ML20149M437

Forwards Response to NRC Insp Rept 50-443/87-07.Safety Implications of Biofouling & Stated Unresolved Issues Discussed

ML20149M437
Person / Time
Site:	Seabrook
Issue date:	11/12/1987
From:	Tracy S EMPLOYEE'S LEGAL PROJECT
To:	Kane W NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION I)
Shared Package
ML20149M429	List: ML20149M426 ML20149M437 ML20149M442 ML20149M455 ML20149M491 ML20149M504
References
NUDOCS 8802260044
Download: ML20149M437 (22)

Text

_ - - - _ _

r -

~

Empl:yee's legal Project P.O. Box 633

• Amesbury, MA 01913 (617) 388-9620 November 12, 1987 William Kane, Director Division of Reactor Projects U.S. N.R.C.

631 Park Avenue King of Prussia, PA 19406

Dear Mr. Kane,

Enclosed please find the Employee's Legal Project response to IGC inspection report nurber 50-443/87-07. I look forward to neeting with your 3 staff on the outstanding issues sometime in the near future. I will be in touch with you or Tom Elsasser t.y phone to arrange a time for that meeting.

Sincerely, j

% J/

vt44 ;t %4<t t (f' Sharon Tracy ff Project Director I/

8802260044 880218 gDR ADOCK 05000443 9

PDR

Employce's legal Project P.O. Box 633 Amesbury, MA 01913 (G17) 388-9620 RESIONSE E NUCLEAR RFEULATORY CDMMIwION REIORT 50-443/87-07 November 12, 1987 l ,

I l

I i

1 l

l l

I

Empi:yee's Legal Project P.O. Box 633 Amesbury, MA 01913 (617)388-9620 RESIONSE E NUCLEAR RMULATORY COMMISSICW REIORT 50-443/87-07 co:Imns Introduction....................................................Page 1 Response........................................................Page 1 Appendix A Safety Irplications of Biofouling at Seabrook Station. . . . . . . . . . .Page 10

\ Appendix B, Part 1 h Unresolved Issues Raised by ELP in September, 1986..............Page 16 Appendix B, Part 2 Unresolved Issues Raised April,1987, and Thereafter. . . . . . . . . . . .Page 19 i

l I

l l

l l

A 1

\

h RESPONSE 'IO NUCLEAR RD3ULA'IORY COMMISSION REIORT 50-443/87-07 November 12, 1987 v/ .

e

, IlfIRODUCTION J

Nuclear Regulatory Comission (NRC) report 50-443/87-07 was written in response to allegations made by former Seabrook Station work.ers who saw unsafe conditions while they were working at the plant. It also answers questions raised by the earlier imC Report 50-443/86-52, which dealt with many of the same problems. tis Enployee's Legal Project (ELP) response

, addresses the technical explanations given by the imC in the same order

, they appear in tac report 50-443/87-07. W e numbered sections designate  ;

i references to the lac report; the lettered sections are the ELP responses. ,

In several cases, the ELP is still analyzing information, and is deferring response in those cases until analysis is conplete.

Issues raised by the ELP but not addressed in NRC report 87-07 are listed in Appendix B. Other issues listed there require further discussion '

because the allegers were not satisfied with tac conclusions, n ose allegations the tac believes reflect "procedural" problems, that is failures in quality control, quality assurance, document control, design control, and technical training, are slated for discussion at a meeting between the imC and the ELP. <

Biofouling, discussed in both the body o'f this report and in Appendix &

is a contention under litigation before the tac by the New England Coalition on thclear Pollution.

7 s

t 4

RESIONSE 'IO NRC REPORT 50-443/87-07 d

) Section,1.3, paragraph 3, discusses programatic weaknesses (poor quality assurance / quality control, technical training problems, document control, 1

and design control), saying deficiencies in these areas were corrected l prior to NRC report 50-443/87-07, issued in August, 1987. Although these -

issues are slated for further discussion, there are several points ELP will j make here.  !

A. Ongoing problems reflected in current tGC inspection reports up to

October,1987 show that even if the programatic deficiencies were t corrected before August,1987, the problems caused by those deficiencies O, are now built into to the plant and are continuing to become evident.

te programatic deficiencies themselves are still continuing and are still i

i 1

being dismissed as isolated incidents by the tac.

y/ B. Because of past deficiencies in docunent control and design control, documentation used extensively by the tac in assessing the state of the  :

plant is not reliable enough to reach conclusions about the plant's safety.

C. As an exanple, see Section 2.31. The tac discovered that the seismic analysis nodel for the Waste Process IMilding did not take into account the structure's "as-built" arrangements. It was only through the imC's in-depth design review of the h% that this discrepency was discovered. tis design review was done only on particular selected structures. Other such dircrepencies in design documents may never be discovered without an in-depth investigation. The imC has cited plant management repeatedly during the construction process and in the latest SALP for design deficiencies ,

and failures to correct recurring problems. i Section 2.1 he NRC concluded cold pulling in the condenser piping did not occur since the piping is fixed only on one end, and cold pulling is

, defined as related to piping fixed on both ends.

A. W e condenser piping Doug Richardson observed being cold pulled is connected on one end too an expansion joint, and was cold pulled so it could be welded to the condenser wall, ne pipe is open on one end, but it is welded to the condenser wall, thus is fixed on both ends. his fact was related to the imC at the April 20, 1987, meeting, and also should have been evident on construction drawings.

B. Doeg Richardson referred to Condenser A, not Condenser 8 (which is how d

the stenographer at the April 20, 1987 neeting between the ELP and the NRC heard and transcribed it). @ is misinterpretation may have caused some confusion. On design and construction docunents, including isometrics and PIDs, this system was designated as part of the extraction piping system, I including the 13th stage piping.

i a Section 2.2 he NRC answered the question of when the CM drawings were issued. The concern was that certain piping in the CBA was field run, that is installed without design drawings y A. ne design of the CBA was changed a nunber of times, as noted by the imC when Engineering Change Authorizations were examined.

3 B. Were were no construction drawings for specific pipes in the CBA system  ;

until Doug Richardson drew them, for exanple, pipes shown on Pullman-i Higgins isometric drawings nunbers 1-CBA-SP-01 through 1-CBA-SP-09. nat ,

pipe was field run, and the lines were disassembled and reinstalled at least twice, i

i section 2.3 tis item deals with Doug Richardson's contention he was not properly trained for work he was doing as an "as-builder" for Pullman-1 CN Higgins. Apparently the tac is saying his job did not require he be trained to ANSI standards. i j s i 2 i i

A. For his work for United Engineers and Constructors, Doug Richardson was K)

/

V' certified by that co many as a Level 2(R) Visual Inspector under UE&C procedure (FACP) Nunber 13, a provision to meet ANSI requirements N45.2 for l

training of inspection personnel nat training was required by UE&c.

B. When he worked for Pullmn-Higgins, prior to being egloyed by UE&C, doing essentially the same job,'Doug Richardson was not trained to the ANSI standards required by UE&C. His concern is that he was not properly trained for the work he performed for Pullen-Higgins, and that cther people working for Pullmn-Higgins may not have been properly trained.

C. Since "as built" information is essential in engineering analyses of safety systems, proper training of as-builders is an inportant safety ,

consideration.

D. As builders working for Pullman-Higgins did write nonconformance reports, and therefore functioned in an inspection capacity J

E. We NRC seems to suggest Doug Richardson was a drafter, when in fact he

was an as builder.

l section 2.4 %is section addresses concerns that various buildings hwe irreparable cracks leaking groundwater,

~

j 2.4.3.1 %e NRC examined the interior wall of the containnent, and saw no water seepage. ne cracks in the containment concrete are to be expected, >

but they are not leaking water.

, A. %e steel liner around the inside of the containment wall would prevent I any observation of scepage into or through the containment wall.

2.4.3.2 Cracks leaking water in the squipment vault, Primary Auxliary Building (PAB), the Waste Process Building (WPS), and the electric cable tray tunnel are unresolved issues. We effects of the leaks are not currently a safety concern to the NRC, but if they are uncorrectable, the l 1eaks would ecocern the NRC. I i  !

j A. Since there is yet no proof the repairs will work, then the leaks must  !

be considered a safety concern in the decision to activate Seabrook i j Station. Some of the systems in these areas ray become radioactive after i l plant activation. Historically, repairs to operating plants have not been

aggresively pursued.

t j B. Se failure of the waterproof merbrane was not addressed in 87-07.

C. tut that the dewatering program used during construction has been discontinued, the leakage problem appears to be worsening %is would

, account for the rash of new leaks described in 87-07.

l D. %e chemical tests of seepage through the walls may tell whether the 2

i j) reinforcement bar is rusting. However, given that concrete is a porous material, water need not have seeped all the way through the wall to be corroding the rebar, J

I

. . E. %e use of Vandex to repair the leaks and its ineffectiveness were not addressed. W at the corrective measures are said to be "generally effective" is not a wholehearted endorsement of the leak repair procedures,

especially since the NRC says additional attention is continually needed to control the problem.

~

i Section 2.5 NtC explanation of "cadweld splices."

! Section 2.6 through section 2.13 nese sections deal with the allegation

] that sediment clogged fire protection system pipes, reducing a 12-inch

diameter pipe to.4 inches. We NRC answers questions raised in previous

discussions between ELP and NRC on this issue

tests observed by the NRC j and when, tests used on the system, location of screens removed to check the system for biofouling and debris,.etc. W e NRC notes that between February and July, 1986, unlined pipes and elbows in the fire punp house i were disassenbled and taken outside to clean out MIC deposits, but tests -

show the problem was corrected.

4 A. What was observed by the alleger was a serious sediment problem.'%e sediment overlays and protects the bacteria causing microbiological 1y

. induced corrosion, or MIC. Recent NRC reports on the MIC problem at nuclear i plants nationwide discuss the great difficulty in detecting and treating the biofouling problem, particularly MIC.

]

B. NRC reports of inspections at Seabrook Station (April and August) on degraded heat exchangers, welds, tubes, and deformed baffle plates reveal a

! serious biofouling problem at Seabrook Station, a problem with grave safety

! consequences. For exanple,= sudden activation of cooling water systems in an emergency can dislodge biofouling material's and other debris and clog cooling systems when they are most needed. Coninon cause failure is not unusual since normal and backup systems are frequently affected at the same time, i

j C. For note detail on biofouling (bivalves, tubeworms, MIC-causing

bacteria, etc.) and the safety problems it is causing at Seabrook Station, see Appendix A.

section 2.14 h e alleger said a cigarette fell into an electrical conduit '

starting a fire. Four or five gallons of water were required to extinguish

it. %is was the second time, for report 87-07, that the NRC attenpted to i find the conduit based on maps drawn by the alleger. Unable to find the l 1 conduit in question, the NRC relied on tests of the cables to conclude l there are no further concerns, j l

l

A. Because of the difficulty in identifying the location of the conduit,

I there is no guarantee the a@ropriate system was examined.

I.

1

section 2.15 %e alleger was < oncerned that inadequate separation of the  !

3 main feedwater and emergency feedwater systems makes them susceptible to i earthquake and fire. Se NRC notes, "We licensee recognized the  !

j susceptibility of the emergency feedwater system to loss frcun a conmon j j I l 4 )

i  !

4

fire and requested a deviation from the requirement for independence of the

, alternate shutdown capability for the fire area of concern."

A. Rather than correcting the ~ design flaw, tne utility was granted a waiver  !

of the requirement to separate the systens. %e purpose of the requirement l

, was to prevent both systems from being disabled by the same fire, so at least one would be available in an emergency which included a fire.  ;

B. In granting the waiver, the imC is apparently relying on the

^

availability of the startup feedwater puq in case of failure of both emergency feedwater punps. %ere are a number of weaknesses in this notion.

1

% e startup feedwater punp and its associated piping is apparently not of safety grade construction. It is located in the turbine building which tray

~

not be seismically designed. %e startup feedwater pum power source is also not seismically qualified. Tb switch the startup feedwater pu@ to a seismically qualified power system, someone cust leave the control room to  ;

throw two circuit breakers. To align piping systems for emergency feedwater i use, operators nust leave the control room to open a valve. We suction ,

line for the startup feedwater punp is positioned so 10 percent of the  :

dedicated emergency feedwater capacity of the condensate storage tank is -

4 I

not available. We startup feedwater punp cannot be operated frmt the '

remote safe shutdown panels, t C. There is no fire protection system in the emergency feedwater pvq l' house.

section 2.16 The imC states the contention that heat exchangers (feedwater l heaters) were installed with a slope when they are supposed to be '

level was not substantiated. .

i j A. No present corment.

)

Section 2.17 tio immediate safety concern exists for leaking cracks in the I Waste Process Building (WPB) and the equipment vault which can't be l

repaired. See Section 2.4.

4 I Section 2.18 % e NRC defined cold pulling and addressed the repeated [

allegations that cold pulling occurred on a fairly frequent basis despite  ;

prohibitions against the practice. Cold pulling "is the practice of pulling j

or jacking of piping to correct misalignment at the closure joint." 2e tBC concluded that the cold pull issue presents no concerns about the

, reliability of piping systems. ,

t l A. By this definition, and as cbserved by Doug Richardson, the 13th stage

drain line in condenser "A" was cold pulled. See Section 2.1.

B. Cold pulling prior to 1982 was not addressed. A 1982 self-initiated l report by the licensee refers to a nunber of nonconformance reports written j because of cold pulling.

4 C. Although the imC says prohibitions against cold pulling existed, 1

i 5

l

X training in this prohibition is not mendoned except for quality control

/ personnel. %us construction personnel, the people who engaged in this (g} practice, were apparently not trained to avoid cold pulling. We many allegations about this practice have come from construction pesonnel. ,

D. %e tmC bases the statement that there was no problem with cold pulling on a study of the use of Deatman clanps in 70 piping systems. Yet the repeated allegations of cold pulling cite the use of chainfalls and come-  ;

alongs. %ose practices were not included in the study and have yet to be addressed, although the ELP has raised this concern in the past. ,

E. %e tEC states that UE&C prohibitions against cold pulling were changed

! in 1986 because they were viewed as overly restrictive. However, this does j not address the concern that the prohibitions, when in effect, were l

repeatedly violated.

! I F. % e IRC says "...it is intuitive that long spans of piping can acconodate cold pull well in excess of the originally specified 1/8 inch. .

However, this refers to piping using tenporary support. Cold pulling problems raised by ELP refer to permanent situations which permanently l stress the pipe. Stress under operating conditions is also not taken into l

account.

section 2.19 he ELP raised the concern that there is grit in the valves of

( the service water system. We tmC said that such grit, if present, would I

have a minimal effect on the valve sealing surfaces.

1 l A. A May 18, 1987 tmC inspection report noted failure in a valve due to wear in the valve body liner. It did not mention the cause of wear. '

B. A July 28, 1987 tac inspection report mentioned a generic problem with seat damage to 30 Fischer valves in the service water system, described the repair procedures, but made no mention of the cause.

C. It should be noted that pipe lining material was recently found in a failed primary conponent heat exchanger, section 2.20 One hundred poor wolds in the service water system are not a ,

cause for further review. i A. We tGC based their conclusions on statistical speculation rather than physical evidence. W e lines in question are buried and filled, and were not actually inspected. % e physical evidence that the welds are bad is not easily physically accessible.

B. In the 87-07 report, the imC did not mention an tec Office of Investigation (OI) report written when this issue was first raised by the alleger in 1984. % e OI reported on an interview with the alleger's supervisor who said the 100 welds were questionable, but could not be

/G inspected because they were buried. %e alleger gave this report to the imC (V ) engineers conducting the 87-07 investigation, one of whom had done the 1984 investigation, but knew nothing of the OI report.

6

1 C. %e piping syctem with the 100 bad welds was downgraded in safety requirements when the bad welds were discovered so they would not have ,to +

be repaired. We downgrading was not addressed, l D. W e tac inspector examined field weld documents for four welds "typical" of those described by the alleger and found them to be fine. However, previously stated ELP concerns with document control render conclusions based on licensee documents questionable, ,

E. Ongoing problems with MIC were not raised in the analysis of the .

severity of the problem of the 100 bad welds. %e bacteria causing MIC (which has reduced projected piping life spans from 30 to 5 years due to corrosion) is particularly attracted to poorly welded areas. See Appendix h*

Section 2.21 Rusting of reinforcement bar due to groundwater leaks in  ;

various buildings. See section 2.4. [

l l Section 2.22 his refers to a semantic difference in locating an area of r

! service water piping.

l l A. No present comment.  ;

\ ,

I i d Section 2.23 %is refers to questions regarding a test on piping cited by the tac.

~

A. No present coment.

i

Section 2.24 Wis refers to a radiological test of steam generator nozzels. '

I r

A. No present coment.

I Section 2.25 According to the alleger, colum bases for the four reactor coolant punps were set further from the reactor than called for in the design, thus possibly overstressing the pipe at the punp and at the reactor. % e NRC concluded the modifications would not overstress the pipe.

A. No present coment.

Section 2.26 A flooding incident which affected the containment spray punp l was identified in a nonconformance report and corrected. _

i '

A. No present coment. f i

Section 2.27 he spring supports failed, and the shafts bent on the punps 7 ,

L

. f-

.m for the boron recovery and liquid waste system, as the alleger said. %is  ;

r occurred sometime prior to Decenber 23, 1986 when a request for design

(%\/

, services was made. A design change is being processed but .is not yet  ;

conplete, ,

l A. We alleger left his job at Seabrook Station in June of 1986 and noted  !

this failure occurred at least several months before he left. Although the failure occurred almost one year prior to report 87-07, a design change had

, not been conpleted, and repairs had not begun. -

} B. Although the utility claims the plant is ready to operate, one of the l systems Wich is inoperable' handles liquid radioactive waste.

Section 2.28 Wols and other debris left in piping during construction was not as great a problem as described by the alleger.

l

-t A. No present coninent. l t

i Section 2.29 he allegation that the wrong guage duct work was installed in

the HVAC system was not substantiated. We safety classification of the i

ductwork was upgraded and stiffeners were added.  !

t A. W e problem was apparently rectified, i t

i O Section 2.30 he alleger said MIC degraded parts of valves in the fire protection system throughout the turbine buiding. We NRC cited I documentation showing 8-inch and 10-inch volves in that system were i disassenbled and cleaned.  !

A. We cause of the problem with the valves was not noted. l I B. Presumably this system was flushed and tested shortly after construction

was conpleted in the winter of 1983-1984; thus any valve problems occurring [

l two and a half years later would not be expected to be due to constructioon j debris, but rather to an ongoing fouling problem. ,

t i

4 ,

Section 2.31 he alleger said the design of the Waste Process Wilding was I

] inadequate. W e NRC performed a plant design inspection in early 1984 and  :

selected the Waste Process Wilding for review. W e NRC found "the seismic  !

analysis model for this structure did not take into account the 'as-built' l arrangements of the structure. We licensee reanalyzed the structure after  !

l the NRC discovery, and found the WPB "needed modification to resist changed  !

a loads and stresses." Modifications were conpleted in mid-1986.  !

j I 3

A. It appears it was by chance the NRC decided to review the WPB structure,  !

and discovered the problem. %e licensee's quality assurance / quality l l control programs designed to detect such problems did not uncover the  !

structural flaws. We NRC has repeatedly stated that the licensee's

). procedural problems in such areas as quality assurance / quality control have [

]' g had no significant inpact on the quality of construction. Yet this is a clear exanple of such an inpact. In a nunber of other cases, there have l

i 8 j

I

i

, , been flaws in the design of the plant, for instance, the cooling towers, f

! the weak beams in the service water punp house, and A!EE large-bore pipe [

supports, so the design problems can be assumed to be fairly widespread.  ;

i Problems with the pipe supports were discovered at another plant, and then i the supports at seabrook station were reevaluated and found deficient. - *

B. Since the structural problem was discovered by the NRC, not by the utility, many similar problems may still exist undetected, i

i l

i k

i >

l I

-i l

I  !

t l

, f t

j l

1 r

l l

l I

l

)

I I

i l

i l

4 6

9

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

- .. . - . x =

a m.d a, m 4 MirAS*MmnMAASE4-mmn&M&A 4 4 = A .B. e s D JmmM6,-== -'B-M&MAMM= Jan---^= Amm u mex m emm-e mm mumm 4am,aAM m, m o m uMm .A-s-.Musmmm+a--- -a, m. - +

i I

1 1 ,

i t

t i 4

4 4

1 6

k  ?

r I

i t 4

1

, APPENDIX A -

i l SAFETY IMPLICATIONS g BICODULI!G M SEABROOK ETTATION 1

• F 4' ,

t g l

I i I 1

l I

l I I

! I i ,

s i i 1 l l j

e 1

I I

,i k

l 10

SAFETY IMPLICATIONS OF BIOfWLItG AT SEABROOK STATION I. ~

( " h)

A bacteria is eating away at piping and other equipment in Seabrook Station's safety cooling systems. Current evidence shows this is an ongoing problem, and that treamnt methods are not effective. The bacteria cats organic mterials in the water around it, and gives off sulfuric acid as a waste product. %is microbiologically induced corrosion (MIC)) and other i forms of biofouling (clams, mussels, oysters, slime) are serious problems at 70 percent of power plants in the U.S.

On April 30, 1987, a tube leak in the "A" train heat exchanger for Seabrook Station's Primary Conponent Cooling Water (PCCW) system led to the discovery of 40 more partially degraded tubes. Wey found the baffle plate in the upper channel head was so deformed that some tube outlets were exposed to inlet pressure. ne cooling water, normily diverted by the baffle plate through the heat exchanger tubes, instead flowed directly from the inlet to the outlet, bypassing the heat exchanger ne be"es of some l tubes were pitted, and some tubes were blocked with "marine deoris" and piping liner materials.

% ree months later, in August, the utility discovered several welds were damaged in the "B" train of the RHR system. %e utility cobbled together one PCCW system by using parts of both the "A" and "B" PCCW trains, a method the lac questioned in very strong language @e utility did not I report this jerry-rigged system to the imC, rather the tac inepec.or 1 -s discovered the situation during a review of the plant 109 7

i  !

(,) According to the imC reprt on the discovery at Seabrook, one train of the PCCW is necessary to support cperation of the residual heat renoval (RHR) system. %e RHR system renoves decay heat from the system. Seabrook Station s safety Evaluation Report states that the PCCW and the service water systems asscciated with the RHR are required for a safe shut down of i the nuclear plant. Because there is coly one heat exchanger in each of the

! "A" and "B" trains of the PCCW, when tha heat exchanger was lost in the "A"

! train, there was no back up system in effect. W e whole FCCW system is l built to seismic category 1 safety standards and runs continuously in all operating nodes. %us, this biofouling problem poses serious questions about the ability of the plant to operate safely.

l Biofouling and the particular safety problems it causes are not unique to Seabrook Station. Identical problens at other nucear plants were caused by MIC and mrine debris. Only in the last several years has the nuclear l industry and the t&C realized that MIC creates a serious safety hazard, and l that current treatment methods don't work.

l l Awareness of the safety consequences of biofouling was thrust on the j industry in 1980 at Arkansas Nuclear One (Ato) units 1 and 2, and caused l enough consternation for the NRC to generate stacks of reports on i biofouling's safety significance. Corrosion, silt, and clamshell debris l plugged the service water supply lines in the high pressure injection system, and clams clogged the inlets to the cor.tainmnt cooling units. We i

[_T tGC labeled the ANO incident "significant" (meaning fairly terrifying) l l

(") because it was a "ccnnon cause failure of redundant safety cooling syst ems," and because the utility's usual detection devices did not show l i

11 l 1 i l

r the failure had cccurred.

. Comon cause failure of redundant systems means that the' "defense in [

depth" concept the nuclear-industry refers to when it claims accidents  ;

won't happen, did not work. %e primary and backup systems were all oUt of camissien. Cnly renoval by hand was effective in getting rid of the biofouling.

4

%e ANO incident stinulated a flurry of activity to kill the creatures '<

causing the problem.-We tac asked all nuclear plant owners what the situation was at their plants, and what they planned to do about it.

Seabrook Station reported the environment there is teeming with nussels.

I

'Iwo years later, ANO 2 again showed serious synptoms of biofouling when, j during testing of a containment fan cooler heat exchanger, the flow of cooling water dropped by two-thirds in five minutes. Obviously, actions taken in 1980 to deal with the biofouling problem had not worked,  ;

At North Carolina's Brumwick Units 1 and 2, in 1981, the baffle plates in residual heat renoval Mm) heat exchangers were displaced because of i biofouling, allowing the cooling water to bypass the heat exchangers. '

Again, this was a loss of the primary and backup heat removal systems.

Wat same year, a heat exchanger in Pilgrim's safety-related Reactor  :

Building Closed Cooling Water (BBCCW) system was knocked out when baffle .

plares deformed, again due to biofouling.

More recently, in 1985, the Millstone 1 nuclear plant in Connecticut had .

g an emergency shut down, or scram, because corrosion products in the  !

j feedwater system traveled through the reactor core into the main steam feed line. Palo Verde 2 in Olifornia had a large nunber of leaking welds in the Essential Spray Pond Piping r.ystem caused by MIC. A 1987 NRC report said 2

MIC has caused severe p1ttics corrosion, and through wall pitting in piping [

and heat exchanger tubing aft er one month. We report said piping designed i to last 30 years nay last only 5.

An NRC report to Cor4 is on abnormal occurrences noted that fire

! protection systems are a prime candidate for biofouling. It also said that 1 since punps used after a Loss of Coolant Accident to recirculate water are usually outside the containment, punp seals degraded by biofouling would

let radiation escape to the atmosphere. (

A 1985 NRC report on the safety consequences of biofouling states the

connon cause nature of the problem means that it can eliminate all RHR punp coolers, leaving no backup system. If RHR cooling is Icst during normal 4

shutdown, a plant can stay in hot standby while the problem is repaired,

However, the report said, if the loss of the RHR system occurs during an i

accident, there could be severe consequences. For exanple, the report said

, the tree Mile Island accident could have been nuch more serious if it had ,

i suffered frcn this sort of biofooling. Also, since the problem affects usually inactive backup systens, it is very likely degradation of those 1

systems due to biofouling may first be noticed when they fail after being  ;

called on in an accident. '

he MIC-causing bacteria is part of an ecosystem which establishes itself naturally in power plant cooling systems where the water is slow d

12  !

flowing. We ecaystem is made up of sediment, which precipitates out of

• m) both salt and frosh water and lines cooling water piping; and bivalves like Asian clams, American oysters, and nussels which enter the system as larvae and attach themselves to pipe walls.

Under the sediment, in an oxygen-free environment, lives the bacteria.

It is called sulfate-reducing bacteria, or SRD. SRB is black, slimey, and .

smells like rotten eggs. We sediment is made of silt, nud, algae, fungi, bacteria, and other organic materials. It forms a hardened surface,-

! protecting the bacteria from almost all disturbances, including i

chlorination methods used at Seabrook Station to eliminate bivalves.

! According to a 1977 report by Normandeau, a cogany under contract to j Seabrook Station, the MIC problem was considered ten years ago. 'At the time, the proposed solution to biofouling was thermal backflushing of the

, cooling systems every two weeks and intermittent chlorination. Even though 1 the plant was designed to acc m odate thermal backflushing, the utility, in testing that method, found it could cause serious damage to the plant. As reported by the Environmental Protection Agency, the utility said:

j 1. We rapid tenperature change of water could cause thermal shock to the j tunnel walls, resulting in structural damage, or spalling.

2. We rapid flow reversal could cause severe water hamer shock and damage

! the intake valves in the punp bays.

j 3. Too rapid a reversal of the flow could upset the steam turbine-electric 3 generator systems and create station instability.

l k

4. Every time the plant is backflushed, it would reduce power generation by

]j half.

Spalling would split off layers parallel to the wall surface; water

hamer can also disable feedwater systems, causing a Ioss of Coolant

Accident; and station instability has a very discomforting ring. So, the i utility argued, to avoid wrecking the plant in their efforts to remove i biofouling organisms, continuous chlorination was the only workable method.

! We EPA agreed, and the cocpany began chlorinating in 1983. %e EPA also

gave permission to thermal backflush up to four times per year, with the

! cption of more backflusing with special EPA permission.

tree years later, Raymond Lavoie, an electrician working on site, saw pipes in the fire sprinkler system so cicgged with sediment that a twelve inch pipe was reduced to an inside diameter of four inches. '.'ne NRC i admitted MIC had been a problem in that system, but said the utility had

! taken the system apart, cleaned it, and tests in 1987 showed that the I problem was under control.

Corrosion and sediment detection is very difficult. When the istC stated ,

in 1986 that there was no longer a problem with MIC in the fire protection  !

, system, they cited the fact that the water, when tested through the system,  !

i was clear. However, another report, written for the NRC in 1987, said l j \ sedimentation has been a problem even where the water appears visibly  !

j clean. Sediment cannot be detected by measuring water flow velocity because  !

i \

?  :

} 13

(

a

r3 the sediment only builds up to a flow velocity it can tolerate, and then

( ) ceases growth. %us, the water velocity remins the same even though the U pipe diameter is considerably reduced. We report also said visual inspection is the most effective way to detect MIC, but it requires -

disrantling of equipment, and is very labor intensive. When such inspections are done on radioactive plants, it reans rote radioactive exposure to workers as well.

At an @ril,1987, meeting with the rmC, Lavoie reported that a formr Seabrook worker told him prts of the gate valves in the fire protection system in the turbine building had to be replaced because they had been eaten by MIC. %e imC admitted that 6 and 8-inch valves in that system had to be removed and cleaned, but did not say why the action had to be taken.

So, three years after the utility began their chlorination problem, sedimnt and MIC were still inhabiting Seabrook Station. And the rest recent reports about the problems in the PCCW show that the methods the utility is using to eliminate the biofouling problem are clearly ineffective.

mis is not surprising since Dr. Daniel Pope of Rensselaer Polytechnical Institute, and under contract to the utility, said in testirony before the EPA that continuous chlorination will not rerove MIC, even if doses of chlori..e ten times that allowed by the EPA are used. Since rarine debris was also found to be clogging the PCCW system in 1987, continuous

/' N chlorination is apparently not effective against rollusks, either.

i  !

'd tbt only is the chlorination method ineffective against biofouling, but the imC, in a 1987 report to Congress, said the reliability of chlorination systems is an industry-wide problem. Also, chlorine itself induces corrosion. A large number of safety-related corpaneats had to be replaced because of chlorine corrosion at the Fort Saint vrain nuclear plant in Colorado, according to a 1987 study.

Dr. Pope also said corrosion can be a particular problem when the bacteria sets in improperly welded areas of pipes. In 1984, and again in 1987, the imC evaluated the effects of 100 irproper welds in the service water system, a problem raised by a welder kho worked on the system.

Because the piping system in question is buried, the imC's inspection consisted of a pperwork review and interviews with utility officials. In deciding the welds were not a problem, the imC did not take into account the effects of MIC on the bad welds.

We biofouling problem at Seabrook Station poses numrous questions about the plant's ability to operate safely. Corrosion and clogging of piping systems due to biofouling can create a risk to the public safety.

Solutions to the problem are quite inadequate. Chlorination does not work, and it also corrodes piping @ernal backflushing might erradicate biofouling, but could also wreck the plant, gm It secca the utility and the lac do not believe safety problers caused by biofouling should stand in the way of the nuclear plant's activation.

(

V) We tiew England Coalition on tNelear Pollution, an legal intervenor against Seabrook Station in the imC's licensing process, has a contention pending 14

~

t l

on the issue of biofouling and its safety significance. However, historically, NRC comnissioners either are not told about problems like this, or ignore the information, before they make a licensing decision.

Clearly, the plant should not receive a low power operating license,  !

especially while the PCCW is cobbled together in a fashion questionable '

even to tEC inspectors. Also, since the biofouling problem became known to '

the utility only after conponents of the PCCW failed, and since MIC in  :

particular is very difficult to detect, the problem mout likely exists in  !

systen. which the utility has not examined. Given the fact that poor welds  !

will aggravate safety problems, and numerous former seabrook Station  !

enployees have said there are many bad welds at the plant (a contention the NRC and the utility continu) to deny), an independent investigation of the plant must be undertaken to resolve these questions before the tac cccmissioners consider further licensing.  ;

?

f L

k 1

i i

e

[

l l

I i

l f

I 15 i

i i

i -1 l

! l 1

l I

I l

I i

i 1

APPh21 DIX B, f

i PART 1: UNRESOLVED ISSUES RAISED SEPIWSER,1986 l

PART 2: ISSUES RAISED APRIL,1987 AND WEREAPITR l

!O l

I O

16

W3 APPI2iDIX B, PART 1 i )

( ,/ Since discussions tegan between the lac and the ELP in the fall of 1986, the NRC has deferred addressing problems involving "procedural" issues.

Syncpsized below are allegations raised by Seabrook Station e m loyees which either fall into this prccedural category according to the tac, or are issues the allegers feel are still cpen to question. We procedural questions include problems in quality assurance, quality control, document contrcl, design control, and tech.dcal training. We apparent failure of these procedures has safety inplications for Seabrook Station.

i

1. Concrete was poured on frozen concrete in containment.

2. Erpty cans, bottles, and other debris, including a welding machine, were discarded in contain,ent concrete.

3. Technical training records do not exist prior to April,1985, preventing assessmnt and verification of training. Some people wre untrained and som were retrained.

4. Writt.en procedures and instructions were a primry training tool. Safety related construction procedures were written in a-biguous language, leaving the interpretation up to the reader.

5. Some welders were trained on the spot, and so e were irproperly trained, p I Scre welds, inaccessible for testing, were never checked.

(V 6. mgineers and tradespeople worked 18 to 20-hour shifts because work was chronically behind schedule.

7. Painters did quality control checks of other painters' work, a violation 1

of federal requirements of quality control "organizational independence."

8. People who reported safety problems suffered personal harassrent.

9. Retur, wire, and other debris were thrown into an electric generator on the second floor, north side, of the equipcent vault.

10. Start-up check offs were done carelessly.

11. Security was slack: an individual walked past the explosives detector with gunpowder in their pocket; guards scoked in the cpen doorway cf the fuel storage building. Unauthcrired pccple had access to restricted areas.

J 2. A TI-10 procedure we irplerented to eliminate inspection steps.

13. Procedures for equiprent installation and other construction practices ,

were viclated. When and if thosc vicletiens sere discovered, rather than being corrccted, the procedures were frequently rewritten to allow the )

violations to stand. l

[ 'N 14. Electricians were not prcretly trained.

\"l 15. Trainers and engineers gave classes which were not adeguate to the 17 l

y I, m

'needs 'of those 'oeing. trained.

1 16. Tracking of blueprints was inpossible, and drawing revision control was j ineffective. ,

1

17. Quali*y control and quality assurance slipped during the last few ye.=rs

] of construction tere was no QA/GC nn third shift, and none for concrete on second shift. .

\ 18. There was cheating on tests.  !

i 19. Paint thinner was spilled on electrical cables, possibly causing long-term degradmtion.  :

20. te reactor was filthy and had wooden ladders and debris in it. General  !

! practices at seabrook station were slovenly.

l- 1. Inproper welds were done by untrained welders. Bottlen of argen used in  !'

. *1 ding were contacinated with moicture but used anyway.

l 22. We control buildits air conditioning system (CBA) refrigerant lines j lack separation, thus could fail at the same time. l I 23. We emergency feedwater system is sup,, lied from a sitsle tank which  !

i also serves as cond .t.*.e storage for the main steam feedwater system. An i

adequate supply of wet 'o the reactor in an emergency cannot be assured i j since a dual system a n . lied by one source.  :

24. Prohibited work practices such as cold pulling and incorrect weld  !

l identification were used throughout the plant. -

3 i

I I I f i

l a r J

j t

1 1 i 4

1 1 n.

j  !

) f j '

1 i l l 'l f

I  ;

l i  :

18  !

i

p APPENDIX B, PART 2 L/I he following is a synopsis of a series of affidavits and statements .

given the NRC in April,1987, and thereafter. Many of the issues raised in these documents were not dealt with'in NRC report 50-443/87-07.

1. Dravo shop welds in the turbine building were defective and uncorrected.

Calls into question the vendor quality assurance program.

2. A Pullman-Higgins welder's affidavit states inadequacies in welding and pipe work were comon. Welds weren't properly identified, weld inspections were inadequate, and training for welders was inadequate. Many welds were done without preheating, creating porosity in the welds. Installation mistakes occurred because blueprints were frequently incorrect. @ ousands of arc strikes occurred. Pipe and pipe supports were assembled using the wrong materials: when the proper material couldn't be located according to the required nunber, other material would be used after the identification nunber was ground out and re-scribed. Sledgehamers and come-alongs were used to force pipes into place. <

3. A former Perini enployee who worked in the document control department states blueprints wcre not updated, co-workers were untrained, didn't know how to read blueprints, and put incorrect numbers on.olueprints. A quality assurance person in charge of CAD welding for containment appeared to be always drunk, and consistently reported incorrect figures.

4. A former laborer says blueprints did not match the as-built plant, and

'v blueprints were destroyed in the blueprint room.

5. A former ironworker states in an affidavit that work done by inexperienced "permit" workers frequently had to be redone. He saw concrete poured @en the temperature was too low, creating a cold seam. Blueprints were very difficult to interpret. Design problems and inexperienced workers led to serious cost overruns.

6. A frs e inspector says there were an exceptionally large nunber of "accw v ts" engineering dispositions to change the plant's design to reile wnat had been built, particularly toward tne end of~ construction.

He believes this was to speed up construction and to save money. He saw cracks in the equipraent vault leaking water. He believes there is exposed rebar in the cooling tunnels, and the tunnels have voids in the concrete, and places where the concrete is too thin.

s . -

j

7. Phone conversations with four:former Suclity Assurance engineers i revealed: a. problems with docenent traceability, installation of inproper '

hardware, lack of weld safety, inability to trace materials back to l vendors, and harassnent when raising safety questions. b. People hired to '

inspect work they had performed, large nwbers of Nonconformance Reports voided when procedures were rewritten to accept the nonconforming condition, materials traceablity problem, and equipment renunbered to g conform to specifications. c.-Inspector forced by a supervisor to cancel a Non-Conformance Report on a procedural violation. .

d. All large bore pipe

,s 19

. welds may be welded beyond the pipe thickness maximum, inproperly certified welds, poor qualf.ty assurance, e. A former UE&C manager knows of

\. massive destruction and theft of documents during the 1984 Reduction In.

Force. ,

8. An' anonymous phone caller revealed the cooling tower concrete was poured in two layers, and the layers are not structurally attached to each other; ,

a weld rejected by an inspector was pencilled in with graphite by the welder and then passed inspection.

9. A former carpenter states there was an incident of cold pulling in the middle of 1983.

., 10. A former carpenter knows of a two-by-four board left in the missile shield of Unit I containment after concrete was poured.

11. A experienced nuclear worker, a quality control inspector, saw an 18-inch crack in the core barrel for the Unit 1 reactor. He believes this was never corrected.

12. While working for an equipment vendor, an individual saw the supervisor falsify certification on materials being sold for Seabrook Station. 'Ihe ,

certification was supposed to have been signed by the manufacturer, not the ,

vendor.  !

r

\ ,

l n

1 E

t i

i O

10

' _