ML062650095

LTR-06-0470 - Robert Leyse E-Mail Re Tube Replacement for Main Condenser at Columbia Generating Station

ML062650095
Person / Time
Site:	Columbia
Issue date:	09/19/2006
From:	Leyse R - No Known Affiliation
To:	Trever K Office of Nuclear Reactor Regulation, State of ID, Dept of Environmental Quality
References
LTR-06-0470
Download: ML062650095 (29)

Text

OFFICE OF THE SECRETARY CORRESPONDENCE CONTROL TICKET Date Printed: Sep 21, 2006 09:53 PAPER NUMBER:

ACTION OFFICE:

LTR-06-0470 EDO LOGGING DATE: 09/20/2006

-7o -.

A AUTHOR:

AFFILIATION:

ADDRESSEE:

SUBJECT:

ACTION:

DISTRIBUTION:

LETTER DATE:

ACKNOWLEDGED SPECIAL HANDLING:

Mr. Robert Leyse ID Mrs. Kathleen Trever, ID Dept of Environ. Quality Tub replacement for main condenser at Columbia Generating Station Appropriate RF 09/19/2006 No Immediate release to the public via SECY/EDO/DPC EDO DEDURS DEDR DEDIA AO NOTES:

FILE LOCATION:

ADAMS DATE DUE:

DATE SIGNED:

7eVý;

CHAIRMAN

- Tube replacement for main condenser at Columbia Page 1 CHAIRMAN - Tube replacement for main condenser at Columbia I

Page 11I From:

<Bobleyse @aol.com>

To:

<kathleen.trever@deq.idaho.gov>

Date:

Tue, Sep 19, 2006 8:17 PM

Subject:

Tube replacement for main condenser at Columbia Kathleen:

Five years ago we had e-mail exchanges regarding the Columbia Generating Station. You contacted the NRC and they correctly responded that Columbia was well aware of the situation at the River Bend BWR in Louisiana (fouling of fuel elements). Of course, as usual, the NRC (and Columbia) did not elaborate.

The other day I was crawling around GOOGLE and I stumbled across the following:

www.energy-northwest.com/downloads/Main%20Condenser.pdf and www.energy-northwest.com/downloads/Main%20Condenser%2OAddendum%201

.pdf I do not believe that Columbia intended for these to hit the fan, but they are out there.

So, to finalize. INL is the center for Nuclear Power Technology in the USA.

The INL experts, independent of EPRI, should look into the situations that are outlined in these two documents. My belief is that Columbia has postponed tube replacement in the main condenser for too long.

In the hands of experts the two documents will speak for themselves. So, please let me know what INL thinks about all this.

The Chairman, NRC should insure that NRC is aware of the two referenced documents and that they are placed into the PDR.

Bob Robert H. Leyse bobleyse@aol.com P. O. Box 2850 Sun Valley, ID 83353 (208) 622-7740 CC:

<Chairman @ NRC.gov>

c:ktemp\\GW}O0001.TMP Mail Envelope Properties (45108885.F5A : 21: 3930)

Page 1 I

Subject:

Creation Date From:

Created By:

Tube replacement for main condenser at Columbia Tue, Sep 19, 2006 8:16 PM

Bobleyse@aol.com Recipients nrc.gov OWGWPO02.HQGWDO01 CHAIRMAN CC deq.idaho.gov kathleen.trever Post Office OWGWPOO2.HQGWDOO1 Route nrc.gov deq.idaho.gov Files MESSAGE TEXT.htm Mime.822 Options Expiration Date:

Priority:

ReplyRequested:

Return Notification:

Concealed

Subject:

Security:

Size 1285 2727 5465 Date & Time Tuesday, September 19, 2006 8:16 PM None Standard No None No Standard Junk Mail Handling Evaluation Results Message is eligible for Junk Mail handling This message was not classified as Junk Mail Junk Mail settings when this message was delivered Junk Mail handling disabled by User Junk Mail handling disabled by Administrator Junk List is not enabled Junk Mail using personal address books is not enabled Block List is not enabled

I-Columbia Generating Station Main Condenser By W. Scott Oxenford, VP Technical Services This document summarizes longstanding performance issues related to the design and operation of the Main Condenser at Columbia Generating Station and solutions to those challenges.

The following categories summarize the issues in introductory level detail:

1. System Components Overview

2. Condenser Leakage

3. Columbia's Condenser

4. Columbia Historical Actions

5. Can Columbia Eliminate Condenser Leakage by Eliminating Debris?

6. Industry Data and Experience

7. Solutions Addendum 1 has been included to highlight costs associated with condenser leakage.

Section 1: System Components Overview The Purpose of the Main Condenser The main condenser is a key component in the closed-loop system that transfers energy from the reactor to the turbine, in support of creating electricity.

The condenser's primary function is to take steam exhaust from the main turbine and return it to a liquid form.

The liquid, called condensate, is highly purified water.

The condensate is preheated and pumped back to the reactor pressure vessel where energy in the form of heat is added to convert the condensate back into steam.

See Figure 1 for a simplified diagram (page 9)

Condenser Configuration The Columbia condenser has three main sections:

1) The steam space, into which the turbine exhaust and other steam sources discharge.

2) The cooling section, where steam passes over brass water pipes filled with cold circulating water, causing the steam to condense back into water (condensate).

3) The circulating water interface consisting of waterboxes, tubesheets, and thousands of tubes. Chemically treated Columbia River water is circulated through special brass tubes, removing heat and transferring it to the Cooling Towers.

CGS Main Condenser White Paper Revision 1, June 2006 Page 1 of 22

I-Cooling Towers The cooling towers provide a water source for the circulating water system and transfer heat to the environment. Evaporative losses are replaced with water from the Columbia River. By nature, the cooling towers act to concentrate debris and impurities. Acid and other chemicals are added to reduce secondary system fouling and corrosion.

It is paramount to prevent the chemical additives and raw water/organic materials from degrading reactor coolant quality.

Section 2: Condenser Leakage What Happens During a Small Condenser Tube Leak?

Due to pressure differences, small amounts of chemically treated raw cooling water (circulating water) are transferred to the pure reactor grade water in the condensate system.

This water is then run through filters, removing some of the introduced impurities. Impurities that get through the filters then go to the reactor vessel, where they begin concentrating. The concentration takes place as the pure water boils into steam and the impurities are left behind.

As condenser leakage increases, the filters become less and less effective at removing the impurities.

Limitations to the Columbia Design Some Boiling Water Reactors (BWRs) were designed with, or subsequently added, additional filters in their condensate systems. Called deep bed demineralizers, they are located between the filters like those at Columbia and the reactor pressure vessel. The deep bed demineralizers enhance filtration and water quality prior to entering the reactor. They allow continued safe operation with much greater condenser leakage than is possible with Columbia's design.

Additionally, the filters are effective at removing copper, which will be discussed later.

Results of a Small Condenser Leak Even a small condenser leak has negative consequences for Columbia, including:

o Filters must be changed more frequently to keep the water as pure as possible.

Changing filters twice as often (frequently required) increases the cost of the filter media (resin) and the associated disposal cost (radioactive waste that needs to be buried). Each of these operations and disposal maneuvers also impact labor costs and employee dose (radiation exposure).

o Once a leak is large enough to locate, the plant is reduced to approximately 60%

power to pinpoint and repair the leakage. Pinpointing and repairing leakage from multiple, small locations, is especially difficult. Each repair entails unplanned CGS Main Condenser White Paper Revision 1, June 2006 Page 2 of 22

5-_

generation losses, employee exposure, personnel safety hazards, and increased labor costs.

See Figures 2A-B (pages 10-11) o Water quality (chemistry) within the reactor degrades. This can:

1) Result in unplanned power reductions or mandatory shutdowns due to exceeding chemistry limits.

2) Increase activation of impurities, which increases radioactive contamination and exposure throughout the plant.

3) Increase the susceptibility of the reactor vessel and internals to cracking, increasing the likelihood for costly repairs.

4) Disturb the corrosion layer on the fuel and reactor internals. Impacting the fuel corrosion layer can lead to fuel damage. Fuel damage can result in plant de-rate, unplanned refueling outages, increased dose rates and employee exposure, and increased stack release rates to the environment.

Section 3: Columbia's Condenser Admiralty Brass Material Columbia's condenser tubes, like many original condensers, were fabricated from admiralty brass. Admiralty brass is made primarily of copper, with the second largest constituent being nickel. It was selected for its excellent heat transfer efficiency and inexpensive cost relative to other suitable materials.

Susceptibility to Mechanical Wear Admiralty brass is more susceptible to damage than the other contemporary condenser materials like stainless steel and titanium. For example, plastic tie wraps have caused leaks in our condenser when they became lodged at the inlet end of condenser tubes and, moved by water flow, wore holes in the soft metal tubes. Titanium is approximately 6.5 times harder and stainless steel is about 3 times harder, making them less susceptible to debris induced damage.

Likewise, steam leakage from exhaust lines into the condenser has been known to wear through tubes, leading to rapid increases in condenser leakage and prompt shutdown of the plant to protect primary system chemistry.

Copper and Fuel Even slow wear of the soft condenser tube material adds copper to the condensate.

Columbia demineralizers are not designed for mechanical filtration, the best method for removal of copper. Based on that and the lack of deep bed demineralizers, Columbia is classified as a 'high copper plant'.

Copper's substantial negative impacts on Reactor fuel integrity were identified in the early days of Boiling Water Reactors. A phenomena, known as Crud Induced Localized CGS Main Condenser White Paper Revision 1, June 2006 Page 3 of 22

J

Corrosion (CILC), is caused by copper entering the reactor, attaching itself to the fuel corrosion layer, and causing localized corrosion and high temperature areas on the fuel cladding. The eventual outcome is often loss of clad integrity and long axial splits of the clad. That allows fission products to spread throughout the plant and ultimately cause increased release rates to the environment. CILC has rendered large quantities of fuel unusable, costing tens of millions of dollars and extended reduced power operation at some units.

See Figure 3 (page 12)

The industry has substantially lessened, but not eliminated, CILC failures by removing copper from their condenser materials or adding deep bed demineralizers. Columbia has done neither, leaving us susceptible to CILC fuel failures. Columbia has carefully selected fuel cladding to minimize the risk of CILC failure. However, the only real way to rule this failure mechanism out is to remove the source of copper completely.

In addition to fuel impacts, copper is also implicated in the trapping of cobalt in the corrosion layers on all reactor internals. This increases overall plant radiation levels and dose to our employees.

Early Condenser Damage Poor chemistry control in the early years of Columbia's operations caused corrosion of the condenser tubes. One result is a phenomena, called dezincification, which caused pits in the condenser tube metal.

The pits remain and provide initiation sites for localized corrosion and subsequent tube leaks/failures. Based on this, Columbia cleans and "eddy current tests" one-third of our condenser system per outage.

With our transition to two-year cycles, Columbia needs to start conducting eddy current tests of all the condenser tubes each outage, starting with R1 8 in 2007. The $700k per outage testing cost could be substantially reduced with improved condenser material condition.

Section 4: Columbia Historical Actions Columbia management conducted a condenser replacement study in 1996. No action was taken on the study results based on unfavorable payback expectations and extended outage time for condenser tube replacement. A key factor at the time was the uncertainty of plant license extension.

Our concerns resurfaced in 1999 following fuel failures at the River Bend plant (admiralty brass condenser with deep bed demineralizers).

We closely tracked the River Bend cause analysis. River Bend fuel corrosion had high copper levels, but the failures were attributed to high iron levels in their corrosion layer. Based on Columbia being a low iron plant, no action was taken.

See Figure 4 (page 13)

CGS Main Condenser White Paper Revision 1, June 2006 Page 4 of 22

During 2001, Columbia found that some of its fuel had thicker than expected oxide layers. A root cause team, with industry expertise, studied the previous operating cycle and fuel scrapings. The fuel had higher than normal levels of copper and iron deposits.

Concerns over probable fuel damage were high.

Spallation (similar to concrete spallation where material falls off) was identified on Columbia fuel.

See Figure 5A, B, C (pages 14-16)

The root cause was determined to be poor demineralizer performance, coinciding with a chemical intrusion due to condenser system leakage. Had the condenser not leaked this challenge to the fuel would not have occurred.

In 2003, another copper reduction study was initiated that included consideration for deep bed demineralizers or removal of the admiralty brass. The recommended solution was not completed due to external cost pressures.

Columbia's management has thoroughly reviewed options for managing ongoing condenser challenges.

On each occasion, continued operating risks were accepted instead of taking action, primarily to avoid costs and extended outage length.

In addition to these studies:

" Columbia has increased the size of our demineralizers to improve filtering efficiency.

Determined the suction screen on the circulating water system was not properly seated, allowing some debris to pass.

Columbia has corrected this and implemented a long-term fix.

" Columbia rebuilt three cooling towers, upgrading the plastic fill and lattice, while removing all plastic tie wraps at $2M per tower. Three towers remain original.

• Columbia has improved our foreign material controls to reduce debris getting into circulating water and subsequently the main condenser.

" Columbia has improved waterbox drainage to allow faster and more complete draining for repairs.

" Columbia staff is looking at screen options in the upcoming outage to further reduce debris entry into the main condenser.

Section 5: Can Columbia Eliminate Condenser Leakaqe by Eliminatinq Debris?

In any raw water system, debris elimination is a challenge that plant designers are faced with. Columbia is blessed with a relatively clean water source in the Columbia River.

The Tower Make-up System takes water from the middle of the river through screens.

Columbia also has screens at the intake to the Circulating Water Pump Pits. These screens are designed to be small enough to prevent plastic tie wraps and larger debris from passing. Despite this, items pass through the screens or are in the system from CGS Main Condenser White Paper Revision 1, June 2006 Page 5 of 22

historical operation. Additionally, the screens must be manually raised for cleaning with the plant in operation which allows debris entry.

Cooling towers, due to their design purpose of transferring heat to the environment, are open and susceptible to items being blown in, dropped in from animals, and dropped in during work under adverse conditions.

Additionally, the extreme weather variations, water flows, chemical additives, and ice build-up cause corrosion and damage that create debris. Any raw water screen system can reduce debris intrusion, but none appear to be 100% effective.

See Figures 6A-D (pages 17-20)

Finally, in 2003, INPO shared a Significant Experience Report on debris intrusion. The document shares operating experience with screens becoming plugged and causing loss of pump suction. This is an issue Columbia has experienced from algae build-up on plant restarts. Reduction in screen opening size increases the likelihood of plugging.

In May of 2004, a root cause analysis was performed to prevent debris related condenser leaks. Remaining actions from that study are planned for implementation in the upcoming outage. Columbia's actions to date have reduced the quantity of debris in the condenser, and should improve more with an improved screen system. However, debris intrusion will always be an issue to some extent.

It should be noted that the root cause 'does not address tube leaks caused by steam impincqement or long term flow induced erosion or other service related tube damage such as dezincification and stress corrosion cracking'.

These failure modes were specifically excluded for the purpose of focusing on debris-related leaks, which caused the condenser leak triggering the root cause analysis.

Section 6: Relevant Industry Data Current Main Condenser Material Of the 34 US BWRs:

16 have stainless steel condensers.

10 have titanium condensers.

4 have a combination of admiralty brass with stainless steel or titanium. These sites use the less damage-susceptible materials in the highest risk areas.

Only 4 have admiralty brass condensers, including Columbia.

Our data shows at least 13 of the BWRs have re-tubed their condensers. Nearly all had admiralty brass and went to a different material.

Most occurred in the 1980's and 1990's.

See Figure 7 (pages 21-22)

CGS Main Condenser White Paper Revision 1, June 2006 Page 6 of 22

Current Plants with Deep Bed Demineralizers Of the 34 US BWRs, 20 utilize deep bed demineralizers.

Of the 8 US BWRs containing some admiralty brass in their condenser, only two operate without deep bed demineralizers to address copper. They are Columbia and Vermont Yankee. Vermont Yankee went commercial in 1972. It is a small 593 MWe BWR recently bought by Entergy.

Vermont Yankee is planning condenser tube replacement as part of license renewal.

Two of the US BWRs with a combination of admiralty brass condensers and deep bed demineralizers are Limerick Units 1 and 2. Following major CILC related fuel damage on Unit 1, Limerick added deep bed demineralizers. This option was selected over re tube because it was factored into the original design, with available space in their Turbine Building.

From this section, the case for change, based on copper alone, is strong. We are one of only two operating BWRs that have admiralty brass condensers without deep bed demineralizers.

The other BWR, Vermont Yankee, contains 8%

stainless steel tubes.

Utilities invested in their facilities to reduce risk. At this point we do not know if their investments passed a business case payback analysis or if action was taken to eliminate the large downside risk, regardless of payback.

Political Landscape Over the past several years, top focus areas of Chief Nuclear Officers have been Security, Fuel Reliability, and Materials Degradation.

In the area of materials degradation, the industry established a program called BWR Vessel Internals Protection (BWRVIP) in 1994 so we could self-regulate, rather than cause the NRC to regulate us.

The program provides research, inspection requirements, program requirements and independent audits to ensure the industry is protecting reactor pressure vessels and internals. Failure to protect these important components can have downside risks not only to individual stations, but the nuclear industry as a whole. We are currently failing to meet the BWRVIP guidance on copper in the reactor coolant, which provides a spotlight on Columbia due to having one of the more significant program deviations.

Peer pressure to eliminate long-term deviations is growing.

Section 7: Solutions:

Deep bed demineralizers alone are not a preferred solution. They will reduce copper and impurities in the reactor coolant, but condenser leakage will continue to be a chronic problem and copper impurities will remain at a lesser amount.

Unplanned downpowers and radiation exposure for condenser repairs will continue, but less frequently. Resin usage and radioactive waste will increase due to the large size of the CGS Main Condenser White Paper Revision 1, June 2006 Page 7 of 22

z deep bed demineralizers.

We also anticipate increased Security staffing due to an additional building to house the deep bed system.

On April 21, 2006 we entered into an agreement with Sargent & Lundy to perform a Feasibility Study on the main condenser. They recently did similar work for four Exelon BWRs and the Fort Calhoun Station.

This will entail a comprehensive analysis of Columbia's history and that of the industry, resulting in recommendations to ensure long-term reliability of the main condenser. The study will be complete prior to the FY08 budgeting cycle. Engineering, design, and procurement are expected to start in FY08, with installation of some or all of the modification in R19 (FY09).

Condenser material replacement is clearly the preferred solution to eliminate leaks and copper sources, ensuring long-term reliability of Columbia's fuel, reactor vessel and internals.

For maximum protection and defensive strategy, installation of deep bed demineralizers in conjunction with condenser tube replacement is another solution.

This is not currently under consideration. However, as the industry gains operating experience in fuels and materials degradation, the Columbia staff will stay abreast and take action as appropriate.

CGS Main Condenser White Paper Revision 1, June 2006 Page 8 of 22

ZCooling Tow ers]

T

/11 -

FIGURE 1 Circulating Water System

-3 air flow Wind

[

L~.

44.

I CW Basi I

CW S re n V

-J Tubes Tubes S

iCondensr Tube D

Sheet #2 E

B IR Tube S

Sheet #3 I

Tube Sheet #4

- Tube Sheet #1 Outlet Intermediate Inlet CGS Main Condenser White Paper Revision 1, June 2006 Page 9 of 22

FIGURE 2A, Main Condenser Waterbox Entry into the condenser waterboxes at power is a challenge to personnel safety. Single isolation valves (some eight feet in diameter) provide worker protection from system pressure. Entry is through small manways as shown in the picture.

CGS Main Condenser White Paper Revision 1, June 2006 Page 10 of 22

FIGURE 2B, Main Condenser Waterbox This shows a worker inside the condenser waterbox.

CGS Main Condenser White Paper Revision 1, June 2006 Page 11 of 22

FIGURE 3, Fuel that has undergone CILC related failures.

NOON=-=&

,L-CGS Main Condenser White Paper Revision 1, June 2006 Page 12 of 22

i FIGURE 4, Failed fuel due to crud and accelerated corrosion.

This fuel was exposed to anomalous primary coolant chemistry, resulting in a heavy oxide layer and accelerated corrosion in 1999. It was fresh fuel.

-~

IJ~ 4

(

dl

'ci V 7 I

CGS Main Condenser White Paper Revision 1, June 2006 Page 13 of 22

FIGURE 5A, One Cycle Fuel in expected condition.

The middle fuel rod has been brushed to remove the oxide layer.

CGS 1-Cycle Bundle (Before Chronic Condenser Leak) 1:

A2 CGS Main Condenser White Paper Revision 1, June 2006 Page 14 of 22

FIGURE 5B, One cycle fuel with it's oxide layer impacted by chemical contamination.

Nodule formation has begun.

CGS 1-Cycle Bundle (After Chronic Condenser Leak)

-. ~,

'"~1 CGS Main Condenser White Paper Revision 1, June 2006 Page 15 of 22

FIGURE 5C, Fuel that has been in the reactor for four cycles, following chemical contamination in the last cycle. Nodule formation and spallation is evident.

CGS 4-Cycle Bundle (After Chronic Condenser Leak)

CGS Main Condenser White Paper Revision 1, June 2006 Page 16 of 22

FIGURE 6A, Cooling Tower As illustrated, the cooling towers have many openings to the environment that allow debris entry.

NORMAL.

MULTIBLATZ PAN OPERAThrG WATER LEVIL DIEIRIUTI.ON FLUME OISTRISBIXIM BASIN PERIMETER WINO ICREEN I*-

c..w INLET PRECArT R

LOUVERS TO'C.W.

TOWYER SY1MMETRICAL ABOUT C PUMV INLET FLUME 6-90931.4 LT JAN 19N.II CWITPMIJ FIGURE 4. COOLING TOWER CROSS SECTION CGS Main Condenser White Paper Revision 1, June 2006 Page 17 of 22

FIGURE 6B, Cooling Tower.

Large fans and vertical louvers allow debris entry pathways.

CGS Main Condenser White Paper Revision 1, June 2006 Page 18 of 22

FIGURE 6C, Cooling Tower Inside a drained Cooling Tower. Workers are careful to remove debris prior to returning it to service.

ii CGS Main Condenser White Paper Revision 1, June 2006 Page 19 of 22

FIGURE 6D, Cooling Tower.

This drained Cooling Tower demonstrates how open to the environment they are. When in service, water fills these passages and falls to the basin below.

I II I '4 I

41.

CGS Main Condenser White Paper Revision 1, June 2006 Page 20 of 22

FIGURE 7, FERMI Condenser Replacement Article INDUSTRY NOTES PROCESS/MANUFACTURING ILIIES Sleam gra na-n Preplanning, analysis key to condenser retubing Desabdon of condensers and the poota Ws!

for 5hed ouases or plant od sWrce

'tzs onsntd dhat ulte continualy ovrl.

mate the possible need for condenser powerlats has been pz Jv~si

d. bedtwettr 2n5%;5=u*

ie lb-seamm-generoar corrosion tn pressut imd-w w

(PWlR

) plants and foa meembi corrosion In bolllg-wa r-li ta (s*Jp.

51c***s~s*

1-0=1 Ust A, a taWK pln or Loemt Edison Co. is located on Lake Erie. a freivah-body of velatively good quality The pbst has a closed circulating-w r

ypm *CMWS) that tedes on be Itk rhr akeup. TIM CCWS comprises a pond.

Aive &aczhing-water pumps. an Admia.

tqyonw d, siogte s condlenser and two nama-daft colioe lower~ General Radoaclw ontmbtllo o t~e mfac an hionnuurmaresdamnanded qiedal precauftris o hper-iA polecton W.IJ t ot rtd O

m i"trie*t-to conntl katdwter.-copper buildup Is one prevuen qiaproach I-rd C Pron. the WtIlts condenser proec MAnAgC4 sates dug i

opper to levels down ft0.2 ppb adverly affects demlnerllzert na times -an Increases mequuments for red waste pm essing and waste d<posallsor ag. Rmova-l of die mof o,=eo denser Is Dectsuber I8. and the modill cation was aeted for die saflocg oulag RRM schedul ftr Much 1991. A ftr Mal prj=e aralatio was art up In Jm awy 1990 is stuy, dy etal and implement fte mogjlcgku the staff lmetioulg assa complcety self-reliant nti both before and during the outage. In a ndion to Proc.

key staff m*mbers were John HonkAa, condenser project engineer, and John O'Dou*a W.enser fe Oiee With at kiss of gneating capacity a key concern, the Can order of business focused on the lenth of tde oun*&, t die past trumbig of BWR pants of similar size and condenser-tube sumber has required from 7S to 110 days-liss for condensers with significansly fewer tubes or R*prial-condenser-eshe replacements.

1uaseleged e

u ovlde. a 65-day aggressive eoaslriic-ion schedule and incentive. detailed prepla*ning Using previous historical data. end Input from Fermi 2. a utility consultant deel aped a construction schedule allowing 75 days fom entry int waxerbo.= to com pletion ofthe Concluding clzsalating-water Wit Allowing Sr pve-o*-age w*erk we of sparat stand-alone fcilities. and sharp focusing of the job to minimize h&rerer CD0cs ftm Other outage activitii". the pro joct team shortened the schedule to 6S day Ambitious as it may have seemed.

fth requiremetnt was incorporated into the project specification. so bWdd.wer pte pae to prescu detailed 65-ay acheftiles at the prebi meeting.

In accordance with utility policy for major contrats, the team established a flixed-price nontract for ratubing the am deser and developed a detailed specifica tiou, the heatt of the contract. 7his didnd Preclud the reqirernent of prices hor spe ific ac'vities.manpower Woading a m[ie stone schedule, and a detailed activity schedule. A detailed reference bid. mwas over. was prepared to provide a hbais hr exehiding low bids horn contractors with margnal experience This mraegy rusulted in two very good proposals, and the award decision was bated primarily on the cotractorz man agemen

-m and its experience In work.

leg together on similar pjects. Stccs of the approach is measmued by the fct that die project was completed aed of sched ule with a quality product and only 3% in additons tom d origind contract plice.

such an achlevemeom clearly, could not have been accomplished without a thorough analysis of the problems Involved, evaluation of the available options and associated limitations, a deaed and tcallstic project schedule and complete agreement on job-specific activi des-eli1 predicated an creation of a dose knit eram with a inilled qppmnach.

Iis was accomplished over a period of nine months starting In Jul 199D. when the contract was awarded to United Meg sees & Conmtcorsamlyt*l, Philadel phia. Pa. and astmzatioana Interfacing commenced. Neoulning in September. the general eotra:tor and Its aubcontrac tots-Heat Eb.chager Systems Inc.

Boton. Mass, &Ai Csanoo-Slir. P

-ilade phis, Pa-woesk closely with the project tam to ptI ;

a detaled schedule, locate providers of goods and servles and con duct the accessery analyses related to tobetbeet stability. sube,'besheet-joint.

strength, and prot:ective-coting evalua fions. So dose an smooth were the lIter actions smoog involved personnel during this WeWo that, by December 1990. ftey CGS Main Condenser White Paper Revision 1, June 2006 Page 21 of 22

• wee no longer idemti6ed by their individu

.a organizations but a the Condenser Pro ject Team.

Prelect e0lments. The primary -old erazions of the project were selection Oflde tube material and the replacement method.

The ieplacement material would have to equal the ptnt's performance experience with Admiralty brass: operin Close to design rating with fewer than 1% of the tubes plugged and fewer than 1% with 50% wan thinning. Othe ntalteish-leted cancern were suscepuMilly to corrosion.

erosion. and fouling at the imal surface.

materials evauated" Included 1ypcs 304 and 316 (austenitic).

Type ALAC (-super' ausienitic), TYPe 439 crfrxic) and Sea-Qe (euper" frth 1c) stanless steel$, and titanium. Tbe mssgnit sees were ao considered suit able because of the aced for water-a1de hyup procedogs Type 439 was eimne because of susceptibilit to pitting and tervice corrosion. Although the costs of the snPer" stainless seels and titanium were comparable tianium was; selected hpowerpta OR m

Snwservice. In mcreover, has beeo sccessful Replacement althuatives were jebundling. retubing only, and rebing with new nihesbaets. Replacemenz of tubes s grups (bondles)--loue with gr-ea suc cess at several Scandinavian unclear plnatts-was not possible at Fermi 2 because of the plato layout and interfer ence. This required hand removal of tubes, negating schedule advantages offered by modula repla-cL Despi* e galvnic lnCoUmPaslfiY, I* wa decided so tain at existing carbon steel tubecheets, primarily because the cou denser h ente outlet waterboxes, which would have bad to be dismantled to install titanium outlet tobesheets. The inlet tubesheets could have been replaced by removing the inlet wetbome and uting the tubesheet-to-em a-df weld. But this would have had to be replaced by a mechanica joint. introducing a potential foreskage Analysis of tubesheet and tube-joint loads indicated no significant change would resul frAm rembibg with tian*um*

$oint-streogth tests were conducted to determine the acceptable tube-rolling torque muder the allowable 1340-5I90-lb foces expected at 55 sig. the design pres tare of the citrc-wat" syssm Five 38-hole mockup tubeseets wee made for pullout tests, two of hem poxy-coated to rel cate the Intended resnbed-condenser mbesheer end-product. About 10% of the titanium tubes tested wer coated with LTocne prior o wlin. Tt results dictated rolong to, ques of 10.5 and II ft-lb at the Inulet and oudetrespectively. DBaed on the wat-caws pullout forcls. dese torque val lan produced a nibe-so-ndesliertjoint with a safety factor of two.

AddItlonI atnalyses were related to condenser uplift. cathodic protectiONi tube vibration. ad effects on circulating-water flow. Stuctural stability was a factor because the full complement of 22-BWG titanium robes would be over one-million pounds lighter than their predecessors.

Analysis showed that resulting uplift loads could be accommodated by the existing foundaitons: anchorage. and structure.

Tests to deemine the'need for cathodic protection because of metal dissimilarities cocluded that the subesheet could be pro cted from galvanic corrosion by costing aone, without use of a sacsicial anode or levessd-Iretmu cathedic proceed Corosion Calculations based on inea smeesentt made on ftst assemblies placed in the circuleting-wate pump house Mfu mind expected corrosion rates below 10 mUflz4 Because other approaches wee ats-proahbitive, and assuming no coating Imperections or failures, it was decided that a thick Ailm coating with high impact rsistc and excellen flexure. cathodic di~shoodmet=

and dileltc-acuimnth peep attics would provide adequate tubesheet Tube-vbIrUon analysis indicated that the ansupported span length oft Uhnnei' wailed titanium tubes would have to be reued o prevent virAtion-Induced fall uses This dictastd the need for staking the tubes. counfiming the eperlence at other plants. Thbes in the air-cooler sections were not included because of tbe far Iowea velocities.that they would be subjected to and the amovat of work involved In installing sakes In these shrouded sections.

Type 304 stainless was chosen over other possible" staking materials for its smooth sudace finish, car*son resistance.

and ea of installation. A dimpled stake design was selected because it offered a locking cpabfi.

The U partt developed for full bundle staking required about 36.0stakes ofvaying length.

Anbass of crcula*ti-water low id ca*d linle inpact on tube cleanliness, con denser pressure. or net generation would result firm the reduced tube velocities ina titanium-tube coudenter-from 7M ftse to 6.53 ft/sec with five pumps operating.

firm 6.8 o 5,80 ft/sec with four pumps.

Velocities were judged adequate to main min tube cleaniess factor at an estimated 90%. No appreciable Increase In nat gen stion would eccur with five pumps until the circuladntg-wavter inlet temperature reaches 7*or hWghem Remaining uneeflalntlaa bad to do with the condition of circulating-water valves, Inlet and cutlet tubesheet. con denser staun side. and suppon plates. PFo aible concerns lnclded valve leakag,. sup port-plate bowing, and microbiologically Influenced corrosion (MIC) of tubeshects.

The valves were clend. inspected, and adjusted for seating during svead krced outages. Tubesheets were also inspected and special tube-worktooling determined.

Stearn-side condition was established and damaged components were fabricated before the start of RF07-The presence of ust nodules indicated areas of MWC.

Selection of the tubesbeet coating required extensive research into the need for surface preparation before application of the coaing. This was dictated by reports indicating premature coating failures were likely In MIC-affected are if corrosion and bacterial colonies were aot removed before the coating was applied. Possible solutions, ranging from ozonation to mangement of steam chambers and flush ing of tubeshect surfaces with potable wa*r, appeared cost-and labor-intensive; limited documentation was available to provide bacteria counts before and after treatment to verify the effectiveness of efforts at eradication.

Accordingly. a treaument process was developed that would not impact the schedule nor be toxic to personnel or the environment-It Involved hydroblastinSg.

spraying with hydrogen peroide, sad blasting, end washing with methylethyl ketoxime (MEK). A mockup tubeshe was prepared, Admiralty brass tubes were rolled In. and the assembly was hamsed in circulating water on the pump scuod aide for six weeks, allowing buildup of a slime and corrosion layer imila to that developed on the condenser tobesheets.

Examination of dre assembly before stat meat tvealed the presecaf of torrosive d-producing bacteria The examInaston was repeated after each step of the eradication procedure. The end result was that 94% to 99% af the bac teds populato was IlMed by kydroblast ing, and another 2% by the pride wash.

Following the M.EK wash, total kill was 97% to 99%. This analysis enabled the elimintoa of MM from the procedure.

Edamt Implemnsting the procedure during the outage. water samples wer tak from MIC-affected areas for batel dats. Com parion to samples taken after hydtnbas ing-allowiag time for bacterial growvth-shwe a 75.2% reduction in the bacterial colony count the reduction achieved by hydroblasting followed by peroxide apraying was 97.i%.

The balance of the detail pe-otiutge planning and activities called for by the retubing specification proceded on or ahead of schedule. As a result, the 6,000, tube condenser was retubed, coated. and tested In a period of 62.5 days with scar lawles woaganshup. The wock was com pleted 25 days ahead of chedule and 20D%

under the budgeted cost-all the more notable for rviAng been achieved despite complicating conditions: contamnated tbesheems radloactive tubing, and the use of special clothing to protect penonnel fm contamination at the watesbox fves (see photo).

IBel Strauss to~ru U

.A#, Ista CGS Main Condenser White Paper Revision 1, June 2006 Page 22 of 22

Columbia Generating Station Main Condenser, Addendum 1 W. Scott Oxenford, VP Technical Services This addendum augments the Columbia Generating Station Main Condenser white paper, providing a summary of economic impacts of condenser leakage.

The analysis is organized in two parts for clarity. The first provides an average cost per event or month. The second provides a cost breakdown over the current operating cycle, from June 2005 through May 2006.

Historical Perspective Columbia has suffered eleven shutdowns and nine reduced power evolutions to address main condenser leakage since it began operation in 1984. The number of events speaks to the chronic nature of this costly operational challenge.

Condenser related shutdowns and down-powers would have been even more frequent if it were not for repeated plant shutdowns, extended economic dispatch periods due to river flows, and annual operating cycles. Each of those operational attributes provided opportunities to perform condenser repairs reducing the potential for even more condenser related shutdowns.

Direct Cost Impact Chemistry Control Columbia's condensate filter demineralizers are changed more frequently to minimize impurities reaching the reactor pressure vessel. The demineralizers are coated with a powdered resin. The costs of increased resin use, shipment, and disposal of the associated radioactive waste is included. Additionally, the circulating water system is operated to reduce the level of impurity concentration, requiring increased chemical treatment.

$124,000/month Average chemistry-related cost for each month Columbia operates with a condenser leak.

$1,030,000 Aggregate chemistry-related cost for this operating cycle (June 05 to May 06).

Tube Plugging Evolution A tube plugging evolution is performed at reduced power and takes about three days.

Detailed planning, oversight, and around the clock coverage limit lost generation. The actual tube plugging activity involves isolating a section of the CGS Main Condenser White Paper Addendum 1 Revision 1, June 2006 Page 1 of 4

condenser, draining it with multiple pumps, cleaning the tubes, identifying the leak(s) and inserting plugs. The evolution involves Columbia staff level of effort, overtime, and the use of contractors.

$350,000/evolution Average incremental direct costs associated with one tube plugging evolution.

$1,400,000 Aggregate cost of the four tube-plugging evolutions this operating cycle (June 05 to May 06).

Indirect Cost Impact Radiation Exposure Columbia's main condenser location exposes personnel to radiation during repairs.

Keeping radiation exposure 'as low as reasonably achievable' is everyone's responsibility.

Radiation exposure is closely monitored by Energy Northwest, the Nuclear Regulatory Commission, and the Institute of Nuclear Power Operations. A non outage month at a top performing boiling water reactor is less than 2 Rem collective radiation exposure, with a total for the year of around 30 Rem. Adding one condenser repair evolution jeopardizes the annual goal of 30 Rem or less.

2.262 Rem/evolution Collective radiation exposure to Columbia staff and contractors, per tube plugging evolution this operating cycle (June 05 to May 06).

9.048 Rem Collective radiation exposure to Columbia staff and contractors this operating cycle (June 05 to May 06).

Replacement Power Columbia conducts tube plugging evolutions at 65% power. Based on typical duration, a tube plugging evolution costs us the equivalent of one day of full power operation. Estimating the value of power of $32.45/megawatt hour produces the following indirect costs.

$1,000,000/evolution Estimated cost of replacement power per tube plugging evolution.

$4,000,000 Estimated cost of replacement power for tube plugging evolutions this fiscal year (July 05 to June 06)

CGS Main Condenser White Paper Addendum 1 Revision 1, June 2006 Page 2 of 4

Costs for 2001 Heavy Oxide Layer Analysis Columbia's staff had extensive fuel rod corrosion related concerns in the 2001 timeframe. A highly experienced industry team was formed to determine the root cause and initiate corrective actions. Condenser leakage was identified as the root cause.

$1,750,000 Cost associated with fuel scrapings, investigation, and research into heavier than expected fuel oxide layer.

One corrective action from the above investigation was to alter Columbia's chemistry controls to favor fuel protection over radiation source term mitigation.

The strategy was followed until the current operating cycle. Oxide formation on the fuel during this test period was normal, thereby validating the root cause determination. Favoring fuel protection created very high radiation source term at Columbia, resulting in increased staff radiation exposure. The impact is difficult to quantify, but very real. Columbia's radiation exposure performance is in the worst quartile in the industry. Columbia is currently the worst plant based on source term measurements. To counteract this, a chemical decontamination of key piping is scheduled for our upcoming outage.

$1,700,000 Approximate cost of chemical decontamination in the upcoming outage to improve radiation source term.

Replacement Power Estimates Since 2000 Lost power generation (associated with main condenser leakage) since January 1, 2001 is equivalent to more than 12.5 days of full power operation. With an estimated value of power of $32.45/megawatt hours, the total cost is substantial.

$12,500,000 Estimated value of replacement power associated with Columbia main condenser leakage events since January 1, 2001.

Conclusion Various conclusions can be drawn from the above data. In simple terms, the author draws the following underlying conclusion, upon which others can build.

'The design and resulting poor performance of Columbia's main condenser has produced many direct and indirect costs and continuously challenged our ability to achieve operational performance levels expected of U.S.

nuclear power plant operators.

CGS Main Condenser White Paper Addendum 1 Revision 1, June 2006 Page 3 of 4

The chronic nature of Columbia's main condenser problem is unlikely to change substantially without addressing the condenser tube material.

Failure to address this issue will increase the risk of fuel damage. That fact alone is ample reason to replace the main condenser tubes. Replacement is also in the best interest of efficient financial operation of the plant as evidenced by the costly history of our present condenser equipment."

CGS Main Condenser White Paper Addendum 1 Revision 1, June 2006 Page 4 of 4