Northern States Power Company, Prairie Island Nuclear Generating Plant, Units 1 and 2, License Nos. DPR-42, DPR-60 and SNM-2506, Issuance of Director's Decision Under 10 CFR 2.206

NUCLEAR REGULATORY COMMISSION

[Docket Nos. 50-282, 50-306, and 72-10]

Northern States Power Company, Prairie Island Nuclear Generating

Plant, Units 1 and 2, License Nos. DPR-42, DPR-60 and SNM-2506,

Issuance of Director's Decision Under 10 CFR 2.206

Notice is hereby given that the Acting Director, Office of Nuclear

Reactor Regulation, has issued a Director's Decision concerning a

Petition dated June 5, 1995, filed by the Nuclear Information and

Resource Service and the Prairie Island Coalition Against Nuclear

Storage (Petitioners) under Sec. 2.206 of Title 10 of the Code of

Federal Regulations (10 CFR 2.206). The Petition requested that Prairie

Island Units 1 and 2 be immediately shut down and the operating

licenses be suspended until the issues raised in the Petition could be

resolved. The Petition was based on alleged problems with cracking of

the Prairie Island steam generator tubes and reactor vessel head

penetrations, use of the transfer channel between the reactor core and

the fuel pool during unloading and loading of dry cask storage units,

and use of the Prairie Island crane.

The Acting Director of the Office of Nuclear Reactor Regulation has

determined that the Petition should be denied for the reasons stated in

the ``Director's Decision Under 10 CFR 2.206'' (DD-96-21), the complete

text of which follows this notice. In reaching this decision, the

Acting Director considered the concerns expressed by the Petitioners in

letters to the NRC dated June 21, 1995, February 19, 1996 and March 13,

1996. The decision and the documents cited in the decision are

available for public inspection and copying in the Commission's Public

Document Room, the Gelman Building, 2120 L Street, NW, Washington, DC,

and at the local public document room located at the Minneapolis Public

Library, Technology and Science Department, 300 Nicollet Mall,

Minneapolis, MN 55401.

A copy of this decision has been filed with the Secretary of the

Commission for the Commission's review in accordance with 10 CFR

2.206(c). As provided therein, this decision will become the final

action of the Commission 25 days after issuance unless the Commission,

on its own motion, institutes review of the decision within that time.

Dated at Rockville, Maryland, this 27th day of November, 1996.

For the Nuclear Regulatory Commission,

Frank J. Miraglia,

Acting Director, Office of Nuclear Reactor Regulation.

DIRECTOR'S DECISION UNDER 10 CFR 2.206

I. Introduction

On June 5, 1995, the Nuclear Information and Resource Service and

the Prairie Island Coalition Against Nuclear Storage (PICANS), now

known as the Prairie Island Coalition (Petitioners), filed a Petition

pursuant to Section 2.206 of Title 10 of the Code of Federal

Regulations (10 CFR 2.206) requesting that the Nuclear Regulatory

Commission (NRC) immediately suspend the operating licenses for Prairie

Island Nuclear Generating Plant, Units 1 and 2, operated by Northern

States Power Company (NSP or Licensee).

II. Background

As a basis for their request, Petitioners presented four concerns

which are summarized as follows: (1) The Prairie Island steam

generators are suffering from tube degradation and may rupture unless

proper testing is conducted and corrective actions are taken; (2) the

Prairie Island reactor vessel head penetrations (VHPs) have stress-

corrosion cracks which, if not found and corrected, may result in a

catastrophic accident involving the reactor control rods; (3) plans for

loading and unloading of dry cask storage units in an emergency, which

include storage of irradiated components in the fuel transfer canal,

were not properly reviewed by NRC and do not satisfy NRC requirements;

and, (4) the physical integrity of the Prairie Island crane used to

lift the dry cask for Prairie Island's spent fuel requires physical

testing and a safety analysis before future crane use following its

handling of a heavy load for an extended period of time.

By a letter dated June 19, 1995, the Director of the Office of

Nuclear Reactor Regulation (NRR) denied the Petitioners' request for

immediate suspension of Prairie Island Units 1 and 2 licenses. The

Director stated that the NRC staff's review of the Petition did not

identify any safety issues warranting immediate action at the Prairie

Island Nuclear Generating Plant. The Director also stated that the NRC

staff would issue a Director's Decision addressing Petitioners'

concerns within a reasonable time.

PICANS submitted a letter to the Chairman of the NRC dated June 21,

1995, which reiterated the concerns raised in the Petition and

requested an evening public hearing within the vicinity of the Prairie

Island facility. In a July 12, 1995, response, the NRC staff informed

PICANS that an evening public hearing was not warranted at that time

but that the request would again be considered at the time of issuance

of the Director's Decision.1 PICANS was further informed that the

concerns raised in the June 21, 1995, letter would be addressed in the

Director's Decision.

---------------------------------------------------------------------------

\1\ For the reasons set out in the cover letter transmitting

this Decision, the NRC staff has again determined that an evening

public hearing is not warranted.

---------------------------------------------------------------------------

On February 19, 1996, Petitioners filed an addendum to their

Petition raising further concerns regarding steam generator tube

cracking and requested that Prairie Island, Unit 1 not be allowed to

return to operation until

[[Page 64542]]

certain inspections of steam generator tubes was conducted. In a March

1, 1996, response, the Director of NRR denied Petitioners' request for

action concluding that no safety issues warranting immediate action had

been identified.

On March 13, 1996, Petitioners submitted another addendum to the

Petition raising additional concerns regarding steam generator tube

cracking at Prairie Island and again requesting that the NRC require

that Prairie Island, Units 1 and 2 be placed in mid-cycle outages for

the purpose of steam generator tube inspections. Petitioners further

requested an informal public hearing if the NRC determined that such

testing need not be conducted.

In an August 21, 1996, response, the Director of NRR concluded that

the addendum did not raise any safety issues warranting immediate

action and that an informal public hearing was not warranted at that

time.

Petitioners' concerns are addressed below. In addressing these

issues, I have considered the concerns expressed by the Petitioners in

the letters of June 21, 1995, February 19, 1996, and March 13, 1996.

III. Discussion

A. Steam Generator Tube Degradation

The steam generators used at pressurized water reactors (PWRs) are

large heat exchangers that use the heat from the primary reactor

coolant to make steam in the secondary side to drive turbine generators

which generate electricity. The primary reactor coolant flows through

tubes contained within the steam generator. As the coolant passes

through the steam generator tubes, it heats the water (i.e., secondary

coolant) on the outside of the tubes and converts it to steam which

drives the turbine generators. Steam generator tubes made from mill-

annealed alloy 600 have exhibited a wide variety of degradation

mechanisms. Such material has been used in a number of steam generators

at commercial nuclear facilities, including the steam generators at

Prairie Island Units 1 and 2. These degradation mechanisms include

mechanically induced (e.g., fretting wear, fatigue) and corrosion-

induced (e.g., pitting, wastage, and cracking) degradation.

Steam generator tubes constitute a significant portion of the

reactor coolant pressure boundary. As a result, the structural and

leakage integrity of the boundary is important in ensuring the safe

operation of the plant. A loss of steam generator tube integrity has

potential safety implications, as noted by the Petitioners, namely, (1)

the loss of primary coolant which is needed to cool the reactor core

and (2) the potential for leakage of radioactive fission products into

the secondary system where their isolation from the environment cannot

be ensured. As a result of the importance of this portion of the

reactor coolant pressure boundary, NRC has regulations on maintaining

the structural and leakage integrity of the steam generator tubes. The

overall regulatory approach to ensuring that steam generators can be

safely operated consists of the following:

(1) Technical specification requirements to ensure that the

likelihood of steam generator tube rupture events is minimized,

including

(a) Periodic inservice inspection of the tubing,

(b) Plugging or repair of tubing found by inspection to be

defective, and

(c) Operational limits on primary-to-secondary leakage beyond which

the plant must be shut down.

(2) Analysis of the design-basis steam generator tube rupture event

to demonstrate that the radiological consequences meet 10 CFR Part 100

guidelines.

(3) Emergency operating procedures for ensuring that steam

generator tube rupture events can be successfully mitigated.

Steam generator tube degradation can be detected through inservice

inspection of the steam generator tubes. These inspections are

generally required by a plant's Technical Specifications which specify

the frequency and scope of the examinations along with the tube repair

criteria. In the 1970s, wastage (i.e., general tube wall thinning) and

denting (mechanical deformation of the tube) were the dominant

degradation mechanisms being observed. These degradation mechanisms

were readily detectable with the bobbin coil inspection method and were

effectively controlled or eliminated, in part, by improvements in water

chemistry. Stress-corrosion cracking (SCC) emerged in the mid-1980s as

the dominant degradation mechanism affecting the steam generator tubes.

SCC can be oriented axially along the tube or circumferentially around

the tube, or can consist of a combination of axial and

circumferentially oriented cracks. SCC that has an axial orientation

can be detected with a bobbin coil probe. The capabilities of the

bobbin coil inspection method at detecting axially oriented cracks

depend on such factors as the location of the cracking, interfering

signals, and the data analysis procedures.

Circumferentially oriented SCC emerged as a significant problem

affecting the industry in the late 1980s. The bobbin coil probe is

generally insensitive to such cracking (i.e., circumferential SCC); as

a result, locations susceptible to circumferential SCC may need to be

examined with techniques other than the bobbin coil. Historically,

probes such as the motorized rotating pancake coil (MRPC) probe have

been used to detect circumferential SCC at locations susceptible to

such degradation. Recently, more advanced probes (e.g., Zetec Plus-

Point probe which contains a plus-point coil) have been used.

Deficiencies have been identified in certain utility inspection

programs for detecting SCC, particularly circumferentially oriented

SCC. Potential deficiencies include using inappropriate probes for

inspecting locations susceptible to circumferential cracking, not

optimizing the test methods to minimize electrical noise and signal

interference, and not being alert to plant-unique circumstances (e.g.,

dents, copper deposits) which may necessitate special test procedures

found unnecessary at other similarly designed steam generators or not

included as part of a generic technique qualification.

Even though deficiencies in eddy-current inspection programs have

been identified, operating experience indicates that steam generator

tube integrity can be maintained at a plant when appropriate eddy-

current data acquisition (including probe selection) and data analysis

procedures are used, when the data analysts have been properly trained,

when the intervals between inspections are determined based on the

inspection findings, and when the operating environment of the steam

generator tubes is controlled (e.g., water chemistry control). Adequate

tube integrity has historically been achieved at plants through

inservice inspections that involved the use of bobbin and MRPC probes.

In some instances, operating intervals were shortened between

inspections to ensure tube integrity.

Nevertheless, inspection findings at the Maine Yankee Atomic Power

Station in 1994 and 1995 raised concerns that large circumferential

cracks could develop over the course of an operating interval or that a

large number of circumferential cracks may be present if a facility was

not using appropriate inspection techniques. As a result of these

inspection findings, the NRC staff issued Generic Letter (GL) 95-03,

``Circumferential Cracking of Steam Generator Tubes,'' on April 28,

1995,

[[Page 64543]]

which: (1) Requested affected licensees to evaluate recent experience

(including the Maine Yankee experience) concerning the detection and

sizing of circumferential cracks and the potential applicability of

this experience to their plants; (2) on the basis of the results of

this evaluation, including past inspections and the results thereof,

and other relevant factors, requested affected licensees to develop a

safety assessment justifying continued operation until the next

scheduled steam generator tube inspections were performed at their

plants; and (3) requested that licensees develop and submit their plans

for the next steam generator tube inspection as they pertain to the

detection of circumferential cracks.

Subsequent to the issuance of GL 95-03, the Petitioners made the

following requests with respect to steam generator tubes at Prairie

Island Units 1 and 2: Request (a)--``That all steam generator tubes in

Prairie Island Unit 2 be given a full length inspection utilizing the

more comprehensive and proactive battery of tests employed at Maine

Yankee during NSP's 1995 outage. Petitioners specifically demand that

the Zetec Plus Point Probe and any state of the art, eddy current probe

for corrosive cracking be employed at Prairie Island 2 during Outage 17

scheduled to end June 15, 1995.'' Request (b)--``That if the Zetec Plus

Point Probe and any state of the art probe are not employed during the

mid-June 1995 outage, then reactor Unit 2 be taken immediately off-line

until such time these specific Zetec Plus Point Probe and any state of

the art, eddy current probe for corrosion cracking are completed.''

Request (c)-- ``That Prairie Island Unit 1 immediately be placed into a

mid-cycle outage to perform the NRC requested actions outlined in

Generic Letter 95-03. In addition, all Unit 1 steam generator tubes be

inspected through the use of the Zetec Plus Point Probe and any state

of the art, eddy current probe for corrosion cracking.''

NSP submitted its response to the generic letter for Prairie Island

Units 1 and 2 by letter dated June 27, 1995. As discussed below, the

information submitted provides no indication of an active

circumferential crack mechanism at the Prairie Island units, nor does

it suggest any significant concern regarding the potential for large,

undetected circumferential cracks at these units.

The Prairie Island Unit 2 steam generators were last inspected in

June 1995. This inspection included a 100-percent, full-length

inspection with the bobbin probe. In addition, a 100-percent inspection

was performed with a combined MRPC/Plus-Point probe from the hot-leg

tube end to 3 inches above the tubesheet. Most row 1 and 2 U-bends were

also inspected with the MRPC/Plus-Point coil. The bobbin probe is

appropriate for performing the general-purpose, full-length inspection

of the tubing because of its capability to detect flaw geometries

exhibiting an axial component (e.g., corrosion thinning and wastage,

mechanically induced wear, pitting, and axial cracks). The bobbin

inspection was supplemented by inspections with a combined MRPC/Plus-

Point probe to provide enhanced sensitivity to detecting cracks. These

inspections encompassed the areas of axial crack activity with the

bobbin coil probe and, in addition, the locations most vulnerable to

circumferential cracking with the MRPC/Plus-Point coil.

NSP reports that the Prairie Island Unit 1 steam generators were

last inspected in January 1996. This inspection included a 100-percent

full-length inspection with the bobbin probe, except for rows 1 and 2

U-bends. Rows 1 and 2 U-bends were examined with MRPC/Plus-Point. All

hot-leg tubes were examined with rotating probe technology (including

Plus-Point) from the tube end to 6 inches above the top of the

tubesheet. All sleeves were examined full length with the Plus-Point

rotating coil.

In addition, NSP's response to the generic letter addressed, in

part, each of five locations at which circumferentially oriented

degradation has historically occurred in Westinghouse steam generators.

These locations are places where there is significant axial stress

associated with variations in tube geometry and include (1) tube

expansion transition areas, (2) dented top-of-tubesheet locations in

partial roll-expanded tubes (described below), (3) dented tube-to-tube

support plate intersections, (4) small-radius U-bends, and (5) sleeve

joints. Significant axial stress would contribute to the development of

circumferential cracking.

Regarding the first and second categories, the tubes at Prairie

Island are roll expanded over only the lower portion of the tubesheet

depth (i.e., partial roll expansion). NSP reports that the incidence of

circumferential cracks at expansion transitions where the tubes have

received a partial-depth expansion has been negligible industry-wide.

For Prairie Island Unit 1, the 100-percent MRPC/Plus-Point inspection

in the tubesheet regions in January 1996 did not find any

circumferential indications in the in-service tubes. Similarly, for

Prairie Island Unit 2, the MRPC/Plus-Point inspections in the tubesheet

regions did not identify circumferential indications.

With regard to the third category, circumferential SCC at dented

tube support plate intersections has only been reported at a limited

number of plants. In addition, dented regions have exhibited both axial

and circumferential SCC with axial SCC typically being the more

frequently observed degradation mechanism. Axial SCC at dented

locations can be detected with the bobbin probe. Although NSP has not

reported performing MRPC or Plus-Point examination at the support

plates, it has examined 100 percent of these locations using a bobbin

probe and has not reported any axial cracking. Not detecting any axial

cracking gives confidence that widespread circumferential SCC is not

occurring.

Regarding the fourth category, SCC in the small-radius (row 1 and

some row 2) U-bends has been extensive in Westinghouse steam

generators. This cracking has been predominantly axial, with only

isolated instances of non-axial cracks. NSP reports that the small-

radius U-bends are routinely inspected with the MRPC. In January 1996,

the licensee inspected 100 percent of rows 1 and 2 U-bends on Prairie

Island Unit 1 with the MRPC/Plus-Point and found no indications. The

June 1995 inspections at Prairie Island Unit 2 with the MRPC/Plus-Point

probe looked at the majority of small-radius U-bends, and found one

axial and no circumferential indications.

Regarding the fifth category, during the January 1996 inspection in

Unit 1, all in-service and new sleeves were examined full length with

Plus-Point. Indications were found in the upper sleeve weld region of

61 ABB Combustion Engineering welded tubesheet sleeves. These

indications were characterized as single or multiple circumferential

indications or volumetric indications. All of these sleeved tubes with

circumferential indications were removed from service by sample removal

and/or plugging. The volumetric indications were evaluated and

indications located within the pressure boundary were plugged. No

sleeves are installed in Unit 2. Sleeves were installed in Unit 1 to

address forms of tube degradation (e.g., axial cracking and

intergranular attack) other than circumferential cracking.

In response to the large number of indications identified in the

upper sleeve welds of ABB Combustion Engineering welded tubesheet

sleeves during the January 1996 Unit 1 outage, the NRC staff held

discussions and meetings with the Licensee to determine

[[Page 64544]]

the root cause of the indications. NSP pulled five sleeve/tube samples

during the outage to perform metallurgical analysis on and determine

the root cause of the indications. Four of the removed tubes contained

circumferential indications and one contained a volumetric indication.

NSP started up Unit 1 on March 3, 1996, and committed to perform a mid-

cycle outage to perform additional inspections unless the results of

the metallurgical analyses from the pulled sleeves indicated that

additional inspections would not be warranted.

ABB Combustion Engineering performed the metallurgical

examinations, with third-party review by the Electric Power Research

Institute. The results showed that the sleeve weld indications were not

service induced. Instead, they were original fabrication flaws that

were the result of faulty cleaning of tube surfaces prior to welding.

The examinations of the tube samples revealed the sizes of the flaws

were such that the structural integrity of the welds was not

compromised. None of the flaws showed any indication of having

propagated in service. Since the indications were not service induced,

the NRC staff agreed that a mid-cycle outage to perform further

inspections was not necessary.

ABB Combustion Engineering is currently revising its topical report

on sleeving to incorporate improved cleaning techniques prior to

installation of sleeves, in order to prevent such flaws in the future.

NSP plans to submit an amendment to the NRC for review to adopt the

revised ABB Combustion Engineering topical report prior to installation

of CE sleeves.

After GL 95-03 was issued, additional information from inspections

performed at Maine Yankee and the destructive examination of several

tubes removed from Maine Yankee became available. This additional

information appears in NRC Information Notice 95-40, ``Supplemental

Information Pertaining to Generic Letter 95-03, `Circumferential

Cracking of Steam Generator Tubes'.'' This information led to the

conclusion that the tubes with the largest indications at Maine Yankee

continued to exhibit adequate structural integrity at the time they

were found. This was attributable, in part, to the crack morphology as

discussed in the Information Notice. As a result, adequate tube

structural integrity was ensured for the operating interval between

inspections, even though the MRPC probe, rather than the Plus-Point

probe, was used during the earlier inspections.

As mentioned above, the safe operation of the steam generators is

ensured by performing inspections and repairing defective tubes,

limiting the operational leakage through the steam generators,

analyzing a design-basis steam generator tube rupture event to

demonstrate acceptable radiological consequences, and having

appropriate emergency operating procedures in place. As discussed

above, the staff believes that the inspection probes used during the

May 1994 and June 1995 outages at Prairie Island Units 1 and 2,

respectively, were adequate to provide reasonable assurance of tube

integrity. In addition, NRC requires an operational leak rate limit to

provide reasonable assurance that, should a leak occur during service,

it will be detected and the plant will be shut down in a timely manner

before rupture occurs and with no undue risk to public health or

safety.

Therefore, on the basis of (1) the fact that appropriate steam

generator tube inspections have been performed, (2) monitoring of

primary-to-secondary leakage is being conducted, and (3) the fact that

appropriate emergency operating procedures are in place, the NRC staff

has concluded that the Petitioners' request for the shutdown of Prairie

Island Units 1 and 2 until full-length tube inspections are completed

using the Zetec Plus-Point probe and any state-of-the-art eddy-current

probe should be denied.

B. Vessel Head Penetration (VHP) Cracking

The Petitioners contend that the VHP's at Prairie Island Units 1

and 2 are likely to have stress-corrosion cracks which, if not found

and corrected, may result in a catastrophic accident involving reactor

control rods. The Petitioners also contend that VHPs in PWRs in France,

Belgium, Switzerland, and Sweden are cracking and that French data

indicate that the cracking mechanism will not necessarily produce a

detectable leak prior to a break that would initiate a serious

accident. The Petitioners further contend that failure of a VHP could

cause the ejection of a control rod drive mechanism (CRDM), resulting

in a loss of control of the reactor and/or a serious leak that could

not be isolated and thereby could induce a loss-of-coolant accident.

The Petitioners request immediate, full inspection of all VHPs in Units

1 and 2 for cracking using state-of-the-art eddy-current testing. The

Petitioners also request that NRC immediately suspend the operating

licenses of both units until the VHPs are inspected.

This same issue has been the subject of a recent Director's

Decision under 10 CFR 2.206 issued by the Director of NRR. See All

Pressurized Water Reactors, DD-95-2, 41 NRC 55 (1995). There, the NRC

staff concluded, after reviewing the information referred to by that

Petitioner, that the likelihood of the formation of circumferential

cracks is small, the likelihood of forming small axial cracks is

higher, and that leaks would develop before catastrophic failure of a

VHP would occur. This would result in the deposition of boric acid

crystals on the vessel head and surrounding area that would be detected

during surveillance walkdowns. The Petitioners contend that this

conclusion is not supportable as French data indicate that the cracking

mechanism will not necessarily produce a detectable leak prior to a

break that would initiate a serious accident.

The NRC staff's review of the French data does not support the

Petitioners' contention that a crack would not be detected due to

leakage prior to catastrophic failure. Topical reports submitted to and

reviewed by the NRC staff indicate that cracks in the CRDM VHP's would

need to grow well above the reactor vessel head before reaching a

critical size that would lead to the catastrophic failure of a CRDM

VHP. The portion of the crack above the head would leak well before the

critical size is reached.

The circumferential crack at the French reactor was very small

relative to the size flaw that would jeopardize structural integrity.

Furthermore, the circumferential crack initiated from the exterior of

the VHP which is more susceptible to circumferential cracking. This

situation occurred after a small axial throughwall crack leaked. Thus,

it is expected that leakage would be detected long before significant

circumferential cracking could occur. Of the numerous VHP inspections

in Europe, Japan, and the United States, no additional cases of

circumferential cracking have been observed. The members of the

Westinghouse, Babcock & Wilcox and Combustion Engineering Owners Groups

through Nuclear Energy Institute submitted acceptance criteria for both

axial and circumferential cracking to the NRC for review and approval.

The acceptance criteria were partially accepted by the NRC staff. The

criteria for axial cracking were accepted as proposed. The criteria for

circumferential cracking were rejected. Any circumferential cracks

found must be reported to the NRC staff for disposition. If VHP

cracking violated the above acceptance criteria, the NRC staff would

review the Licensee's plan for monitoring or repair of the crack.

[[Page 64545]]

Finally, a foreign reactor developed extensive circumferential

cracking in VHPs as a result of two major demineralizer resin ingress

events in the early 1980s. The NRC staff issued a request for

additional information to NSP on September 25, 1995, to determine if

any similar resin ingress events had occurred at Prairie Island. The

Licensee responded to the NRC staff on October 24, 1995, that there

have been no resin ingress events at Prairie Island.

The NRC staff has closely monitored VHP cracking experience in the

U.S. and abroad and has reviewed extensive evaluations of VHP cracking.

The evaluations and operating experience indicate that it is highly

unlikely that significant circumferential cracks could develop and that

there is significant margin between the flaw sizes that would result in

detectable leakage and the flaw sizes that would jeopardize structural

integrity. Thus, the staff has concluded that VHP cracking is not a

safety concern at this time. To assure that VHP cracking continues to

be properly monitored and controlled, the NRC is in the process of

preparing a Generic Letter requesting addressees to describe their

program for ensuring the timely inspection of PWR CRDM VHPs and other

VHPs. This letter was issued for public comment on August 1, 1996.

Accordingly, the requests made by the Petitioners for the shutdown

of the Prairie Island units and inspection of the VHPs with enhanced

inspection techniques is denied. As explained above, the NRC staff has

concluded that no substantial health and safety issues have been raised

by the Petitioners.

C. Unloading of Dry Cask Storage Units

Spent fuel discharged from a reactor core is stored on site in a

spent fuel pool prior to transfer to the U.S. Department of Energy

(DOE) for final deposition. Typically, one-third of a reactor core is

discharged every refueling outage (approximately every 18 months in the

case of each of the Prairie Island units). The Licensee concluded

several years ago that it would reach maximum capacity in its spent

fuel pool in 1994, prior to availability of a DOE repository for

storage of spent fuel. To support the need for continued storage of

spent fuel at the reactor site, the Licensee applied to NRC for a

license to store spent fuel in an onsite independent spent fuel storage

installation (ISFSI). NRC issued Materials License No. SNM-2506 to NSP

on October 19, 1993, for receipt and storage of spent fuel at the ISFSI

on the site of the Prairie Island Nuclear Generating Plant. Materials

License No. SNM-2506 allows NSP to use the TN-40-type casks for storage

at its ISFSI. The TN-40, a metal cask system, is designed to store 40

PWR spent fuel assemblies in each cask. Dimensions of the cask (with

protective cover) are 202 inches high with an outside diameter of 103.5

inches. A loaded TN-40 storage cask weighs 109.3 metric tons.

On April 28, 1995, a public meeting was held in Red Wing,

Minnesota, to present NRC inspection findings related to dry cask

storage activities at the Prairie Island plant. Questions were raised

by members of the public as to how the Licensee would unload the spent

fuel in a dry storage cask, if it became necessary, i.e., would there

be enough empty fuel racks in the spent fuel pool to accommodate

unloading of the cask.

In a letter to the NRC dated May 3, 1995, the Licensee submitted a

plan for unloading the TN-40 cask in response to the questions raised

at the April 28, 1995, meeting. In that letter, the Licensee stated

that some of the fuel racks in the spent fuel pool contain nonfuel-

bearing components, which could be relocated to a temporary location in

the fuel transfer canal. Alternatively, it may be possible for the

components to be stored temporarily in the TN-40 cask, should it become

necessary to unload a cask. In the latter case, even though the TN-40

cask being returned to the spent fuel pool may no longer be qualified

to hold spent fuel, it quite possibly could still safely hold

irradiated nonfuel-bearing components.

The Petitioners raised issues concerning compliance with 10 CFR

50.59 and the need to make changes to Technical Specifications in order

to use the fuel transfer canal for nonfuel-bearing components under the

Licensee's plan. Petitioners also stated that 10 CFR 50.59 requires a

safety analysis and amendment to the operating license with a public

hearing whenever a change occurs in Technical Specifications for spent

fuel pool and reactor transfer canal use. Petitioners further stated

that a safety analysis is essential when a Technical Specification

change occurs.

The need for a change to the Technical Specifications and the

process to be followed under 10 CFR 50.59 are two separate, but

related, issues. With regard to the Prairie Island Technical

Specifications, the plan proposed by the Licensee in its letter of May

3, 1995, for dealing with the need to unload a cask, would not involve

a change to Technical Specifications because Technical Specifications

do not address use of the fuel transfer canal nor do they address

movement of nonfuel-bearing components within the spent fuel pool.

Prairie Island's Technical Specification 3.8 specifies operating

limitations associated with fuel-handling operations and core

alterations only. Further, the fuel transfer canal is not classified as

a reactor safety system. The fuel transfer canal provides no protection

for the reactor, nor does it mitigate the consequences of a postulated

accident to the reactor. The fuel transfer canal is a component of the

fuel storage and fuel handling systems, which is considered a plant

auxiliary system rather than a reactor safety system. As use of the

fuel transfer canal in the Licensee's plan does not involve a change to

the Technical Specifications, an amendment for this reason would not be

required and the opportunity to request a public hearing with regard to

a Technical Specification change would, therefore, not arise.

With regard to Sec. 50.59 of Title 10 of the Code of Federal

Regulations, that provision allows a Licensee to make changes to its

facility and procedures as described in the Final Safety Analysis

Report (FSAR) without prior approval from NRC, provided a change in

Technical Specifications is not involved (which, as described above, is

met in this instance) and an unreviewed safety question does not exist.

Before moving the nonfuel-bearing components to temporary storage racks

in its fuel transfer canal, NSP would need to determine if this use of

the transfer canal changes the facility or procedures as described in

the FSAR. If NSP determines that a change has been made to the facility

or procedures as described in the FSAR, then a safety evaluation

pursuant to 10 CFR 50.59 is required to be performed by the Licensee.

If a Technical Specification change were needed (not the case as

discussed above), or an unreviewed safety question existed, NRC review

and approval would be required. Otherwise, the Licensee could make the

modifications without prior NRC approval. Licensees submit a list of

modifications that were performed under 10 CFR 50.59 without NRC

approval to NRC annually.

The Licensee did not fail to comply with the requirements of 10 CFR

50.59 by presenting a plan for retrieval of fuel from a cask, which

included an option to place nonfuel-bearing components in the fuel

transfer canal. At the time a cask unloading is deemed necessary, the

Licensee can evaluate the specific modifications needed to implement

the plan and determine whether 10 CFR 50.59 is applicable.

When applying for the license, NSP performed an accident analysis,

in its Safety Analysis Report, as required by

[[Page 64546]]

NRC regulations.2 In its Safety Evaluation Report dated July 1993,

the NRC staff reviewed the Licensee's accident analysis and determined

that ``Dose equivalent consequences, from a single cask, to any

individual, from direct and indirect radiation and gaseous activity

release after postulated accident events, are less than the 50 mSv (5

rem) limit established in 10 CFR 72.106(b).'' Additionally, in its

Environmental Assessment, dated July 28, 1992, the NRC staff assessed

the accident dose at the Prairie Island site boundary as: ``a small

fraction * * * of the criteria specified.* * * '', and found that:

``These doses are also much less than the Protective Action Guides

established by the Environmental Protection Agency (EPA) for

individuals exposed to radiation as a result of accidents;* * *''

Because it has been shown that the dose equivalent from a single cask

to any individual from postulated accident events is not in excess of

the levels required for taking protective actions to protect public

health, the NRC staff considers that a time-urgent unloading of the TN-

40 cask is not a likely event.

---------------------------------------------------------------------------

\2\ The Licensee analyzed accidents classified as Design Events

III and IV, as described in ANSI/ANS 57.9, ``Design Criteria for an

Independent Spent Fuel Storage Installation (Dry Storage Type).''

Design Event III consists of that set of infrequent events that

could reasonably be expected to occur during the lifetime of the

ISFSI. Design Event IV consists of the events that are postulated

because their consequences may result in the maximum potential

impact on the immediate environs. Included among the scenarios

considered under Design Event IV was a loss of confinement barrier

leading to an immediate release of radioactivity.

---------------------------------------------------------------------------

Even if such an unlikely accident occurred and the Licensee

determines that corrective actions may need to be taken to maintain

safe storage conditions, options are available. This may include

returning the cask to the auxiliary building and/or the spent fuel pool

for repairs. Once the cask is in the spent fuel pool, it does not

necessarily have to be unloaded to maintain safe storage conditions. In

addition, the Licensee may have other options available to cover this

unlikely contingency including temporary storage of spent fuel in a

spare storage cask or use of an existing certified transportation cask.

The Licensee would have time to consider these, and other available

options, in such an unlikely event.

Petitioners also raise an issue concerning the necessity to offload

both the entire reactor core and a TN-40 cask simultaneously. NRC has

no requirement for licensees to maintain the spent fuel capacity to

offload the entire core at once. Prairie Island normally offloads only

one-third of the core during refueling outages. If NSP determines the

need to offload the entire core during a refueling outage, NSP can

install temporary fuel racks in the cask laydown area in the spent fuel

pool. Therefore, a cask could not be unloaded for the short time that

temporary racks are installed in the cask laydown area. The staff does

not view this as a problem for two reasons. First, the probability that

a cask would require unloading at the same time a full-core offload is

in process is extremely small. Second, in the event it became necessary

to unload a cask, fuel could be placed back into the reactor vessel and

the temporary fuel storage racks could be removed. As discussed above,

time-urgent unloading of a TN-40 cask is extremely unlikely. The cask

could then be unloaded after the cask laydown area was cleared of the

temporary fuel storage racks.

In addition to assuring that a TN-40 cask could be unloaded if

necessary, the Licensee's plan also provides assurance with regard to

spent fuel retrievability. Subpart F of 10 CFR part 72 provides general

design criteria for ISFSIs and monitored retrievable storage

installations. Section 72.122 sets overall requirements and 10 CFR

72.122(l) provides for retrievability of the fuel and states: ``Storage

systems must be designed to allow ready retrieval of spent fuel or

high-level radioactive waste for further processing or disposal.'' The

NRC staff concluded in a May 5, 1995, letter to the Licensee that the

ability to unload a TN-40 cask if necessary in accordance with the

Licensee's plan would satisfy this fuel retrievability provision.

Finally, Petitioners state that the wrong NRC department reviewed

and approved NSP's plan for retrievability of irradiated fuel. The

Office of Nuclear Material Safety and Safeguards (NMSS) is responsible

for licensing and regulating all issues under 10 CFR part 72, including

issues related to the design requirements for ISFSIs. Therefore, NMSS

is the correct NRC office to review whether the licensee's plan met 10

CFR 72.122(l). As discussed above, the Licensee's plan does not involve

a Technical Specification change. Accordingly, NRR review of such a

change would not be required. If, upon implementing its plan, the

Licensee determined that a safety evaluation pursuant to Sec. 50.59 was

required, NRR review and approval would be required only if an

unreviewed safety question existed.

With regard to the requests made by the Petitioners, there is no

basis for suspending NSP's operating licenses for the Prairie Island

units until a safety analysis is completed, reviewed, and approved by

NRC, and until NSP's licenses are amended and public hearings have been

held. If NSP plans to implement a specific plan to utilize the fuel-

transfer canal which changes the facility or procedures as described in

the FSAR, then an evaluation pursuant to 10 CFR 50.59 would be required

at that time, which would not require prior NRC approval unless an

unreviewed safety question exists or a change to Technical

Specifications is required.

D. Auxiliary Building Crane

Petitioners contend that a recent incident at Prairie Island on May

13, 1995, involving the crane used to lift the dry cask for Prairie

Island's ISFSI, requires physical testing and safety analysis before

future crane use. The incident resulted in the crane holding the

123.75-ton cask above the surface of the reactor pool for 16 hours. The

Petitioners assert that the incident could have caused metal fatigue

within the crane's structure and the cables attached to the crane.

Also, Petitioner Prairie Island Coalition asserts in its June 21, 1995,

letter to the Chairman of the NRC that the crane, its cable, and its

cable mechanisms were not designed to withstand holding nearly a

maximum load for 16 hours.

The Prairie Island auxiliary building crane was upgraded in 1992 in

accordance with the provisions of Topical Report EDR-1(P), ``Ederer

Nuclear Safety-Related Extra Safety and Monitoring (X-SAM) Cranes.''

The crane is designed and tested in accordance with the NRC staff's

guidance as outlined in NUREG-0554, ``Single-Failure-Proof Cranes for

Nuclear Power Plants,'' and NUREG-0612, ``Control of Heavy Loads at

Nuclear Power Plants.''

The staff evaluated the design of the auxiliary building crane and

the lifting device for the cask as part of its review of the dry cask

ISFSI. This crane system is designed so that a single failure will not

result in the loss of the capability of the system to safely retain the

load (this design is known as single-failure proof). The crane is

designed to handle a rated load of 125 tons and is capable of raising,

lowering, and transporting occasional loads, for testing purposes, of

25-percent higher than the rated load without damage or distortion to

any crane part. All parts of the crane that are subjected to dynamic

strains, such as gears, shafts, drums, blocks, and other integral

parts, have a safety factor of 5 (i.e., they are designed to lift 5

times the design rated load). The hook has a

[[Page 64547]]

design safety factor of 10 and was subjected to a 200-percent overload

test followed by magnetic particle inspection prior to initial

operation. Protection against wire rope wear and fatigue damage are

ensured by scheduled inspection and maintenance. The special lifting

device used for cask movement is designed to support 6 times the weight

of the fully loaded cask and was subjected to a 300-percent overload

test by the manufacturer. The lifting device undergoes dimensional

testing, visual inspection, and nondestructive testing every 12 months

(plus or minus 25 percent).

A single-failure-proof crane, such as the crane at Prairie Island,

that has become immobilized by failure of components while holding a

load, is able to hold the load or set the load down while adjustments

or repairs are made. Safety features and emergency devices permit

manual operation to accomplish this task. Two separate magnetic brakes

are provided as well as an emergency drum band brake. Each magnetic

brake provides a braking force of at least 150 percent of rated load.

The emergency drum brake assures that the load can be safely lowered

even if power is lost to the crane. Because of the large design margins

and the ability to withstand a failure of any single component, the NRC

staff does not postulate a load drop from a single-failure-proof crane.

After the incident on May 13, 1995, the Licensee temporarily

removed the crane from service for testing. The Licensee and the crane

vendor performed testing on the crane to analyze the event and assure

the crane was operable. The Licensee's analysis of the May 13, 1995,

incident found the problem to be an improperly calibrated load cell (a

load cell is a device that measures the load being lifted by the crane

and provides input to an overload-sensing device). It was determined

that the actual load was less than what was being sensed by the

overload-sensing device. The function of the overload-sensing device is

to stop the operation of the crane when the load reaches a

predetermined value. This prevents loading the crane beyond its rated

load by maintaining loads within the design working limit, thereby

maintaining safety and the physical integrity of the crane system.

Since the design-rated load of the crane was not exceeded during

the incident, there is no reason to assume that the crane cannot

continue to operate safely. Even if the rated load had been exceeded,

an analysis would be needed to determine how much the rated load was

exceeded and if that amount is significant. When cranes are built,

manufacturers conduct proof tests at a load above rated load. The proof

test for this crane was 25 percent higher than the 125-ton design-rated

load for the main hoist (i.e., the proof test was 156.25 tons).

With regard to the Petitioners' comment about metal fatigue, metal

fatigue is a condition that results from cyclic stress. Cyclic stress

is produced by repeated loading and unloading. The crane is designed to

handle all loading and unloading cycles during the life of the plant,

including construction and operating periods. A single static

(constant) load such as the load in question, does not produce the

cyclic stress that causes metal fatigue. The Petitioners' contention

that it was never contemplated that the Prairie Island polar crane hold

a load of 123.75 tons inches above the surface of the reactor pool for

16 hours is incorrect. The contemplated failure mechanism of a single-

failure proof crane is to hold the load safely at any location until

the load can be safely moved. Because of the large design margins, the

length of time that a design-rated load (or a load less than design

rated) is on the hook of a single-failure-proof crane is

inconsequential.

With regard to cable and cable mechanisms (also known as the

reeving system and lifting devices), the crane is provided with a

balanced dual reeving system with each wire rope capable of supporting

the maximum critical load (if a load being held by a crane can be a

direct or indirect cause of release of radioactivity, the load is

called a critical load). The hydraulic load equalizing system allows

transfer of the load to the remaining rope, without overstressing it,

in the event of a failure of one rope. Protection against wire rope

wear and fatigue damage are ensured by scheduled inspection and

maintenance.

In conclusion, NRC agrees with the Licensee in its determination

that the cause of the incident was an incorrectly calibrated load cell.

This cause was documented in NRC Inspection Report 95-006, issued June

27, 1995. NRC has determined that the Licensee met the design and

testing requirements established in industry standards for the control

of heavy loads such as a dry storage cask, that the overload-sensing

device worked as designed, and that no safety issue was involved in the

Licensee's use of the auxiliary building crane and associated cask

handling equipment to move the cask. Therefore, the Petitioners'

requests for suspension of NSP's licenses for the Prairie Island units

until physical testing and safety analyses can be performed on the

crane are denied.

IV. Conclusion

Petitioners requested an immediate suspension of NSP's licenses for

Prairie Island Units 1 and 2 until corrective actions of potentially

hazardous conditions would be taken by NSP and NRC with regard to

issues identified in the Petition. The institution of a proceeding in

response to a request for action under 10 CFR 2.206 is appropriate only

when substantial health and safety issues have been raised. See

Consolidated Edison Co. of New York, (Indian Point, Units 1, 2, and 3),

CLI-75-8, 2 NRC 173, 176 (1975), and Washington Public Power Supply

System (WPPSS Nuclear Project No. 2), DD-84-7, 19 NRC 899, 923 (1984).

I have applied this standard to determine if any action is warranted in

response to the matters raised by the Petitioners. Each of the claims

by the Petitioners has been reviewed. The available information is

sufficient to conclude that no substantial safety issue has been raised

regarding the operation of Prairie Island Units 1 and 2. Therefore, I

conclude that, for the reasons discussed above, no adequate basis

exists for granting Petitioners' requests for immediate suspension of

NSP's licenses for Prairie Island Units 1 and 2.

A copy of this decision will be filed with the Secretary of the

Commission for the Commission to review in accordance with 10 CFR

2.206(c).

As provided by this regulation, this decision will constitute the

final action of the Commission 25 days after issuance, unless the

Commission, on its own motion, institutes a review of the decision with

that time.

Dated at Rockville, Maryland, this 27th day of November, 1996.

For the Nuclear Regulatory Commission.

Frank J. Miraglia,

Acting Director, Office of Nuclear Reactor Regulation.

[FR Doc. 96-30949 Filed 12-04-96; 8:45 am]

BILLING CODE 7590-01-P