ML20037B190

From kanterella
Jump to navigation Jump to search
Safety Evaluation Supporting Amend 23 to License DPR-2
ML20037B190
Person / Time
Site: Dresden 
Issue date: 01/06/1978
From:
Office of Nuclear Reactor Regulation
To:
Shared Package
ML20037B187 List:
References
NUDOCS 8009080578
Download: ML20037B190 (20)


Text

A SAFETY EVALUATION IN SUPPORT OF EXTENDING THE ECCS EXEMPTION AND THE DATE FOR COMPLIANCE

(]

WITH THE COMMISSION'S ORDER RELATED TO REACTOR PROTECTIVE SYSTEM MODIFICATIONS COMMONWEALTH EDISON COMPANY DRESDEN UNIT H0. I DOCKET NO. 50-10

1.0 BACKGROUND

The Dresden Nuclear Power Station Unit No. I was designed and constructed in the late 1950's and was issued Facility Operatino License No. DPR-2 on October 14, 1960. Dresden Unit No.1 is the first nuclear power plant licensed for commercial operation.

Subsequently, the Nuclear Regulatory Commission (formerly the Atomic Energy Commission) has ' developed and adopted new regulatory requirements and has provided guidance to the nuclear industry in the form of codes and standards which identify acceptable methods to comply _ with the Commission's regulations.

The Commission has identified significant areas at Dresden No.1 which recuire modifications to the facility as it was originally constructed.

These areas include the installation of a High Pressure Coolant Injection System, the nodification of the Reactor Protection System, and the installation of an automatic fire protection system.

These modifications have been determined neces-sary by the Commission to provide substantial additional protection -

for the health and safety of the public and to meet specific recuire-ments of Commission regulations promulcated subsecuent to the issuance of the Dresden Unit No.1 Operating license in 1960.

2.0 INTRODUCTION

2.1 ECCS Exemption Extension The Commission directed installation of Emergency Core Coolina Systems (ECCS's) for all power plants in July 1971 under its policy statement regarding Interim Acceptance Criteria.

The acceptance criteria for these systems were finalized with the issuance of 10 CFR 50.46 and Appendix K.

To date, all operatina facilities have completed installation of an ECC5 exceot Dresden Unit No. 1.

In this case, as in several others, the Commission granted a series of variances and exemptions which allowed continued coeration while backfitting of an ECCS was in progress.

8009080 b h

J The last such exemption relating to Dresden Unit No.1 was aranted

~

Jto Commonwealth. Edison on August 21, 1975, to allow operation until December 31, 1977.

By letter dated July 8,1977, the Commonwealth Edison Company (CECO) requested that the expiration date of the Commission's August 21, 1975 Exemption from certain requirements of 10 CFR F0.46. be extended to. December 31, 1978.

2.2 June 23,1976 Order Modifications By the same July 8,1977 letter, Comnonwealth Edison also reauested that the date for compliance with the Commission's June 23, 1976 Order for Modification of License be similarly extended to December 31, 1978.

This order required that Commonwealth Edison submit to the Comnission, by September 1976:

1.

Aiditional information relating to the efficacy of the Dresden Unit No.1 containment spray systen.

2.

Details of proposed modifications that are necessary to inake the design of plant protection systems conform with the reouirements of Sections 4.2, " Single Failure Criterion,"

and 4.6, " Channel Independence" of IEEE Std. 279-1968.

3.

Detailed analyses of all portions of plant protection systems, including the proposed modifications, which show that the modified systems meet the single failure criterion.

4.

Results of analyses and/or tests which demonstrate that all instrumentation and electrical equipnent essential to safety can function in the environment during and following an accident.

If satisfactory results are not obtained, describe the modifications necessary to assure that all instrumentation l

and electric equipment essential to safety can function in the environment.

t Commonwealth Edison has complied with itens 1, 2 anc a by their submittals of September 30, 1976, and November 15, 1976.

Accordinoly, these requirements are beino deleted fron License DPR-2.

This deletion is an administrative action and does not inolve a signifi-cant hazards consideration.

The Order further required.that Conmonwealth Edison complete their proposed reactor protection system modifications, fire protection system installation and environmental qualification of electrical equipment and instrumentation important to safety, by December 31, 1977.

Commonwealth Edison has been unable to complete these nodifications and has requested an extension to permit time to complete these requirements of the Order.

' 3.0 EVALUATION We have reviewed Commonwealth Edison's July 8,1977 request IReference 1) for extension to determine whether the public interest warranted the extension, and to assure that the health and safety of the public would not be adversely impacted.

In our review, we examined:

1.

The loss of coolant accident analysis and the performance caoabili-ties of the existing plant systens upon which reliance nust be placed to assure that the performance requirements of 10 CFR 50.46 will be met; 2.

Commonwealth Edison's progress in comolying with the Conmission's August 21, 1975 Menorandum and Order of Exemption and June 23, 1976 Order for Modification of License; 3.

Modifications to _the facility proposed by Commonwealth Edison to provide added assurance that operation beyond December 31, 1977 does not adversely impact the health and safety of the public; and 4.

The capability of the facility to withstand random single failures in the reactor protection system without completion of all of the modifications required by the June 23, 1976 Order, for the additional period of operation requested by Commonwealth Edison's July 8,1977 exemption request.

3.1 Loss of Coolant Analysis We have evaluated the loss-of-coolant accident (LOCA) analyses submitted by Commonwealth Edison Company for Dresden Unit Ho. I with the cresent core loading (Re,"crences 2 and 3).

We also reviewed the safety equipment assumed to function in the LOCA analyses.

1 1

1 3.1.1 ECCS-LOCA Model and Analysis Results The analyses were performed utilizing a slightly modified version of the General Electric Company Non-Jet-Pump Boiling Water Reactor Evaluation Model (the model).

The model generally has been reviewed and approved as meeting all requirenents of Appendix K to 10 CFR 50.46.

The modifications that were made to the codel to accommo-date certain unique features of Dresden 1 (principally the dual-cycle primary system and safety system credit for operation of the feedwater system and emergency condenser) have not been given final staff approval.

However, after a preliminary review of the modified model as applied to Dresden 1, we conclude that the model is reasonably conservative ~and can be used for the ourpose of evalu-ating this exemption request and determining whether or not the plant meets the performance requirements of 10 CFR 50.46.

  • ll analyses using this model were performed assuming 102% of licensed core power and Maximum Average Planar Linear Heat Generation Rate (i.e.,

MAPLHGR, a measure of local power) allowed by current Technical Specific tions unless otherwise stated.

We conclude that if credit is assumed for operation of certain equipment as specified below, Dresden 1 meets all performance recuire-ments of 10 CFR 50.46 (related to peak cladding tenperature, local oxidation, core-wide metal-water reaction, coolable aeometry, and long term cooling) when the reactor is operated according to current LOCA-related Technical Specifications, e.g., Maximum Average Planar Linear Feat Generation Rate (MAPLHSR) and power-to-void ratios.

3.1. 2 Review of Safety Eouipment Needed to Meet Performance Recuirements The staff's major review effort on the exemption extension request concentrated on determining whether or not the ecuionent assumed to operate in the LOCA analyses would in fact be operable following a LOCA. We reviewed pertinent information including Pioing and Instrument. Diagrams (P & ID's) of all vital systens, obtained updated P & ID's where necessary, reviewed CECO's rescanses to numerous staff questions regarding the equionent's cesien, reliability and operating history, and finally several staff members spent two days at the plant site conductiqc a first-hand

- 5 ~-

exanination of the vital ECCS safety equipment.

The conclusions and assessments contained within the following list of vital ecuipment are based'on all of the above review activities.

Tne eouipment that must be assumed operable folloveing a LOCA to meet the performance requirements of 10 CFR 50.46 is:

3.1.2.1 The Core Spray- (CS) System. This system takes water from the Fire System, and uses any two out of three 50*e capacity pumps to supply water to the ring spray header which distributes the water over the core to provide cooling. Rated flow for this systen can be achieved only after the core has been depressurized to below 140 psig due to low pressure desian of the systen.

The core spray system is the only system capable of providing short term cooling in the specific case of a LOCA having a break location which precludes the possibility of reflooding the core (e.g., bottom breaks).

It employs a single-active-failure-proof piping network with parallel active components so that it is vulnerable only to passive failure of non-parallel portions of piping. A considerable amount of such non-parallel piping exists between the CS pumps and the core spray header, approximately 200 ft.

This run of piping is restrained aaainst seisnically induced motion and has been subjected to periodic in-service in tions.

No defects have been detected in this pipino and its failure is not likely.

3.1.2.2 The Post Incident (PI) System.

This system consists of two pumps (with one CS pump on standby as backup if either PI pump should fail) and an active-single-failure-proof piping network to provih the plant's only long term core cooling capability for non-reflood-able breaks.

The PI system takes water from the containment sunp and pumps it through the PI heat exchangers (which are cooled by-water from the Fire System) into the single-active-failure-proof CS system described above, where it is distributed over the core.

Af ter cooling the core, the flow then goes out the break back to the containment sump, completing the closed loop cooling systen.

Like the CS system, the PI system is capable of cooling the core only after the core has been depressurized.

-C-3.1. 2. 3 Tne Fire System.

The CS and PI systems each require v.ater from the Fire Systen to perform their safety functions.

The ri re Systm nro-vices the necessary net-positive-suction head to the CS pumps (10 psia is required) and it provides the secondary cooling water to the PI heat exchangers.

The Fire System is a site-wide system inter-connected to Dresden 2/3, containing multiple sources 'of water (in-ciucing several screen wash pumps, two diesel fire r" s, and ser-vice water pumps) and several alternate piping routes which could be used to supply the necessary water.

The principai vulnerability of the system is possible damage to its buried pipino network, for example, due to an earthquake.

Even with the new above-around flexible hose connection from certain pumps (Dresder. I screen wash and diesel fire pumps) to the CS suction header, this vulnerability still is present since a sianificant amount of existing pioina cannot be isolated from the new flexible connection.

That is. flow nicht no cut oostulated breaks in such existing piping instea: ef to the CS pumps.

The licensee has estimated that the probability of a seismic event occurring at jhe Dresden site which could cause slicht damaqe to be low (about 5x10~ per year).

The staff judges the likelihood of nore severe earthquakes which could cause failures in essential safety system is to be even lower.

Consequently, we rely on the low proba-tility of an earthquake during the extension period in order to allow credit for availability of the rire System.

3.1.2.4 The Feedwater (FW) System. The FW systen functions #cilowinn a LOCA by injecting high pressure, relatively cold water into the core.

This causes the primary system to deDressuri20.

Such depressuriza-tion is essential tollowing certain size and locatic-breaks for the low pressure core spray system to function in time tc prevent 10 CFR 50.46 performance criteria from being exceeded.

The D-1 plant does not have the capability to relieve core cressure either autenatically or manually by openino relief valves.

Continuous, uninterrupted operation of two feedwater pumps and their associated condensate pumps is required to assure that performance criteria of 10 CFR 50.46 are met.

This requires the availability of offsite power, plus the assumption that no single failure will aisable the FW system.

These are p-inciple provisions of the exemtion whose extension is being reqJested.

The probability is cc sidered hiqh tnat offsite power will be continuously available.

-e have conducted a preliminary review of the FW system, its control systen, and the operating history of the FW system, and we conclude that there is reasonable assurance that the FW system will not exnerience a single failure following a LOCA, for the following reasons:

1.

Modifications have been nade to lock out the power supply to the only motor operated valve in-the system which cannot be byoassed and which could disable the FW system if it were to drive to the closed position; 2.

The FW system automatic controller only operates one valve, which will fail as-is with loss of power (nornally 80% open when the core is operating at full power);

3.

The level sensing instrumentation on the FW control system is a delta-pressure switch, which essentially neasures collapsed water level, i.e., the true amount of coolant mass in the core.

That is, the'FW-system controller is not likely to sense a swollen (high) level in the primary system if such a swollen level were to occur due to pressurization followinq a LOCA.

Therefore, tne FW automatic controller is not likely to cause the FW control valve to drive closed, with the_ attendant possi-bility of failure in the closed position, which would delay FW system availability until the valve could be re-opened or byoassed:

4.

Experience during past scrans due to turbine trio (similar to a small break LOCA as far as the core and the FW controller is concerned) have resulted in initial water level decreases due to void collapse caused by the pressure increase, this causes the FW valve to remain at its former setting or drive to a nore open position, further reducing the probability of FW systen unavailability due to a valve failinn in the closed position following a LOCA.

3.1.2.5 Emergency Condenser (EC).

The EC is a large tank of water with tWo submerged independent tubing bundles.

Opening ~one valve in either of the bundles allows primary system stcan to rise into one of the submerged tubing bundles from the primary system; the primary steam is condensed within the bundle, and falls by gravity back into the primary system.

The EC thus aids the FW System in deoressurization of the crimary system for smaller breaks.

so that the low pressure CS system can function.

Credit for only one of the two systems (1/2 EC) is assumed in showing compliance with performance criteria of 10 CFR 50.46, since single failure of either inlet valve could cause 1/2 of the EC to fail (The assumption of this single-failure is not precluded by the exemption whose extension is renuested).

3.1.3 LOCA Analyses Assuming Deoraded Ecutoment Availability Although operation of all eauipment specified above is necessary to provide compliance with all oerformance criteria of 10 CFR 50.46, the following conclusions can be made with respect to LOCA analysis results assuming certain degraded equipment perfomance.

Exceot for any deviations specifically noted below, all results are from analyses which use the same modified ECCS nodel and inputs discussed earl ier.

Since the CS, PI, and Fire systems can each tolerate any active single failure with no resulting perfomance degradation as discussed above, and since a single failure in the EC has already been assumed in the 27alysis results, these conclusions are confined to degraded perfomance of the FW system.

If the FW system is lost simultaneous with the LOCA dgto loss of offsite power, then Commonwealth Edison (CECO) states that one l

FW pump will autonatically restart within 30 to 50 seconds powered by the new on-site diesel generator. The conservatism of these tir'es will be confirmed by preoperational testing performed by CECO.

Assuming these times, then LOCA analyses, assuming one FW pump is regained in 30 seconds and 50 seconds, show peak-clad temperatures -

(PCT's) of 2317*F and 2414*F, respectively, for the limiting break size.* These PCT's are beyond the performance requirenents of 10 CFR 50.46, but considering the conservatisn believed to be in the modified ECCS model as applied to Dresden 1, they provide sone indi-cation that core melt would be avoided under these degraded condi-tions, should they occur.

CECO has also provided a calculation for the case where the FW systen is lost simultaneo FW system credit).p)with the LOCA and never regained (i.e., no For this extreme case, two additional modifications were made to the nodified ECCS model** and one chance was made to the input-The 0.15 to 2.0 ft.

break areas are liniting when FW delays are assumed:

larger breaks depressurize quickly.through the break without dependence upon the FW system; smaller breaks never uncover or remain uncovered for a shorter time before the FW system-and 1/2 of the EC depressurize the core and allow operation of the CS system.

+*l)

Decay heat eaual to 1.0 times "ANS" value, compared to 1.2 tines "ANS" value required by Appendix K.

The NRC staff is aware of data to support use of "1.0 times ANS"; this change in " Appendix X" require-ments may be approved at some future date.

2) Credit for 154% of the Emergency Condenser (EC).

That is, analysis did not assume single failure cf 1/2 of the EC, and did assume credit for measured test data that shows C nerfornance to be Sa# better than cesign values (which design values w"re usea in all other calcu-lations reported).

9_

assumptions to the model.***

These changes make the result less conservative than others reDorted, and somewhat discontinuous with respect to thcse other results.

However, the results are believed to be still conservative, and they do result in a predicted PCT below 2400*F, again providing some indication that core melt would not occur under these extremely-degraded conditions, should they occur.

3.1.4 Conclusions Related to LOCA Analyses As stated above, we conclude:

1.

That the ECCS-LOCA model used for the analyses provided is reasonably conservative and can be used for the purpose of determining whether or not the plant meets the performance requirenents of 10 CFR 50.46.

Employing this model, we have concluded that the perfornance requirements of 10 CFR 50.46 will be met following a costulated LOCA.

2.

That there is reasonable expectation that the equionent assumed; il to function in the ECCS-LOCA analyses, which show full compliance 12 with the performance requirements of 10 CFR 50.46, will in-fact function following a LOCA, ano that this expectation has been somewhat improved by eouipment chances made since granting of the original exemption.

3.

That even if some of the. equipment needed to assure full compliance with performance requirenents of 10 CFR 50.46 does not function, there is still reasnn to believe that core nelt would not occur.

Analyses were perforned assuning a derated maxinum allowable local power consistent with the plant's current Technical Specifi-cations, which are required to account for possible degradatirn of core spray flow due to steam redistribution effects.

Dependinc on how " flat" the core can be operated, the local power linics cause an effective total core power derate of 20; to 301 Credit for effect of this "derate" on local poser is assumed in tais analysis.

(As noted, all other analyses reported above aio not assume credit for this derate.)

Y

s._

3.2 Caoability of the Reactor Protection System to-Withstand Sinale Failures The' Reactor Protection System (RPS) at Dresden Unit tio.1 is basically a two channel system (Channels A and B).

Within each channel, two or more relay contacts for each plant paremeter monitored are connected in series (A1, A2, etc.).

If any one sensor controlling one of the relay contacts within a channel senses a trip condition then the channel trips. The output

~

of each channel is connected to one of two scram solenoid valves in series for each control rod.

Both of the scram solenoid valves for an individual control rod must be opened to scram the rod.

Therefore, the RPS channels A&B must trip before a scram is initiated.

This arrangement provides a high degree of protection from both a nuisance trip due to a spurious signal or a f ailure to scram all of the-control rods as a result of random equipment functional failures.

However, as noted above, due to the physical

. arrangement of the RPS components, a single gross physical failure of certain terminal blocks, instrument racks, sphere penetrations and wire ways, or a hot short around certain relay contacts, could prevent a scram from one or more of the monitored plant parameters.

In order to make an assessment of the extent to which the plant is vulnerable to the above listed failure modes, members of the staff visited the plant on October 3,1977, to meet with the licensee and physically inspect the RPS.

At'this meeting, the staff reviewed detailed schematic diagrams and analyses of the potential failure modes that had been performed by the licensee. Based on these discussions and on our review of information previously submitted by the licensee in its letter dated December 10, 1975, we were able to determine that the following design features represent the most significant vulnerability of the RPS to single failures:

1.

Channel A and B i * 't sensors in some cases are mounted on the same instrumen. rack and are not separated by space or barriers (e.g., all the high reactor pressure sensors are mcunted on rack ER-1).

2.

Channel A and B input sensors in some cases are wired on the same local term nal block and/or are in the same caole (e.g..

i all the sensors for the primary steam isolation valve closure trip and the low condenser vacuum trip are wired on a terminal block at the turbine stand).

i t

_ _ _ _ _ _ _ - 3.

Channel A and B input sensor wiring is at least partially routed in common cable trays and uses common containment sphere penetrat;ons (e.g., cables for all the Peactor Vessel Level, Primary Steam Drum Level, Reactor Pressure and RPS Scram Backup trips use the same penetration).

4 All the Channel A and B individual trio parameter and scram solenoid bus circuit wiring is terminated in the RPS Auxiliary Panel ( AP-5) located in the control room.

The scram solenoid buses are terminated at this panel approximately 41/2 inches from the Channel A and B power sources on the opposite side of a wiring grill.

A hot short of the scram solenoid bus to the A or B circuit power supply could prevent an automatic or manual scram of all the control rods.

5.

Both Channel A and B RPS scram solenoids must he deenergized to scram the reactor.

A hot short on either scran solenoid bus cable between the scram solenoid bus fuse cabinet in the containment sphere and Panel AP-5 in the control room could prevent a manual or automatic scram of all the control rods.

i The scram solenoid bus-cables share both penetrations and cable trays with sources of the 125Y a.c. power necessary for a hot short that could prevent a scram.

Itens 1 through 3 above are of lesser significance than items 4 and 5 because they are examples of potential-single failures that could prevent a scram from some, but not all, of the plant parameters monitored.

These circuits are deenergized to trip.

Therefore, failures are likely to be in the safe direction with the exception of hot shorts.

In the case of hot shorts, several shorts with no degraded voltage would be required simultaneously in different circuits to prevent a scram by all the sensed parameters.

There are ten plant parameters monitored (e.g., high containment sphere pressure, high neutron flux) with a minimum of two separate sensors per channel which could produce a channel trip when the reactor is in the run mode.

The probability of a single failure, during the extension period, that could disable a sufficient number of RPS trip input parameters to cause a significant fuel damage is accentably loa.

The types of single failures (i.e., hot shorts) that could result from the conditions described in items 4 and 5 are of much creater significance than the f ailures described in items 1 through 2, because these single hot shorts could prevent a scram of all the control rods from all of the plant parameters nonitored.

Even though the consequences of such failures are very significant, the

t ' probability of their occurring is acceptably-low as. dis;tssed in the following paragraph.

= -

n the case of item 4, the terminal blocks subject to not shorts are on opposite sides of a wiring grill and are covered by flat insulating shields which leave very litt'e bare conduc. tor surface exposed as a mechanism for hot shorts.

Any shorting conductor would have to be U-shaped, about 41/2 inches wide, and fit into a _ space about-1/2 inch square at each end, or have sufficient energy to physically damage the insulating shields, terninal blocks and steel wiring grill so as _ to brina the snorting circuits toaether without at the same time grounding-them.

Within the wirinc grill, it is possible that the cables subject to hot shorts are in direct contact with each other separated only by the insulaticn c* the conauctors.

Based on our physical exanination of these circuits at the plant, we have concluded that the only shorting cechanism for these cables that could be considered credible during the extension period would be a fire.

Since the wirino grill on Panel AP-5 is located in the control roon, the operators would be immediately aware of a fire and could initiate a manual scram before the fire burned long enough to create a hot short.

Even if the postulated fire were not controlled, or the reactor was not manually scrammed, the short would most probably eventually go to ground and cause a reactor scram because the wirinq orill and panel are nade of metal and are grounded.

In order to discuss the failure mechanism-in item 5 it is necessary to further-describe the reactor control rod scram solenoid circuits.

The control rods are divided into three aroups at the scram solenoid fuse cabinet located in the containment sphere.

Each control rod in a group is controlled by two scran solenoids in the group that must be deenergized to scram. Each of the three aroups is energized by.a circuit between panel AP-5 located in the control room and the scram solenoid fuse cabinet.

The three circuits for all three groups of the separate A or B scram solenoids are in a common cable jacket along with the common wire which is arounded.

Therefore, there are two cables; one cable for each set of scram solenoids or 3, that traverse the plant area from the containnent-sphere thrauch the schere penetrations and the cable tunnel to the control rocm in close proximity to sources of hot shorts which could prevent a scram of the reactor.

t The. failure mechanisms for hot shorts for these cables would have to be scmething that could penetrate both insulating,iackets around

- a scram solenoid bus circuit and the potential source of shorting

_ power, and at the same time bring these cables into contaqt with each other without also coming into contact with a grounding source such as the common wire carried within the scram solenoid bus cables.

- The only failure mechanism that we have been' able to identify that could be considered credible during the next extension period would be a fire. This concern will be resolved by the installation of the fixed automatic' fire suppression system proposed by CECO.

Du ing the additional exemption period of operation proposed.

- without an automatic fire suppression system, we have determined that the installation of a fire detection system that alerts the operator of a fire in the cable tunnel supplemented by manual fire suppression capability for the entire cable tunnel and a pro-cedural requirement to manually scram the reactor if a fire is detected, coupled with a modified scram solenoid circuit which assure shorts to ground, provide adequate assurance that the reactor can be safely shut down in the event of a fire -in the cable tunnel.

Until these additional measures are implemented to prevent potential hot shorts of the scram solenoid circuit caused by a fire in the cable tunnel, Commonwealth Edison will supp'ement the fire pro-tection in this area with an hourly fire watch.

- Regarding environmental qualification of electrical eauipment, the equipment that must function in the LOCA environment has been reviewed to assure that proper operation would be expected.

Both the licensee and staff are continuing to evaluate this area' to assure that other safety related equipment needed to function under -

accident induced environmental condition will not fail due to those environmental conditions.

On December 21, 1977, the staff requested CECO, as part of the Systematic Evaluation Program,'to examine the environmental qualification of all electrical equipment inside and outside of containment that must function for any Design Basis Event such as a LOCA. The staff expects to complete its review of this issue within the next three months; if during this time any deficiencies are uncovered appropriate action will be taken to ensure protection of the public health and safety.

L 3.3 Conclusion Relating to Interim Operation Prior to Comoletion of Fire Protection and Environmental Modifications As discussed above, CECO has implemented interim fire protection measures to assure that the redctor protection system is safeguarded against cable tunnel fires and has verified that all electrical com-ponents needed to actuate the emergency core cooling system are l

qualified to operate in the LOCA environment.

We have reviewed these measures taken by the licensee and conclude that their implerentation increases the margin of safety during the interim period until October 31, 1978, provides sufficient assurance that the health and safety of the public is not endangered by such operation, and is acceptable to the Regulatory staff.

3.4 Compliance With the Reouirement of the Existing ECCS Exemption and the June 23,1976 Order 3.4.1 ECCS Modificatio,ns Commonwealth Edison has completed the detailed desien of the Dresden high pressure coolant injection (HPCI) system and has submitted the design to the Commission by letter dated October 16, 1975.

In response to questions generated by the Commission's review of CECO's design report, CECO has supplemented its design report by additional information dated July 26, August 31, and October 20, 1976.

The Regulatory staff review of the design report and its supplements is continuing.

Commonwealth Edison currently has completed activities leading to the purchase of approximately 90% of the equipment and installation services for the HPCI system.

CECO projects a most probable cot.pletion date for the HPCI system project of February 1979.

Construction of the building designed to house the HPCI system is now scheduled for completion in March 1978.

This completion date has been delayed.

The design delay was principally associated with the need to enlarge the HPCI building over the preliminary design.

The construction delay was attributed to problems with discovery of slightly contaminated soil and the harsh 1976/1977 wi n ter while constructing the HPCI building.

The projected completion schedule above includes these delays.

The presently pacing item for installation of the HPCI system is equip-ment delivery for HPCI pump motors, service water and fuel oil transfer pumps and fire protection equipment.

This equipment is expected to be available in August,1978.

Installation and pre-operational testing

__-------m-----------------___.______._.___

_ is expected to be completed by December 31, 1978.

The licensee proposes to shutdown in the Fall of 1978, decontaminate the reactor coolant system in Hovember,1978, and connect the HPCI system and other related I

equipment.

3.4.2 Reactor Protection System and Fire Protection Order Commonwealth Edison has complied with the requirements contained in Paragraph 2.C(3)a of our June 23,1976 " Order for Modification of License" which dealt with CECO's submittal of design information related to their proposed modifications.

However, due to equipment procurement problems and the desire to complete these modifications concurrently with the HPCI installation outage af ter the primary system decontamination CECO has been unable to complete its modification by December 31, 1977, as specified in the June 23,1976 Order.

The completion of the modifications to the reactor protective system will involve modifications in high radiation areas.

Some of these areas are scheduled to be decontaminated in the Dresden chemical cleaning outage presently scheduled for November,1978.

CECO's modifications to be carried out in the lower radiation fields.

The Regulatory staff concurs with CECO's intent of reducing the exposures associated with these modifications by coroleting the modifications after the Dresden i decontamination has been completed.

In regard to the length of extension needed by the licensee to complete this project, the staff does not agree that a need exists to extend the exemption past the proposed decontamination shutdown in the Fall of 1978.

lhis is because an ECCS is not required for plants not operating at power.

Therefore, the extension is being granted only until October 31, 1978.

The NRC staff in an effort to assure that the modifications identi.fied in the August 1975 and June 1976 Orders are implemented as quickly as possible, has imposed a license condition that requires Ceco to submit a monthly report of the progress in com-plying with these Orders and other related activities such as decon-tamination which could impact CECO's schedules.

With regard to the modifications to the fire protection system required by the staff's June 23, 1976 Order, CECO has been unable to install a fixed automatic fire protection system as reauired.

. In the interin period of coeration until the fixed aut o.atic system is installed Connonwealth Edison has installed a 'iq detection system in the cable tunnel and has extended the coverage of a manually operated fire hose station to provinc complete coveraae of the cable tunnel.

3.5 Interin Facility Modifications and Procedural Chances 3.5.1 Chances Proposed by the Licensee In order to increase the reliability of the systems relied uocn to mitigate the consecuences of a Loss-of-Coolant Accident at Dresden Unit No, I, Connonwealth Edison has proposed eiaht interin changes to the facility and its operating procedures.

The eight chances are:

1.

Electrically disable feedwater valve MD 130 in the full open position to prevent a f ailure of this valve fron rendering the feedwater systen inonerable.

2.

Modify the abnornal operating procedures to instruct the enerator to check the feedwater control valve nosition durina a LOCL and to open bypass valve MO 50 if the Control valve has f ailed.

3.

Install and utilize a 345/138 KV transformer to connect the Dresden Unit No.1 138 KV systen with the Drescen Unit ':cs. ?

and 3 345 KV syster.

This enhances the reliabilitv of the off site power supply to Dresden Unit No.1.

4.

Remove the plastic face from each of the drun level switches.

This will cualify these switches to nerform their ECCS function in a LOCA environnent.

5.

Review the sphere hiah pressure and drun level sencors, switches, and cables which initiate the ECCS to determine that they can function in a LOCA environment and nake nodifications as necessary.

6.

Durina the Summer.19i7 refuelinn. 3 leak test of the ncimary ccolant 50undary was performed in accordance with t^e 3rrlicable editicn of the ASME Roiler and Dressure Vessel Code, Section MI.

This test provides further assurar.ce of the hiah decree cf inte:rity of the primary coolant Soundary ninion.

. 7.

To aug ent the hiqhly reliable off site pmice and the existinn on site ciesel cenerator, a backup diesel aenerater will bn available at the plant site by December 31, 1977.

This diesel generator will provide a redundant source of on site oower for long te-n cooling functions.

8.

The fire protection systen will be nodified such that lonc term cooling can be accomplished without relying on portions of its underground piping.

This modification involves providing fittines in the fire protection system piping near the core spray and post incident pumps and in the cribbouse.

These fittings will provide a neans to connect piping or hose betwaen the fire protectior, systen pumps and the eneroency core coolino systens in the event the permanent connection via the under-grour.d Dining system is uravailable.

We have reviewed these changes and conclude that they will provide additional assurance that tne Dresden Unit No.1 Emergency Core Cooling Systen will function as required during a LOCA and we are amending Facility Operating License DPR-2 to require that these modifications nust be completed to permit operation af ter December 31, 1977.

i..?

Additional ".hanges Identified Durina NRC Review In additicr. to the eight changes proposed by Comnonwealth Edison in its July 8,1977 extension request, our review has identified certain other nodifications which we require to provide a high degree of assurance tnat the ECCS will function as required.

These additional changes have been discussed with the Commonwealth Edison staff and they have agreed to inclement then.

The channes are:

1.

The Drescen Unit No. I station batteries shall be restrained to withstand seismic loads.

2.

Electrical cabinets containing controls for essential or bacton ECC5 e;uipment shall be restrained for seismic loads.

3.

The een "ackup diesel Generatnr shall be operable and nust be cacaS e of starting automatically and ooerating one feedsator pu o a c condensate pung within 50 seconds of a loss of of fsite poeer.

l

9

. 4.

Tne new flexible hose connecting the ?re'. den Uni t No. 1 fi re m,-

ed the suctica of the core spray syste, s W ! y connecter anc remain connected throughout the extension period.

5.

All missing protective covers and terninal board i n sul a ti nq shields shall be replaced within the reactor orotc.ction system and ECCS electrical cabinets and any extranecus itc~ stored in these cabinets shall be renoved.

6.

Five smoke detectors shall be installed in the cable tunnel to detect cable fires.

The reactor will be manually scranmed innediately uoan detection of a fire.

7.

The existing reactor protection systen surveillance procedure shall be nodified to require that the RPS functional canability te verified weekly at the scran solenoid fuse cabinet in the containment rather than fron the control roon.

8.

Rewire scram solenoid circuit with shielried cable and institute an hourly fire watch until rewired.

In our review of the facility modifications proposed by Connonwealth Edison, we have identified additional modifications and procedural chances which we consider necessary to provide additional assurance that the ECCS will function as required.

Me conclude that with the licensee's proposed nodification supplemented by these additional nodifications identified by the staff, the margins of safety at Dresden Station will be increased and that ooeration darino the extension until October 31, 1978, is acceptable.

'.0 FINDING RELATED TO GRANTING OF A FURTHER EXEMPTION We have reviewed Conmonwealth Edison's July 8.1977 suhmittal and the affidavits attached thereto which provided CECO's explanation for the delay in completion of the modifications, and have con-cluded t-hat CECO has shown good cause why the ECCS exaration expiration date and the date for compliance with tne June 23, 1976 Order should be extended to permit continued operation of the Drescen clant while continuina to modify the f acility.

l Cur cerclusinn is based unon our deternination that delavs encountered ay CECO have been due to itens v!hich t ey have been h

unable to credict such as procurement difficulty, the presence of the contaminated soil underlyina the construction site and the harsh winter weather that slowed their schedule.

l

_ lo -

The af fidavits filed by Ccnnonwealth Edison further dennnstrate that the unavailability of Dresden Unit No. i beyond Or conbor 31, 1077 will require the use of un to 83 million gallons of t' vel cil at a cost of approximately 26 nillion dollars.

The bulk of this cost would be passed directly to Connonwealth Edison's consumers.

Thus, the NRC staff concludes that the extension of time fro, December 31, 1977, to October 31. 1978, to permit the completion of required modifications at Dresden Unit No. I would be in the pu5lic interest given the absence of any undue risk in continued oneration.

5.0 ENVIRONMDlTAL CONSIDERATION We have deternined that the amendnent does not involve a change in effluent types or total enounts nor an increase in power level and will not result in any sinnificant environnental innact.

Havino nade this determination, we have further concluded that the anendment involves an action which is insignificant from the standpoint of environmental inpact and oursuant to 10 CFR s51.5(d)(4) that an environmental inpact statement or negative declaration and environ-mental impact appraisal need not be prepared in connection with the issuance of the amendment.

c.0 CONCLUSION We have concluded, based on the considerations discussed above, that:

(1) there is reascnable assurance that the health and safety of the public will not he endangered by coeration in the crocosed manner, (2) such activities will be conducted in conoliance with the Commission's regulations and the issuance of this amendment will not be inimical to the connon defense and security or to the health and safety of the public, and (3) that the exenstion is authorized by law and will not endanger life or procerty or the comnon defense and security and is otherwise in the public interest.

We have also concluded that in relation to the additional license conditions required by the staff and the deletion of nrevious license conditions that have been complied with, that:

(1) because the amendment does not involve a significant increase in the probability or consequences o' accidents previously considered and does not involve a significant decrease in a safety naroin, the amendment does not involve a significant hazaros consideration, (2) there is reasonable assurance that the health and safety of the public will not be endangered by oneration in the nroposed manner, and t3) such activities will he conducted in compliance with the Commission's reculations and the issuance of the anendnent will not be inimical to the conmon defense and security or to the health and safety of the cuhlic, ate: January 6, 1978 n

REFERENCES 1.

1;RC l'enorandum and Order in tne matter of Commonwealth Edison Company, August 21, 1975.

2.

"Eetter to B. Rusche, f RC, fron P. L. Bolger, Commonweal th Edison Ccnpany,

~

with attached Dresden Unit No.1 LOCA Analysis, July 31, 1975.

3. to letter to E. Case, ! RC, from R. L. folger, Commonwealth l'

Edison Company, The XN-1 Heatup Analysis iupplement to Dresden Unit No. l' LOCA Analysis, August 3,1977.

4.

Letters to D. K. Davis, NRC, froa M. S. Turback, Connonwealth Edison Conoany, dated flovember 11,15 and 28,1977, i

L

.