ML093200312

Summary of Category 1 Public Meeting with Indiana Michigan Power Company and NRC Staff to Discuss Responses to Generic Letter 2004-02 Requests for Additional Information

ML093200312
Person / Time
Site:	Cook
Issue date:	11/20/2009
From:	Beltz T Plant Licensing Branch III
To:
beltz T, NRR/DORL/LPL3-1, 301-415-3049
References
GL-04-002, TAC MC4679, TAC MC4680
Download: ML093200312 (80)

Text

UNITED STATES NUCLEAR REGULATORY COMMISSION WASHINGTON, D.C. 20555-0001 November 20,2009 LICENSEE:

Indiana Michigan Power Company FACILITY:

Donald C. Cook Nuclear Plant, Units 1 and 2

SUMMARY

OF OCTOBER 14, 2009, CATEGORY 1 PUBLIC MEETING TO DISCUSS RESPONSES TO GENERIC LETTER 2004-02 REQUESTS FOR ADDITIONAL INFORMATION (TAC NOS. MC4679 AND MC4680)

On October 14, 2009, a Category 1 public meeting was held between representatives of Indiana Michigan Power Company (I&M, the licensee) and the U.S. Nuclear Regulatory Commission (NRC) staff from NRC Headquarters, One White Flint North, 11555 Rockville Pike, Rockville, Maryland.

The purpose of the meeting was to provide an opportunity to resolve any remaining concerns related to the licensee's proposed response to requests for additional information (RAI) associated with Generic Letter (GL) 2004-02, (Agencywide Documents Access and Management System Accession No. ML091490421), and the licensee provided supplemental information for NRC staff review and comment. This was the 5th public meeting - the 2nd held at NRC Headquarters - between the NRC staff and I&M to discuss the proposed responses to the RAls associated with GL 2004-02 for the Donald C. Cook Nuclear Power Plant (CNP). is a list of meeting attendees. is a meeting handout (slide presentation) provided by the licensee.

At the conclusion of the meeting held on August 26, 2009, the NRC staff questioned the licensee regarding strainer head loss testing and the loading analysis of the main and remote strainers inside the containment sump.

In the October 14, 2009, meeting, the licensee described plant modifications designed to improve the performance of the containment sump and also discussed the margins in its evaluation for addressing GL 2004-02.

The licensee further stated that it removed a significant amount of the problematic insulation and other materials that could result in sump clogging. The licensee installed a vented containment sump and level instrumentation to support operator actions, should they be necessary, to mitigate a reduction in flow to the sump.

At the October 14, 2009, meeting, the licensee's presentation described how analysis and testing bounded the actual conditions at CNP regarding the evaluation of sump performance submitted for GL 2004-02. The NRC staff concurred with the licensee's position that there is sufficient margin to overcome uncertainties in the flow split (debris loading) between the main and remote strainers, and in the strainer head loss testing itself. However, the staff was still unclear as to how those margins would be reflected and maintained in the CNP licensing basis.

The licensee agreed to clarify the CNP licensing basis.

- 2 The NRC staff and licensee agreed that a final submittal date for the RAI response would be February 15, 2010.

Please direct any inquiries to me at 301-415-3049, or Terry.Beltz@nrc.gov.

~.IZ&/i,v+--

Te~eltz, Senior Project Manager Plant Licensing Branch 111-1 Division of Operating Reactor Licensing Office of Nuclear Reactor Regulation Docket Nos. 50-315 and 50-316

Enclosures:

1. List of Attendees

2. Meeting Handout cc w/encls: Distribution via Listserv

LIST OF ATTENDEES OCTOBER 14, 2009, MEETING WITH INDIANA MICHIGAN POWER COMPANY TO DISCUSS REQUESTS FOR ADDITIONAL INFORMATION ASSOCIATED WITH GENERIC LETTER 2004-02 FOR THE DONALD C. COOK NUCLEAR PLANT, UNITS 1 AND 2 Indiana Michigan Power Company Nuclear Energy Institute Stewart Bailey Michael Scott John Lehning Kevin O'Connor Michael Scarpello Paul Leonard John Butler Steve Smith William Knous (Alion Science &Technology)

Paul Klein Nathan Mar (Alion Science &Technology)

Matthew Yoder Christopher Hott Ralph Architzel Terry Beltz Meeting Handout

~'ND'ANA IiIiIM'CH'GAN POWER" Aunit ofAmerican Electric Power NRC - Donald C. Cook Nuclear Plant Public Meeting Containment Recirculation Sump Performance October 14, 2009 1

Overview

• This presentation provides supporting information for those RAls that could not be resolved in previous meetings:

Debris Split / Flow Split questions:

• 5, 6a, 6b, 6 closing, 14, 17 Chemical Effects head loss & bump-up factor:

• 13, 16b Radial Decay of Pressure for Marinite Testing:

• 2a Installed Configuration of Cal-Sil

• 4 Cal-Sil Erosion

• 7d Pressurizer compartment breaks

• 25b 2

Overview

• The two main issues of discussion today are:

Was the testing that was performed to establish the design basis recirculation sump strainer system head loss sufficiently conservative with regards to the debris and flow distribution to the main and remote strainers?

Was the established chemical effects bump-up factor sufficiently conservative with regard to the methodology used to determine this factor?

3

Safety Case for Reasonable Assurance

• Margins and conservatisms associated with the two principal topic areas will be further discussed within those topics

• Additional margins and conservatisms exist as described in the document provided as a handout for this meeting 4

Key Efforts Undertaken to Resolve GSI-191 and Provide Reasonable Assurance

• Strainer area increased from 85 ft2 vertical flat screen to 1972 ft2 complex design (pockets)

• Strainer assemblies significantly separated within containment flow and debris fields

• Significant debris reduction efforts to remove problematic debris

• Extensive testing to determine bounding response of recirculation sump strainers to a LOCA event

• Installed recirculation sump level instruments to provide Operators with warning of excessive strainer blockage and provided procedural guidance for reducing demand (flow) on strainer to mitigate The results of these actions demonstrate that reasonable assurance exists that the Cook design and installation of the recirculation sump strainer system will ensure the required core and containment cooling functions exist and will be maintained 5

Section View Recirculation Sump Strainer System

-000, Vent

.. 1****

000' EleY. 614 fl. - 2 9/16 in.

Contaiment wall Etev. 603 ft.. 11 318 In.

.~ < < q Level SWllctl Sel POint'\\.

Elev. 601 rt *91n. }Mam Stra"Trier.

Conlainment Floor 900 sq. ft.

Elevation 598 ft.. 9 318 in~

4 In. Curb 6

+-'

c CD E

c.ro

+-'

c o

()

Lo Cf)......

.- c..

L-(])

..c u

(])L O(])

...... e 3:Cf) o(])

-0 U-I L

e(])

._ e 00*

~~

(J)

co

Q)c:.m...

en Q)..... o E

Q)

~

So... o

+-' a.

Q)

(,)

So...

(])

+-'

c

'+o c o

+-'

(,)

(])

(J)

RAls for Flow &Debris Distribution RAI5 Degree of uniformity in debris distribution between the main and remote strainers

• RAI6a Reduced flow resistance on main strainer during pool fill resulting in less flow to the remote strainer

• RAI6b A more representative analytical model of head loss at the main strainer during recirculation would likely result in significantly larger flow and debris fractions arriving at the remote strainer

• RAI 6 Closing The flow distribution between the two strainers would be more uniform...this overestimate of flow and debris transport to the main strainer appears non conservative

• RAI 14 Provide information that the test methods did not result in non-conservative head loss results or provide information that shows the potential non-conservatism of these practices were offset by other conservatisms contained in the test protocol

• RAI 17 Provide an evaluation of the sensitivity of overall system head loss to various debris loads split between the main and remote strainers 11

RAls for Flow & Debris Distribution

• The previously listed RAls contain a common theme:

Was the testing that was performed to establish the design basis recirculation sump strainer system head loss sufficiently conservative with regards to the debris and flow distribution to the main and remote strainers?

The information presented today will demonstrate that the testing was sufficiently conservative

• An analysis of the debris quantities used for strainer head loss testing and the effects of varying debris distribution between the main and remote strainers on the overall system head loss has been performed. The results of this analysis are contained in the next presentation by ALIGN Science &

Technology.

12

NRC... Donald C. Cool<

Nuclear Plant Public Meeting D.C. Cool<

Containment Recirculation Sump Performance Deb i /Flow Split RAI Support October 14, 2009 A L ION SCIENCE AND TECHNOLOGY

~

with your needs.

Introduction

• Sump Strainer System Design

• Debris Split

• Effect on the System Head Loss

pump suction sump Sump Strainer System Design crane wall main, **__

straine Remote pump

..- Ductwork remote suction Side View

-.. strainer Plan View Remote ductwork remote straIner sump pit The D.C. Cook sump strainer system is distinctive in that it is comprised of two separately positioned strainer arrays which both drain into the sump pit. The main strainer drains directly into the sump pit, while the slightly larger remote strainer funnels through a ductwork from outside the crane wall which then drains into the sump pit.

~

Aligned with vniJlr n,gn't.

Sump Strainer System Design HL-Rem.ote HL-Waterway Strainer QR~

~

Q Q----.

QM _.~.

QM /

lk-Main St:rain.er

• Through the main strainer flow path the only source of resistance is the debris bed which forms on the strainer.

• . Through the remote strainer flow path, resistance is encountered both from the debris bed which forms on the strainer, and from waterway losses within the ductwork.

• These waterway resistances include those from the remote strainer plenum, the duct work linking to the sump pit, and the exit losses from the duct work to the sump pit.

o Aligned with your needs.

Sump Strainer System Design HL-Remote HL-Waterway Strainer QR~

(-

Hr. = KQ~

2gA2 Q -----+

Q----.

Where:

Qht.".w,.,.

HL IS the head loss across the strainer K a constant "K factor" determined for the screen QM/

Q IS the flow rate across the screen g IS the gravitational constant A IS the area of the screen.

lit-Main Strainer

• The head losses created by each of these resistances are proportional to the square of the flow rate through the corresponding branch of the sump strainer system (QM and QR) for turbulent flow.

• The sum of these two flow rates is always equal to the total system flow rate (Qt= 14,400 gpm).

• An increase of flow through the remote strainer (QR) will always cause a corresponding decrease in flow through the main strainer (QM) and vice versa.

'" Aligned with your Sump Strainer System Design DOBS Case Total Quantity of Debris Available at the Sump Strainers 350.00 ~,--_.

~

329.81bs I

I I

300.00 +1-------1 I

I available at strainers I

I 250.00 +1---

~

200.00 +1

~

tV o

~

.~ 150.00 +1------1 LJ CI>

C 100.00 +1------1 17.77Ibs Total Fibrous Debris 50.00 1 I

available at strainers

,------J I

0.00+1------'

Similar to flow rate, there is a defined and finite quantity of debris produced during the design basis accident which subsequently transports to the strainers. This debris splits and accumulates on the two strainers, but will never exceed the total quantity transported to the strainers.

~

Aligned with VOllr nl_11:.

Sump Strainer System Design DGBS Case Total Quantity of Debris Available at the Sump Strainers 35lUJO r-- _.. _._-

HLW HLR 329.81bs 300.00 +1----1 Total Particulate Debris available at strainers QR~

~

25000 +1----1 Qt~

~

Qt ---+

200.00

~ a tl

~ 150.00 QM----'

c..

QM/

100.00 +I---~

17.77lbs Total Fibrous Debris 50 00 +I----J available at strainers HLM 000+1---

HL\\f== H LR + H Lw== HIS Physically, the head loss across the remote strainer flow path must equal the head loss across the main strainer flow path. This head loss is also known as the system head loss.

• As the debris accumulates on the strainers, the flow split between the main and remote strainer will change to ensure that the head losses across the two branches always remain equal.

Sump Strainer System Design DGBS Case Total Quantity of Debris Available at the Sump Strainers 350.00TI--------------------------------------------,

300.00 +1---

250.00 +1---

~

~ 200.00 1---

i:..

o"

,!! 150.00 +1----.

C

.c 100.00 1---

SO,OO +1---

0.00+1---

329,81bs Total Particulate Debris available at strainers Total Fibrous Debris 17.77 Ibs available at strainers

• The debris bed is formed overwhelmingly by particulate debris.

The low quantity of fiber is insufficient to provide the structure necessary to create large bed thicknesses.

Primary Question Does the head loss testing represent a conservative debris split to reflect the highest achievable head losses?

Debris Split DGBS Case Tested Debris Quantities Vs. Quantities Available 1600 r-Quantity of Particulate Debris Tested 1400 1200

_ 1000

~

Total Quantity Particulate of

.~

Debris Available at the

l 800 Strainers- -_.

o I/J Total Quantity of

~

Fibrous Debris o

600 Available at the Quantity of Fibrous Strainers Debris Tested 400 200

\\

o Fibrous Debris Debris Type The quantity of particulate debris used in testing was over 4 times that actually available to accumulate on the strainers.

• The quantity of fibrous debris used in testing was nearly double that actually available to accumulate on the strainers (178% of the available quantity).

Particulate Debris

Debris Split DGBS Case Tested Particulate Debris Quantites Per Strainer Vs. Quanitities Predicted 1000.-----

Quantity of Particulate 900 Debns Testecf onMain Strainer Quantity of Particulate 800 Debris Tested on Remote Strainer 700 Cil g.

~

600 Quanity of Debris Quantity of Debris

~..

0

~

Predicted at Main Predicted at III 500

.c Strainer Remote Strainer Q)

C

.2l..

400

'3

'f..

u tL.

300 200 100 0

Remote Strainer

• Debris Quantity Tested (Scaled) III Quantity Predicted at the Strainer The debris was predicted to accumulate primarily on the main strainer.

The particulate debris loads tested were quadrupled, but the predicted ratio of debris split was maintained.

Main Strainer

.hTI

~~?~

Remote Strainer Debris Interceptor ebris Split Plan View Lower Containment

• The debris was predicted to accumulate primarily on the main strainer due to differences in the transport paths to two strainers.

• The main strainer is within the crane wall and will begin accumulating debris immediately.

• The remote strainer is outside the crane wall and will not begin accumulating debris until the start of recirculation.

6 Debris from the break within the crane wall can transport in the pool to the main strainer.

Debris transporting to the remote strainer must exit the crane wall through the debris interceptor and the flow holes.

Debris transporting to the remote strainer must then traverse the majority of annulus to reach the remote strainer.

Debris Split 1000 900 800 700

'iil

g.

~

i:

600 C1l

J a rJl 500

.0 Gl C

Gl 10 400 "5

<.> t:

C1l C.

300 200*

100 0

DGBS Case Tested Particulate Debris Quantites Per Strainer Vs.

Revised Quanitities Predicted (Equal Debris Distribution per Square Foot Strainer Area)

Quantity of Particulate Quantity of Debris Predicted at Main Debris Tested on Main Strainer Strainer fv1ain Strainer Quantity of Particulate

Debris Tested on Remote Strainer Quantity of Debris Predicted at Remote Strainer Remote Strainer

• Debris Quantity Tested (Scaled)

• Quantity Predicted at the Strainer Equal debris loading of the available debris per square foot of strainer area results in debris loading for both the main strainer and remote strainer well below the quantities tested.

Debris Split DGBS Case Tested Particulate Debris Quantites Per Strainer Vs. Quanitities Predicted 1000 Quantity of Particulate 900

---neoriSlesfea-on Mam Strainer 800 Quantity of Particulate 700

[Je6rrs--TestecfonRemOte Cil

~

Strainer

~

600

'E Quantity ofb-ebris 0"

Predicted at

.~

500 Remote Stra-iner

.0 CI>

C

~

'3'"

400

.S! t:

a. '"

300 200 100 0

The amount of particulate debris predicted to accumulate on the strainers could be applied in totality to the remote strainer and still remain below the quantity tested.

Main Strainer Remote Strainer

,_ Debris Quanti!y Tested (Scaled) II Actual Quan~ty Available at the Strainer

_T *.,.

DebrisSPlit DGOS C,.. Te.led P,rtl,ul,1e Deb,l. Qu,n'lt.. P" S,,,ln,,v*. Revl..d Qu,nltltl.. Predl'led 1000 Quantity of Particulate 900 DebrIS TestedorfMain Strainer 800 Quantity of Particulate Ii) 700 Debris Tested on Remote

.0 Strainer Z'

600

~

ltl

l a III 500

.0 Q)

C Q)

'lii 400

i 1::

ltl

0.

300 200 100 0

Quantity of Debris Predicted at Hemore Strainer rvlain Strainer Remote Strainer I_ Debris Quantity Tested (Scaled) II Revised Predicted Debris Quantity Available at the Strainer This would make the amount tested on the main strainer significantly conservative as with this debris split there would be nothing to accumulate on the main strainer.

Debris Split DGBS Case Tested Particulate Debris Quantites Per Strainer Vs.

Total Quantity Available at the Strainers 1000 Quannty ofParflculafe Debris Tested on Main 900

-Strainer 800 700

'iil

§.

~

~

600 otarQuantity of Particulate co 0

l Debris Available the Strainers Quantity of

.~

500

.c PartlculateDeoris C.,

Tested on Remote 10 400

i Stratner

'f co Q.

300 200 100 0

Remote Strainer This margin is even more prevalent on the main strainer, where the debris quantity tested exceeds double the total quantity of debris available.

Thus quantities exceeding the total quantity of debris available were tested on each strainer simultaneously.

rvlain Strainer I_ Debris Quantity Tested (Scaled) _ Revised Predicted Debris Quantity Available at the Stra;~~r*'

Quanity ofDeoris Predicted at Main Strainer DGBS Case Fibrous Debris Quantities Debris Split 0-'----

Remote Strainer

• Debris Quantity Tested (Scaled)

• Revised Predicted Debris Quantity Available at the Strainer This margin is also present in the fibrous debris quantities, though the difference is not as great as is present in the particulate debris quantities.

Main Strainer 20 18 16 14 Ii)

~ 12 i:

III o

~

III 10

~

Q)

C

~

8 o ii u::

6 4

2 Quantity Quantity of Fibrous of Fibrous

-DeBfis

--Deoris Tested on Tested on Main Remote Strainer Strainer

-Quantity of Debris

_. -Predrctedl at Rel11Gte Strainer I

Debris Split DGBS Case Tested Fibrous Debris Quantites Per Strainer Vs.

Revised Quanitities Predicted (Equal Debris Distribution per Square Foot Strainer Area) 20 I

Quantity Quantity 18 I

-ofFibrous ofFibrous Debris Debris 16 I

---'festea-on -- -

LTeste<ton Main Rerrote 14 I

-Strainer

-'--Strainer Ii) g

~ 12 Quanity

~

l 0 '"

of Debris Quantity

.!!! 10

--Predicted-offiebrts

.Q Gl c

at Main Predictedi IJl

l 8

-Strainer

-at 0...

.Q Rerrote u::

Strainer 6

4 2

0 tvlain Strainer Remote Strainer

• Debris Quantity Tested (Scaled)

• Revised Predicted Debris Quantit~ Available at the Str~

However, equal debris loading of the available debris per square foot of strainer area still results in debris loading for both the main strainer and remote strainer below the quantities tested.

Debris Split DGBS Case Tested Fiber Debris Quantites Per Strainer Vs.

Revised Quanitities Predicted (All Available Debris on the Main Strainer) 20 18 Quantity of Fibrous Total 16 I

-Quanfity-Debris of Tested on 14

-Main I5foOs Quantity 1 Strainer Debris of Fibrous

~ 12

-l)ebris c

III a

j Tested on

.!!1 10

..c...

Remote c

Ql Strainer VI 8

j 0...

..c iI:

6 4

2 0

• Debris Quantity Tested (Scaled)__Actual Quantity Available at the Strainerl Even though the fibrous debris quantity margin is not as great as that present in the particulate debris, the quantity of fibrous debris tested on the main strainer still exceeds the total quantity of fibrous debris available.

Main Strainer Remote Strainer

~Aligned with.._..- _........

Debris Split Is the debris split conservative?

The debris split represents the best estimate of a prototypical debris split.

The quantity of debris tested was comprised of more than 400% of the particulate debris, and nearly 180% of the fibrous debris actually available at the strainers.

This margin of additional debris conservatively accounts for any difference from the testing debris split.

If testing were to be re-performed the quantities of debris would be significantly lower regardless of the debris split.

o with your needs.

Effect on the System Head Loss We next determine the effect of changes to the debris split on the overall system head loss.

Effect on the System Head Loss Testing Head Loss Curves For Main Strainer DebrisfTest Row Variation 9 I I

MS 240%

of Total Available Debris 8

7 0' 6 N

I:,

.:: 5 III

.3 III 4

"0 III Q>

~

MS 180%

I: 3 of Total Available Debris 2 -,

MS 120%

lOf Total Available Debris o

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

Percent of Test Flow through the Main Strainer

• MS 120% Debris

• MS 180% Debris... MS 240% Debris In testing, the effect of varied flow rates on the head loss across the main strainer from debris beds comprised of 120%, 180%, and 240% of the total available debris were found

Effect on the System Head Loss DGBS Case Debris Quantities Tested on the Main Strainer Vs. Quantities Available 1000.00 Quantity of Particulate Debris Tested on 900.00 the Mairi-Sfraifler 800.00 700.00 Ul 600.00

-TofarOuanfityofPar1iculafe

~

Debris Available at both c:

Ol 500.00

l Strainers (J

III

..c Quantity of Fibrous Gl 400.00 c

DebiTs-Tested on

-T.0.. taoI Quantity oj the Main Strainer Fibrous Debris 300.00

-Available at botH!

Strainers 200.00 100.00 0.00 Particulate Debris Debris Type For ease of understanding, it is worthwhile to express all debris quantities in terms of the sum quantity of debris available at both strainers Fibrous Debris

Effect on the System Head Loss HXXI.OO

!XXI.OO

&:JO.OO 700.00

~

.Q

~ 600.00 o 500.00

~

0::

.Q..

C

-;;; 400.00

(;

30000 20000 10000 0.00 Total Debris Quantity Tested on lI1e Main Strainervs. Total Debns Quantity Available 1000 900 800 700

.Q 600

.~ c 500 0

Total Debns Quantity =893 Ills

.Q..

400 (257% of tte Total Available Quantity) 300 200 100 1-1 0

Debris Tested on the Mlin Strainer P<:tual Debris Available I_ Partx:ulate Debris Tested _ Fibrous Debrs Tested. PartX:ulate Debris Avai~bI. at the Slralners.!ilrous Debris Avai~ble at the Slrainers' Debris Loads Tested Vs. Actual Quantity of Debris Tested (CCI Test 12) 120% oITolal Available Debris 180% oITotal Available Debris 240% oITotal Available Debris Actual Quantity of Debris Available Debris Quantities Tested fa Actual Quantity. 50% of Tested_~uanti;;. 75% of Tested Quantity.100% of Tested Quantity As a simplification, the particulate and fibrous debris quantities are added together to ease in the expression of all debris quantities as a percentage of the total debris quantity available at the strainers.

In this expression, the debris quantities tested on the main strainer were equal to 120%, 180%, and 240% of the total debris quantity available.

Effect on the System Head Loss Testing Head loss Curves For Main Strainer DebrisfTest Row Variation 9 I I

rv1S 240%

of Total Available Debris 8

7 0- 6 N

I:

I:§. 5 ~-

til

.3 til 4

"'C ell IV rv1S 180%

I: 3 ----

of Total Available Debris 2

MS 120%

~

lOf Total Available Debris o

• rv1S 120% Debris _ rv1S 180% Debris... rv1S 240% Debris Thus, the relation between head loss and flow rate through the main strainer is known for debris beds comprising 120%, 180%, and 240% of the total debris available at both strainers.

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

Percent of Test Flow through the Main Strainer

Effect on the System Head Loss Testing Head Loss Curves For DebrisfTest Row Variation 9

RS 240%

of Total Available Debris I

~

3*

IV1S 240%

i ofAtal Available Debris 8

7

~6 2

.=-

5 III

.3 III 4

IV1S 180%

"tl ltl ClI of Total Available Debris

I:

2 MS120"/o of Total Available Debris

~ - 5?:S...---==:s.

=

o.

~. :s;;:~ -

I 0%

20%

40%

60%

80%

100%

120%

Percent Test Flow through the Main Strainer

• MS 120% Debris _ IV1S 180% Debris A IV1S 240% Debris. RS 120% Debris _ RS 180% Debris A RS 240% Debris The corresponding curves for the remote strainer can be overlayed on those of the main strainer.

Increasing flow through the main strainer reduces flow through the remote strainer, and vice versa.

120% of Total Available Debris on Mlin Strainer RS 240%

of Total Available Debris

=6,1687x2-12.337x +6,1687 1

1, 2,

~

~

Effect on the System Head Loss Testing Head Loss Curves Row Variation Testing Head Loss Curves Row Variation 120% Debris on Main Strainer 1240% Debris on Remote Strainer 180%Debris on Main Strainer 1180% Debris on Remote Strainer 9 r-,- - - - - - - - - - - - - - - - - - - - - - - - - -,

8 MaxilTlJm Head Loss Achievable for 240% of Total Available Debris on Remote MlxilTlJm Head Loss Achievable for Strainer

-6

-6 o

180% of Total Available Debris on Remote

t:

t:

Strainer

.~ 5

.= 5 180% of Total Available Debris on Mlin

~180%

~4

~4 Strainer RS 180%

anoTaI Available-cebris y=1.7226x2 - 3.6611x +1.9385 of Total Available Debris

"'C

"'C y=2.3842x2+0.254x

~3

~3 I

~ ~120%

I of Total Available Debris o

+---.

y=

+

0.6166x2 0.5974x 0%

20%

40%

60%

80%

100%

120%

0%

20%

40%

60%

80%

100%

120%

Percent Test Row through the Main Strainer Percent Test Flow through the Main Strainer

• MS 120% Debris! RS 240% Debris II ~ 180% Debris I RS 180% Debris The intersection of any two curves represents the maximum head loss achievable for those debris loads By plotting the maximum head loss for multiple curves with the same total debris load we can determine the effect of the debris split on head loss

o Aligned with your needs.

Effect on the System Head Loss System Head Loss Vs. Debris Split for Debris Load Equivalent to 360% of Total Available Debris 0.70 " ----------~-------------------

0.60 +1-----------------------------------------1

~

I o50t-------------------------~IL-'

o I

I N;OM

~

~

-I "g 0.30 I

\\

CI>

180% on Main Strainer

~

180% on Remote Strainer 0.20 I 120% on Main Strainer 240% on Remote Strai 0.10 I

==---.-T 0.00 I 1

1 1

1 1

I 0%

60%

120%

180%

240%

300%

360%

Percentage of Total Available Debris on Main Strainer I

I I

I I

)

I 360%

300%

240%

180%

120%

60%

0%

Percentage of Total Available Debris on Remote Head Loss increases with percentage of debris load allotted to the main strainer Why?

AligneCl wlth'-you ffect on the System Head Loss 1.---- Waterway Losses HL-Re:mote Strainer

~

HL \\Vaterway

'8. ----.

QR -----.

Q Q ----.

QM QM Hr.-:?v1am Strainer

• As more debris is allotted to accumulate on the main strainer, the flow rate increases through the remote strainer flow path and reduces through the main strainer to maintain equivalent system head loss.

Increased flow rate through the remote strainer path increases head loss from the waterway.

Increased debris on the main strainer causes increased flow through the remote strainer path. Increased flow through the remote strainer path causes increased system head loss due to the presence of the ductwork/waterway.

1000.00 Effect on the System Head oss System Head Loss Vs. Debris Split ooes Case Debris Quantmes Tested on the Main Strainer Vs. Quantities Available for Debris Load Equivalent to 360% of Total Available Debris 0.70 r,-------------------------

Quantity of Particulate Debris Tested on 900.00

---

therv1ain-Strainer 0.001

i 600.00 700.00

~I 0

! 600.00 Total Quantity of Particulate

~ OAO I

£

~

Debris Available at both

~

~

500.00

--straineTs--"-

~

o o

..l

~

Quantity of Fibrous

~

-:° 30 1

~

400.00 Debrlstest-ed-on--- Total Quantity of x

the Main Strainer Fibrous Debris 300.00

\\

Available alboth 0.20 I 120% on Main Strainer Strainers 240% on Remote Stra, 200.00 0.10 I 100.00

~OOI I

0.00 0%

60%

1200/,

180%

240%

300%

360%

Fibrous Debris

/

"7' I

.... Ii

==--.-r I

~

j

\\

/

Particulate Debris P.rcent~g. of Total Available Debris on Mt3ln Str.tin.r Debris Type I

I I

I I

I I

360%

300%

240%

180%

120%

60%

0%

Percenug. of Total A....all03bl. Debris on Remote System Head loss increases as the debris split favors loading on the main strainer due to the waterway losses associated with the remote strainer flow path.

Any event which would result in more debris reaching the remote strainer would only reduce system head loss by relieving the debris available to load on the main strainer.

The quantity of main strainer debris tested exceeds the total available quantity of debris

J\\lignea wiin'you Effect on the System Head Loss DGBS Case Debris Quantities Tested on the Main Strainer Vs. Quantities Available 1000.00 Quantity of Particulate Debris Tested on I

900.00 the Main Strainer-800.00 70000 -.

lil 600.00 g

Total Quantity ofParticulate

~

Debris Available at both c:

500.00

- Strainers 0

~..

III Quantity of Fibrous 400.00 0

Debris Tested on Total Quantity of the Main Strainer Fibrous Debris I

300.00 Available at bot Strainers 200.00 100.00 000 Particulate Debris Debris Type The most conservative debris split is 100% of the available debris accumulating on the main strainer.

The testing used in the design basis already used more than 100% of the available debris on the main strainer (with additional debris accumulated on the remote strainer).

Fibrous Debris

g. Aligned with voulr nl!ed!~.

Conclusion

• The debris quantity tested exceeds and bounds the most conservative debris split, making the testing debris quantities conservative.

• If retesting were to be performed, the realized head losses would be significantly lower regardless of the debris split.

RAls for Flow & Debris Distribution

• The ALIGN presentation provided an analysis that demonstrates that if testing had been performed with the expected plant debris quantities, the resulting system head loss would have been significantly lower regardless of the flow and debris split between the main and remote strainers. The primary reason for this can be seen in the table below.

Total Tested Particulate DEGB Ibs Total Available Particulate DEGB Ibs Total Tested Fibrous DEGB fe Total Available Fibrous DEGB te Total Tested Particulate DGBS Ibs Total Available Particulate DGBS Ibs Total Tested Fibrous DGBS te Total Available Fibrous DGBS te 1799.41 760.67 13.211 7.42 1367.09 329.86 13.211 7.40 Available Particulate per Unit Strainer Area Ibs/W 0.973 0.411 0.739 0.178 Available Fibrous per Unit Strainer Area ft3 Jft2 0.007 0.004 0.007 0.004 14

RAls for Chemical Effects & Bump-Up Factor

• RAI-13 Reflective Metallic Insulation (RMI) debris bed in front of the strainer could result in non-conservative head loss values

• RAI-16b Higher non-chemical debris head loss prior to chemical addition could affect the calculated bump-up factor and a higher particulate to fiber ratio could result in a lower increase in head loss following chemical addition 15

RAls for Chemical Effects & Bump-Up Factor

• I&M acknowledges that the RMI debris bed present during testing would not be expected in the plant but is considered to have minimal impact on the resulting chemical effects head loss This testing was not determining an overall strainer head loss since only the main strainer portion (most heavily loaded with debris) was tested The RMI was initially in the bottom of the test flume which allowed the rest of the debris materials to reach the strainer pockets As a result of the design of the floor of the test flume, an upward lift was created which repositioned the RMI debris bed over several hours, after the particulate and fibrous debris bed had formed in the pockets

- It should also be noted that RMI will capture other debris sources regardless of where the RMI may be in the containment pool for which no credit is taken

• Review of other industry tests that used RMI as part of their test sequence resulted in the following observations (with eel strainers):

A high fiber plant had a significant increase in head loss following chemical addition

- Two tests for low fiber plants had a slight increase in head loss in one test and a slight decrease in head loss in an identical test 16

RAls for Chemical Effects & Bump-Up Factor

• As can be seen in the picture below, the chemical precipitant fully penetrated the RMI debris bed to interact with the fiber/particulate debris bed in the pockets 17

RAls for Chemical Effects &Bump-Up Factor

• A further evaluation was performed of the plants that have performed testing with CCI pocket strainers. The table presented below provides key points from this review:

Plant Strainer Opening Size (in.)

Tested Fiber Bed Thickness (in.)

Tested Flow Rate per Unit Strainer Area (gpm/ff)

Debris Only Head Loss (in. H2O)

Post 100%

Chem.

Add Head Loss (in. H2O)

Chemical Precipitate Added to I Injected in Test Loop Increase Factor (Bump-up)

A 0.083 0.053 4.72 20.9 26.9 2100 ppm AI 480 ppm Ca 220 ppm Si 1.3 B

0.083 0.091 2.41 7.2 20.1 1.138 kg NaAISbOa 0.566 kg AIOOH 0.477 kg Ca3(P04)2 2.8 Cook (DEGB) 0.083 0.10 10.5 32.04 45.96 1600 ppm AI 2700 ppm Ca 3800 ppm Si 1.43 Cook (DGBS) 0.083 0.10 10.5 53.16 81.6 1600 ppm AI 2700 ppm Ca 3800 ppm Si 1.53 C

0.083 0.3 2.16 97.23 117.43 2.577 kg NaAISi30 a 0.749 kg AIOOH 1.21 D

0.083 0.9 2.07 13.05 57.0 2.961 kg NaAISi30 a 0.599 kg AIOOH 4.37 E

0.063 0.069 4.70 19.2 38.4 5.058 kg NaAISi30 a 7.779 ka AIOOH 2.0 F

0.063 0.134 4.15 10.68

< 96 (1)

> 4.29 kg NaAISi30 a

< 8.99 G

0.063 0.03 1.56

< 12.0

< 42 2.96 kg NaAISi30 a 3.5 H

0.063 0.238 0.58 4.01 40.1 1.398 kg NaAISbOa 10 I

0.063 Information not readily available (1) Predicted Maximum Head Loss 18

RAls for Chemical Effects & Bump-Up Factor

• The plants in the section of the table above the gray line are those with strainer openings the same size as Cook

• Of those plants, the increase factor is generally shown to be related to the fiber bed thickness for thinner fiber beds, with one exception One of the plants formed calcium phosphate precipitate which resulted in a slightly higher increase factor than Cook

- This precipitate has been shown to cause significantly higher head losses across strainers as compared to other precipitates

• As demonstrated in the table, there is some variability associated with determining the increase factor as a result of chemical effects

• If Cook had performed testing with the actual quantity of fiber in containment (+10 % margin), the fiber bed thickness would be ~

0.053 inches

- Comparing this value to Plant A in the table shows that the tested head loss increase factor for Cook is consistent with Plant A 19

RAls for Chemical Effects & Bump-Up Factor

• Cook has conservatively applied an overall system head loss increase factor of approximately 2.5 above the tested debris only head loss value This was done to account for uncertainty in test methodology including the quantity of debris that could be expected to participate in strainer head loss Tested System Head Loss (ft H2O)

Licensing Basis System Head Loss (ft H2O)

Increase Factor DGBS 0.82 2.09 2.55 DEGB 1.046 2.67 2.55 The stated head loss values are the 68°F normalized values Based on the assumed increase in head loss above the debris only values, an appropriate bump-up factor for chemical effects exists 20

RAls for Chemical Effects & Bump-Up Factor

• I&M judges that the approach used for developing the bump-up factor was reasonable and conservative

• A highly compacted debris bed limits the flow paths through the bed resulting in a debris bed that would be more susceptible to the effects of chemical precipitate addition

• The chemical precipitates resulted in an approximate 50% increase in head loss across the strainer

• Decreasing the particulate to fiber ratio (since there is very little fiber) would result in a more porous bed and resultant lower head loss following chemical addition The particle size for the chemical precipitates is significantly smaller than the size of the particulate debris sources expected to exist in containment

- With insufficient fiber to weave the debris bed together, the addition of chemical precipitates will not be able to create a significant increase in head loss 21

Margins Available for Debris/Flow Splits and Chemical Effects Bump-Up Factors

• For the DEGB, the Alternate Analysis criteria of Section 6 of NEI 04-07 was utilized with the necessary qualified equipment installed to alert the Operators, and the necessary procedural controls in place to reduce head loss across the strainers while maintaining licensing basis core and containment cooling functions

• An increase factor of ~ 2.5 was applied to the calculated strainer system head loss value obtained from testing with quantities of debris significantly in excess of those available within containment that contribute to strainer head loss (See Table on next slide) 22

Actual Actual DEGB Quantity DGBS Quantity Debris Type I Units I Test Available Margin Test Available I Margin Quantity at Both Quantity at Both Strainers Strainers lbs 307.665 298.82 77.227 74.94 Ibs 0.188 0.1894 0

0 lbs 1.5228 1.534 1.1285 1.136 Ibs 1.52 1.536 0

0 Ibs 203.585 207.4 3.8 3.84 Ibs 0.57 1.82 0.57 0.57 lbs 19.712 16.9 19.712 16.9 lbs 78.416 74.4 78.416 74.4 8.32 16.12 8.32 16.12 Ibs 4.212 3.4 4.212 I

3.4 Ibs 995.2 38.88 178.5 99.67 1799.41 760.67 Latent Fiber 13.125 7.33 13.125 7.33 Fire Proof Ta e Fines 0.057 0.0576

.057 0.0456 Ice Stora e Ba Fibers 0.0273 0.026 0.0273 0.026 Ice Storage Bag Liner 0.000236 0.00022 0.000236 0.00022 Shards ft3 Pieces of Work Platform ft3 0.0021 0.002

~~I 0.0021 I 0.002 Rubber

-n.,~_<~"

____... _.* ~-",'..<fl' 13.2116 7.41582 15.79848 I 13.2116 I 7.40382.

23

Margins Available for Debris/Flow Splits and Chemical Effects Bump-Up Factors

• Additional margin exists due to conservatisms taken for testing and test results Strainer system head loss values were normalized to 68°F which is below the expected lower temperature of 100°F. At 100°F, the head loss would be== 30%

less The flow rate assumed for testing was== 1000 gpm greater than the conservatively determined maximum flow rates for both trains of ECCS and CTS taking suction on the recirculation sump. Reducing flow by this quantity would result in an== 20% reduction in head loss Tested System Head Loss (ft H2O)

System Head Loss Expected (ft H2O)

Licensing Basis System Head Loss (ft H2O)

Increase Factor DGBS 0.82 0.46 2.09 4.54 DEGB 1.046 0.59 2.67 4.53 Additional margin would exist if strainer testing was performed with the expected debris quantities which would result in a lower system head loss.

24

Conclusions for Debris/Flow Splits and Chemical Effects Bump-Up Factors Issues

• As provided within this presentation, significant margins exist for demonstrating that the results of the strainer testing that was performed provide substantial basis for establishing:

I&M has demonstrated reasonable assurance that the installed recirculation sump strainer system will perform its required design function of providing the necessary core and containment cooling in the unlikely event of a LOCA.

25

Additional RAls With Open Questions

• RAI2a Radial decay of pressure associated with nozzle size used for destructive testing of Marinite board material

• RAI4 Installed configuration of jacketing system on Cal-Sil installed at Cook compared to the configuration of Cal-Sil tested at OPG

• RAI7d Comparison of the flow velocities in the Cal-Sil erosion test loop compared to the velocities that exist within Cook's containment pool

• RAI25b Quantity of debris that would be generated from a small break in the pressurizer enclosure compared to the quantity that would be generated for a DGBS 26

RAI2a

• One of the generic questions associated with two-phase jet testing of materials is the effect of the radial decay of pressure away from the jet centerline

• For Cook, the failure mechanism that resulted in debris generation of Marinite board debris was the physical deformation of the unsupported and unrestrained cable tray section that was tested

- As the cable tray deformed, a lever was applied between the face edges of the Marinite that was attached to the cable tray

• NUREG/CR-6772 established a destruction pressure of 64 psi (a ZOI of==:: 3D) for Marinite which is the pressure at which damage starts to occur

• The Marinite that is closest to a bounding break location is==:: 40 27

RAI2a

• As can be seen in the pictures on the next two slides, the cable tray that was tested at a ZOI of ~ 3.4D was deformed to the point that further destruction could not reasonably occur

• For conservatism, the quantity of debris generated from the breaks that resulted in debris generation (a ZOI of less than 5.5D) was applied to all Marinite installations out to a ZOI of 9.80

• The total quantity of Marinite available for debris generation is just a small fraction of the total particulate debris sources available (~

0.1%)

I&M judges that the effects of radial decay had insignificant impact on the total quantity of Marinite debris assumed to be generated 28

0)

N

~

en

~

CD L

S CD CO

~

eo LI CD

-..c eo o c

o Q)

~

c Leo

~

o

('I')

+-'

en

~

L (J.)

I I '

<:(

~

eo LI (J.)-.0 eo

()

c o (J.)

+-'

c Leo

~

RAI4 For jacketed insulation within containment, the governing engineering specification requires that banding (20 mil) with seals (crimp lock clamp device) be placed no more than 12" apart except for foam insulation installations which requires a maximum 6" spacing

• The OPG testing utilized a maximum spacing of 8.25"

• The failure mode during the OPG testing was shearing of the aluminum jacket adjacent to the banding

• Cook uses stainless steel jacketing in containment which has a substantially higher shear strength than the aluminum

• The increased spacing will not result in a significant increase in the quantity of Cal-Sil pieces generated following a LOCA due to the increased strength of the materials

• The pictures on the following slide show typical installations in Cook's containment 31

~

o o

U

+-'

ro c o ro

+-'

-ro

+-'

en c

c o

+-'

ro

J en c-CJ)

I ro U

RAI7d For RAI 7d, ALIGN will present additional information to support this item.

During previous interactions with the NRC, we had stated that we would perform an analysis of the test loop for comparison to the plant conditions. We have subsequently determined that this analysis may not be necessary based on an evaluation of the available information.

33

NRC - Donald C. Cook Nuclear Plant Public Meeting Containment Recirculation Sump Performance Support of Erosion Parameter Comparison October 14, 2009 A L ION SCIENCE AND TECHNOLOGY

Erosion Parameter Comparison Results

• The test velocity was >2 (-2 to 3) times greater than the average pool velocity for the non transporting portions of the pool that the erosion factor was applied to.

• The pool TKE is insignificant (3 to 4%

) as compared to the pool kinetic energy

• The test kinetic energy is >4 (4 to 13) times the pool kinetic energy.

• The fact that the samples obstruct a portion of the test flow area causing higher velocities is additional conservatism.

• Although the exact TKE level in the tests is not known, the velocity is high enough for turbulence to occur.

g ***~i~~With your needs.

Test Flow Test Flow Non-Transport Test Flow Velocity +

Velocity/Non-Average Pool Velocity 30% From Transport Velocity Blocking Average Pool Screen Velocity Material & Size ft / sec ft / sec ft / sec Cal-Sil Small 0.11 0.40 0.52 364%

Marinite Small 0.11 0040 0.52 364%

222%

Non Transport Turbulent K' rEI T t FI KE I Test Flow + ITKE Pool/KE I KE TesUKE Average Pool me Ie nergy es ow 30%

KE PIP I (TKE) 0, 00 00 KE Material & Size ft"2 / see"2 ft"2 / see"2 ft"2 / see"2 ft"2 / see"2 Cal-Sil Small 0.0061 0.0002 0.0800 0.1352 4%

1322%

Marinite Small 0.0061 0.0002 0.0800 0.1352 4%

1322%

Cal-Sil Large I

0.0162 I

0.0005 I

0.0800 I

Marinite Laroe 0.1352 I

3%

I 494%

Average velocity in non-transport regions for small pieces of Cal-Sil and Marinite E=Integral/Area = 1.0ge - 1 VelMog

Average velocity in non-transport regions for large pieces of Cal-Sil and Marinite VelMog E=Inlegral/Area= 1.84e-l

TKE for non-transporting small pieces of Cal-Sil and Marinite E=Integral/Area =2.45e - 4 tke 0.490 0.327 0.163 0.000

~

~

TKE for non-transporting large pieces of Cal-Sil and Marinite E=Integral/Area =5.32e-4 tke 0.610 0.407 0.203 0.000

AJigr'ed with your needs.

Erosion Parameter Comparison Conclusion From Slide 2

• The test velocity was >2 (-2 to 3) times greater than the average pool velocity for the non transporting portions of the pool that the erosion factor was applied to.

• The pool TKE is insignificant (3 to 4%) as compared to the pool kinetic energy

• The test kinetic energy is >4 (4 to 13) times the pool kinetic energy.

• The fact that the samples obstruct a portion of the test flow area causing higher velocities is additional conservatism.

• Although the exact TKE level in the tests is not known, the velocity is high enough for turbulence to occur.

• Therefore, it can be reasonably concluded that the flow conditions in the erosion test were conservative without doing extensive additional analysis to calculate the test TKE.

RAI25b For RAI 25b, the analysis of the quantity of debris generated within the pressurizer enclosure as a result of a single sided break of a 6" pipe and double ended 4" pipe is ongoing The primary debris source that exists at these break locations is Cal Sil insulation Due to the relatively short distances associated with ZOls of concern, it is expected that the total quantity of debris will be significantly less than the quantity from the DGBS The analytical information for this RAI will be available to support timely response to the RAls 34

Conclusion I&M has demonstrated through rigorous and extensive analysis and testing installation of significant modifications in the Cook units implementation of programmatic controls to maintain the debris source term within necessary limits and the use of significant margins and conservatisms That reasonable assurance exists that the recirculation sump strainer system and all interconnected components will function to satisfy the required functions of core and containment cooling in the highly unlikely event of a LOCA 35

Q).!!2

>..c co+-'

..c c

J 0 o

c

~o.

~en

t= en c :J C\\-

J U

t/)

t

.!!2 (9

0-0 Z

Q..

o z

Q...c o

Q) en

Q) 0) l o

..c CO

.....J

+-' 0) t/)

L.. C U

W Q Q)

J 0

J O+-'

a

~en

~ :J c-o CO Q)

..c -0

+-'.-

Q)

Q) > :J

>een Q.. en.

- 2 The NRC staff and licensee agreed that a final submittal date for the RAI response would be February 15, 2010.

Please direct any inquiries to me at 301-415-3049, or Terry.Beltz@nrc.gov.

IRA!

Terry A. Beltz, Senior Project Manager Plant Licensing Branch 111-1 Division of Operating Reactor Licensing Office of Nuclear Reactor Regulation Docket Nos. 50-315 and 50-316

Enclosures:

1. List of Attendees

2. Meeting Handout cc w/encls: Distribution via Listserv DISTRIBUTION:

PUBLIC RidsRgn3MailCenter Resource PKlein, NRRlDCI/CSGB LPL3-1 RlF CTucci, NRR EGeiger, NRRlDSS/SSIB RidsAcrsAcnw_MailCTR Resource SBailey, NRRlDORL JLehning, NRRlDSS/SSIB RidsNrrDorlLpl3-1 Resource MScott, NRRlDSS/SSIB SBagley, EDO Region 3 RidsNrrPMDCCook Resource SSmith, NRRlDSS/SSIB JSavoy, NRRlDSS RidsNrrLATHarris Resource MYoder, NRRlDCI/CSGB BUn, RES RidsNrrDssSsib Resource WJessup, NRRlDE/EMCB RArchitzel, NRRlDSS/SSIB RidsOgcRp Resource ADAMS Accession Numbers' ML093200312 OFFICE DORULPL3-1/PM DORULPL3-1/LA DSS/SSIB/BC(A)

DORULPL3-1/BC NAME TBeitz THarris SBailey RPascarelli DATE 11/16/09 11/17/09 11/18/09 11/20109 OFFICIAL RECORD COPY