• • UNITED STATES OF AMERICA

NUCLEAR REGULATORY COMMISSION

In the Matter of

COMMONWEALTH EDISON COMPANY

(Dresden Nuclear Power Station, Units 1 and 2; Quad Cities Nuclear Power Station, Units 1 and 2)

) )

~ )

~ EXEMPTION

I.

Docket Nos. 50-237, 50-249, 50-254, and 50-265

Commonwealth Edison Company (CECo, the licensee) is the hold€r of

Operating License Nos. DPR-19 and DPR-25, which authorize operation of

Dresden Nuclear Power Station, Units 1 and 2; and Operating License Nos.

DPR-29 and DPR-30, which authoriie operation of Quad Cities Nuclear Power

Station, Units 1 and 2. These licenses provide, among other things, that

Dresden artd Quad Cities are subject to ell rules, regulations, and Orders of

the Commission now or hereafter in effect.

Dresden Station is comprised of two boiling water reactors at the

licensee's site located in Grundy County, Illinois. Quad Cities

Station is comprised of two boiling water reactors at the licensee's site

located in Rock Island County, Illinois.

I I.

By letter dated November 12, 1991, the licensee requested an exemption

from certain Type B (local leak rate) testing requirements of Appendix J to

·10 CFR Part 50, fora total of 83 two-ply containment penetration expansion

.. • • -2-

bellows at four reactor units. This is because the bellows design is such

that they cannot be properly tested to satisfy Type B testing requirements,

barring replacement with bellows of a different design.

On November 19, 1990, during the Quad Cities, Unit 1, Cycle 11 Refueling

Outage, the licensee performed a local leak rate test (LLRT) on the Orywell

Ventilation Penetration· X-25. bellows. This was done in the normal way, by

pressurizing the small space between the two plies of the bellows assembly

by using a test tap made for that purpose. The measured leakage rate was

4.3 standard cubic feet per hour (scfh). After the LLRT was performed, a

significant amount of maintenance and new construction work was performed in

the area surrounding the bellows asseinbly. This included extensive maintenance

on a valve located in-line with the bellows (valve 1-1601-23) and installation

of a new penetration (X-109, Reactor Vessel Level Instrumentation Lines)

directly above X-25. After the completion of this maintenance and construction,

a new LLRT was performed on the X-25 bellows, with a measured leakage rate

of 6. scfh. Approximately .two days later, the primary containment integrated

leak rate test (ILRT) was performed. While the containment was at pressure,

application of a soap solution to the surface of the X-25 bellows indicated

three cracks ranging in length from 0.187 inch to 1.7 inch, and a large

number of small pin-hole cracks. The ILRT was successfully completed with

the leaking bellows in its as-found condition. Following the ILRT, an

·. add·itional LLRT was performed on the bellows, and the results matched the

previous LLRT leakage rate. A soap solution was applied to the bellows

assembly during this LLRT and showed only a few small leaks.

• •• -3-

Next a "special" LLRT was performed in an effort to quantify actual

leakage from the bellows. A steel plate was welded to the vent line inlet

which is located inside the drywell. The bellows were pressuriZed through

a.threaded hole in the plate and a leak rate test was performed on the·

entire penetration. The soap solution indicated a large leak with many

small leaks similar. to that encountered during the ILRT. A leakage rate of

137 scfh was measured.

With the validity of the LLRT in question, the licensee tried to

determine the sensitivity of the LLRT procedure to detect and quantify

leaks. A 0.25-inch hole was drilled through the two bellows from the outer

diameter to the inner diameter in the convolute adjacent to the LLRT taps

on the bellows. A LLRT was performed and resulted in a small increase in

leakage (from 6 scfh to 7 scfh)~ A second hole w~s drilled and the LLRT

was repeated~ The me~sured leakage was 8 scfh.

These tircumstances indicated that the current method used to p~rform

a LLRT on two-ply containment penetration bellows could identify leakage,

but could not quantify the extent of the leakage. The licensee formally

notified. the Commission of these findings by letter dated March 27, 1991.

The bellows assembly for penetration X-25 at Quad Cities Station is

typical of two-ply- bellows for other containment penetrations at both Quad

Cities and Dresden Stations. These assemblies are original plant

equipment, which were manufactured and installed in the late 1960s and

early 1970s. These flexible metallic bellows are constructed with two

plies of austenitic type 304 stainless steel which are formed together into

cytindrical corrugated bellows"_elements._ .. This design_ configuration is

• • -4-

typical of bellows penetrations which are used at all units at Dresden and

Quad Cities Stations. The investigation conducted by the licensee, which

- included discussions with the supplier and an independent analysis at

Argonne National Laboratory, revealed that the forming process can bring

_ the plies into contact, thereby limiting the flow of the local leak rate

test medium (inert gas or air) between the inner and outer plies. The X-25

LLRT and ILRT test results indicated that leakage can be detected under

these conditions;. however, the leakage cannot be accurately quantified.

In order to achieve full compliance with Type B testing requirements,

both Dr~sden and Quad Cities would be requited to replace all two-ply

containment penetration bellows with a testable bellows design. The cost

of replacement of two-ply bellows assemblies is projected to be

approximately $400,000 per assembly, which would mean between approximately

$7.1 million and $9.5 million per unit for both Dresden and Quad Cities

Stations. The total cost of bellows replacement is projected to be

approximately $33.35 million.

I I I.

In lieu of an Appendix J Type B test on the applicab-le two-ply

conta'inment penetration bellows assemblies at Dresden and Quad Cities

Stations, the licensee proposes to implement the following testing program: -.

1. All two-ply bellows will be locally pressurized with air (between the plies) at a pressure of Pa. The leakage rate will be meas~red in accordance with station procedures. If leakage is less than 0.5 scfh, the bellows assembly will be considered to be intact and no further testing on that bellows assembly is necessary.

2. Tf the leakage rate :is greater than or equaLto -0.5 s_cfh, t_hen_ the bellows assembly will be locally pressurized at the test

.~ • • -5-

taps with helium (between the plies) at a pressure of Pa. The outer ply will then be tested for the presence of helium with a helium sniff detector. If no helium is detected, the integrity of the outer ply will be considered to be intact, and no further testing on that bellows assembly is necessary.

3. ·If helium leakage is detected through the outer ply, then the inner ply will be tested for the presence of helium. If no helium is detected, the integrity of the inner ply will be considered to be intact, and no further testing on that bellows assembly is necessary.

4. If.helium is detected through both the inner and outer plies, then the protective covers will be removed, and the outer ply will be examined by penetrant and/or snoop testing. All observed flaw indications will be measured and mapped. Bellows assemblies which indicate leakage through both plies will not be considered to be intact.

5. All crack indications will be evaluated by the [licensee's] Nuclear Engineering Department (NED) and the current and projected leakage rate will be calculated. The NED review will include a structural assessment of the bellows with regards to critical flaw size.

6. Upon completion of the two-ply bellows testing program, a Type A ILRT test will be performed to verify primary containment integrity.

7. All two-ply bellows assemblies which demonstrate leakage through both· plies will be replaced during the subsequent refuel outage, unless Commonwealth Edison Company provides justification for continued operation greater than one operating cycle. · · · ·

Thi.s Exemption and associated testing program is requested for each non­

testab le two-ply bellows assembly (original design). Upon replacement with

a t~stable bellows· assembly, that bellows will no longer be included .in the

Exemption and will be required to be tested in accordance with the normal

Type 8 test program. Similarly, if a method is developed which insures a . .

. valid Type B test on one or more bellows assemblies, those bellows will

also b~ excluded from the Exemption and will be reqOired to be tested in

accordance with the normal-Type B test prograr:l.

r' • • -6-

This testing program is intended to assure that at least one ply of a

two-ply bellows is intact and that overall containment leakage is within

its allowable limit as shown by Type A testing. The Type A test is

essential to this program, because it is the only test available that can

properly quantify the bellows' .leakages,· albeit not individually. This is

especially important f~r those bellows which are known to leak but will

not be replaced until after another cycle.

It is also important to b~ assured that a leaking bellows will not

degrade excessively during the period that ~nds with its replacement. The·

licensee examined the X-25 bellows assembly ·from Quad Cities and determined

that the crack mechanism was transgranular stress corrosion cracking

(TGSCC). The licensee stated that this mechanism, which has caused . .

previous bellows assembly ~eterioration at Dresden and Quad Cities, is

normally characterized by the slow development and propagation of.cracks.

The X-25 penetration deteriorati-0n is unique in the licens~e's experience

since the bellows appeared to exhibit' a large. increase in leakage during

one operating cycle based upon the potential impact of 137 scfh on the ILRT

results. This large amount of leakage would have had a noticeable impact

upon ILRT results during previous outages (although it would not have led

·to the failure of the ILRT). This significant leakage increase may have

occurred as a result of maintenan_ce work associated with the replacement of

a valve. which is directly in-line with the bellows. During the replacement . '· ~

:of the in-line valve (1-1601-23), which is i6cated approximately 12 inches

from the bellows ass_embly, excessive force was used to remove the valve •

. Resultant torsional and/or.translational for.ces may have7caused an

• • -7-

accelerated growth of existing TGSCC in the bellows. The metallurgical

investigation also identified the presence of several corrosive species

which contribute to TGSCC. These included chlorides, flourides, and sulfides.

The origi~al form of this material could not be determined and the method

substance deposit is therefore unknown.

The licensee has performed a fracture mechanics evaluation, detailed

in their submittal, which concludes that substantial structural margin

exists to ensure that, during one operating cycle, catastrophic failure

should not occur for bellows ass.emblies with cracks and holes of the type

and size that would be detected by the proposed surveillance procedure.

Recent tests at Dresden, using the proposed procedures, have detected

flaws of the anticipated size in several bellows assemblies. The staff

questions, however, whether the LLRT with air could detect all significant

flaws, since the flow of air between the plies is so restricted. In order

to insure that the air test and associated threshold of 0.5 scfh would,

with sufficient sensitivity, detect leakage from a bellows assembly, th€

licensee performed a one-time helium flew rate validation test on the two­

ply bellows assemblies at Dresden Station during October 1991. This

validation consisted of two separate pressurizations and leak rate

·measurements, one with air at Pa,_and one with helium at Pa. By

pressurizing with helium, the licensee would then be able to detect if an . . .

obstruction between the plies was preventing air from reaching a leak. The

. helium flow rate compared favorably with the expected flow rates of helium

calculated for turbulent and laminar flow conditions, knowing the measured

, air-flow rates.

• -8-

Therefore, the initial air leakage test described i11 the proposed test

program is a valid means to test for the presence of leaks in two-ply

·bellows assemblies. This in turn validates the ability of the proposed

testing program {which includes a Type A test) to verify the integrity of a

bellows assembly, and insure that primary containment leakage is less than

0.75 La.

The staff finds that the proposed testing program will detect bellows

assemblies with significant flaws and result in replacement of flawed

assemblies within one operating cycle, during which period there is

reasonable assurance that the bellows assemblies will not suffer excessive

degrudaticn. If the 1icensee should propose to \-Jait lcnger than one cycle

to replace any belows assembly, the staff must evaluate and approve the

request at that time.

IV.

Accordingly, the ·Conunission has determineci pursuant to 10 .CFR 50.12(a){l)

and {a){2)(ii), that (1) the Exemption from Appendix J is authorized by law,

will not present an undue risk to the public health and safety, and is

consistent with the common defense and security, and (2) application of

the regulation. in this particular circumstance is not necessary to achieve

the underlying purpose of the rule. The Commission concludes the testing

and replacement program for the containment penetration bellows assemblies

is an ·acceptable alternative to the Appendix J Type D testing requirement.

Accordingly, the Conuni ss ion hereby grants the Exemption from Appendix J ..

'!· • • -9-

Pursuant to 10 CFR 51.32, the Commission has determined that the

granting of this Exemption will have no s_ignificant impact on the quality

·of the human environment (57 FR 4651).

This Exemption is effective upon issuance.

Dated at Rockville, Maryland this 6th day of February 1992 ·

FOR THE NUCLEAR REGULATORY COMMISSION

·;Jma·a~ Bruce A. Boger, Director Division of Reactor Projects III/IV/V Office of Nuclear Reactor Regulation

• • ENCLOSURE B

BASIS AND JUSTIFICATION PROPOSED REVISION TO EXEMPTION FROM 10 CFR 50 APPENDIX J

DRESDEN STATION AND QUAD CITIES STATION

1.0 . BACKGROUND

NRC staff approved an exemption to 1 OCFR 50, Appendix J Type B Local Leak Rate Test (LLRT) requirements for Dresden Units 2 & 3 and Quad Cities Units 1 & 2 via letter from B.A. Boger to T.J. Kovach dated February 2, 1992 (see Enclosure A). The exemption allows ComEd to apply special testing techniques in lieu of performing a test which meets Type B requirements for certain containment penetrations equipped with two-ply bellows which, at the time, .were unable to be tested in strict conformance to the Appendix J criteria. The special testing techniques, as specified in the exemption, included a sequence of air and helium based LLRTs for each affected penetration .and performance of an Integrated Leak Rate Test (ILRT) per Appendix J Type A requirements upon completion of the bellows testing during each refuel outage. For clarity throughout the remainder of this submittal, those portions of the special testing techniques which involve air and heliuin LLRT applications will be referred to as the "special testing program" and those portions which involve the ILRT will be referred to as the ."Type A test".

2.0 ACTIONS SUBSEQUENT TO ORIGINAL EXEMPTION

Since the approval of the original exemption, ComEd has applied the special testing requirements to two-ply bellows and performed the Type A tests during seven refuel outages. The results of these applications are summarized in Table 1. At the time that ComEd proposed the exemption, a Type A test was required every outage in accordance with License (Technical Specification 3/4. 7 .A.2 ) and Appendix J criteria for determination of ILRT test frequency. Based upon completion of two consecutive successful Type A tests since the original exemption was approved, a Type A test is no longer required each outage for Quad Cities Units 1 & 2.

Since the· original bellows leak was found at Quad Cities Station Unit 1 (containment penetration X-25) during the Cycle 11 refuel outage (November 1990), ComEd has not discovered any large bellows leaks at either Dresden Station or Quad Cities Station. Due to the discovery of very small leaks using the special testing program described in Enclosure A, the bellows listed in Table 2 have since been modified. These bellows have been removed from the list describe·d in the original exemption (Enclosure A) and are now on standard Type B testing schedules.

In addition to performing the special testing program and Type A tests (as described in Enclosure A) for two-ply containment penetration bellows during refuel outages at Dresden Station and Quad Cities Station, ComEd initiated research efforts to develop new technology which would allow ComEd to perform valid Type B tests on the existing ~~ll_ov~1 s as_~~mbly.

-1-

·' • • ~omEd investigated various testing alternatives. From this investigation, ComEd has identified several viable methods for executing a valid type B test on the subject bellows. One method involves the addition of a bellows test enclosure equipped with leaktight seals. During a local leak rate test, the space between the new cover and the outside of the bellows assembly is pressurized, thus allowing a valid Type B test on the bellows. Covers such as this have been designed and successfully tested. A second method involves installation of a rubber boot inside the drywell to form a seal between the drywell atmosphere and the bellows. During a local leak rate test, the space between the boot and the bellows' inner-ply is pressurized, thus allowing a valid type B test. This configuration has recently been successfully tested at a vendor facility. Finally, in some cases it may be feasible to weld a cover plate inside the drywell to provide a seal between the process pipe and the drywell atmosphere. This approach was used to determine the actual leakage of the X-25 penetration during the development of the original exemption (Enclosure A). As with the rubber boot application, the space between the plate and the bellows' inner-ply is pressurized, thus allowing a valid type B test.

In addition to the application of the above techniques, ComEd has, in some cases, implemented the complete replacement of the existing two-ply bellows assemblies with a completely new, redesigned, and testable two-ply bellows (see Table 2). ComEd will utilize any one, or a combination, of the above options in order to resolve all two-ply bellows assemblies which exhibit leaks through both plies, as identified by the special testing program. The choice of which option(s) to use will depend upon a number of factors including, but not limited to, the condition of the bellows, the penetration geometry, and the availability of materials. ComEd will also continue to research other alternatives in order to more efficiently and effectively meet Appendix J, and license requirements for the two-ply bellows.

3.0 REGULATORY BASIS FOR THE PROPOSED REVISION TO THE EXEMPTION

Pursuant to 1 Q CFR 50.12(a), Commonwealth Edison Company is requesting a revision to an approved exemption from the requirements of 10 CFR 50 Appendix J, Section lll.D.2(a) for two-ply containment penetration bellows at Dresden and Quad Cities Stations. This requirement states:

"Type B tests, except tests for airlocks, shall be performed during reactor shutdown for refueling, or other convenient intervals, but in no case at intervals greater than 2 years."

.The requested revision to the approved exemption from Appendix J requirements is appropriate under the current circumstances based upon the criteria established in 10 CFR 50.12.a.2(iii). This regulation requires the presence of special circumstances in order for the Commission to consider granting an exemption. The regulation states that special circumstances are present whenever compliance would result in undue hardship or costs which are significantly in excess of those contemplated when the regulation was adopted, or significantly in excess of those incurred by others in similar situations.

Based upon historical bellows testing-experience, Com Ed has demonstrated· that the· · performance of a Type A test every outage (as required by the originally approved

-2-

. i • ! ,J • • exemption) is not necessary in order to meet the intent of Appendix J for non­leaking two-ply bellows. ComEd has developed a set of alternative options for resolving bellows leaks which preclude the need for a Type A test to be performed each refuel outage. Consequently, ComEd now believes the costs to perform the Type A test every outage are significantly in excess of those contemplated when the regulation was adopted, and significantly in excess of those incurred by other utilities in similar situations.

Given the current schedule for refuel outages at Dresden and Quad Cities Stations and accounting for the results of recent type A tests (which controls the frequency of type A testing as required by Appendix J), the immediate benefit from approval of this revision to the exemption will be elimination of the requirement to perform a type A test during the next refuel outage for Quad Cities Units 1 & 2. The estimated cost of a Type A test, as described in NUREG-1493, "Performance-Based Containment Leak-Test Program", Draft Revision 2, dated March 31, 1994, is $1 .89 million. The estimated cost savings which will be incurred by ComEd by applying one of the alternatives versus performing an unnecessary Type A test are represented in Table 4. ComEd expects to achieve additional savings through the elimination of additional Type A tests as the recent trend in positive Type A test results is carried forward at Quad Cities Units 1 & 2 and Dresden Units 2 & 3.

This request for a revision to the approved exemption will apply to the non-Type B testable, two-ply containment penetration bellows listed in Table 3 for Dresden Station and Quad Cities Station. This proposed revision eliminates a requirement that has now been determined to provide little or no safety benefit, yet requires significant resources to perform. The proposed revision consists of a modification of the requirement to perform a Type A test every refuel outage. Specifically, ComEd proposes· inclusion of the following alternatives to the current requirement to perform · a Type A test every refuel outage:

"Upon completion of the two-ply bellows special testing program, the following act;ons shall be taken to address any two-ply bellows which have been identified as leaking through both plies:

A) . All bellows which leak through both plies shall be tested in accordance with Type B requirements to ensure license limits are met prior to return to service.

-OR-

BJ A Type A ILRT test shall be performed to verify primary containment integrity. All two~ply bellows assemblies which demonstrate leakage through both plies shall be replaced or subjected to a valid type B test to demonstrate license limits are met prior to return to service from the subsequent refuel outage, unless ComEd provides justification for continued operation greater than one operating cycle. "

-3-

j '' "' • • This proposed revision to the previously approved exemption will ensure that the leakage from leaking two-ply bellows is detected and adequately quantified, either individually or in aggregate, to ensure that license limits are met (containment penetration leak rate acceptance criteria are specified in Technical Specification 3/4.7.A.2 for Dresden Units 2 & 3 and Quad Cities Units 1 & 2).

The proposed revision to the exemption and associated special testing procedure is applicable for each non-testable two-ply primary containment bellows assembly. Upon replacement, modification or adaptation of the additional methods which allow the existing configuration to be tested in accordance with type B requirements , that bellows will no longer be included in this exemption, and will be required to be tested in accordance with the Type B test program.

4.0 JUSTIFICATION FOR THE PROPOSED REVISION TO THE APPROVED EXEMPTION

ComEd's request to revise the previously approved exemption from Appendix J Type B test requirements for two-ply bellows at Dresden Station and Quad Cities Station (Enclosure A) is based upon the ability to identify leaking bellows, and the assurance that the leakage from a leaking bellows will be quantified, either individually or in aggregate, to ensure compliance to license limits. The ability to detect and quantify bellows leakage is based upon the following insights gained during testing of two-ply bellows at Dresden Station and Quad Cities Station during recent refuel outages (D2R13, D3R12, D3R13, Q1R12, Q1R13, Q2R11, and Q2R12):

• There is minimal probability for the occurrence of a large leak in a two-ply bellows;

• The special testing program is effective for identifying small leaks in two-ply bellows;

• The Type A test is ineffective for identifying small leaks in two-ply bellows; and

• More cost effective alternatives have been developed for quantifying the leakage.

4.1 THERE IS MINIMAL PROBABILITY FOR THE OCCURRENCE OF A LARGE LEAK IN A TWO-PLY BELLOWS

The data in Table 1 demonstrate that, over the last seven refuel outages, there has been a relatively small percentage of two-ply bellows which actually leak through both plies. Furthermore, each of the identified leaks has been significantly smaller than the X-25 bellows leak which was the subject of the current exemption.

ComEd has evaluated the relative risk for the occurrence of a large bellows leak. - -·This evaluation- considered'the following:~

-4-

· .. "' .... •

• The cause of the original bellows leak (containment penetration X-25) at Quad Cities Station;

• The configuration of the X-25 containment penetration as it relates to the configuration of other bellows penetrations;

• Historical bellows testing data from Dresden Station and Quad Cities Station;

• Modification of the X-25 (and similar) bellows; and

• Replacement I modification of all bellows with leaks through both plies.

Based upon this evaluation, it is ComEd's position that the relative risk for occurrence of a large leak in a two-ply bellows is minimal.

The large leak found in the X-25 bellows is believed to have occurred as a result of maintenance work associated with the replacement of a valve which was directly in-line with the bellows. The resultant torsional and/or transnational

· forces may have caused an accelerated growth of existing Transgranular Stress Corrosion Cracking (TGSCC) in the bellows.

Contributing to the stresses associated with maintenance activity is the configuration of the X-25 bellows. This bellows configuration is actually ·a process pipe, as opposed to a connection of the process pipe to the containment liner. There are only two pathways like this in each unit. All of the other bellows in each unit connect the process pipe to the containment liner. This configuration is less susceptible to stresses which can accelerate existing cracking mechanisms. In addition, subsequent to the original exemption, all bellows with the same configuration as X-25 have been modified to allow for performance of a valid Type B test.

Since approval of the current exemption, ComEd has either replaced, modified, or scheduled the replacement/modification of all two-ply bellows which have been identified with leaks through both plies. These actions, in conjunction with the continued application of the special testing program, further reduce the risk of any large bellows leakage developing between regularly scheduled Type A tests (i.e. Type A tests performed per Technical Specification and App~ndix J frequency requirements).

4.2 THE SPECIAL TESTING PROGRAM IS EFFECTIVE FOR IDENTIFYING SMALL LEAKS IN TWO-PLY BELLOWS

Based upon testing experience gained since the approval of the exemption (February 1992), ComEd has determined that the special testing program described in the exemption (Enclosure Al is a very sensitive method of

. "detecting.bellows leaks. -T=his testing experience-is described,in Table-1. _Had. these procedures been in place prior to the exemption, ComEd would have

-5-

identified the Quad Cities X-25 bellows leak, prior to the Type A test. During the seven refueling outages since this special testing procedure was implemented, ComEd has identified 16 bellows penetrations with leaks through both plies. In each case, the leaks have been significantly smaller than the X-25 bellows leak.

Since the exact magnitude of leakage from each of these 16 bellows could not be determined, bellows with even extremely small leaks have been replaced. In addition, the Type A tests performed every outage since approval of the current exemption have never found a bellows leak which was undetected by the special testing program. Therefore, this procedure has the ability to detect leaks much smaller than would be detected by the Type A test.

Based upon this historical test experience and test data, ComEd has concluded that the special testing program is a highly effective method for identifying small leaks in two-ply bellows assemblies.

4.3 THE TYPE A TEST IS INEFFECTIVE FOR IDENTIFYING SMALL LEAKS IN TWO­PL Y BELLOWS

The Type A test is designed to determine whether the total leakage rate of air from containment is in excess of an allowed leakage limit (La). Based upon the measured leak rates of less than 0.01 La (as determined by the special testing program), ComEd has determined that various bellows at Dresden Station and Quad Cities Station were defective, and subsequently replaced these bellows. The Type A test is not sensitive enough to accurately and repeatably detect leaks that are this small.

The Type A test is validated by demonstrating agreement within 0.25 La, with a known, imposed leak. This uncertainty is 25 times the values that ComEd was able to measure with the special testing program. In addition, the Type A test does not differentiate between leakage that is coming from a bellows or leakage that is coming from the containment isolation valves and structure which are at the same time being challenged.

Therefore, based upon the relative sensitivity of the Type A test compared to the special test procedure, and historical test results, ComEd believes that the Type A test is unable to detect typical leaks from a two-ply bellows assembly.

4.4 MORE COST EFFECTIVE ALTERNATIVE METHODS HAVE BEEN DEVELOPED FOR QUANTIFYING LEAKAGE

ComEd recognizes the value of the Type A test for ensuring compliance to license limits in cases where a known leaking bellows is to be left in an as­tound condition until the next refuel outage. However in the revised exemption request, ComEd has proposed a set of alternative approaches (including, but not limited to, continued reliance on the Type A test) which can be applied to ensure that leakage from a bellows is-quantified. -Application of any one of-c these alternatives will ensure that, as a minimum, the intent of Appendix J type

-6-

''

B testing requirements is met. A comparison of the costs associated with each of the various options is included in Table 4.

4.5 REPLACEMENT AND MODIFICATION COSTS

As previously discussed, ComEd has identified the inability to quantify leakage of two-ply primary containment penetration bellows at Dresden Station and Quad Cities Station in accordance with 10 CFR 50 Appendix J Type B test requirements utilizing standard testing methods and currently installed hardware. In order to achieve full compliance with these requirements, both Dresden and Quad Cities Stations would be required to replace or modify all two-ply primary containment penetration bellows. As an alternative, ComEd has developed a set of special methods for testing these bellows. These methods ensure that the intent· of Appendix J type B testing requirements is met.

The cost of replacement or modification of two-ply bellows assemblies is projected to range between approximately $1.4 million and $11.5 million per unit for both Dresden Stati'on and Quad Cities Station. A detailed cost analysis

· for the replacement or modification of two-ply bellows is described in Table 5. The total cost of bellows replacement/modification is projected to range between approximately $6. 7 and $33.35 million.

ComEd believes that this expenditure of resources is not justified for the additional level of safety that it would provide. This is based upon ComEd's proven special testing program (see Enclosure A), which ensures that all leaking two-ply bellows are identified; and the proposed revision to the Type A test requirement (see Section 2.0 of this Enclosure), which ensures that the leakage from any leaking two-ply bellows is quantified.

5.0 SUMMARY

ComEd requests NRC staff approval of a revision to a previously approved exemption to the requirements of 1 OCFR 50, Appendix J. The exemption is still required to preclude the application of significant resources to the upgrade of containment penetrations for the sole purpose of allowing testing methods to strictly conform to Appendix J requirements. The revision to the exemption allows for application of more cost effective alternative approaches for meeting the underlying intent of the rule. These alternatives have been developed based upon and are justified by the insights gained from application of the previously approved exemption over the course of seven refuel outages at Dresden and Quad Cities Stations.

-7-

OUTAGE

D2R13

D3R12

D3R13

01R12

01R13

02R11

02R12

TABLE 1

HISTORICAL TWO-PLY BELLOWS TEST RESULTS SPECIAL TESTING PROGRAM

#BELLOWS TESTED

24

23

17

16

15

17

17

AND TYPE A TEST

#BELLOWS LEAKS IDENTIFIED VIA SPECIAL TESTING PROGRAM

4·

5

0

2

2

1

2

-8-

#BELLOWS LEAKS IDENTIFIED VIA TYPE A TEST

0

0

0

0

0

0

0

TABLE 2

REPLACED OR MODIFIED TWO-PLY BELLOWS ASSEMBLIES

DRESDEN STATION AND QUAD CITIES STATION

DRESDEN STATION

UNIT 2

X-113 RWCU X-125 Drywell Vent X-149A A Core Spray X-1498 8 Core Spray

UNIT 3

X-1078 8 Feedwater X-111 A A Shutdown Cooling

X-125 Drywell Vent X-138 Standby Liquid Control

X-1498 8 Core Spray

QUAD CITIES STATION

UNIT 1

Replaced with a Type 8 testable two-ply bellows Replaced with a Type 8 testable two-ply bellows Replaced with a Type 8 testable two-ply bellows Replaced with a Type 8 testable two-ply bellows

Replaced with a Type 8 testable two-ply bellows Replaced with a Type 8 testable single ply bellows and cover Replaced with a Type 8 testable two-ply bellows Replaced with a Type 8 testable single-ply bellows and

· cover Replaced with a Type 8 testable single-ply bellows and cover

X-12 Shut Down Cooling Supply Replaced with a Type 8 testable two-ply bellows X-23 R8CCW Supply Replaced with a Type 8 testable single-ply bellows· and

X-25 Drywell Vent X-26 Drywell Return

UNIT 2

X-14 Reactor Water Cleanup

X-25 Drywell Vent X-26 Drywell Return

cover Replaced and made testable Made testable

Replaced with a Type 8 testable single-ply bellows and cover Made testable Made testable

-9-

•ti ,.. • W:

Unit 2 X-105A X-1058 X-105C X-105D X-106 X-107A X-1078 X-108A X-1098 X-111 A X-111 8 X-116A X-1168 X-123 X-124 X-126 X-115A X-130 X-144 X-147

Unit 3 X-1058 X-105C X-105D X-106 X-107A X-108A X-109A X-111 8 X-113 X-116A X-1168 X-123 X-124 X-126 X-128 X-147

· X-149A

TABLE 3

DRESDEN STATION

Containment Penetrations Subject

To Request for Revision to Existing Exemption

A Main Steam Line 8 Main Steam Line C Main Steam Line D Main Steam Line Main Steam Line Drain A Feedwater Line 8 Feedwater Line Isolation Condenser Steam Line Isolation Condenser Return A Shutdown Cooling 8 Shutdown Cooling A LPCI Injection 8 LPCI Injection R8CCW Supply R8CCW Return Drywell Return HPCI Steam Supply Standby Liquid Control Inlet CRD Return Reactor Head Spray

8 Main Steam Line C Main Steam Line D Main Steam Line Main Steam Line Drain A Feedwater· Line Isolation Condenser Steam Line Isolation Condenser Return 8 Shutdown Cooling Reactor Water Cleanup A LPCI Injection 8 LPCI Injection R8CCW Supply R8CCW Return Drywell Return HPCI Steam Supply Reactor Head Spray A Core Spray

-10-

Unit 1

X-7A X-78 X-7C X-7D X-8 X-9A X-98 X-10 X-11 X-13A X-138 X-14 X-24 X-47

Unit 2

X-7A X-78 X-7C X-7D X-8 X~9A

X-98 X-10 X-11 X-12 X-13A X-138 X-16A X-23 X-24 X-47

• TABLE 3 (continued)

QUAD CITIES STATION

Containment Penetrations Subject

·To Request for Revision to Existing Exemption

A Main Steam Line 8 Main Steam Line C Main Steam Line D Main Steam Line Main Steam Line Drain A Feedwater Line 8 Feedwater Line RCIC Steam Supply HPCI Steam Supply A LPCI Injection 8 LPCI Injection Reactor Water Cleanup Suction R8CCW Return Standby Liquid Control

A Main Steam Line 8 Main Steam Line C Main Steam Line D Main Steam Line Main Steam Line Drain A Feedwater Line 8 Feedwater Line RCIC Steam Supply HPCI Steam Supply Shut Down Cooling Supply A LPCI Injection 8 LPCI Injection A Core Spray R8CCW Supply R8CCW Return Standby Liquid Control

\'

-11-

'· i.: • •. i..

*

**

***

• TABLE 4

COMPARISON OF ESTIMATED COSTS

FOR APPLICATION OF ALTERNATIVE METHODS

METHOD

ADD BELLOWS TEST ENCLOSURE

TEST WITH BOOT (where possible)

TEST WITH PLATE (where possible)

TYPE A TEST

REPLACE WITH TYPE B TEST ABLE

$ PER IDENTIFIED LEAKING BELLOWS *

100k

50k

50k

378k - 1.89M**

380k - 577k***

COSTS ARE APPROXIMATE AND DO NOT INCLUDE COSTS TO APPLY SPECIAL TESTING PROGRAM WHICH IS REQUIRED AS A PREREQUISITE FOR EACH ALTERNATIVE.

BASED UPON $1,890,000.00 PER TEST DIVIDED BY THE NUMBER OF HISTORICAL LEAKING BELLOWS ( 1-5 PER OUTAGE - SEE TABLE 1).

COSTS VARY DEPENDENT UPON FINAL DESIGN CONFIGURATION SELECTED (SEE TABLE 5).

-12-

.. -~ . \:.~~, ._.IJ.. • •

TABLE 5

PROJECTED COSTS - BELLOWS REPLACEMENT/MODIFICATION

Based upon the following circumstances, ComEd has revised the original cost estimate (as described in Enclosure A) for replacement and/or modification of the applicable two-ply bellows.

• ComEd underestimated the replacement cost of a two-ply bellows in Reference (a). The stated cost was $397,000 per bellows. The new cost is now known to be

\ . closer to $577,000 per bellows.

• Since 1991, ComEd has developed new methods for the replacement of two-ply bellows which are more economical than the 1991 replacement methods.

• The number of bellows which need to be replaced has decreased since 1991. Some bellows have been replaced and others have been made testable. These bellows are now outside of the scope of this exemption.

A. Projected Cost of Two-Ply Bellows Replaced with the Same Type (per penetration)

1. Engineering 2. Material and Fabrication 3. Onsite Mechanical Construction Support 4. Bel_lows Supplier Installation

Total Cost (per penetration)

$30,000 $70,000 $300,000 $177,000

$577,000

B. Projected Cost of Two-Ply Bellows Replacement with a Single-Ply Bellows and a Testable Cover

$380,000

C. Projected Cost of the Addition of a Bellows Test Enclosure $100,000

D. Total Cost per Unit

The total cost will depend upon whether method A, B or C is used. At this time, ComEd believes that a combination of all three methods will need to be employed in order to successfully address all of the remaining untestable bellows. For this reason a range of possible costs are given for each unit.

UNIT

Dresden 2 Dresden 3 Quad Cities 1 Quad Cities 2

# OF PENETRATION

20. 17 14 16

COST PER UNIT ($1000's) $2,000 to $11,540 $1, 700 to $9,809 $1,400 to $8,078 $1,600 to$ 9,232

- Total Replacement Cost·($1000's)· , - $ 6,700-to-$38,659·

-13-