pressures greater than 0.40 to ensure that the minimum negative pressure of 0.25 in, w.g. is maintained.

The negative pressure of 0.25 in, w.g. can be maintained with a single fan operating. In addition, surveillance procedure, SP 2609E, section 4.3, is performed on a refueling frequency to ensure the required negative pressure is obtained relying only on the operation of AC-130 with AC-1 open. This surveillance verifies the ability to maintain the building negative pressure per the design requirements.

Page 16 of 22

Attachment 3 L ICAVP RFURAI Response Form Continuation Sheet

(

Response ID: M2-IRF-00481 RFURAI Number: N/A AR Number: N/A DR Number: DR-0027 (Parsons) AR Number: 97024996 CR Number: M2-97-2294 l M2-DRT-0002'7 (NU)

Disposition Continuation: Item 3: (continued) '

The current FSAR section 6.7.4.1 addresses the previous basis for failure of AC-1. UIR 2224 and ACR M2-96-0788 were initiated to update this section of the FSAR to reflect the current basis as described in Memo MP2-DE-96-0458. It should be noted that FSARCR 95-MP2-42 had already provided the appropriate information by adding FSAR Section 5.3.5 which addressed the current single failure reliability basis for dampers AC-1 and AC-11 including the addition of damper AC-130.

CONCLUSION:

Based on the above discussion, NU has concluded that the issue reported in Item 3 of Discrepancy Report DR-0027 does not represent a discrepant condition.

, ITEM 4: The Inability to Maintain Minimum Negative Pressure with One EBFS Fan l! Operating, AC-1 Open, and AC-130 Projected to Open at Less Than the Design Negative Pressure of 0.25 in. w.g.; The Reliance on Damper AC-130 Alone for Make- i up Air for Containment Cleanup Actions.

1) If a LOCA occurs while purging the enclosure building, supply fan F-23 and damper AC-1 are automatically stopped and closed. EBFS fans F25A/B start and run drawing air from the enclosure building. Normally, main exhaust fans F-34A/B/C would continue to operate, damper AC-8 would go to its open position, and AC-11 would close which would isolate the main exhaust fan suction. This is the expected response for the system (s).

The DR contends that gravity damper, AC-130, could open at less than the required negative pressure,

-0.25 in, w.g.. The DR postulated this condition based on an analysis which assumed five pounds of force applied to the counter balance arm could open the damper with one EBFS fan operating. The five pound applied force was indicated in surveillance procedure SP2609E as an operator aid in determining the approximate manual force needed to open the damper and assure no damper binding.

Surveillance Procedure SP2609E, Enclosure Building Filtration System Testing - Refueling, verifies operability of the EBFS. To test AC-1, AC-130 must be manually opened. The procedure Note in section 4.1.13 and 4.2.14 of SP2609E identifies an approximate " manual" force required to hold the suction damper AC-130 open with the use of the counterbalance lever arm. The note was included in the

' procedure to assist the operator in determining what method to use to manually open the damper for testing and it is not a design requirement.

l

{

Page 17 of 22 l

Attachment 3 l ICAVp

! RFI/RAI Response Form Continuation Sheet

(

Response ID: M2 IRF 00481

! lRFI/RAI Number: N/A AR Number: N/A l DR Number: DR-0027 (Parsons) AR Number: 97024996 CR Number: M2-97-2294 M2-DRT-00027 (NU) i Disposition Continuation: Item 4: (continued) 4

2) When containment is purged using the EBFS containment cleanup flow path, one EBFS fan F-25A o.' B draws air from containment. The fresh air supply utilizes the flow path from the supply system withcut fan F-23 operating.

The DR contends that during containment purge, using the EBFS containment cleanup flow path, damper AC-130 is relied upon alone to open by the differential pressure created by the EBFS fan for the supply of makeup air.

During containment purge using the EBFS containment cleanup flow path, AC-1 is manually opened. If the differential pressure created by the EBFS fans is greater than approximately 0.40 in. w.g., damper AC-130 will automatically open to supplement the air supply to containment as governed by the amount of back pressure. The main supply of air during this purging evolution is provided from the enclosure building as damper AC-3 is manually positioned in the "mid" location. Procedure OP2314B, step 4.1.12b, sets up this flow path.

i l

GENERAL COMMENT

S Two general comments were made at the end ofItem 4 of the DR relating to HVAC calculation review and evaluation of RAI-468 information. The text of this portion of the DR appeared to be more of a

, narrative explaining the basis for the ICAVP auditor review rather than explanation of a physical discrepancy. However, the following is a response to these comments:

Calculation Review - the DR contends that some HVAC calculations reviewed were found to have a generic problem in that the calculations were out of date and have several analytical problems. A weakness with HVAC calculations in general was previously identified by NU. Engineering Work Request, EWR #2-96-105, was authorized to review and revise these calculations. This task is currently underway. M2 HVAC calculations have been identified, reviewed, and indexed. Currently, the deficient l QA calculations are being corrected, the Non-QA calculations will follow.

l l Page 18 of 22

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

Attachment 3 ICAVP RFI/RAI Response Form Continuation Sheet Response ID: M2 IRF-00481 RFI/RAINumber: N/A AR Number: N/A DR Number: DR-0027 (Parsons) AR Number: 97024996 CR Number: M2-97-2294 M2-DRT-00027 (NU)  ;

Disposition Continuation: Item 4 (continued)

Evaluation of RAI-468 Information - the DR contends that the Surveillance Procedures do not test the system with two fans operating. Thus, the capability of the system to achieve and maintain the building minimum design pressure, when damper AC-1 is in the open position, has not been verified. Assurance that the building integrity pressure limit is not exceeded when the damper is in the closed position has not been demonstrated.

Based on the installation and testing of gravity damper AC-130, located upstream of damper AC-1, a flow path frota the supply ducting is not considered cc ile. Hence, verification of the minimum design pressure is not necessary. Assurance that the budding integrity pressure limit is not exceeded with the damper in the closed position has been previously identified and is addressed in DR Item 1.

ITEM 5: UIR 3129 Conclusions and Corrective Action - NU did not Recognize the Potential

( Safety Significance of the Consequence of Breaching the Enclosure Building Integrity.

The DR contends that based on the lack of testing and inadequate tem design calculations identified in UIR 3129, NU did not generate a CR. It is postulated that NU u.J not recognize the potential safety significance of the consequence of breaching the enclosure building integrity due to a perceived negative pressure greater than the limit of-2.0 in. w.g.. Maintaining the structural integrity of the enclosure building is essential in order to take credit for the filtered, elevated release path used to meet 10CFR100 release limits.

The DR contained a partial listing of the engineering documentation used to support the EBFS system licensing and design bases. Existing ACRs/CRs, LERs, and modifications, as well as preliminary assessments were evaluated and used in the PI-7 Graded System Review Packages to support the system LB/DB. During the PI-7 review, a number of UIRs and CRs were prepared to address issues that could not be substantiated.

When these UIRs were dispositioned, assessments were made regarding their significance and if the finding had already been addressed on an existing ACR/CR. If follow-on work was necessary (i.e.

calculations, modifications, etc.) before the exact impact of a potential finding was known, the UIR dispositioner performed assessments and/or preliminary calculations to determine if the finalization of the problem resolution would identify a significant condition. In the case of UIR 3129, the review of the EBFS design, calculations, test data, existing LERs, and ACR/CRs indicated that the existing design would meet its design basis but additional calculations and completion of existing corrective actions l were necessary to completely document the support of the DB.

Page 19 of 22

Attachment 3 i ICAVP

{

RFI/RAI Response Fonn Continuation Sheet

{

Response ID: M2-IRF-00481 RFI/RAI Number: N/A AR Number: N/A I DR Number: DR-0027 (Parsons) AR Number: 97024996 CR Number: M2-97-2294 M2-DRT-00027 (NU)

Disposition Continuation: Item 5: (continued) l Enclosure Building integrity was not considered an issue because fan performance was not considered sufficient to reach a negative pressure which would challenge the enclosure building design as described in item 1 and 2 above. Single failure vulnerabilities were also addressed in past assessments and documentation, the results of which were deemed applicable to this case. Based on engineering i documentation, a CR was not deemed necessary at the time of the UIR resolution.

All UIRs were prepared, dispositioned, reviewed, and ultimately approved by the Expert Panel (EP) in I accordance with PI 14, " Configuration Management Plan Project Administrative Instruction". One of the purposes of the expert panel approval was to evaluate reportability and operability issues, ensure CRs were issued as required, and all appropriate correction actions were defined.. In fact, during the EP ,

approval for UIR 3129, the preparer /dispositioner was questioned as to why a CR had not been issued to I address the fmdings. Based on the UIR preparer's response, the EP was satisfied with the existing documentation and his assessments that a significant condition did not exist. In the case of UIR 3129, a CR was.not necessary until the implementation of the corrective actions showed otherwise. On this l basis, the UIR was dispositioned without the issuance of a new CR.

DR-0027 was issued by the ICAVP contractor as a Level 1 DR. As further substantiation to the significance and reportability of the issues raised in DR-0027, CR M2-97-2294 was issued to document this potential discrepancy independent of the Parsons Significance Level and if pre-discovered by NU (NU discovered DR issues are not processed separately through the NU " corrective action system"). The associated CR review subsequently conducted and presented to M2 PORC for approval did not identify any reportable conditions.

CONCLUSION:

Based on the above discussion, NU has concluded that the issue reported in item 5 of Discrepancy Report DR-0027 does not represent a discrepant condition.

l

\

l Page 20 of 22 l

-  :::=- ~ <

Attachmrnt 3  ;

ICAVP RFI/RAI Resnonse Form Continuation Sheet i

Response ID: M2-IRF-00481 RFI/RAI Number: N/A AR Number: N/A DR Number: DR-0027 (Parsons) AR Number: 97024996 CR Number: M2-97-2294 M2-DRT-00027 (NU) i References (Previously Transmitted):  ;

1) Startup Field Report No. HV-81, enclosure Building Filtration,3/27n5.

2) MP2 LER 94-040-02, Ventilation Design Deficiency Affecting Enclosure Building Integrity,9/11/95.

3) Memo M2-DE-96-0458, AC-1 and AC-11, MP2 Enclosure Building Secondary Containment Integrity Single Failure Deficiencies,9/9/96.

4) PDCR 2-041-95, Rev. O, Containment and Enclosure Building Purge System Damper Modifications.  !

5) UIR 2224, The EBFS is not single failure proofin accordance with FS AR Chapter 6.7 with respect to l 2-AC-11.

6) ACR M2-96-0788, FSAR not updated with respect with 2-AC-11 single failure requirements.

l 7) SP 2609E, Rev 6, Enclosure Building Filtration System Testing - Refueling.

8) ARP 2590H, Rev. 2, Alarm Response for Control Room Radiation Monitor Panels, RC-14

9) OP 2314G, Rev.12, Enclosure Building Filtration System.

.0) OP 2314B, Rev.16, Containment and Enclosure Building Purge.

I1) PIR 2-95 126, Design Flow in Main Exhaust /Ctmt Enclosure Building Purge Preliminary Review, dated 2/9/95,

12) Memo DE2-95-0472, AC-1 and AC-11, MP2 Enclosure Building Secondary Containment Integrity Single Failure Deficiencies - Engineering Analysis,6/6/95.

13) Memo DE2-95-0543, Containment and Enclosure Building Purge System Single Failure Problems - AC-1 and AC-11,6/14/95.

14) Memo NE-95-SAB-225, Single Failures of EBFS and Their Impact on Public Safety,5/25/95.

l 15) Calculation 2-ENG-174, Rev. O, Air Flow Through a 4" Hole From the Enclosure Building at Design Pressure.

16) UIR 3129 with action tracking assignments (ARs).

New References (Attached) 17). Millstone Unit 2 Pre-Operational Test T2314GP, Rev 1,7/9D5 New References (Not Attached)

18) NU Calculation 97EBF-02000-M2, (in preparation), " Enclosure Building Inleakage and Negative Pressure" l

Page 21 of 22

Attachment 3 ICAVP RFI/RAI Response Form Continuation Sheet Response ID: M2 IRF-00481 RFI/RAI Number: N/A AR Number: N/A DR Number: DR-0027 (Parsons) AR Number: 97024996 CR Number: M2-97-2294 M2-DRT-00027 (NU)

Ccnclusion Continuation:

Item 1: NU has concluded that the issue reported in Item 1 of Discrepancy Report DR-0027 is a discrepant  !

condition previously identified by NU in UIR No. 3129 and is a Significance level 3. UIR No. 3129 identified i that a calculation or procedure does not exist to verify the enclosure building upper limit negative pressure of 2 in, w.g.. Corrective Actions consisting of creating a new calculation, evaluating the enclosure building integrity  !

verses the new calculation results, and review the need for a new surveillance procedure to test both fans were initiated; reference AR 97019618, assignments 01,02, and 03.

Item 2: NU has concluded that the issue reported in Item 2 of Discrepancy Report DR-0027 does not represent a discrepant condition. NU believes that the original plant licensing basis does not require damper AC-11 to be subject to single failure criteria. This is based on the lack of redundancy in the system design and research of the licensing basis documentation. In addition, probability analysis has been performed which indicates that based on the relatively short amount of time that the enclosure building is being purged, the probability of ,

occurrence of the single failure for damper AC-11 is low and the risk to public safety was determined to be l legligible. The postulated single falure question for damper AC-11 was previously reported via LER 94-040.

Item 3: NU has concluded that the issue reported in Item 3 of Discrepancy Report DR-0027 does not represent a discrepant condition. A new counterbalanced gravity damper, AC-130, has been installed upstream of damper AC-1 which has been designed and is tested to remain closed at a negative pressure of 0.25 in. w.g.. This new damper will provide the necessary isolation of the supply air flow path so that the Enclosure Building Filtration Fans will be able to draw the required negative pressure of 0.25 in. w.g..

Item 4: NU has concluded that the issues reported in Item 4 of Discrepancy Report DR-0027 do not represent a discrepant condition. First, the five pound applied force was provided in surveillance procedure SP2609E to assist the operator in determining what method to use to manually open the damper for testing and it is not a design requirement. Testing of damper AC-130 shows that it opens at a negative pressure of approximately 0.40 in w.g.. Second, the normal flow path for makeup air for containment purge using the Enclosure Building Filtration System flow path is not through damper AC-130 but through AC-3. AC-3 is manually positioned during the test procedure to allow enclosure building air to be used for makeup.

Item 5: NU has concluded that the issue reported in Item 5 of Discrepancy Report DR-0027 does not represent a discrepant condition. The Enclosure Building integrity was not considered an issue because a preliminary review indicated that the fan performance was not considered sufficient to reach a negative pressure which would challenge the enclosure building design. Single failure vulnerabilities were also addressed in past assessments and documentation, the results of which were deemed applicable to this issue. Based on this

(

2ngineering documentation, the engineering staff and the UIR expert panel, who approves the UIR resolution, l concurred that the issue was not safety significant and a CR was not deemed necessary.

Page 22 of 22

. a.--

Attachment 2 ICAVP Rgulatus_ Form Response iD: M2-IRF-01635 RFI/RAI Number: N/A AR Number: N/A DR Number: DR-0027 (Parsons) AR Number: 97024996 CR Number: M2-97-2294 I M2-DRT-00027 (NU)

. System Number /Name or Program ID/Name: 2314G/EBFS

Subject:

Enclosure Building Filtration & ContainmendEnclosure Building Purge System Design.

This discrepancy is rated as Significance Level I by Parsons.

Background:

Discrepancy Report, DR-0027, reports the following discrepancies conceming maintaining the Enclosure Building Filtration Region (EBFR) at a negative pressure.

(Continued on Page 2) g Continuation Disposition: This response provides additional infonnation to the initial DR-0027 response, M2-IRF-00481.

(Continued on Page 7) j ,

g Continuation

Conclusion:

This response provides additional information to the initial DR-0027 response, M2-IRF-00481. NU has concluded that DR-0027, has identified a condition previously discovered by NU which i requires correction. NU considers the issues identified in DR-0027 to be a Significance Level 3.

(Continued on Page 9)

EContinuation ,

Preparer: ~ -

Date: 8

c. scully /

Technical Review: _. Date: 4 X t- ff Unit Lead Review: .

bbn Date: /I ,'p '

F. Mattioli Technical Review Team Concurrence:- .

yre MD '

Date: bO/ N '

%an i

NU ICAVP Project Manager Approval: A e Date: 5 / 9f J. F/uge_re f

NU CMP Director Approval: Brez Date: b'7 I R. Necci Pa e 1 of 10

Attachment 3 ICAVP

( REl/RAI Response Form Continuation Sheet Response ID: M2-IRF-01635 RFI/RAI Number: N/A AR Number: N/A DR Number: DR-0027 (Parsons) AR Number: 97024996 CR Numbcr: M2-97-2294 M2-DRT-00027 (NU)

BACKGROUND CONTINUATION:

Attachment 1 provides Parsons comments to the initial response to DR-0027, M2-IRF-00481 as presented on 1/19/98.

Item 1: Breach of Enclosure Building Integrity due to Execss Negative Pressure (EHFS Fans Operating)

In the event of an emergency condition, an Enclosure Building Filtration Actuation System (EBFAS) signal will start the two EBFS fans F-25A & F-25B. The fans will run until shutdown by the plant operators. Damper 2-AC-1 (EB air supply isolation damper) will close upon receipt of the emergency signal, if not in the normally closed position. FSAR Section 6.7.2.1 states that 2.0 in.

i w.g. is the maximum difTerential pressure that the enclosure metal siding can sustain and still maintain its leak-tight characteristics. Since the EBFS does not have pressure control provisions to prevent exceeding the building maximum pressure limit, a potential exists for breaching the integrity of the enclosure building when two fans operate with damper 2-AC-1 closed.

Item 2: Breach of Enclosure BuildinFJnteerity Due to Excessive Neeative Pressure (Matin Exhaust Fans Operatine and Exhaust Damper 2-AC-11 Openedh if a CIAS occurs while purging the Enclosure Building, the purge supply fan F-23 and damper AC-I are automatically stopped and closed, respectively. The Enclosure Building purge exhaust damper AC-8 remains open. The main exhaust fans will continue to operate and, if damper 2-AC-11 fails open, draw air from the Enclosure Building until the fans are turned off manually following a Unit 2 Stack high radiation alarm. The EBFS is also activated automatically and both fans operate.

Damper AC-1 is a pneumatic damper. The sudden closure of this damper while the main exhaust fans are exhausting air from the Enclosure Building, could cause a sudden increase in negative pressure in the building. The design exhaust rate from the building is 32,000 CFM and the operating pressure in the main exhaust plenum is - 5.5 in, w.g. (Dwg 25203-26057). This pressure is significantly higher than the -2.0 in.wg maximum pressure limit for the building.

Page 2 of 10

Attachment 3 ICAVP

( RFI/RAI Response Form Continuation Sheet Response ID: M2-IRF-01635 RFI/RAI Number: N/A AR Number: N/A DR Number: DR-0027 (Parsons) AR Number: 97024996 CR Number: M2-97-2294 M2-DRT-00027 (NU)

When the main exhaust fans are operating, the air exhausted from the Enclosure Building (via exhaust damper AC-11)is mixed with exhaust air from the other buildings prior to discharge to the Unit 2 stack. It is possible that the main exhaust fans will continue to operate together with the EBFS fans. The negative pressure induced by the main exhaust fans in the building is a back pressure to the EBFS fans and will cause the EBFS fans to operate to the left of the their combined fan curve, thus, increasing the building negative pressure.

An analysis of this potential breaching of Enclosure Building leak-tightness integrity does not exist.

NRC Safety Evaluation Attached to MP2-DE-96-0485 (Reference I.1) addressed the failure of non-safety damper AC-11 from the perspective of releases via the main exhaust path. NNECo

( committed to perform certain operator actions to shutdown this release path following receipt of a high radiation signal. However, the Safety Evaluation did not consider the potential for excessive negative pressure in the Enclosure Building due to damper AC-11 remaining in the open position.

The closure of damper AC-11 may be necessary to ensure Enclosure Building integrity.

Item 3: Two Fan Operatine Capacity below Design In-leakage with Damper 2-AC-1 In Open Position:

Consider the case for which a LOCA occurs with damper 2-AC-1 in the open position. In addition, consider the single failure ofdamper 2-AC-1 as failing in the open position. For this scenario, the design building in-leakage is 2560 + 8400 = 10,960 CFM. Since the in-leakage exceeds the design capacity of one fan (9000 cfm)it is concluded that both fans must operate to achieve the design 0.25 in.wg negative pressure. Using FSAR Figure 6.7-3 to estimate the two fan operating capacity indicates that the system may not be capable of handling the in-leakage. Thus, the design minimum

-0.25 in. wg building pressure may not be achieved.

Page 3 of 10

Attachment 3 ICAVP I EFI/RAI Response Form Continuation Shert Response ID: M2-IRF-01635 RFI/RAI Number: N/A AR Number: N/A DR Numbcr: DR-0027 (Parsons) AR Numbcr: 97024996 CR Number: M2-97-2294 M2-DRT-00027 (NU)

Item 4: Inability to Maintain Minimum Neestive Pressure with One EHFS Fan Operatine and AC-1 OpatElla According to the NRC Safety Evaluation Attached to MP2-DE-96-0485 (Reference 1.1):

"...In the event of a LOCA or MSLB during purging, with failure of actuation signal CHl-CIAS, damper AC-1 would fail to close and fan F-25A would fail to start. This combination oflack of isolation (AC-1) and reduced filtered exhaust capability (loss of one train of exhaust / cleanup) would prevent the secondary containment from functioning properly as a fission product cleqnup system for primary containment leakage as the single operating F-25 fan would not have sufficient capacity to establish and maintain the necessary negative pressure in the unisolated Enclosure Building..

(

...Although the licensee claims that corrective action is not required by the original licensing basis, a modification has been proposed to eliminate the AC-1 vulnerability. A gravity damper would be installed as shown in the drawing. It would be weighted such that operation of purge fan F-23 opens it, but a -0.25 wg. vacuum due to operation of an EBFS fan would not cause it to open. This action would eliminate the AC-1 single failure condition." ,

1 According to the NOTE 2 under 4.1.13 of SP 2609E,"Approximately 5 pounds of force applied to l F-23 suction damper, AC-130, counter weight lever is sufTicient to open the damper." The damper size is 47" x 47". For conservatism, use half of the damper area as the effective area that is subjected to a differential pressure. Thus, a pressure differential of about 0.13 in. wg. will open the

)

i gravity damper. According to drawing 25203-26057, the static pressure in the vicinity of AC-130 (

pressure point 13)is -0.11 in.wg. Test data (attached to Reference B.13) shows that one EBFS fan operating can create higher negative pressures than 0.25 in, wg. (-0.35 to -0.75 in.wg).

In addition, during contaimnent purge using the EBFS containment cleanup flow path, AC-1 is l manually opened, but the purge supply fan must not be started (refer to 4.1.12 of SP 2314B). Thus, AC-130 is relied upon to open by the differential pressure created by the EBFS fan for makeup air during containment cleanup.

Page 4 of M

Attachment 3 1CAVP

( RFI/RAI Response Form Continuation Sheet Response ID: M2-1RF-01635 RFI/RAI Number: N/A AR Number: N/A DR Number: DR-0027 (Parsons) AR Number: 97024996 CR Number: M2-97-2294 M2-DRT-00027 (NU)

The above contradicts the AC-130 performance requirement as stated in the AC-1 resolution. AC-130 will open at less than the design negative pressure of 0.25 in. wg, with the purge supply fan F-23 shutdown.

Item 5: UIR 3129 Conclusions and Corrective Action The CMP via UIR 3129 recognized the need for a new analysis / calculation to provide the system operating curve and operating procedure to test both fans operating simultaneously. The UIR Recommended Disposition Details are repeated below:

"1. AR 97019618-01 is written for CMP to evaluate need for additional surveillance or test based

( on AR 97019618-03 analysis results.

2. AR 97019618-02 is written for CMP to verify that EBFR siding to sustain and still maintain leak-tight characteristics at upper limit of 2 inch of w.g. negative pressure.

l

3. AR 97019618-03 is written for CMP to generate new calculation showing system operating curve with one fan operating and also, when two (2) fans operating in parallel. Upon completion evaluate for AR 97019618-01."

The UIR Final Disposition is repeated below:

i

" Expert Panel: AGREES with Recommended Disposition Details. AR 97019618-01, CMP to evaluate need for additional surveillance or test based on AR 97019618-03 analysis results. AR 97019618-02, CMP to verify that EBFR siding to sustain and maintain leak-tight characteristics at 2 inch of w.g. AR 97019618-03, CMP to generate new calculation showing system operating curve with one fan and also, when two (2) fans operating in parallel. Upon completion evaluate for AR 97019618-01."

It is clear from the above that damper 2-AC-1 concerns were not recognized by the UIR. The need I

for a test is subject to the analysis results and not mandated. Mandated testing is required since, at best, an analysis is subjective for the EBFS. Testing for in-leakage rate was not addressed,

- therefore, degradation of the building leak-tightness characteristics can not be monitored.

Page 5 of 10 I

J

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

Attachment 3 ICAVE

, RFI/RAI Response Form Continuation Sheet Response ID: M2-IRF-01635 RFI/RAI Number: N/A AR Number: N/A DR Numbcr: DR-0027 (Parsons) AR Number: 97024996 CR Number: M2-97-2294 M2-DRT-00027 (NU)

The UIR (section 1, item 1) states "...two (2) fans in operation must be capable of maintaining a negative pressure in the EBFR less than the upper limit of 2 inches w.g." The UIR (section 2, item

1) states " System performance calculation for the fans are inadequate." UIR section 2, item 4 states "There is no procedure to test these two fans operating simultaneously and record a maximum negative pressure developed in the EBFR."

Maintaining the structural integrity of the EB is essential in order to take credit for the filtered, elevated release path used to meet 10CFR100 release limits. However, the system design calculations and testing program are inadequate to demonstrate that the system meets its design requirement. However, the UIR states as conclusion 1 (Section 2),"This UIR has been determined not to require a CR and has not identified a potential safety significant condition." The Final Disposition Section of the UIR did not contradict this statement. This conclusion is inconsistent with the information presented in the UIR. NNECo UIR-3129 did not recognize the potential safety significance of the consequence of breaching the Enclosure Building integrity.

Jlasis for Sienificance Level 1:

One of the safety functions of EBFS is to collect and process potentially radioactive airborne particles and gases in the EBFR following a LOCA and limit the site boundary radiation doses to the 10CFR100 requirements. Due to the lack of supporting calculations and/or test procedures /results the ability of the EBFS to perform its primary safety function cannot be assured.

Discrepancies identified may:

1. Breach Enclosure Building integrity due to excess negative pressure (items 1 & 2), and

2. Fail to maintain the minimum required negative pressure with design building in-leakage (items 3

&4).

NNECo UIR-3129 did not recognize the potential safety significance of the consequence of breaching the Enclosure Building integrity.

Page 6 of 10 s _

Attachment 3 ICAVP

( RFI/RAI Response Form Continuation Sheet Response iD: M2-IRF-01635 RFl/RAI Number: N/A AR Number: N/A DR Number: DR-0027 (Parsons) AR Numbcr: 97024996 CR Number: M2-97-2294 M2-DRT-00027 (NU)

DISPOSITION CONTINUATION:

Item 1: Breach of Enclosure Huilding Integrity due to Excess Necative Pressure (EBFS Fans Operatine)

Calculation 97-EBF-02000-M2, Rev. O," Enclosure Building Inleakage and Negative Pressure,"

dated 12/18/97 calculated the negative pressure in the EB with both the EBFS fans operating in parallel. The calculation results establish that the negative pressure is 0.5 in. w.g. which is below the negative pressure of 2.0 in, w.g. described in the FSAR. Calculation 97-EBF-02000-M2, Rev. O is currently being revised. All required FSAR (section 6.7) changes and procedure changes associated with the calculation results will be made following approval of Revision 1. The response to DR-0426 will address the calculation revision and the associated required FSAR and procedure changes.

As previously stated in M2-IRF-00481, NU has concluded that the issue reported in item 1 of Discrepancy Report DR-0027 is a discrepant condition previously identified by NU, as documented in UIR 3129 and is a Significance Level 3 discrepancy. Significance Level 3 was chosen because the design basis was not fully verified as the formal pressure calculations were not performed.

Item 2: Breach of Enclosure Buildine Inteerity Due to Excessive Negative Pressare (Main Exhaust Fans Operatine and Exhaust Damper 2-AC-11 Opened):

The previous response, M2-IRF-00481, to DR-0027, Item 2, stated that a new calculation will be created to determine the EB performance assuming the failure modes identified in DR-0027, Item 2.

Due to the variables associated with the single failure scenario and the lack of test data a calculation could not be perfonned. Instead, Technical Evaluation, M2-EV-98-0095 was prepared to describe the single failure scenarios associated with the CEBPS Isolation dampers 2-AC-1 and 2-AC-11 and providejustification that the Enclosure Building Filtration System (EBFS), CEBPS and the Enclosure Building (EB) meet their design and licensing basis.

The Technical Evaluation concludes that the 2-AC-1 i single failure scenario condition of having both the MES fans and the EBFS fans drawing down the EB would not impact the EB leak-Page 7 of 10

Attachment 3 ICAVP r

RFI/RAI Response Form Cor111tiuntion Sheet Response ID: M2-1RF-01635 RFI/RAI Numbcr: N/A AR Numbcr: N/A DR Numbcr: DR-0027 (Parsons) AR Number: 97024996 CR Number: M2-97-2294 M2-DRT-00027 (NU) tightness integrity based on the original qualification testing of the EB. As previously stated in M2-IRF-00481, NU has concluded that the issue reported in Item 2 of Discrepancy Report DR-0027 does not represent a discrepant condition. NU considers the single failure scenarios associated with 2-AC-1 and 2-AC-11 beyond the original licensing and design basis of the CEBPS and no further corrective actions are required.

Note: The required FSAR changes associated with the single failure scenarios are addressed in UlR 2224, UIR 3367, and ACR M2-96-0788. In addition, all required FSAR (section 6.7) changes and procedure changes associated with the calculation change results will be made following approval of the calculations. The response to DR-0426 will address the revision to Calculation 97-EBF-02000-M2, Rev. O, and the associated required FSAR and procedure changes.

Item 3: Two Fan Operatine Capacity below Desien In-leakane with Damper 2-AC-1 In Open Positioni ,

The previous response, M2-IRF-00481, to DR-0027, item 3, stated that the 2-AC-1 single failure scenario was eliminated by implementation of PDCR MP2-041-95. Surveillance testing per SP2609E verifies that the required negative pressure will be maintained with one EBFS fan operating and 2-AC-1 open.

Calculation 97-EBF-02000-M2, Rev. O, calculated the inleakage into the EB to be 8,700 cfm for one EBFS fan operating. The calculated inleakage value does not match the inleakage value provided in the FSAR (section 6.7). Calculation 97-EBF-02000-M2, Rev. O is currently being revised. All required FSAR (section 6.7) changes and procedure changes associated with the calculation results will be made following approval of Revision 1. The response to DR-0426 will address the calculation revision and the associated required FSAR and procedure changes.

NU has concluded that the issue reported in Item 3 of Discrepancy Report DR-0027 is a discrepant condition previously identified by NU, as documented in UIR 3129 (calculation) and UIR 3367 (FSAR section 6.7 and 5.3.5). NU considers Item 3 a Significance Level 3 discrepancy based on the FSAR changes. The EBFS is capable of performing its intended function as verified by surveillance testing per SP2609E.

Page 8 of 10

l Attachment 3 LCAYf

( itFI/RAI Resnonse Form Continuation Sheet

i Response iD

M2-1RF-01635 RFI/RAI Number: N/A AR Number: N/A DR Numbcr: DR-0027 (Parsons) AR Numbcr: 97024996 CR Number: M2-97-2294 M2-DRT-00027 (NU)

Dems 4 and 5 i i

No additional response. Response provided in M2-IRF-00481 concluded that Items 4 and 5 do not represent discrepant conditions CONCLUSION CONTINUATION:

Ite m 1 As previously stated in M2-IRF-00481, NU has concluded that the issue reported in item 1 of

(

piscrepancy Report DR-0027 is a discrepant condition previously identified by NU, as documented in UIR 3129 and is a Significance Level 3 discrepancy. The response to DR-0426 will address the revision to Calculation 97-EBF-02000-M2, Rev. O and the associated required FSAR and procedure changes. 4 Item 2 Technical Evaluation M2-EV-98-0095 concludes that the 2-AC-1I single failure scenario would not impact the EB leak-tightness integrity. As previously stated in M2-IRF-00481, NU has concluded that the issue reported in Item 2 of Discrepancy Report DR-0027 does not represent a discrepant condition. NU considers the single failure scenarios associated with 2-AC-1 and 2-AC-11 beyond the original licensing and design basis of the CEBPS and no further corrective actions are required.

Item 3 NU has concluded that the issue reported in Item 3 of Discrepancy Report DR-0027 is a discrepant condition previously identified by NU, as documented in UIR 3129 and UIR 3367. NU considers l

Item 3 a Significance Level 3 discrepancy based on the required FSAR changes. The EBFS is capable of performing its intended function as verified by surveillance testing per SP2609E.

s l

Page 9 of 10

l l

Attachment 3 ICAVP j RFI/RAI Response Form Continuation Sheet Response 1D: M2-1RF-01635.

RFI/RAI Numher: N/A AR Number: N/A DR Number: DR-0027 (Parsons) AR Numbcr: 97024996 CR Numbcr: M2-97-2294  ;

M2-DRT-00027 (NU)

Items 4 and 5 No additional response. Response provided in M2-IRF-00481 concluded that Items 4 and 5 da not represent discrepant conditions KD'ACHMENTS:

1. Technical Evaluation, M2-EV-98-0095, Rev. O, " Single Failure of Dampers 2-AC-1 and 2-AC-11, impact on EB Integrity"

( -

l l

Page 10 of 10

l i'

Docket No. 50-336 B17444 i

l i

J Attachment 4 Millstone Nuclear Power Station, Unit No. 2 Discrepancy Report DR-0312 4

d '

)

i 1

l l

c 4

September 1998 I

. ~. .. _ . , . _ , . . _ - . . -~ =. - ~ . . ~ . . - . . .

M1.A:s W w a -$ sA.- rruci.,etga,a nat w eLaC-o m-tbd-dbtf3 1 U: P Ul>*At PRGE: .$d PARSONS POWER GROUPINC. ~

ICAVP MILLSTONE UNIT 2 2675 Morgantown Road, Reading, PA 19607 DISCREPANCY REPORT t (610) 855-2000

• FAX: (610) 855-2509

=

DR NUMBER: D R-0312 DR TITLE:

Water in Diesel Oil Storage Tank (T47A) Could Enter Diesel Oil Supply Tanks REVISION: 3 ISSUE DATE: 1/27/98 ORIGINATING GROUP: TierI SIGNIFICANCE LEVEL: 1 DISCREPANCY

]

References:

1. AOP 2560 Storms, High Winds, and High Tides

2. FSAR 8.3 Emergency Generators

3. OP 2346B Diesel Fuel Oil System

4. SP 2672 Sampling and Inventory ofDiesel Oil Storage Tank, T47A

5. 25203-29032 Diesel Oil Storage Tank T-47A t

6. 7604-M-125 Miscellaneous Shop Assembled Tanks

7. ARP 2590F Emer8ency Operating Procedures and Guidelines, Alarm Response for Control Room Panel C-08

8. Figure 7.3 "10CFR50.59 Safety Evaluation Screening", prepared to support PDCR 2-108-92, dated 8/27/97.

Q 9. MP2-DG DBDPackage-Diesel Generator l 10. NRC Letter Docket / License: 50-336/DPR-65 l 11. OP 2346A EmergencyDiesel Generators

! 12. SP 2613E Diesel Generator Fuel Oil Sampling

13. FSAR 4.2.2 Codes Adhered To

14. MP 2721J Periodic Inspection ofUnit 2 Tanks

15. 91-BOP-813-ES, Rev. 3, MP2 EDG Operating Time with 24,000 Gallons ofDiesel Fuel Oil

?

Available at a Continuous Rated Load of2750 kW Fuel Consumption L. 16. IPEEE December 29,1995, MP2, Response to Generic Letter 88-20, Supplements 4 and 5, Individual Plant Examination for External Events - Summary Report

Background:

l i

Diesel Oil St,orage Tank T47A is a 25,000-gallon underground, heru' unt.1 cylindrical tank. The tank is ~

single-wall, carbon steel coated with "bitumastic mill undercoat" (exterior) and no interior coating. The tank was installed around 1972. A cathodic protection system (rectifierPO29A, zine anodes, and test station) was added after October 1974.

Transfer Pumps P-47A and P-47B are installed in individual 24-inch diameter steel riser pipes connected to nozzles on the top of the tank. Bottom of tank - l'-0" (plus or minus) depending upon which dre. wing Page 1 of15 DR-0312. DOC ll i

i 1

,c.w m u c. u-vetsN Attu2:u<a mo-tAA $m lu:Fuwt Pm- 34 is used. Top of tank = 11' (plus or minus). Grado - 14'-8". Top of pump manways is 22' (plus o minus). Other connections extending to or above grade are fill, sample, vent, and level pipes. S

[

The storage tank is not seismically qualified nor flood-protected. A sufficient quantity of oil is conta in the Diesel Oil Supply Tanks to operate the diesel (s) for approximately 7 days. Each Supply Ta (13,500-gallon nominal capacity) must contain 12,000 gallons. See Ref. 2. Pump (s) P-47A and P-47

" start (s) automatically when the respective level decreases below 95% and stops when tank level increases above 95%." See OP-2346B. (Setpoint Change #2-88-022, in process, changed mimmum to 93%.)

"Two EDGs operate for 8 hours9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br />, then one EDG operates for the remaining length of time. Fuel consumption is 3.6 gpm for each E9G at 2750 KW. During a LOCA with a LNP (loss of normal p the two day tanks (Supply Tanks) are cross-connected." Ref.15, dated April 1997. (The current license states one EDG for one hour instead of 8 hours9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br /> with the other EDG continuous). The supply tanks are cross-connected via valve 2-FO-83, " Fuel Oil Supply Header Cross-tie" Discrepancy:

Item 1:

Failure Mode Water entering the underground Diesel Oil Storage Tank (T47A) will be automatically pumped into the Diesel Oil Supply Tanks T48A and T48B.

With water entry into the Diesel Oil Storage Tank from a single failure, during a LOCA in an LNP, it would take less than 3 hours3.472222e-5 days <br />8.333333e-4 hours <br />4.960317e-6 weeks <br />1.1415e-6 months <br /> to pump 6 inches ofwater into each Supply Tank. The fuel supply pipes a 6 inches above the bottom of the supply tanks. Water will shut down both EDGs a short time later.

i (with the currently licensed scenario 1.e. one EDG for one hour, it would take up to six hours, assumin the tanks contribute equally, for the continuously running dicsci to shut down. At that time the second EDG would not restart or ifit starts it would not run for an extended length of time.)

If only one EDG were to started, it would run for 3 hours3.472222e-5 days <br />8.333333e-4 hours <br />4.960317e-6 weeks <br />1.1415e-6 months <br /> before automatically shutting down. At that time the other EDG would be started. It too would run for 3 hours3.472222e-5 days <br />8.333333e-4 hours <br />4.960317e-6 weeks <br />1.1415e-6 months <br /> before automatically shutting down.

Water Sources Water sources are surface runoff, groundwater (normal or elevated), or flood water. Probable Maximum Precipitation can ceuse surface flooding between 15.5 to 16.2 feet MSL (mean sea level), Ref 16.

Normal groundweter level is elevation 5 feet MSL (halfway up the tank). The containment, turbine, and auxiliary buildings a.e protected from floodir.g. The design flood produces water levels up to elevation 18.1 feet MSL, stillwa:er and 25.2 feet MSL, wave runup, Ref 16 for PMH(Probable Maximum .

Hurricane). According to reference 1, the EDGs will be operated during flooding up to and above a water level ofelevation 22'.

Page 2 of 15 DR-0312. DOC

w, w .m c.w e nu e mte,wr mu.sium mu-tm-m iuinutht Nu Water Entry Point i

(

We could not find documents that showed the Grade connections (fill, sampling) to be floodwater static head (approximately 7 feet). In addition, we could not find documentat the Pump manways or vent pipe could not be damaged by floating debris during floodin find information conceming tank leakage under static head of flood water, ponded rain groundwater. The Tank is single-wall steel and Refs. 8 and 9 imply that the tank may already storage tank low level setpoint was lowered to reduce le tks," Ref. 8, and "In 1987, the storage tanks was raised," Ref. 9, page 3.4-1-17. For the purposes of elanfying this discussion w assumed the vent pipe is damaged by floating debris during flooding, breaking a pipe w an opening into the tank through which water can enter.

Detection of Water in Storage Tank SP 2672 requires the storage tank f47A inventory to be checked weeldy and fuel quality Water infiltration can occur between inventory checks or during site flooding. "A quarterly approx.10 gallons of dicsci fuel from the bottom of the storage tank to remove any moisture

- accumulation is being accomplished at present "(by AWO see Ref.10). " Underground Tank Volumetric Test" is performed, frequency not given (see Ref.14). " DIESEL OIL STOR TANK LVL, LI-70

06) has setpoint of 20%. "IF level is low, NOTIFY Operations Technician to order fbel" (see Water Pumped to Supply Tanks by Transfer Pumps Diesel Oil Supply Tanks are automatically maintained level between 93% and 95 % full by automa level control operation by the oil in the tanks. AOP 2560 does not address the shutdown of transfer pumps P-47A and P-47B during/ or after flooding. The Transfer Pumps take suction at approximate '

inches above the bottom of the Storage Tank. There is no continuous monitoring for water in the Die i

Oil Stomge Tank. The only reason we could fmd for the Transfer Pumps to be removed from automatic J opention is in response to high level alarm (97%) at windows B32 and B33 on Control Room Panel C-  !

08. In this case the pump supply breaken are opened to prevent tank overflow, Ref. 7. >

Item 2:

Tank Elevation Discrepancies Elevation of Diesel Oil Stcrage Tank T47A is recorded at three different elevations. The elevations are i shown below:

Source Daenment lank Bottom Elevation (MSL)

SK-M-305,7604-M-75 Pumo Mountino Plate Elev. (MSL) l '-0" 21'-6" 25203-28406-29 2'-0" Not Shown Calculation Y-T O'-0" 22'-0" 25203-29032 Distance from bottom of tank to pump mounting plate = 20'-6" i

' Page 3 of 15 DR 0312. DOC

e ur u a i . es a esa s.e te.ust f11LL5 8Lst oAu-uaa c:>tr> I U: t UUut><t_ tut!db Signific:nca Level:

g "To provide a reliable onsite source of auxiliary power if the preferred source is lost, the unit onsite emergency generators. They are redundant, independent and separate, and ar other than that described.". FSAR 8.3.1.1. Automatic pumping from tank T47A removes the

" independent and separate" provisions required by the licensing basis documents for th .

Diesel Oil Supply Tanks are not " redundant" when operated as they are currently. A s in tank T47 A) will cause loss of both EDGs.

Since4.both Level EDGs would shut down this is a Level 1 discrepancy. The tank elevation discre Jon A. Winterhalter Tier 1 J/27/98 _

Orieinator Group patt EVALUATION

@ BASIS VALID BASISINVALID - CLOSED 0 PREVIOUSLYIDENTIFIED BY NNECo - CLOSED Basis valid.

M.J. Akins 01/28/98 Group Lead Date REVIEW AND APPROVAL Reviewed

E.A. Blocher 01/28/98 Deputy Project Dimetor Date Approved: D.L. Curry 01/30/98 Project Director Date i

Forwarded to NNECo, NEAC, and NRC: 01/30/98 Posted to WWW: 02/08/98 Date Date i

I Page 4 of 15 DR-0312. DOC

(

" ~~ ~

7 ; ,,g

,gggggg-][]j ~ ~ ]Q],[ [m

SUMMARY

OF NNECo PROPOSED CORRECTIVE ACrlON Responsr received from NNECo on 05/06/98

{ Disposition:

Itern I Disposition:

Water in Diesel Oil Storage Tank (T47A) Could Enter Diesel Oil Supply Tanks

} NU has concluded that the condition described as Item I in Discrepancy Report DR-0312 does n discrepant condition. NU has evaluated each of the design basis scenarios that NU has concluded c

issue relative to the diesel generators and the potential for failure of the units due to water intrusion underground Diesel Oil Storage Tank. We note that in each instance, the license and design bases ar

are, however, procedural improvements that can and are currently bemg incorporated as a result of the

evaluations associated with this DR. These changes, which are a direct result ofNU's evaluation of the discrepancies identified by Parsons in this discrepancy report, decrease the likelihood that the diesel g could be adversely impacted by any postulated failures associated with the evaluated scenarios. Our review beyond the specific set of conditions identified by Parsons, to consider the potential for a seis

• the underground storage tank without failure to the fuel transfer system, and the subsequent transfer ofwat laden fuel to the Diesel Oil Supply Tanks. A seismic evaluation of the underground storage tank uti

Generic Implementation Procedure (GIP) techniques for seismic evaluation of equipment in operatin (developed for resolution of USI A-46 Program) has been completed. That evaluation concludes that the tank

} proper will survive the seismic event, but that the connections to the tank at the top of the tank may be de

]

j Further discussion as to the impact of this degradation is presented in the detailed write-up that follows. The procedural improvements noted above also enhance Millstone 2's response to the seismic event, providin

)- positive assurance that neither of the diesels would be adversely impacted.

i A brief discussion of pettinent issues relating to Parsons' assumptions utilized in describing the physical fuel storage system and failure assumptions is in order, prior to NU's detailed discussion, as follows:

9-

• Parsons notes that the underground storage tank is not flood-protected - We note that, as discussed in detail later, the tank was designed to maintain leak tightness for the maximum water levels associated with the Proba Maximum Hurricane (PMH). This is evidenced by the placement of the single open-to atmosphere connectio the tank; i.e., the vent pipe with flame arrestor, which is at approximate elevation (+)25'-0", a point which higher than the maximum wave runup level. The issue of floating debris and the potential impact of that d will be discussed in detail in the body of our response. We do agree that, from a vulnerability to floating deb standpoint, the vent appears to be the weakest link in the tank and related piping system.

• Parsons notes that the two Diesel Oil Supply Tanks are cross-connected - The two dedicated Diesel Oil Supply Tanks are cross-connectable, but are not normally cross-connected.

Since there is confusion noted as to MP 2's design basis for the length of run time for both diesels follo LNP, we note that the current license basis run time for the two units is 24 houra Refer to A'mendment No. 212 t

o Facility Operating License No. DPR-65 dated January 23,1998. Subsequent to that time, one unit would be secured as deemed appropriate by the operators and those in the Emergency Onsite Facility.

• Parsons notes the issue of" water entry into the Diesel Oil Storage Tank from a single failure", on page 2 of DR. NU notes that, without a mechanism that differs from the day-to-day chronic age-related degradation of the tank present, we consider the tank a passive device, which failure does not need to be postulated in the short Page 5 of I5 DR-o312. DOC i

I

, m, w .~ c , . m .

. ~ . . - .uu_mue om-en- - iu:euwt mm I

term and which can be deemed incredible in both the long and short terms due to the lack ofmoving pI critical seal or gasketing assemblies. By a mechanism, we refer to phenomena such as a seism a LOCA with a LNP, there is no mechanism since external events, including seismic, are not postulated t concurrent with the LOCA,

)

• Parsons notes that normal groundwater level is at 5 feet MSL in the vicinity of the underground tank-- W  !

note that that level corresponds to a site average and was used in the buoyancy calculation for the Calculation Y-T, Revision 0 titled " Diesel Oil Storage Tank Foundation"), but that the original co were taken in the three locations closest to the tank show the highest recorded groundwater level was at Elevation (+) 0 feet.

• Parsons notes the conflicting information as to the elevation of the bottom of the Tank - We agree thatI are errors in our documentation of this elevation, and confirm that the actual elevation is Elevation (+)2'-7l l

e Parsons notes that the design basis flood produces levels ofstillwater of(+)18.1 feet MSL and wave

(+)25.2 feet MSL -- We correct the wave runup at to be (+)18.9 feet at the east side of the facility wher underground tank is located. (Refer to FSAR Table 2.5-1)

• Parsons notes that the tank "may alreedy leak" and cites two references supporting this conclusion -- We note that this is not the case, and that the confusion on this issue is the result of statements identified by Parso Reference 8. The Diesel Oil Storage Tank low level setpoint was actually lowered to reduce the quanti stored on site. This was done to minimize oxidation and microbial growth affecting the entire stored fuel s in the Diesel Oil Storage Tank. As stated in attached Memo MP-10401, dated 5/21/87 titled NOA 9213 -

" Diesel Generator Fails Test Because ofDegraded Puel" (CR-0487-004)"By allowing this tank to be consumed to a lower level, it would allow a fresher supply ofoil to be available for transfer to the day tanks".

Our evaluation follows:

• Extended Full Power Operation Followed By A Loss Of Normal Power, With Or Without A LOCA:

0 The water sources that need to be considered for this design basis scenario include surface runoff and/or groundwater intrusion into the Diesel Oil Storage Tank. The water intrusion paths into the tank would be the result of a long term, non-acute mechanistic degradation of the underground storage tank and/or its appurtenanc.cs.

Surveillances are performed on the contents of Diesel Oil Storage Tank T47A to ensure that diesel fuel samp drawn from the bottom of the tank contain less than 0.05% water. Procedure SP 2865, Revision 0, " Sampling and Analysis of Diesel Oil Storage Tank"(Attached) provides instructions to obtain a sample from the bottom of Tank T47A for analysis to determine the degree to which water may have accumulated and to monitor color. At least once every 31 days a sample is taken and analyzed to ensure that less than 0.05% ofwater is present in the sample. This surveillance is applicable during all modes of operation when fuel oil is in the diesel oil storage tank. Additionally, Procedure SP2613E, Revision 6, Ch 1-3, " Diesel GeneratorFuel Oil Sampling"(Attached) provides for sampling of the Diesel Oil Storage Tank T47A for off-site analysis. This is completed at least once every 92 days to meet Technical Specification Surveillance Requirements 4.8.1.1.2.b for Tank T47A. Sampling i

s also performed following any deliveries of fuel oil to Tank T47A. This surveillance is performed during all Operational Modes. No evidence of water intrusion into the Diesel Oil Storage Tank has ever been identified.

Page 6 of 15 DR-0312. DOC

\

. - - - . --- -= .. - - - _ . _ -- --_ ______

- , . ..- ~ . .

. ~ . . . . ~

. ~ ... a.un mu2,ium uso-un- m suauom m* w As rtated above, water intrusion paths into Dies:1 Oil Storege Tank T47A for this design basis s as a result oflong term, non-nute mechanistic degradation of the tank. This type of damage corrosion of the tank and/or it appurtenances (fill pipe, pump stands, vents, sampling connectio

(

cathodic protection system wes installed in 1974 to prevent such corrosion from occurring. C systems are designed to prevent tanks from corroding by reversing the naturally occurring electrolytic c produced current at the interface of the tank exterior and surrounding back-fill that can degrade tank impressed current protection system is used to protect the diesel oil storage tank. This impressed cu protection system introduces an electric current into the ground through a series of anodes that are no to the tank. Since the electric current flowing from these anodes to the tank and its appurtenan the corrosive current attempting to flow from the tank and its appurtenances, the anodes are corro the tank. The diesel oil storage tank cathodic protection system is shown on Northeast Utilitie 35028 titled, " Cathodic Protection Off-Gas Pipe, Diesel Oil & Fuel Oil Tanks, TBCCW & RBCCW M Procedure MP2720A3, Rev. 2 Ch. I titled " Cathodic Protection Maintenance" (Attached) sp maintaining and testing of the cathodic protection system The Diesel Oil Storage Tank, re voltage, current, and reference cell voltage are recorded monthly by the Technical Services Depar Corrosion Control Section.

1 The tank was procured under Bechtel Specification 7604-M-125 titled " Specification ofMiscel Assembled Tanks for the Millstone Point Company, Millstone Nuclear Power Station U ilt No. 2" To fbrther reduce the potential for corrosion, the exterior of the tank was subjected to a commercial sand and painted with a coat of bitumastic mill undercoat. The interior of the tank was hand tooled clean in accordance with SPEC-SSPC-SP2-63 and coated with diesel oil. Applicable codes and speci6 cations use procurement of the tank included NFPA No. 30 and Underwriters Laboratories No. 58 - Standards for Underground Tanks for Flammable 1 iquids.

Upon the unlikely event that water would enter the Diese! Oil Storage Tank, it would not be immed pumped to the Diesel Oil Supply Tanks when the transten pumps automatically start. Pumps P47A&B suction taken approximately 11 inches off the tank bottom. Any water entering the tank would settle to the tank bot

( and sump. The bottom eleven (11) inches of tank elevation below the pump suction point will accommodate th

,d accumulation of approximately 1150 gallons of water prior to any water being transferred to the Diesel Oil Supply Tanks. In order to have 1150 gallons or more accumulate between the water testing surveillance identified in SP 2865, an in-leakage rate averaging approximately ).5 gallons per hour over the course of th 31 days would have to occur. This isjudged to be a conservatively high flow rate following the breac tank due to a age-related corrosion induced failure.

The breaching of the tank due to corrosion would have to occur concurrently with a source of either runoffor groundwater being present which could enter the tank. Surface runoff that could enter the tank would b

' a minimum in the area of the Diesel Oil Storage Tank due to grade being covered directly above the tank w concrete pad and curbing system. Areas around the concrete above the tank are covered with a layer of Ground water levels vary greatly depending on circumstances such as seasonal variations or heavy ra Monitoring i of the groundwater levelin the vicinity of the Diesel Oil Storage Tank is not performed. Further nvestigation of the water level ofElevation (+) 5.0 feet identified in Calculation Y-T, Revision 0 titled " Diesel Oil Storage Tank Foundation", shows that this groundwater level was probably used based on information that was provided in Amendment No.1 to the License Application dated 10/27/69, A blanket statement for the entire site is presented in Amendment No.1, Section 6.0 stating "The actual water table is probably at app elevation (+) 5.0". Actual boring logs closest to the location of the Diesel Oil Storage Tank were retrieved fro r

. a ecords. The " Boring Location Plan" is shown on NU Drawing 25203-10005, Revision 0 and the " Boring re presented on NU Drawing 25203-10006, Revision 0. A review of the two referenced drawings shows that borings 2-DH4,2-DH-2, and 2-DH-9 are closest to the location of the Diesel Oil Storage Tant <

DR-o3 t2. DOC s

1 4

a

. , - - - -r. . - - - --

. - ~ - - . -

e,-m m c . uu enunctu.ut rutLdu;Ne estu-tod cNJ TO:FOJGERE NW

- The approximate groundwater elevations ct these locations which were taken in October,1966 were as follows:

Bonng Number Groundwater Level b 2-Dal (-) 7 feet 2-DR2 (-) 4 feet 2-DE9 (+) 0 feet The following must be noted when reviewing the reponed groundwater levels:

i e Reading were taken over 30 years ago.

• Changes to the site have occurred since these readings were taken including the construction of both Millstone Units 2 and 3.

• All three readings would result in the groundwater table being below the bottom of the diesel c?

storage tank by a minimum of two (2) feet and seven (7) inches.

No leakage of either water into the diesel oil storage tank or oil out of the tank has ever been identi6ed. Due additionally to the ramifications ora leak ofdiesel oil to the environme..t, the tank is subjected to a volumetric test every three years to ensure its integrity. The most recent volumetdc testing of Tank T47A was perform on 8/8/96 by Pennoni Associates Inc. EnvironTEL Division. The test results were satisfactory indicating n leakage into or out of Tank T47A. A visual inspection of the tank interior is completed every ten (10) ye Access to the underground tank is made via the 24 inch pump stand nozzles. Finally, we note that this tank is scheduled for replacement with a vaulted unit meeting current environmental standards in 2002 in accordance I with State of Connecticut rules covering underground fuel oil storage tanks.

Original Bechtel Calculation Y-T, Revision 0, titled " Diesel Oil Storage Tank Foundation" provided for an anchorage system of the Diesel Oil Storage Tank T47A to its foundation under the worst case condition of the tank being fully submerged and empty with a saturated backfill. This design ensures no gross failure ofTank

( T47A due to buoyancy effects will occur. This calculation conservatively assumed grotmd water level to be at g the top of the tank and also included a sketch indicating a water level which is the site avera8e. We believe tha this sketch is the basis for Parson's noted reference for the Elevation (+) 5.0 foot groundwater level versus the boring results in the vicinity of the tank noted above.

Based on the above discussion, NU concludes that for the unit trip from full power, with or without a LOCA case, the existing fuel forwarding system design does not present the potential for loss of either of the two (2 diesel generators.

-

• Extended Pall Power Operation Followed By Unit Shutdown In Accordance With The TS Prior To Arrival Of A Probable Maximum Hurricane (PMH) And Associated Design Basis Flood Levels:

NU agrees with the Parson's observation that, given the maximum water level that could be experienced und PMH conditions, there could be water leakage into the underground storage tank by way of the extemally exposed connections to the tank being damaged by floating debris. While these connections are protected to a '

degree from floating debris, that protection does not extend to the elevation that would be required to prote these connections in the worst case high water level event. As stated above, procedural improvements made by NU, which are a direct result of this Parsons identified scenario, will be incorporated to eliminate any potential that the diesel generators could be adversely impacted.

Page 8 of 15

, DR.0312. DOC i

___,m___ _

-~...v.

- . . - -- - = - - -- '- -- - - ~~ ^ ^ ~ ~ ~

. ~ . . . . - . .uw uit mu-- i v. r ui.e re.w l -

It is apparent that design end instaliatIon document 1 tion for the Diesel Oil Storage Tank did consider the PMH

[ llooding efTect as recognized at the time. Connections to the tank were made with seal welds that would prevent inleakage of water. The nozzles on which pumps P47A&B were mounted were provided with gaskets at their i

flanges to prevent in-leakage of water. The fdl pipe connection was provided with a threaded cap. The vent flame arrestor provided a direct opening for water intrusion due to the effects of the PMH. This vent was mounted above the expected PMH flood wave runup level eliminating the potential for water entry path during PMH floodinr3 via that pathway. NU has prepared calculation 98-ENG-02567C2 - Revison 0 titled " Diesel Oil Storage Tank PMH Flood Evaluation of Outer Steel Walf' whien evaluates the Diesel Oil Storage Tanks capability to acconunodate the effects of the PMH's hydrostatic loading and soil loading. The results of the attached evaluation concludes that the tank is capable of accommodating these loadings.

Further evidence of consideration of PMH induced water intrusion to the Diesel Oil Storage Tank is presented in questions as requested in AEC letter, Mr. Karl R. Goller, PWR Branch No. 3, Directorate ofLicensing to Mr.

Donald C. Switzer, President, The Millstone Point CompanA dated December 29,1972. Question 8.20 asked among other questions to "Also describe measures taken to prevent and detect the degradation of the fuel supply by water resulting from the effects of natural phenomena (storms, flood, hurdcane), condensation, and/or poor  ;

oil supply." The response to this question was as follows: 1 (1) There is a provision for a sampling connection on the diesel oil storage tank. Samples of the dicsci oil will be taken and tested at regular intervals to detect any degradation of the oil in the tank.

(2) in order to prevent any effects ofnatural phenomena, the vent flame arrestor on the diesel oil storage tank has been installed well above flood level wave runup elevation.

NU has concluded that the design of the Diesel 0;l Storage Tank for PMH induced flood effects meets the unit's Licensing Basis and Design Basis.

The availability of both Emergency Diesel Generators is required up to and above the Elevation 22 foot flood level. Additionally, the ability to protect a Service Water Pump Motor (one function of which is to cool a diesel g generator) and the diesel generators during the period when water level is above Elevation 22 feet is a design requirement. The following provides " defense-in-depth" evidence that, if the emergency diesel generators were lost to a PMH flood induced scenario like this one postulated by Parsons in this DR, the capability to maintain the plant in a safe shutdown condition would remain.

As described in FSAR Section 2.5.4.2.1, plant personnel can maintain the plant in a safe shutdown condition through the Probable Maximum Hurricane (PMH) when the primary objective is the removal of decay heat. The FS AR describes the incorporation of a steam driven pump and manually-positionable components into the Millstone Unit No. 2 plant design, which provide for decay heat removal without dependence on emergency power from either the offsite supplies or the diesel:

The Condensate Storage Tank provides inventory for a minimum period of 10 hours1.157407e-4 days <br />0.00278 hours <br />1.653439e-5 weeks <br />3.805e-6 months <br />. This is followed by the transfer of a reserve supply for the auxiliary feedwater system from the fire water storage tanks using the associated diesel driven fire pump. As a result, the 10 hours1.157407e-4 days <br />0.00278 hours <br />1.653439e-5 weeks <br />3.805e-6 months <br /> can be significantly extended. The fire water storage tanks are supplied by the city water system which is expected to remain pressudzed by the domestic .

water supply's diesel-driven pumps, thereby providing a virtually unlimited supply of water. The Primary Water Storage Tank capacity following restoration of power to the primary water transfer pumps would also be available ifrequired. NU notes that this method ofdecay heat removal is credited in the units license for those floods resulting in maximum flood levels in excess of 22' MSL; however, we fbsther note that this method will, in practical terms, satisfy the decay heat removal requirement for flood resulting in water elevations of any level lower than that presented by the PMH scenario. We do stress once again, however, that the unit's design and license bases do not credit this method except for the greater than 22' flood level scenario.

Page 9 of 15 DR-n312. DOC l

l l

~

, = .~ .1 1 Tig..i;;1ci,LG om C ,_ ,; 7

-;o. , - '

m.c y NU has concluded that the design and license bases associated with this scenario are satisfied with the currer,t configuration of the fuel tanks. We have further determined that it is prudent to incorporate procedure steps that will result in the operators disabling Fuel Oil Transfer Pumps P47A&B immediately prior to the expected arrival i of a PMH induced flood event. Power would not be restored until some time after the flood waters recede, and the fuel oil contained in the tank has been sampled and determined to be of the required integrity. Millstone Station has the onsite capability to complete the . appropriate set of fuel oil tests to support the determination of acceptability of this supply. The safety related and seismic dedicated Diesel Oil Supply Tanks for each of the diesel generators have sufficient stored oil to support the operation of each of these units for a minimum of two (2) days. This is sufficient time during which an alternative supply can be secured in the event that fuel stored in the underground storage tank has been determined to be unusable.

• Extended Full Power Operation Followed by a Unit Manual or Automatic Trip Due a Seismic Event Without a LOCA:

The diesel oil storage tank T-47A is a non safety-related, non-seismic underground tank containing fuel oil that is tested regularly, if a seismic event occurs, the diesel oil in the diesel oil storage tank can not be relied upon as a qualified source of fuel oil without prior testing of that fuel following the event, and potentially, the utilization of an alternative fuel oil forwarding system. )

Even though this fuel oil source cannot be specifically credited to be available following a seismic event, the possibility exists that the following worst case scenario could occur:

i

1) Diesel Oil Storage Tank T47A remains generally intact and pumps P47A&B remain capable of transferring fuel to the diesel generator day tanks.

2) Diesel oil transfer piping remains intact and Pumps P47A&B remain energized and functional.

3) Groundwater enters the degraded Diesel fuel oil storage tank following the seismic event. It is postulated that water laden fuel could potentially be transferred to the Diesel Generator day tanks.

(

The groundwater table levels in the vicinity of the underground tank following the scismic event may or may not be above the bottom of the tank. If the water table level is below tlae bottom of the tank, no water intrusion into the tank would be expected. This scenario could resuk in diesel oil flowing out from the tank to the surrounding soil. Surrounding backfill could possibly be introduced to the Diesel Oil Storage Tank inventory at the connections at the top of the tank. In-place strainers between the Diesel Oil Storage Tank T47A and Diesel Oil Supply Tanks T48 A&B would remove this sediment, and, in the process, potentially clog the strainers resulting in the termination of the transfer of fuel. The Diesel Oil Supply Tanks would continue to supply the diesels for a  !

minimum of approximately two days each at full power. Operation ofboth units at fbil power for an extended period of time is highly unlikely since there is no LOCA ongoing; however, the Diesel Oil Supply Tanks can be replenished as necessary from offsite or other onsite supplies.  !

The following considerations apply ifthe groundwater table level is at a level which is above the bottom of the diesel oil storage tank:

Diesel Oil Storane Tank intenrity is breached above water table level at the connections - This scenario would not result in water intrusion since the water table level will, in the worst case, be signi6cantly lower. Backfill l could possibly be introduced to the Diesel Oil Storage Tank inventory. Again, as discussed above, in-place

! strainers between the Diesel Oil Storage Tank T-47A and Diesel day tanks T48A&B would remove this sediment. If the sediment eventually clogs the strainers, the supply of fuel oil from Tank T47 would be terminated. The dedicated Diesel Oil Supply Tanks' fuel oil integrity would be maintained. Actions to replenish the diesel fuel oil to the Diesel Oil Supply Tanks would be taken, as appropriate.

Page 10 of 15 DR-0312. DOC i, !

a

,-q' -. -, -

- g;;;7_  ;.,a 7;,;;g;;,,ggg ~ mg.6 - ' m. l _ ~ ' ] , _ , ,, ~]l Diesel Oil Storage Tank integrity is breached below water table levtl - Technical Evaluation M2-EV-98-0083, Revision 0, titlcd StructuralIntegrity ofDiesel Generator Fuel OilStorage Tunk Under Earthquake Loads, provides a basis for concluding that the underground tank proper would survive the seismic event and maintain it's integrity. This attached evaluation, completed using GlP approved techniques, concluded that any failure of the tank is likely to occur at the connections to the tank at its top. Therefore, there would not be in-leakage groundwater since in the worst case,it is significantly below the top of the tank.

While an acceptable design basis is pmvided for this condition, NU has determined that it is prudent to incorporate procedure steps that will result in the operators disabling Fuel Oil Transfer Pumps P47A&B immediately following the occurrence of a scismic event. Power would not be restored until some time after the seismic event, when the fuel oil contained in the tank has been sampled and determined to be of the required integrity.

The above, when combined with

' the approximate 1150 gallon " margin" that exists in the diesel oil storage tank to accommodate minor water intrusion, e

the " margin" that also exists in the Diesel Oil Supply Tanks T48 A&B, and, 4-actions that would likely occur as a resuk of recommendations made to remove power from the pumps by the Technical Support Team which would be assembled in the Emergency Operations Facility, provides reasonable assurance that the transfer ofg oundwater from the Diesel Oil Storage Tank to the Diesel Oil Supply Tanks and then, to the Emergency Diesel Generators would not have occurred, historically. Given  ;

the relatively lightly loaded diesel generators under the postulated scenarios, the fuel consumption rate will be i significantly reduced resulting in additional time for the unit operators or those in the EOF to analyze those i systems important to the maintenance of safe shutdown. Past Simulated Emergency Response Drills at NU have demonstrated a high regard and degree of attention to the integrity of those systems necessary to maintain a supply of AC power.

I NU concludes that, for the seismic event case, the transfer of water-laden fuel to the diesel generator day tanks would not occur due to the maintained structural integrity of that portion of the underground tank system essential to the maintenance of near leak-tightness. Funher, there is sufHeient time available that, should some  ;

amount of water be introduced into the tank, the appropriate expedise would take actions to recommend that the tank be isolated from the dedicated Diesel Oil Supply Tanks.

The Design Basis Scenarios discussed above bound all other cases, including those that could occur with the unit previously shut down.

We note that Diesel Oil Storage Tank (T47A), including transfer pumps (P47A&B), and associated pipin8 h ave, since unit startup, been designated as non-CategoryI systems at Millstone Units No. 2. Diesel oil has been identified as being automatically transferred from the underground diesel oil storage tank to the diesel oil supply tanks.

NU concludes that Issue 1 of this DR does not represent a discrepant condition. Therefore, Sigmficance Levels do not apply. However, as a direct result ofParsons' identification of those issues discussed in this DR, NU is taking actions to provide even greater assurance that the underground fuel oil storage tank, its appurtenances and the manner in which the system is operated will not challenge the basis for inclusion of redundancy and independence in the design of the onsite electricsl power supplies; i.e., the diesel generators.

Page iI of !$

DR-0312. DOC

Fu 6di m 4-e Au r au i.re.x uit mLLSm ble-o n-caos IU Uutke **. : o Itern 2 Disptsitinn:

Tank Elevation Discrepancies The elevation discrepancies reported in the documents listed below will be investigated L)esign documents will be revised, as appropriate, to correct the discrepancies.

Source Document Tank Hottom Elevation (MSL) Pump Mounting Plate Elev. (MSL)

SK-M-305, 7604-M-75 l '-0" 21'-6" 25203-28406-29 2 '-0" Not Shown Calculation Y-T O'-0" 22'-0" 25203-29032 Distance from bottom of tank to pump mounting plate = 20'-6" FSAR 2.5.4.2.5 Invert Elevation l'-8" Field walkdowns have been performed and dimensions were taken to determine the elevation of the tank bottom.

This elevation was determined to be Elevation (+)2'-7" This places the Pump Mounting Plate at Elevation

(+)23'-1" and the flame arrestor on top of the vent pipe above Elevation (+)25'-0" NU concludes that the drawing errors depicting the actual tank bottom elevation constitutee a Significance Level 4 discrepancy.

Conclusion:

NU has concluded that the issues reported in DR-0312 has identified a CONFIRMED SIGNIFICANCE LEVEL 4 Condition that requires correction.

Item 1 - Water in Diesel Oil Storage Tank (T47A) Could Enter Diesel Oil Supply Tanks Based on the above, NU has concluded that the license and design bases for MP 2's emergency onsite power supplies including fuel supplies are met. However, Northeast Utilities is amending procedures such that there is a D greater degtee of assurance that electrical power will be removed from the P47A&B transfer pumps prior to the transfer of p otential water-laden fuel, should these pumps remain operable for any of the evaluated scenarios.

Item 2 - Ted Elevation Discrepancies Based on the above, Northeast Utilities concludes that a Level 4 Discrepancy does exists. Design documents will be revised, as appropriate, to correct the discrepancies.

f l

l Page 12 or 15 DR-o312. DOC

em-m ,n m ai r nuwaam mLL5 un- 6m-uw-em IU4 00GERE htt

• M COMMENT ON NNECo PROPOSED CORRECTIVE ACTION Specific Comments:

Acceptable Leakage Rate:

Under Disposition subheading, " Extended Full Power Operation Followed By A Loss Of Normal Power, With Or Without A LOCA," an allowable leakage rate was discussed. The maximum leak rate was calculated in which accumulated water would not reach the level of the pump suction between the monthly sampling intervals. The calculated rate was 15 gallons per hour and was " judged to be a conservatively high flow rate following the breaching of the tank due to an age-related, corrosion induced failure."

Parsons does not agree that this leak rate is conservative. For discussion purposes, the orifice diameter required to allow a leak of 1.5 gallons per hour was calculated assuming a vertical, sharp-edged circular orifice with an external water head of 2 feet. The required orifice diameter for that leak rate is one millimeter. This is a very small perforation. Alternately, a 1/4 inch diameter orifice under a 3-inch head of water will allow 97 gallons per hour into the tank. For a tank with a diameter of 10 feet and length of 44 feet a one millimeter diameter perforation is not judged to be a conservatively large Icak.

Corrosion-induced failure due to external or internal degradation mechanisms could occur anytime and hasn't been sufficiently addressed. Examples of such mechanisms are external corrosion due to coating aging fhilure/ holidays or higher than normal local electrolytic corrosive cells. Internal corrosion due to microbiological induced corrosion (mic) of the uncoated interior also requires consideration.

As stated by NNECo. the groundwater elevations are not monitored in the tank area so the average and seasonal groundwater elevations are not known. A loose Crade connection could leak under rainfall runoff build-up of several inches on the concrete slab over the tank. Also, asphalt paving is not impervious. Non-flood water sources could come from several paths, none of which can be discounted without a means to monitor / measure.

Calculation 98-ENG-02567C2 Implications:

b Under Disposition subheading " Extended Full Power Operation Followed By Unit Shutdown In Accordance With The TS Prior To Arrival Of A Probable Maximum Hurricane (PMH) And Associated Design Basis Flood Levels," the new calculation 98-ENG 02567C2 was discussed. The tankwas shown to withstand the loading a resulting from flooding above grade levels. The analysis used the new tank wall thickness of 3/8 inch. The maximum calculated Actual / Allowable stresses were shown to vary from 92.9 to 95.6 percent (tables on page 7 of 50). These stresses are based on the unverified assumption that the tank retains its original wall thickness of 3/8 inch. Without the benefit of tank wall thickness measurements, we cannot agree that the tank will resist the forces from flooding.

Even with cathodic protection, corrosion can't be discounted. Internal tank inspections will not discover external tank corrosion. The effectiveness of the cathodic protection system's actual protection of the tank cannot be proven by operational testing of the cathodic protection system itself. The proof of protection would be tank wall thickness measurements. Consider what the Actual / Allowable stress ratios would be if the corroded tank wall was actually 1/4 inch innead of the original 3/8 inch.

Regardless, NNECo's proposed revision to procedures disabling the fuel oil transfer pumps prior to arrival of PMH induced flooding will prevent problems of water transfer from tank failure in this case. Parsons considers disabling of the automatic fuel oil transfer interlock to maintain the licensing basis for" independent and separate" to be a significance level 3 discrepancy.

Page 13 of 15 DR-03 t2. DOC

m u , .m . . . . . ..m...- . .u.w i u,. mu-oo.3 cau, iu.r - mm . u Ganeral Cemments: "

I i

The NNECo Disposition proposed procedural revisions that would minimize the risk association with loss of

( onsite power due to flooding or seismic events. These are low probability, extreme events. Parsons believes that there is a greater risk from corrosion-induced leakage which could occur anytime and we feel that it wasn't sufficiently addressed. The fact remains that the tank. which is a non-safety component, is still connected, automatically controlled makeup pumps, to both of the Emergency Diesel Generator trains. The proposed disposition still violates the " independent and separate" requirements of the licensing documents.

The disposition states that the tank is scheduled for replacement in the year 2002. This concern, therefor finite life span. However, during the time period from 1998 to 2002 the plant is exposed to signi6 cant (and increasing) risk considering of the age of the tank and cumulative effects of corrosion which might be o

Conclusion:

item 1:

We have reviewed the NNECo Disposition and find that item I remains discrepant. In the absence of a wall thickness survey and a means to detect inleakage ofwater between monthly sampling, and the fact that durin I

non emergency (flood / seismic) conditions, the pumps remain on automatic level control, we maintain that this condition is discrepant to Significance Level 1. The reliability of the both EDGs could be degraded by a non-safety component.

Item 2:

This item will be considered closed after the revised documents or Change Notices have been reviewed.  ;

(

i Prepared: M.J. Akins 06/17/98 Group Lead Date Reviewed: E.A. Blocher 06/17/98 Deputy Project Director Date Anoroved: D L. Curry 06/17/98 Project Distetor Date Fo warded to NNECo, NEAC, and NRC:06/1988 and Posted to WWW:06/21/98

[t Date Page 14 of 15 DR-0312. DOC

^

, ~. ;~ ; . .~ . . ;

. . : . . . _3 , ,6m., , e ,_ ;m...m, . . . . - - Mia, 1

I FINAL RESOLUTION

(

Open: Item 1 unresolved.

Item 2 remains open pending review ofdrawing and calculation changes.

Note; DR-0721 also discusses FSAR statements about diesel oil storage tank withstanding flo E. A. Blocher b Deputy Project Director Date

(

l l

DR 0312. DOC

,~2

.. . - ~ -~ - - - - '-- ~^' ~ ^

! . JtN- 19-90 14 : 0@ '- FRCM:PE&CGI MILLSTONE 610-G55-2509 10:FOUGERE PmE:Od PARSONS POWER GROUP INC. 2675 Morgantown Road, Reading, PA l

ICAVP MILLSTONE UNIT 2 19607

( DISCREPANCY REPORT (610) 855 2000

• FAX: (610) 855-2509 l

DR NUMBER: DR-0312 DR TITLE: Water in Diesel Oil Storage Tank (T47A) Could Enter Diesel Oil Supply Tanks REVISION: 2 ISSUE DATE: t/27/98 l ORIGINATING GROUP: Tier I SIGNIFICANCE LEVEL: 1 '

l DISCREPANCY

References:

l

1. AOP 2560 Storms, High Winds, and High Tides

{

2. FSAR 8.3 Emergency Generators l

3. OP 2346B Diesel Fuel Oil System

4. SP 2672 Sampling and Inventory ofDiesel Oil Storage Tank, T47A

5. 25203-29032 Diesel Oil Storage Tank T-47A

6. 7604-M-125 Miscellaneous Shop Assembled Tanks t 7 ARP 2590F Emergency Operating Procedures and Guidelines, Alarm I

( Response for Control Room Panel C-08

8. Figure 7.3 "10CFR50.59 Safety Evaluation Screening", prepared to support -

PDCR 2-108-92, dated 8/27/97.

9 MP2-DG DBDPackage - Diesel Generator

10. NRC Letter Docket / License: 50-336/DPR-65

11. OP 2346A Emergency Diesel Generators

12. SP 2613E Diesel Generator Fuel Oil Sampling

13. FSAR 4.2.2 Codes Adhered To

14. MP 2721J Periodic Inspection ofUnit 2 Tanks

15. 91-DOP-813-ES, Rev. 3, MP2 EDG Operating Time with 24,000 Gallons ofDiese! Fuct Oil Available at a Continuous Rated Load of 2750 kW Fuel Consumption

16. IPEEE December 29,1995, MP2, Response to Generic Letter 88-20, Supplements 4 and 5, Individual Plant Examination for External Events - Summary Report

Background:

Diesel Oil Storage Tank T47A is a 25,000-gallon underground, horizontal' cylindrical tank. The tank is single-wall, carbon steel coated with "bitumastic mill undercoat" (exterior) and tm interior coating. The tank was installed around 1972. A cathodic protection system (rectifier PO29A, zinc anodes, and test station) was added after October 1974.

l Transfer Pumps P-47A and P-47B are installed in individual 24-inch diameter steel riser pipes connected to nozzles on the top of the tank. Bottom of tank = 1'-0"(plus or minus) depending upon which drawing l-Page 1 of 15 DR-0312. DOC .

i

(

.- _. ~ . -- _ _ . . - - . , - . - - . =_ . .~. . . - - ~ . - .

Ju 4-19-98 14 WS FRUM: PEF.CGI f1ILLSTCtE 6m-ms-pe,09 IO:FO WERE I pnpe n3 is used Top of tank = Il'(plus or minus). Grade = 14'-8" Top of pump manways is 22' (plus or minus). Other connections extending to or above grade are fdl, sample, vent, and level pipes. See Ref. 5.

The storage tank is not seismically qualified nor flood-protected. A sufficient quantity of oilis contained in the Diesel Oil Supply Tanks to operate the diesel (s) for approximately 7 days. Each Supply Tank l (13,500-gallon nominal capacity) must contain 12,000 gallons. See Ref. 2. Pump (s) P-47A and P-47B

" start (s) automatically when the respective level decreases below 95% and stops when tank level increases above 95%." See OP-2346B, (Setpoint Change #2-88-022,in process, changed minimum level to 93%.)

"Two EDGs operate for 8 hours9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br />, then one EDG operates for the remaining length of time. Fuel consumption is 3.6 gpm for each EDG at 2750 KW. During a LOCA with a LNP (loss of normal power) the two day tanks (Supply Tanks) are cross-connected." Ref.15, dated April 1997. (The current license states one EDG for one hourinstead of 8 hours9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br /> with the cther EDG continuous). The supply tanks are cross-connected via valve 2 FO-83, " Fuel Oil Supply Header Cross-tie" Discrepancy:

Item 1:

Failure Mode Water entering the underground Diesel Oil Storage Tank (T47A) will be automatically pumped into the Diesel Oil Supply Tanks T48A and T48B.

With water entry into the Diesel Oil Storage Tank from a single failure, during a LOCA in an LNP, it j ( would take less than 3 hours3.472222e-5 days <br />8.333333e-4 hours <br />4.960317e-6 weeks <br />1.1415e-6 months <br /> to pump 6 inches of water into each Supply Tank. The fuel supply pipes ar+

6 inches above the bottom of the supply tanks. Water will shut down both EDGs a short time later.

(With the cutrently licensed scenario i.e. one EDG for one hour, it would take up to six hours, assuming the tanks contribute equally, for the continuously running diesel to shut down. At that time the second l EDG would not restart or ifit starts it would not run for an extended length of time.)

i If only one EDG were to started, it would run for 3 hours3.472222e-5 days <br />8.333333e-4 hours <br />4.960317e-6 weeks <br />1.1415e-6 months <br /> before automatically shutting down. At that i

time the other EDG would be started. It too would run for 3 hours3.472222e-5 days <br />8.333333e-4 hours <br />4.960317e-6 weeks <br />1.1415e-6 months <br /> before automatically shutting down.

Water Sources .

Water sources are surface runoff, groundwater (normal or elevated), or flood water. Probable Maximum -

Precipitation can cause surface flooding between 15.5 to 16.2 feet MSL (mean sea level), Ref 16.

Normal groundwater level is elevation 5 feet MSL (halfway up the tank). The contai=w turbine, and ,

auxiliary buildings are protected from flooding. The design flood produces water levels up to elevation

18.1 feet MSL, stillwater and 25.2 feet MSL, wave runup,'Ref.16 for PMH (Probable Maximum

! Hurricane). According to reference 1, the EDGs will be operated dudng flooding up to and above a i water level of elevation 22'.

1 i

a

'l

,( Page 2 of' 15 l DR-0312. DOC k

__ .__ __ _ . - - . - - -- > - - ~ ^---~' ' "' ~ " " " ~

l JLs4-19-98 14: 07 FROM PESCGI tilLLSTOts? 610-HS5-P5r09 10:FULbERE PRGE:04 Water Entry Point We could not find documents that showed the Grade connections (fill, sampling) to be watertight under floodwater static head (approximately 7 feet). In addition. we could not find documentation to show that i LDS.UMUIR.RunyAys. ply 9ntipe, pot;ljAot by darnaged by, floating debrisi duJ ng flooding. We could not Uno inloimation concermng tank icanage unuct blupc ucou us sivuu wetes, g>ussucu easistau sussutt, us groundwater. The Tank is single-wall steel and Refs. 8 and 9 imply that the tank may already leak. "The i storage tank low level setpoint was lowered to reduce leaks," Ref. 8, and "In 1987, the concern for leaky {

storage tanks was raised," Ref. 9, page 3.4-I-17. For the purposes ofclarifying this discussion we have '

assumed the vent pipe is damaged by floating debris during flooding, breaking a pipe weld, and providing an opening into the tank through which water can enter.

Detection of Water in Storage Tank SP 2672 requires the storage tank T47A inventory to be checked weekly and fuel quality testing monthly.

Water infiltration can occur between inventory checks or during site flooding. "A quarterly pumping of approx.10 gallons of diesel fuel from the bottom of the storage tank to remove any moisture accumulation is being accomplished at present "(by AWO see Ref.10) " Underground Tank Volumetric Test" is performed, frequency not 8i ven (see Ref.14). " DIESEL OIL STOR TANK LVL, LT-7004"(C-

06) has serpoint of 20%. "IF level is low, NOTIFY Operations Technician to order fuel " (see Ref. 7).

l Water Pumped to Supply Tanks by Transfer Pumps '

Diese! Oil Supply Tanks are automatically maintained level between 93% and 95 % full byautomatic level control operation by the oilin the tanks. AOP 2560 does not address the shutdown of transfer

[ pumps P-47A and P-47B during/ or after flooding. The Transfer Pumps take suction at approximately 11 inches above the bo.ttom of the Storage Tank. There is no continuous monitoring'for water in the Diesel Oil Storage Tank. The only reason we could find for the Transfer Pumps to be removed from automatic g operation is in response to high level alarm (97%) at windows B32 and B33 on Control Room Panel C-

08. In this case the pump supply breakers are opened to prevent tank overflow, Ref. 7 j Item 2:

Tank Elevation Discrepancies Elevation of Diesel Oil Storage Tank T47A is recorded at three different elevations. The elevations are shown below:

Source Document Tank Bottom Elevation (MSL) Pumo Mountina Plate Elev. (MSL)

SK-M-305, 7604-M-75 l '-0" 21 *-6" 25203-28406-29 2'-0" i Not Shown Calculation Y-T O'-0" 22'-0" 25203-29032 Distana from bottom of tank to pump mounting plate - 20'-6" l

l I rs DR-0312. DOC Page 3 of 15 i

JtN-19-S8 14: 08 FRt11:fE&CGI MILLSTDE F.t n-FM-25acy TO:FOUGLTE N:Ei5 Significance Level:

(

i "To provide a reliable onsite source of auxiliary power if the preferredwosource is los l

i other than that described." FSAR 8.3.1.1. Automatic pum

" independent and separate" provisions required by the licensing .basis The documen in tank T47A) will cause loss of both EDGs. Diesel Oil Supply Tanks Since Level 4. both EDGs would shut down this is a Level i discrepancy. The tank elevation Jon A. Winterhalter Tier 1 L/27/98 Orieluntor

_ Group gge EVALUATION

@ BASIS VALID 0 DASIS INVALID - CLOSED 0 PREVIOUSINIDENTIMED BY NNECo- CLOSED Basis valid. 'i

.(

M.J. Akins ~

01/28/98

) Group Lead Date REVIEW AND APPROVAL Reviewed: E.A. Blocher 01/28/98 Deputy Project Director Date Approved: D.L. Curry 01/30/98__

Project Director Date .

1 Forwarded to NNECo, NEAC, and NRC: 01/30/98 1 Posted to WWW: 02/08/98 Date Date Page 4 of15 DR-0312. DOC

JLR F t 9-98 t 4: 08 FRGM:fTRrEI MILLSTOrE 6106-N TO Wf1 N N: fE

= _

SUMMARY

OF NNECo PROPOSED CORRECTIVE ACT ON

~

Response received from NNECo on 05/06/98 1

Disposition:

Item 1 Disposition: l Water in Diesel Oil Storage Tank (T47A) Could Enter Diesel Oil Supply 1. .

NU has concluded that the condition described as Item I in Discrepancy Report DR-031)i discrepant condition NU has evaluated each of the design basis scenarios that NU has conclu issue relative to the diesel generators and the potential for failure of the units ndue to water intrusi l

o the underground Diesel Oil Storage Tank. We note that in each instance, the license and design bas are, however, procedural improvements that can and are currently being incorporated as a result of the i '

evaluations associated with this DR. These changes, which are a direct result of NU's evaluation of the discrepancies identified by Parsons in this discrepancy report, decrease the likelihood that the could be adversely impacted by any postulated failures associated with the evaluated scenarios.

beyond the specific set of conditions identified by Parsons, to consider the potential for a the underground storage tank without failure to the fuel transfer system, and the subsequent transfer '

laden fuel to the Diesel Oil Supply Tanks. A seismic evaluation of the underground storage tan Generic Implementation Procedure (GTP) techniques for seismic evaluation of equipment in ope (developed for resolution of USl A-46 Program) has been completed. That evaluation concludes that the proper will survive the seismic event, but that the connections to the tank at the top of the tank may Further discussion as to the impact of this degradation is presented in the detailed write-up that follows. T procedural improvements noted above also enhance Millstone 2's response to the seismic event, provid '

positive assurance that neither of the diesels would be adversely impacted. '

+

A briefdiscussion ofpertinent issues relating to Parsons' assumptions utilized in describing the physical fu storage system'and failure assumptions is in order, prior to NU's detailed discussion, as follows:

1 e Parsons notes that the underground storage tank is not flood-protected - We note that, as discussed in detail later, the tank was designed to maintain leak tightness for the maximum water levels associated with the Probab Maximum Hurricane (PMH). This is evidenced by the placement of the single open-to-atmosphere connectio the tank; i.e., the vent pipe with flame arrestor, which is at approximate elevation (+)25'-0", a point which is higher than the maximum wave runup level. The issue offloating debris and the potential impact of that debri will be discussed in detail in the body of our response. We do agree that, from a vulnerability to floating de standpoint, the vent appears to be the weakest link in the tank and related piping system.

• Parsons notes that the two Diese! Oil Supply Tanks are cross-connected - The two dedicated Diesel Oil Supply Tanks are cross-connectable, but are not normally cross-connected.

Since there is confusion noted as to MP 2's design basis for the length of run time for both diesels follo LNP, we note that the current license basis run time for the two units is 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />. Refer to Amendment No. 212 t

o Facility Operating License No. DPR-65 dated January 23,1998. Subsequent to that time, one unit would be secured as deemed appropriate by the operators and those in the Emergency Onsite Facility.

• Parsons notes the issue of" water entry into the Diese! Oil Storage Tank from a single failure" on page 2 of t DR. NU notes that, without a mechanism that differs from the day-to-day chronic age-related degradation of the l tank present, we consider the tank a passive device, which failure does not need to be postulated in the short l

Page 5 of15 DR-0312.IX)C

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

JtN-19% N:H9 Ha n:PEECGI Nit LSTOIE G10-OS5 E o9 TO: F0 FERF . PFA:o?

i term and which can be deemed inc[e$fible in both the long and shortarterms s and due to t critical seal or gasketing assemblies. By a mechanism, we refer to phenomena su a LOCA with a LNP, there is no mechanism since external events, including seismic, ar concurrent with the LOCA.

• Parsons notes that normal groundwater level is at 5 feet MSL in the vicinity of the unde note that that level corresponds to a site average and was used in the buoyancy calcu Calculation Y T, Revision 0 titled " Diesel Oil Storage Tank foundation"), but that t were taken in the three locations closest to the tank show the highest recorded groundwater lev Elevation (+) 0 feet.

• Parsons notes the conflicting information as to the c!cvation of the bottom of the Tank --

are errors in our documentation of this elevation, and confirm that the actual elevation is E .

(+)25.2 feet MSL -- We correct the wave runup at to be (+

underground tank is located. (Refer to FSAR Table 2.5-1)

+ Parsons notes that the tank "may already leak" and cites two references supporting this conclu that this is not the case, and that the confusion on this issue is the result of statements ide Reference 8. The Diesel Oil Storage Tank low level setpoint was actually lowered to reduc in the Diesel Oil Storage Tank. As stated in attached Memo MP-

" Diesel Generator Fails Test Because of Degraded Fuel" (CR-0487-004) "By allowing this tank t to a lower k.et, it would allow a fresher supply of oil to be available for transfer to the day tanks"

Our evaluation follows

i i

• Extended Full l ower Operation Followed By A Loss OfNormal Power, With Or Without A LOCA:

The water sources that need to be considered for this design basis scenario include surface runoff and/o groundwater intrusion into the Diesel Oil Storage Tank. The water intrusion paths into the tank would be the result of a long term, non-acute mechanistic degradation of the underground storage tank and/or its appurtenances.

\'

Survelibnces are performed on the contents ofDiesel Oil Storage Tank T47A to ensure that d drawn from the bottom of the tank contain less than 0.05% water. Procedure SP 2865, Revis and Analysis ofDiesel Oil Storage Tank" (Attached) provides instructions to obtain a sample from th Tank T47A for analysis to determine the degree to which water may have accumulated and to monit '

least once every 31 days a sample is taken and ana.lyzed to ensure that less than 0.05% of wate sample. This surveillance is applicable during all modes of operation when fuel oil is in the diesel tank. Additionally, Procedure SP2613E, Revision 6, Ch 1-3, "Diesci Generator Fuel Oil Samplin  ;

provides for sampling of the Diesel Oil Storage Tank T47A for off-site analysis. ' This is completed  !

every 92 daya to meet Technical Specification Surveillance Requirements 4.8.1.1.2.b for Tank T47  !

is also performed following any deliveries offbel oil to Tank T47A. This surveillance is perform Operational Modes. No evidence of water intrusion into the Diesel Oil Storage Tank has ever been idenI ,

l Page 6 of 15 DR-o312. DOC l

t m , ._ _ __ _ _ _ _ _ _ _ _ . __ _

JUN - I P'Jd im 10 FROl1:PEACGI NILLSTCrE M H-rm-PW ind UUME PTE' OB mc _ _ -

___m As stated above, water intrusion paths into Diesel Oil Storuge Tank T47A for this m

ould be corrosion of the tank and/or its appurtenances , ..A (fill pipe cathodic protection system was installed in 1974 to prevent such corrosion .

n from syst ems are designed to prevent tanks from corroding by seversing the naturally occur picduced current at the interface of the tank exterior and surrounding back-fill An that can impressed current protection system is used to protect the dieseleoil storage current tank Thi protection system introduces an electric cuttent into the ground through a series of a to the tank. Since the electric cunent flowing from these anodes to the tank an and tha corrosive current attempting to flow from the tank and its appurtenances, an the a 35028 titled, " Cathodic Protection Off Gas Pipe, Diese! . ..

Procedure MP2720A3, Rev. 2 Ch I titled " Cathodic Protection Maintenance"(A maintaining and testing of the cathodic protection system. The Diesel Oil Storage voltage, current, and reference cell voltage are recorded monthly by the Technical Servic L Corrosion Control Section. ,

The tank was procured under Bechtel Specification 7604-M-125 titled " Specificat Assembled Tanks for the Millstone Point Company, Millstone Nuclear Power Station Unit No. 2" To further reduce the potential for corrosion, the exterior of the tank was subjected to a com and painted with a coat of bitumastic mill undercoat. The interior of the tank was hand tooled cle accordance with SPEC-SSPC-SP2-63 and coated with diesel oil. Applicable codes and specific procurement of the tank included NFPA No. 30 and Underwriters Laboratories No. 58 - Standards for '

Underground Tanks for Flammable Liquids.

Upon the unlikely event tiiat water would enter the Diesel Oil Storage Tank, it would not pumped to the Diesel Oil Supply Tanks when the transfer pumps automatically start. Pumps e P47A&B taken approumately 11 inches off the tank bottom. Any water entedng the tank would sett!c to the t and sump. The bottom eleven (11) inches of tank elevation below the pump suction point will accom g accumulation of approximately 1150 gallons of water prior to any water being transferred to the Diesel O Supply Tanks. In order to have 1150 gallons or more accumulat between the water testing surve identified in SP 2865, an in-leakage rate averaging approximately 1.5 gallons per hour over the co 31 days would have to occur. This is judged to be a conservatively high flow rate following th tank due to a age-related corrosion induced failure.

The breaching of the tank due to corrosion would have to occur concurrently with a source ofeither runo groundwater being present which could enter the tank. Surface runoff that could enter the tank a minimum in the area of the Diesel Oil Storage Tank due to grade being covered directly above the ta concrete pad and curbing system. Areas around the concrete above the tank are covered with a Ground water levels vary greatly depending on circumstances such as seasonal variations or he Monitoring of the groundwater levelin the vicinity of the Diese! Oil Storage Tank is not performed.

investigation of the water level ofElevation (+) 5.0 feet identified in Calculation Y-T, Revision 0 title Oil Storage Tank Foundation", shows that this groundwater level was probably used based on informa was provided in Amendment No. I to the License Application dated 10/27/69 A blanket statement for the entire site is presented in Amendment No.1, Section 6.0 stating "The actual water table is probably elevation (+) 5.0". Actual boring logs closest to the location of the Diesel Oil Storage Tank were retrie records. The " Boring Location Plan" is shown on NU Drawing 25203-10005, Revision 0 and the "

l are presented on NU Drawing 25203-10006, Revision 0. A review of the two referenced drawings shows tha borings 2-DH-1,2-DE2, and 2-DH-9 are closest to the location of the Diesel Oil Storage Tank.

Page 7 of t5 DR o312 DOC l

I

JI N-19-9A 14: 11 FPO15 PE9.CGI titLt STOtE 610 fE5 P5m TO ff m ERE rm :09 The approxhnate follows. groundwater elevations at these Incations which were taken in Octo Boring Number Groundwater Level 2-Dil-1

(-) 7 feet 2-Dil-2 (-) 4 feet 2-011-9 (4 ) 0 feet The following must be noted when reviewing the reported groundwater levels:

• Reading were taken over 30 years ago.

• Changes to the site have occurred since these readings were taken including the constr both Millstone Units 2 and 3.

• All three readings would result in the groundwater table being below the bottom of the dies storage tank by a minimum of two (2) feet and seven (7) inches.

No leakage of either water into the diesel oil storage tank or oil out of the tank has ever been id additionally to the ramifications of a leak ofdiesel oil to the environment, the tank is subjected to test every three years to ensure its integrity. The most recent volumetric testing ofTank T47A was p on 8/8/96 by Pennoni Associates Inc. EnvironTEL Division. The test results were satisfactory leakage into or out of Tank T47A. A visual inspection of the tank interior is completed ever Access to the underground tank is made via the 24 inch pump stand nozzles. Finally, we note tha scheduled for replacement with a vaulted unit meeting current environmental standards in 2002 in with State of Connecticut mies covering underground fuel oil storage tanks.

Original Bechtel Calculation Y T, Revision 0, titled " Diesel Oil Storage Tank Foundation" provide anchorage system of the Diesel Oil Storage Tank T47A to its foundation under the worst case c'ondition tank being fully submerged and empty with a saturated backfill. This design ensures no gross failure T47A due to buoyancy effects will occur. This calculation conservatively assumed ground water level t g the top of the tank and also included a sketch indicating a waterlevel which is the site average. We b this sketch is the basis for Parson's noted reference for the Elevation (+) 5.0 foot groundwater level ve boring results in the vicinity of the tank noted above.

Based on the above discussion, NU concludes that for the ud.t trip from full power, with or without a LOCA case,generators.

diesel the existing fuel forwarding system design does not present the potential for loss of either o

• Extended Full Power Operation Followed By Unit Shutdown In Accordance With 1he TS Prior To Arriva A Probable Maximum Hurricane (PMH) And Associated Design Basis Flood Levels:

NU agrees with the Parson's observation that, given the maximum water level that could be exper PMH conditions, there could be water leakage into the underground storage tank by way of the exposed connections to the tank being damaged by floating debris. While these connections are prote degree t from floating debris, that protection does not extend to the elevation that would be required hese connections in the worst case high water level event. As stated above, procedural improvem made by NU, which are a direct result of this Parsons identified scenario, will be incorporated to el potential that the diesel generators could be adversely impacted.

Page is or 15 DR-o312. DOC

Jut 4-19-98 14 : 12 FROMtPEECGI Mit.l5 TOTE 610-Oss-PSm l In:FT W FRF PnGE110 i It is apparent that design and installation documentation for the Diesel Oil Storage Tank d I llooding effect as recognized at the time. Connections to the tank were made with sea inleakage of water. The nozzles on which pumps P47A&B were mounted wereerprovided wi flanges to prevent in-leakage of water, The fill pipe connection was provided with a threaded c flame-arrestor provided a direct opening for water intrusion due to the effects of the PMH. This ven mounted above the expected PMH flood wave runup level eliminating the potential for wa PMH flooding via that pathway. NU has prepared calculation 98-ENG-02567C2 - Revison 0 title Storage Tank PMH Flood Evaluation ofOuter Steel Wall" which evaluates the Diesel Oil Storage T i capability to accommodate the effects of the PMH's hydrostatic loading and soil loading. The resul attached evaluation concludes that the tank is capable of accommodating these loadings.

Further evidence of consideration of PMH induced water intrusion to the Diesel Oil Storage Ta questions as requested in AEC letter, Mr. Karl R. Goller, PWR Branch No. 3, Directorate ofLicen Donald C. Switzer, President, The Millstone Point Company, dated December 29,1972. Question 8.20 asked among other questions to "Also describe measures taken to prevent and detect the degradatio by water resulting from the effects of natural phenomena (storms, flood, hurricane). condensa oil supply." The response to this question was as follows:

(1) There is a provision for a sampling connection on the diesel oil storage tank. Samples of the diesel oi taken and tested at regular intervals to detect any degradation of the oil in the tank.

(2) in order to prevent any effects of natural phenomena, the vent flame arrestor on the diesel oil sto has been installed well above flood level wave runup elevation.

NU has concluded that the design of the Diesel Oil Storage Tank for PMH induced flood effects meets the uhit's Licensing Basis and Design Basis. "

i The availability of both Emergency Diesel Generators is required up to and above the Elevation 22 foot flood level. Additionally, the ability to protect a Service' Water Pump Motor (one function ofwhich is to cool a diesel N generator) and the diesel generators during the period when water level is above Elevation 22 feet is a desig requirement. The following provides " defense-in-depth" evidence that, if the emergency diesel generators were lost to a PMH flood induced scenario like this one postulated by Parsons in this DR, the capability to main the plant in a safe shutdown condition would remain.

As described in FS AR Section 2.5.4.2.1, plant personnel can maintain the plant in a safe shutdown condition through the Probable Maximum Hurricane (PMH) when the primary objective is the removal of decay heat. The FSAR describes the incorporation ofa steam-driven pump and manually-positionable components into the Millstone Unit No. 2 plant design, which provide for decay heat removal without dependence on emergency power from either the offsite supplies or the diescis.

The Condensate Storage Tank provides inventory for a minimum period of 10 hours1.157407e-4 days <br />0.00278 hours <br />1.653439e-5 weeks <br />3.805e-6 months <br />. This is followed by the transfer of a reserve supply for the auxiliary feedwater system from the fire water storage tanks using the associated diesel driven fire pump. As a result, the 10 hours1.157407e-4 days <br />0.00278 hours <br />1.653439e-5 weeks <br />3.805e-6 months <br /> can be significantly extended. The fire water storage tanks are supplied by the city water system which is expected to remain pressurized by the domestic water supply's diesel-driven pumps, thereby providing a virtually unlimited supply ofwater The Primary Water Storage Tank capacity following restoration of power to the primary water transfer pumps would also be available if required. NU notes that this method of decay heat removal is credited in the units license for those floods resulting in maximum flood levels in excess of 22' MSL; however, we further note that this method will, in practical terms, satisfy the decay heat removal requirement for flood resulting in water elevations of any lev lower than that presented by the PMH scenario. We do stress once again, however, that the unit's design and license bases do not credit this method except for the greater than 22' flood level scenario.

l DR-0312. DOC

JtN-19-W 14: 12 .FRON:PERCGI MILLSTn+ 61 n-H%-XO3 TO:FDLEERE (NJ,Ftti 1

i NU has concluded that the design and license bases associated with this scenario are satisfied with the current

' configuration of the fuel tanks. We have funher determined that it is pmdent to incorporate procedure step 1 will result in the operators disabling Fuel Oil Transfer Pumps P47A&B immediately prior to the expected arrival of a PMll induced flood event. Power would not be restored until some time after the flood waters recede, and the fuel oil contained in the tank has been sampled and determined to be of the required integrity. Millstone Station has the onsite capability to complete the appropriate set of fuct oil tests to support the determination of  !

acceptabil'iy of this supply. The safety related and seismic dedicated Diesel Oil Supply Tanks for each of the diesel generators have sufficient stored oil to support the operation ofeach of these units for a minimum of two (2) days. This is sufficient time during wiiich an alternative supply can be secured in the event that fuel stored in i the underground storage tank has been determined to be unusable.

{

e Extended Full Power Operation Followed by a Unit Manual or Automatic Trip Due a Seismic Event Without a LOCA: I The diesel oil storage tank T-47A is a non safety-related, non-seismic underground tank containing fuel oil that is  ;

tested regularly. If a seismic event occurs, tne diesel oil in the diesel oil storage tank can not be relied upon as a  !

qualified source of thel oil without prior testing of that fuel following the event, and potentially, the utilization of an alternative fuel oil forwarding system.

Even though this fuel oil source cannot be specifically credited to be available following a seismic event, the possibility exists that the following worst case scenario could occur:

i .

i

.1) Diesel Oil Storage Tank T47A remains generally intact and pumps P47A&B remain capable '

of transferring thel to the diesel generator day tanks.

2) Diesel oil transfer piping remains intact and Pumps P47A&B remain energized and functional.

3) Groundwater enters the degraded Diesel fuel oil storage tank following the seismic event. It is i j postulated that water laden fuel could potentially be transferred to the Diesel Generator day . I tanks.

g The groundwater table levels in the vicinity of the underground tank following the seismic event may or may not be above the bottom of the tank. If the water table level is below the bottom of the tank, no water inttusion into the tank would be expected. This scenario could result in diesel oil flowing out from the tank to the surrounding soil. Surrounding backfill could possibly be introduced to the Diesel Oil Storage Tank inventory at the connections at the top of the tank. In-place strainers between the Diesel Oil Storage Tank T47A and Diesel Oil Supply Tanks T48A&B would remove this sediment, and, in the process, potentially clog the strainers resulting in the termination of the transfer of fuel. The Diesel Oil Supply Tanks would continue to supply the diesels for a minimum of approximately two days each at full power. Operation of both units at full power for an extended periM of time is highly unlikely since there is no LOCA ongoing; however, the Diesel Oil Supply Tanks can be replenished as necessary from offsite or other onsite supplies.

The following considerations apply if the groundwater table level is at a level which is stbove the bottom of the dicsci oil storage tank: -

Diesel Oil Storace Tank inteerity is breached above water table level at the connections - This scenario would not result in water intrusion ainee the water table level will, in the worst case, be significantly lower. Backfill could possibly be introduced to the Diesel Oil Storage Tank inventory. Again, as discussed above,in-place strainers between the Diesel Oil Storage Tank T-47A and Diesel day tanks T48AAB would remove this sediment. If the sediment eventually clogs the strainers, the supply of fuel oil from Tank T47 would be terminated. The dedicated Diesel Oil Supply Tanks' fuel oil integrity would be maintained. Actions to replenish the diesel fuel oil to the Diesel Oil Supply Tanks would be taken, as appropriate.

Page to of 15 DR-0312. DOC

- - . - , - - . - > - . , -,7 s - y

Jtm 19-9a N:t3 am:rw.cGI Mti t sTrw maowom To: m p N:1P I l

i _-_

Diesel Oil Storage Tank integrity is breached below water table level- Technical Evaluation M2-EV-98 Revision 0, tit 1ed Structuralhotegrity ofDieselGenerator FuelOilStorage Tcmk Under FArthquake Loads, provides a basis for concluding that the underground tank proper would survive the seismic event and mainta I

it's integrity. This attached evaluation, completed using GIP approved techniques, concluded that any failure the tank is likely to occur at the connections to the tank at its top. Therefore, there would not be in-leaka groundwater since in the worst case, it is significantly below the top of the tank.

While an acceptable design basis is provided for this condition, NU has determined that it is pmdent to incorporate procedure steps that will result in the operators disab'ing Fuel Oil Transfer Pumps P47A&B immediately following the occurrence of a seismic event. Power would not be restored until some time after the seismic event, when the fuel oil contained in the tank has been sampled and determined to be of the requ integrity.

' The above, when combined with the approximate 1150 gallon " margin" that exists in the diesel oil storage tank to acconunodate minor water intrusion. l I

the " margin" that also exists in the Diesel Oil Supply Tanks T48A&B, and, actions that would likely occur as a result ofrecommendations made to remove power from the pumps by the Technical Suppon Team which would be assembled in the Emergency Operations Facility, provides  ;

reasonable assurance that the transfer of groundwater from the Diesel Oil Storage Tank to the Diesel Oil - l Supply Tanks and then, to the Emergency Diesel Generators would not have occurred, historically, Given the relatively lightly loaded diesel generators under the postulated scenarios, the fuel consumption rate will be significantly reduced resultlag in additional time for the unit operat' ors or those in the EOF to analyze those' systems important to the maintenance of safe shutdown. Past Simulated Emergency Response Drills at NU '

have demonstrated a high regard and degree of attentbn te the integrity of those systems necessary to maintain a supply of AC power.

l L

NU concludes that, for the seismic event case, the transfer of water-laden fuel to the diesel generator day tanks would not occur due to the maintained structural integrity of that portion of the underground tank system essential to the maintenance of near leak-tightness. Further, there is sufficient time available that, should some amount of water be introduced into the tank, the appropriate expertise would take actions to recommend that the tank be isolated from the dedicated Diesel Oil Supply Tanks.

The Design Basis Scenarios discussed above bound all other cases, including those that could occur with the unit previously shut down.

We note that Diesel Oil Storage Tank (T47A), including transfer pumps (P47A&B), and associated piping have, since unit startup, been designated as non-Category I systems at Millstone Units No. 2. Diesel oil has been identified as being automatically transferred from the underground diesel oil storage tank to the die el oil su tanks.

NU concludes that issue 1 of this DR does not represent a discrepant condition. Therefore, Significance Levels do not apply. However, as a direct result ofParsons' identification of those issues discussed in this DR, NU is taking actions to provide even greater assurance that the underground fuel oil storage tank, its appurtenances and the manner in which the system is operated will not challenge the basis for inclusion ofredundancy cnd l independence in the design of the onsite electrical power supplies; i.e., the diesel generators.

i

{

Page 11 of 15 DR .0312. DOC l

JUH-19-98 14: 14 FRON:PEECGI MILLSTOtF 610-055-2509 . TO:FOUGERE PAGE 13 Item 2 Disposition:

Tank Elevation Discrepancies The elevation discrepancies reported in the documents listed below will be investigated. Design documents will be revised, as appropriate, to correct the discrept eies.

Source Document Tank Bottom Elevation (MSL) Pump Mounting Plate Elev. (MSL)

SK-M-305, 7604-M-75 l '-0" 21'-6" 25203-28406-29 2 '-0" Not Shown Calculation Y.T O'-0" 22'-0" 25203-29032 Distance from bottom of tank to pump mounting plate - 20' 6" FSAR 2.5.4.2.5 Invert Elevation l'-8" Field walkdowns have been performed and dimensions were taken to determine the elevation of the tank bottom.

This elevation was determined to be Elevation (+)2'-7". This places the Pump Mounting Plate at Elevation

(+)23'-1" and the flame arTestor on top of the vent pipe above Elevation (+)25'-0",

NU concludes that the drawing errors depicting the actual tank bottom elevation constitutes a Significance Level 4 discrepancy. ;r ,

Conclusion:

NU has concluded that the issues reported in DR-0312 has identified a CONF [RMED SIGNIFICANCE LEVEL 4 Condition that requires correction.

I

( Item 1 - Water in Diesel Oil Storage Tank (T47A) Could Enter Diese! Oil Supply Tanks Based on the above, NU has concluded that the license and design bases for MP 2's emergencyonsite power supplies including fuel supplies are met. However, Northeast Utilities is amending procedures such that there is a

{ greater degree of assurance that electrical power will be removed from the P47A&B transfer pumps prior to the transfer of potential water-laden fuel, should these pumps remain operable for any of the evaluated scenarios.

Item 2 - Tank Flevation Discrepancies Based on the above, Northeast Utilities concludes that a Level 4 Discrepancy does exists. Design documents will be revised, as appropriate, to correct the discrepancies.

Page 12 of 15 DR-o312 DOC

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

JLket9-w 14:15 FRrnPE&CGI NILI.SirW 610-855-R509 TO:f 006t* PoGE:14 COMMENT ON5 ECo PROPOSED CORRECTIVE ACTION j Specific Comments:

Acceptable Leakage Rate:

i Under Disposition subheading," Extended Full Power Operation Followed By A Loss O Or Without A LOCA," an allowable leakage rate was discussed. The maximum leak rat accumulated water would not reach the level of the pump suction between the monthly sam breaching of the tank due to an age-related, corrosion induced failu ,

i Parsons does not agree that this leak rate is conservative. For discussion purposes, th to allow a leak of 1.5 gallons per hour was calculated cssuming a vertical, sharp-edged circula external water head of 2 feet. The required orifice diameter for that leak rate is one millime small perforation. Altemately, a 1/4 inch diameter orifice under a 3-inch head of water w hour into the tank. For a tank with a diameter of 10 feet and length of 44 feet a one millimeter diam perforation is not judged to be a conservatively large leak.

! Corrosion-induced failure due to external or internal degradation mechanisms could occur anytim been sufficiently addressed. Examples of such mechanisms are external corrosion due to coa failure / holidays or higher than normal local electrolytic corrosive cells. Internal corrosion d!

induced corrosion (mic) of the uncoated interior also requires consideration.

As stated by NNECo, the groundwater elevations are not monitored in the tank area so the averag groundwater elevations are not known. A loose grade connection could leak under rainfall ru several inches on the concrete slab over the tank. Also, asphalt paving is not im;iervious. Non-flood water j

sources could come from several paths, none of which can be diwounted without a means to monitor / measur Calculation 98-ENG-02567C2 Implications:

s

1 Under Disposition subheading " Extended Fuli Power Operation Followed By Unit Shutdown In Accordan!

With The TS Prior To A l Tival Of A Probable Maximum Hurricane (PMH) And Associated Design Basis Flood l Levels," the new calculation 98-ENG-02567C2 was discussed. The tank was shown to withstand resulting from flooding above grade levels. The analysis used the new tank wall thickness of 3/8 inch. The maximum calculated Actual / Allowable stresses were shown to vary from 92.9 to 95.6 percent (t of 50). These stresses are based on the unverified assumption that the tank retains its original wall thic 3/8 inch. Without the benefit of tank wall thickness measurements, we cannot agree that the tank will resis forces from flooding.

Even with cathodic protection, corrosion can't be discounted. Internal tank inspections will not discover extemal tank corrosion. The effectiveness of the cathodic protection system's actual protection of the tank cannot be proven by operational testing of the cathodic protection system itself. The proof of protection woul be tank wall thickness measurements. Consider what the Actual / Allowable stress ratio corroded tank wall was actually 1/4 inch instead of the original 3/8 inch.

Regardless, NNECo's proposed revision to procedures disabling the fuel oil transfer pu'mps prior to arriva PMH induced flooding will prevent problems of water transfer from tank failure in this case. Parsons considers disabling of the automatic fuel oil transfer interlock to maintain the licensing basis for " independent and to be a significance level 3 discrepancy.

i Page 13 of15 DR-o312. DOC

Jt.N-19-98 14 : 16 FRO 1:PEECGI MILLSTOW Fi10-HSS-P969 TO: F ULfGERE PAGE:15 i --

General Comments:

{ The NNECo Disposition proposed procedural revisions that would minimize the risk association w

! onsite power due to flooding or seismic events. These are low probability, extreme events. Parso there is a greater risk from corrosion-induced leakage which could occur anytime end we feel that sutliciently addressed. The fact remains that the tank, which is a non-safety component, is automatically controlled makeup pumps, to both of the Emergency Diesel Generator trains. Th disposition still violates the " independent and separate" requirements of the licensing documents.

The disposition states that the tank is scheduled for replacement in the year 2002. This concern, th finite life span. However, during the time period from 1998 to 2002 the plant is exposed to increasing) risk considering of the age of the tank and cumulative effects ofcorrosion w

Conclusion:

Item 1:

We have reviewed the NNECo Disposition and fmd that item I remains discrepant. In the absence of a w thickness survey and a means to detect inleakage ofwater between monthly sampling, and the non emergency (flood / seismic) conditions, the pumps remain on automatic level control, we maintain that this condition safety is discrepant component. to Significance Level 1. The reliability of the both EDGs could be degraded b Item 2.

This item will be considered closed after the revised documents or Change Notices have been d revis u

(

Prennred: M.J. Akins r -

d 7 -

!f/

Group Lead Date Revs.ewed:

Y&

E.A. Blocher p/pg Deputy Project Director Date Approved: D.L. Curry b~

Project Director Date Forwarded to NNECo, NEAC, and NRC: DN Posted to WWW:

Date Date

, Page 14 of 15 4

( DR-0312. DOC I

98 14: 16 FROM:PE&CGI anILLSTOE 610-83'5-2503 TO:FOtJ3ERE R r5E: 16 FINAL IMSOLUTION Open: item 1 requires NNECo follow-up to address inleakage concerns.

Item 2 remains open pending review of drawing and calculation changes.

-AA>#84-<A

.A. Blocher MJwn Deputy Project Director Date l

l l

I

(

i l

1 r .

I DR 0312. DOC J

. SAN '30-98 12:15 FROM:PE&CGI MILLSTOtE 610-855-2539 TO:FOUGERE PAGE:22 PARSONS POWER GROUP INC.

2675 Morgantown Road, Reading, PA ICAVP MILLSTONE UNIT 2 19607

( DISCREPANCY REPORT (610) 855-2000

• FAX: (610) 855-2509 DR NUMBER: DR-0312 DR TITLE: Water in Diesel Oil Storage Tank (T47A) Could Enter Diesel Oil Supply Tanks REVISION: PRELIMINARY ISSUE DATE: 1/27/98 ORIGINATING GROUP: Tier 1 SIGNIFICANCE LEVEL: 1 l DISCREPANCY l

References:

1. AOP 2560 Storms, High Winds, and High Tides

2. FSAR 8.3 Emergency Generators

3. OP 2345B Diesel Fuel Oil System

4. SP 2672 Sampling and Inventory ofDiesel Oil Storage Tank, T47A

5. 25203-29032 Diesel Oil Storage Tank T-47A

6. 7604-M-125 Misce!!aneous Shop Assembled Tanks I. ARP 2590F Emergency Operating Procedures and Guidelines, Alarm Response for Control Room Panel C-08

8. Figure 7.3 "10CFR50.59 Safety Evaluation Screening", prepared to support PDCR 2 108-92, dated 8/27/97

9. MP2-DG DBDPackage -Diesel Generator

10. NRC Letter Docket / License: 50-336/DPR-65

11. OP 2346A Emergency Diesel Generators

12. SP 2613E Diesel Generator Fuel Oil Sampling

13. FS AR 4.2.2 Codes Adhered To

14. MP 2721J Periodic Inspection of Unit 2 Tanks

15. 91-BOP-813-ES, Rev. 3, MP2 EDG Operating Time with 24,000 Gallons of Diesel Fuel Oil Available at a Continuous Rated Load of 2750 kW Fuel Consumption

16. IPEEE December 29,1995, MP2, Response to Generic Letter 88-20, Supplements 4 and 5, Individual Plant Examination for External Events - Summary Report

Background:

'j Diesel Oil Storage Tank T47A is a 25,000-gallon underground, horizontal cylindrical tank. The tank is single-wall, carbon steel coated with "bitumastic mill undercoat" (exterior) and no interior coating. The tank was installed around 1972. A cathodic protection system (rectifier PO29A, zine anodes, and test station) was added after October 1974 Transfer Pumps P-47A and P-47B are installed in individual 24-inch diameter steel riser pipt , connected to nozzles on the top of the tank. Bottom of tank = l'-0" (plus or minus) depending upon which drawing Page 1 of 4 DR4312. DOC

8AM-30-90 12:26 FROM:PCECGI MILLSTONE 610-055-2509 TO:mlGFRE PAGE:23 is used. Top of tank = 1l' (plus or minus). Grade = 14'-8" Top of pump manways is 22' (plus or minus). Other connections extending to or above grade are fill, sample, vent, and level pipes. See Ref. 5.

The storage tank is not seismically quali6ed nor flood-protected. A sufficient quantity of oil is contained in the Diesel Oil Supply Tanks to operate the diesel (s) for approximately 7 days. Each Supply Tank (13,500-gallon nominal capacity) must contain 12,000 gallons. See Ref. 2. Pemo(s) P-47A and P-47B

" start (s) automatically when the respective level decreases below 95% and stops when tank level increases above 95%." See OP-2346B, (Setpoint Change #2-88-022, in process, changed minimum level to 93%.)

"Two EDGs operate for 8 hours9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br />, then one EDG operates for the remaining length of time. Fuel l

consumption is 3.6 gpm for each EDG at 2750 KW. During a LOCA with a LNP (loss of normal power)  !

the two day tanks (Supply Tanks) are cross-connected." Ref.15, dated April 1997. (The current license states one EDG for one hour instead of 8 hours9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br /> with the other EDG continuous) The supply tanks are cross-connected via valve 2 FO-83, " Fuel Oil Supply Header Cross-tic" Discrepancy:

Item 1:

Failure Mode Water entering the underground Diesel Oil Storage Tank (T47A) will be automatically pumped into the Diesel Oil Supply Tanks T48A and T48B.

With water entry into the Diesel Oil Storage Tank from a single failure, during a LOCA in an LNP, it

-(- would take less than 3 hours3.472222e-5 days <br />8.333333e-4 hours <br />4.960317e-6 weeks <br />1.1415e-6 months <br /> to pump 6 inches ofwater into each Supply Tank. The fuel supply pipes are 6 inches above the bottom of the supply tanks. Water will shut down both EDGs a short time later.

(With the currently licensed scenario i.e. one EDG for one hour, it would take up to six hours, assuming the tanks contribute equally, for the continuously running diesel to shut down. At that time the second EDG would not restart or ifit starts it would not run for an extended length of time.)

If only one EDG were to started, it would run for 3 hours3.472222e-5 days <br />8.333333e-4 hours <br />4.960317e-6 weeks <br />1.1415e-6 months <br /> before automatically shutting down. At that time the other EDO would be started, it too would run for 3 hours3.472222e-5 days <br />8.333333e-4 hours <br />4.960317e-6 weeks <br />1.1415e-6 months <br /> before automatically shutting down.

Water Sources Water sources are surface runoff, groundwater (normal or elevated), or flood water. Probable Maximum Precipitation can cause surface flooding between 15.5 to 16.2 feet MSL (mean sea level), Ref.16.  !

Normal groundwater level is elevation 5 feet MSL (halfway up the tank). The containment, turbine, and auxiliary buildings are protected from flooding. The design flood produces water levels up to elevation 18.1 feet MSL, stillwater and 25.2 feet MSL, wave runup, Ref.16 for PMH (Probable Maximum Hurricane). According to reference 1, the EDGs will be operated during flooding up to and above a water level of elevation 22'.

i Page 2 or4 DR-o312. DOC

I JAN-30-SG 12:16 FROM:PE&CGI MILLSTDE 610-855-2509 TO FOUGERE PAGE E4

Water Entry Point

(

' We could not find documents that showed the Grade connections (fill, sampling) to be watertight under floodwater static head (approximately 7 feet). In addition, we could not find documentation to show that the Pump manways or vent pipe could not be damaged by floating debris during flooding. We could not find information concerning tank leakage under static head offlood water, ponded rainfall runoff, or l groundwater. The Tank is single-wall steel and Refs. 8 snd 9 imply that the tank may already leak. "The storage tank low level setpoint was lowered to reduce leaks," Ref. 8, and "In 1987, the concern for leaky .

storage tanks was raised," Ref. 9, page 3.4-1-17. For the purposes ofelarifying this discussion we have assumed the vent pipe is damaged by floating debris during fleding, breaking a pipe weld, and providing an opening into the tank through which water can enter.

Detection of Waterin Storage Tank SP 2672 requires the storage tank T47A inventory to be checked weekly and fuel quality testing monthly. -

Water infiltration can occur between inventory checks or during site flooding. "A quarterly pumping of approx.10 gallons ofdiesel fuel from the bottom of the storage tank to remove any moisture accumulation is being accomplished at present "(by AWO see Ref.10). " Underground Tank Volumetric Test" is performed, firequency not given (see Ref.14). " DIESEL OIL STOR TANK LVL, LI-7004" (C.

06) has setpoint of 20%. "IF level is low, NOTIFY Operations Technician to order fuel" (see Ref 7). .

I Water Pumped to Supply Tanks by Transfer Pumps l

Diesel Oil Supply Tanks are automatically maintained IcVel between 93% and 95 % full by automatic level control operation by the oil in the tanks. AOP 2560 does not address the shutdown of transfer

( pumps P-47A and P-47B during/ or after flooding. The Transfer Pumps take suction at approximately 11 inches above the bottom of the Storage Tank. There is no continuous monitoring for water in the Diesel Oil Storage Tank. The only reason we could find for the Transfer Pumps to be removed from automatic operation is in response to high level alarm (97*4) at windows B32 and B33 on Control Room Panel C-

08. In this case the pump supply breakers are opened to prevent tank overCow, Ref. 7 Item 2:

Tcnk Elevation Discrepancies Elevation of Diesel Oil Storage Tank T47A is recorded at three different elevations. The elevations are shown below:

Source Document Tank Bottom Elevation (MSL) Pumo Mountina Plate Elev. (MSL)

SK-M-305, 7604-M-75 l '-0" 21*-6" ,

25203-28406-29 2 '-0" Not Shown Calculation Y-T O'-0" 22'0" 25203-29032 Distance from bottom of tank to pump mounting plate - 20'-6" t

h on-o3aooc I

l

[ J W-30-98 32:17 FROM FEtCGI MILLSTONE G10-855-2509 TO:FOUGERE PACE:25 Significcnce Levc.1:

l "To provide a reliable onsite source of auxiliary power if the preferred source is lost, the unit has two

( I onsite emergency generators. They are redundant, independent and separate, and are used for no purpose i other than that described." FSAR 8 3.1.1. Autornatic pumping from tank T47A removes the l

" independent and separate" provisions required by the licensing basis documents for this system. The Diesel Oil Supply Tanks are not " redundant" when operated as they are currently. A single failure (water in tank T47A) will cause loss of both EDGs. I I

Since both EDGs would shut down this is a Level 1 discrepancy. The tank clevation discrepancies are level 4.

Jon A. Winterhalter A Tier 1 1/27/98 Orieinator -

g g EVALUATION

@ BASIS VALID 0 BASIS INVALID - CLOSED 0 PREVIOUSLY LDENTIFIED BY NNEco - CLOSED l

\

l l

M.J. Akins /

78 .

Group Lead Date i

! REVIEW AND APPROVAL C$  %

Reviewed: L A, Blocher 2 Deputy Project Director Date Approved: D.L. Curry /f30fN Project Director Date l

l DR4312. DOC i