ML17139C000

Forwards NRC Review of June 1983 Susquehanna Environ Advocates Study,In Response to 830909 Request Re Util Handling of Equipment Malfunctions of Reactor Safety Sys. Review Did Not Reveal Adverse Trend Requiring NRC Attention

ML17139C000
Person / Time
Site:	Susquehanna
Issue date:	11/23/1983
From:	Dircks W NRC OFFICE OF THE EXECUTIVE DIRECTOR FOR OPERATIONS (EDO)
To:	Specter A SENATE
Shared Package
ML17139C001	List: ML17139C000
References
NUDOCS 8312090040
Download: ML17139C000 (23)

Text

A

>0~ 33 1983 The Honorable Arlen Specter United States Senate Washington, DC 20510 Dear Senator Spect'er:

We appreciate the opportunity to respond to your request of September 9,

1983 regarding a report dealing with the Susquehanna nuclear power plant located in Berwick, Pennsylvania.

The report was prepared by the Susquehanna Environ-mental Advocates (SEA).

In order to be responsive to your request several of the NRC staff, including the NRC resident inspectors located at the plant site, have conducted a tech-nical review of the identified concerns.

We have attempted to be responsive to as many concerns as we can.

It has taken us a little longer than expected to complete this effort and we apologize for any inconvenience this delay may have caused.

We recognize that reasonable persons can reach different conclusions based on the same information.

In this regard we have attempted to provide additional information that may not have been available to SEA at the time that they prepared their report.

Accordingly, in,preparing our response we have attempted to characterize the SEA concerns, listed these in the enclosure and provided clarifying information.

We share the broader SEA concern as to the need for diligent adherence by a licensee to the conditions imposed by the operating license.

In the matter of the Susquehanna facility we note that there have been several problems iden-tified during the startup testing phase of operation.

To some extent the SEA report focuses on such problems.

We acknowledge that it is not uncommon to encounter such problems during the initial startup and testing phase of a nuclear plant.

The cor rective actions taken to prevent recurrence are also critical and our inspection and licensing efforts are oriented toward assuring that the remedial measures are effective.

Our review has not identified any unusual adverse trend that warrants immediate NRC attention beyond that pro-vided by the existing inspection and licensing program.

~ )

I hope our response is helpfu'1 to you.

If we can be of any further assistance, please let us know.

Sincerely, (Signed) William J. Dirckf 8312090040 831123 PDR ADOCK 05000387 U

PDR William J. Dircks Executive Director for Operations RI:DPRP Greenman/cgl 11/18/83

Enclosure:

NRC Review, of the Susquehanna Environmental Advocates Study Issued in June 1983 RI:DPRP RI:DRA RI:RA OC HO Starostecki Allan Hurl ey Dircks ll 21 83 ll 21 83 ll 21 83 il g 6 ll-l -83

• a fl

/Il 'c)C 1'I t V I

(aJ

~ ) f'dail g

~ I '<<'c/

4 1 h

'"J I fCfffr) 1.1 't '<<)I) a a) Ia f,)CI~ /C/al I afa] I)

I,I I)') I. lfl~:,"f

h 1

'<<h I:faf

/" l 1 ) I'

.;II. ':,'rl.f

<<J

/II H' a) I a I I/)

/aa a

c".l) e f >>/<<h I'Q'k I;f

<P II 1 a i II a III

)(clC)

IIII I'

aII /

r I

L.

I "H(t

.1)l f K

K <</

al' a

~ 'l ~

~.

a ~

I <<

K a

tf) fifa l a I~

I)

I, '

) 1 a): I,<<

I

') a )

f'.'Il.f I

) 'll a

K ~

I l

I) 3 la'3 4 I) 1

~

H <<

, I

~ I I I I r

~

Hrf I

I

'ac Vc'I "a

a)fll

, 1 ':

a i;

". i

. Ij.

" <<j II c )I a

<<q

')

4 1<<

I ra P

H

~ ~

a a

')/I J

Distribution:

VStello RDeYoung HDenton LUnderwood Heltemes TMurley RStarostecki EDO 13561 SECY 83-2266

= PDR LPDR Docket Ho. 50-387 Region I Docket"Room OCA GRhoads EMcCabe RI PAO

4 II p

I W

.,)

~

~

1 (30 C

kd n

tl

Enclosure (1)

NRC Review of the Sus uehanna Environmental Advocates Stud Issued In June 1983 A.l SEA Concern Adequacy and timeliness of information available to the NRC and public to assess safe reactor operation is questionable.

The SEA report is limited by the information available to the public.

A.2 NRC View Information available to the NRC and the public includes Licensee Event Report (LER) data plus the publicly distributed NRC reports of inspection of the facility.

NRC inspection reports specifically address the staff's review of LER's.

In addition, occurrences and situations involving potentially significant events are required to be promptly reported to the NRC over the dedicated, emergency noti-fication telephone system (ENS) or "Hot Line" installed at operating nuclear power plants.

In the SEA review, it appears that selected LER's may have been evaluated without also considering the NRC staff's resolution, as documented in the monthly inspection reports.

While some relevant information may not have been available, due to routine administrative processing of some recent inspection, when the SEA study was issued, we believe the information now available could alter the conclusions reached by SEA.

B.l SEA Concern The utility app'ears to have and use wide discretion =in liberally interpreting the terms of its operating license and reporting re-quirements.

For example, no LER's were found for a fire reported to the news media or for spills of radioactive water.

Also, LER's from Susquehanna are almost all submitted on or near the 30 day time limit for such submission.

The public doesn't see LER's until after the NRC mails them to the public document room.

And, the public is almost totally reliant on PPSL for information on the safety of plant operation.

Closer scrutiny of LER's and rigorous followup action by the NRC are urgently needed.

B.2 NRC View The staff has found that licensee reporting of events at Susquehanna has been substantially in conformance with NRC requirements, but there is a continuing concern with the completeness of some reports.

These specific concerns were discussed with the licensee.

A NRC Region I letter dated August 18, 1983 forwarded to the licensee the NRC Systematic Assessment of Licensee Performance (SALP).

That

Enclosure 1

assessment specifically discussed the issue of LER completeness and the SALP report is in the public document room.

Future LER's will continue to be closely reviewed.

The Code of Federal Regulations and the licensee's operating license specify items which must be reported to the NRC.

The SEA report is based on one reporting mechanism (LER's) but does not appear to consider the Emergency Notification System reports (which did report the fire and a premature discharge of radioactive water to the Susquehanna River) or the publicly distributed NRC inspection reports (which include coverage of the radioactive spills identified by SEA).

Such events are discussed further in subsequent sections of this enclosure.

Although an LER may not be submitted until about 30 days after an

event, NRC review of significant issues is not delayed until them.

The two full-time resident inspectors at Susquehanna review licensee activities and records daily, and prioritize their inspection cover-age based on safety significance of the items identified.

Their coverage of facility safety, and that of region-based NRC inspectors, is also documented and routinely distributed to the public document room in inspection reports.

Even if the licensee's LER's were much better, it would be necessary to review the NRC inspection reports to assess the NRC coverage of the events involved.

In addition to LER submittals, the licensee issues press releases on significant events and makes other information available through its "Special Office of the President."

That office does provide license information on events to the public and can be reached at (717) 759-2281.

"This is also a means of providing information to the public well before an analysis and LER are completed, but it is not a

substitute for NRC review of events.

C.l SEA Concern Corrective actions for failed reactor safety components are question-able or non-existent.

C.2 NRC View Corrective actions on specific LER's addressed by SEA are discussed along with the LER in subsequent sections of this enclosure.

We agree with SEA on the need for adequate corrective actions.

For each LER identified by SEA, and for corrective actions in general, the Susquehanna licensee has taken corrective actions which we have evaluated as acceptable.

Enclosure 1

D.2 Standb Gas Treatment S stem (SGTS Problems D.2.a. 1 SEA Concern When both radioactive gas treatment systems failed to operate proper ly upon loss of offsite power, no immediate corrective action was proposed for the inadequate heating of inlet air.

And, no written report of the investigation to preclude future problems could be located.

D.2.a.2 NRC View LER 82-32 discussed the failure of this system to respond following a loss of offsite power.

The licensee identified control circuitry problems which prevented the SGTS from starting.

First, a loss of offsite power to the SGTS radiation monitoring instruments caused a spurious high radiation signal which tripped the lead SGTS fan.

As noted in the SEA report, this problem was addressed by a modifi-cation of the radiation monitoring instruments.

The other problem, insufficient differential temperature across the SGTS heater, tripped the standby SGTS fan.

Initially, the licensee increased the trip delay from 120 sec'onds to 200 seconds to allow more time for the heater to provide a

greater differential temperature prior to tripping of the fan.

Additional testing in Nay of 1983 found more problems in the SGTS heater circuitry.

This was reported by the licensee in LER 83-61 and the plant modifi.cation then made was documented in NRC Inspection Report 50-387/83-12.

That modification changed the temperature controller to fail to a maximum demand position upon loss of power so that, upon subsequent re-energizing, the heater would operate at full capacity to heat the air to the desired temperature.

We consider the licensee's corrective actions on these pro-blems acceptable.

D.2.b.l SEA Concern The plant was operated at full power without either SGTS operable because a maintenance error disabled both systems.

This failure was discovered only when maintenance on the second unit revealed that it was inoperable.

Warning alarms in the control room were not adequately investi-gated.

D.2.b.2 NRC View LER 83-26 discussed the inoperability of this system.

Technical Specifications require that, if both SGTS sub-systems are inoperable, the licensee should initiate action to shut down the'plant within one hour and be in at least a

Enclosure 1

Hot Shutdown condition six hours later.

On this occasion, the licensee violated his Technical Specifications by operating the plant for approximately 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> with both SGTS subsystems inoperable.

The licensee was assessed, and

paid, a civil penalty of $60,000.

The fact that indicators and alarms did provide information to the control room operators and they did not properly assess the significance of the indicators was a significant factor in the NRC decision to impose a civil penalty.

Corrective actions by the licensee included specific training as a result of this event for further reinforcing of the importance of system status and its impact on Technical Specifications.

NRC review of this event is documented in NRC Inspection Report 50-387/83-03.

In this case, the licensee also modified procedural controls to more clearly identify the potential for development of problems like this.

The licensee's corrective actions were found acceptable.

D.2.c. 1 SEA Concern Testing with the plant in cold shutdown resulted in dis-covery that two blank flanges were missing from ductwork connecting the Unit 1 refueling floor to the Unit 2 side of the reactor building, thereby breaching secondary con-tainment.

Because there is an 18 month inteJval between secondary containment flow surveillance tests and the last test was in October 1982, higher than allowed flow rates for.secondary containment integrity may have existed.

A radioactive release on the Unit 1 refueling floor when both SGTS systems were inoperable could have resulted in an immediate impact on public health and safety through escape of radioactive gases from the compromised secondary con-tainment.

D.2.c.2 NRC Yiew LER 83-33 discussed the discovery that flange removal resulted in a flow path from the Unit 1 refueling floor to the Unit 2 Reactor Building.'RC review of this event, documented in NRC Inspection Report 50-387/83-06, concluded that the flanges were removed while the plant was shut down.

The SGTS was not required to be operable then.

Discovery and correction occurred prior to the next reactor

startup, when the SGTS would have been required to be operable again.

This did not have a significant effect on

~

plant safety because the flanges were not removed during operation.

A safety problem could indeed have developed if a radiation problem had occurred on the Unit 1 refueling floor.

But, since the Unit 1 reactor vessel head and drywell head remained in place, and since there was no spent fuel stored on the'refueling floor, there was a

0 Enclosure negligible chance for development of such a problem.

Even if such a problem had arisen, the i'ndicated SGTS flow would have assured a properly filtered exhaust.

The conclusion that the 18 month surveillance of SGTS flow is inadequate is somewhat subjective.

Indicated flows subsequent to flange reinstallation have been consistent with those obtained prior to flange removal.

E. 1 SEA Concern E.2 Both Primary Containment Atmosphere Gaseous Radioactivity Monitoring channels were inoperable during reactor startup.

One stuck "A" channel meter was corrected by a mechanical shock without documen-tation of further investigation.

The "B" channel appeared to suffer from flow blockage.

There was no LER identification of means of preventing future occurrences.

NRC View F.1 Primary Containment Atmosphere Gaseous Radiation Honitor.

LER 83-42 reported that, during a reactor startup on February 25, 1983, both channels of the primary containment atmosphere gaseous radioactivity monitoring system were found to be inoperable.

Such a situation is

. permissible if appropriate compensatory measures are taken.

Plant Technical Specification 3.4.3. 1 permits continued operation for up to 30 days in this situation, as long as two of the required leak detection systems are operable, and as long as grab samples of the containment atmosphere are taken every 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> to compensate for the inoperable gaseous radioactivity monitoring system.

The licensee did not have an operable gaseous monitor for 7 hours8.101852e-5 days <br />0.00194 hours <br />1.157407e-5 weeks <br />2.6635e-6 months <br /> and 20 minutes on February 25, 1983; one grab sample of containment was taken during this time and measured for gaseous activity.

The licensee's actions were in accordahce with the Technical Specifications.

In addition, since this is the only reported occurrence of this problem and over nine months have elapsed since its occurrence, it does not appear to be a problem in need of additional corrective action.

We share the SEA concern that local elected officials have a need to be aware of those activities that can affect public health and safety.

In this

regard, we do meet with local officials on about an annual basis to explain the NRC inspection program and the NRC role in emergencies.

SEA Concern F.2 Independent analysis of reactor safety issues by local governments and close regulatory overview by the NRC is recommended.

NRC View Although our experience shows that local governments are not staffed with the technical resources needed to fully analyze reactor safety

issues, there is a strong local interest in the activities of the

Enclosure NRC.

Me believe that the overview by the NRC, using both resident and region-based inspectors to augment the technical reviewers in our Bethesda offices, is sufficient to assure a complete and independent analysis of safety issues.

Regulatory oversight is provided not only by direct inspection effort but also includes consideration of LER's and other potential problems as generic issues.

Sometimes additional requirements are imposed as a result of this evaluation review process.

G.l SEA Concern G.2 The civil penalty seems inconsistent in view of the number and seriousness of other LER's filed by the utility since the start of the reactor test program in September 1982.

It is difficult to identify a pattern of consistent reasoning in the NRC reaction to this incident.

Without NRC publicly correlating this event with other related events or with the PP&L performance to date, it is difficult to consider the civil penalty as representing a firm or consistent policy towards safety systems at Susquehanna.

NRC View The NRC enforcement policy includes several different means to insure licensee compliance with the regulations.

The nature and number of violations have to be considered in light of the impact.

on safety systems.

Although there may be disagreements as to the seriousness of certain violations, we do not view LER's alone in judging the nature of a problem.

In the particular situation with the civil

penalty, the problem involved inoperability of a safety system beyond prescribed limits and failure to take -prompt compensatory measures.

This problem was principally caused by human error during routine surveillance testing.

In accordance with our enforcement policy, the staff does not generally issue notices of violation or civil penal-ties for events that are beyond the control of the licensee.

For

example, mechanical failures do occur and are addressed in LER's.

However, further evaluation is needed prior to concluding whether a

failure is random or indicative of a more serious problem.

The situation involving the civil penalty is more fully addressed in Inspection Report 50-387/83-03.

H.

Reactor Coolant S stem Recirculation Pum s and Related Events H.l.a SEA Concern R

During startup testing, high water conductivity caused auxiliary boiler flashover.

The change in voltage caused equipment, including the reactor coolant recirculation pumps, to fail.

The recirculation pumps were restored to service after an'nspeci-fied time period.

The LER stated that the ineffectiveness of installed water conductivity meters was well known, and the utility issued a memorandum stressing the importance of inde-

Enclosure 1

pendent testing of conductivity.

Corrective measures included installation of isolation transformers and investigation of acquisition of more accurate conductivity meters.

NRC View H.2. a LER 82-65 discussed a trip of the recirculation pumps on November 28, 1982, due to a voltage dip caused when high con-ductivity water provided a high current path in the electrically heated auxiliary boilers.

The licensee complied with his Technical Specifications and restored the recirculation pumps to operation.

To prevent recurrence, the licensee installed isolation transformers in the power supplies to the auxiliary boilers.

Instead of relying solely upon the installed conduc-tivity cells, the licensee now requires the chemistry techni-cians to sample the auxiliary boiler water and to perform an in'dependent conductivity measurement before starting those boilers.

Loss of recirculation pumps is an event that has been designed for and its safety implications have been reviewed as a

part of the licensing process.

In this case the appropriate compensatory measures were taken.

SEA Concern With the reactor at 75K power, a motor-generator for the reactor coolant recirculation pump shut off.

The pump was returned to operation after an unspecified time.

An investigation committee convened within an hour.

The committee reported no electrical or mechanical corrective action, but instead recommended that the plant be returned to operation so that a data history on five electrical factors could be established.

, H.2.b NRC View LER's 83-12, 83-14, and 83-25 all discussed trips of the recir-culation pump "A" motor-generator.

In all cases, compensatory act'ions required by the Technical Specifications were accom-plished.

The problem was finally isolated to a silicon control rectifier (SCR) which was intermittently breaking down in the voltage regulator circuitry of the recirculation motor-generator exciter.

Since the breakdown of the SCR was an intermittent fault, troubleshooting after each trip did not initially find the problem.

After the January 19, 1983 trip (LER 83-14), the licensee installed temporary recorders in the motor-generator circuitry to monitor voltage.

When the recirculation pump tripped again on February 4, 1983 (LER 83-25), the licensee used these temporary recorders to locate the fault and replaced the voltage regulator card.

Bench-testing of the faulty card located the defective SCR.

This event was reviewed by the NRC and the licensee's actions were found to be acceptable as documented in NRC Inspection Report 50-387/83-03.

The SEA

Enclosure 1

report apparently did not consider LER 83-25, which documents the sequence of recirculation pump trips and actions taken by the licensee.

H.3.a SEA Concern In January 1983, while at 85K power, reactor coolant conduc-tivity increased rapidly.

Power was reduced by 2'5 and was returned to 85%.

Thereafter, water conductivity rose above the maximum allowed.

The cause was contamination by an organic cleaning solution during maintenance.

Temporary administrative actions were proposed.

Final administrative procedures were not implemented until four months later.

In November 1982, high water conductivity had led to a failure of the coolant recircu-lation pumps.

H.3.b NRC View LER 83-16 discussed a January 1983 reactor coolant conductivity excursion.

The licensee's Technical Specifications allow continued operation for 12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br /> with conductivity greater than 10 micromhos per centimeter, and for 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br />'ith conductivity greater than 1 micromho per centimeter.

On January 29, 1983, the conductivity exceeded 10 micromhos per centimeter for 75

minutes, and exceeded 1 micromho per centimeter. for 9 hours1.041667e-4 days <br />0.0025 hours <br />1.488095e-5 weeks <br />3.4245e-6 months <br />.

These values are within the Technical. Specification time limits.

The licensee made temporary changes to his sampling procedures to require sampling for organics in the radwaste water prior to sending it to the condensate storage tank.

Those temporary procedures r'emained in effect until incorporation into the next revision to the procedures.

This is an acceptable means of correction.

The November 28, 1982 water conductivity problem in the auxiliary boilers resulted from pH control additives in the auxiliary boiler heater (auxiliary boiler water does not come in contact with reactor coolant water).

The high conductivity in the reactor coolant resulted from organic cleaning solution collection in the radwa'ste system, transfer to the condensate storage

tank, and entry into the feedwater system as makeup water.

The two events are not interrelated.

I.l SEA Concern In February 1983, with the reactor at 965 power, the flow of water from the reactor feedwater pumps exceeded that specified by the operating license.

Such an excess flow could cause a too rapid change of temperature within the reactor.

This event was reportedly caused by an incorrect'etting on the pump controls.

The problem was corrected by changing step settings.

Enclosure 1

I.2 NRC View LER 83-10 discussed the licensee's actions upon determining that the feedwater pump high-speed stop settings were not conservative

enough, thereby creating the possibility that runout (maximum) flow from the feedwater pumps could exceed the value assumed for the core reac-tivity transient analysis described in the Susquehanna Unit 1 Final Safety Analysis Report (FSAR Section 15.1.2).

Actual flow did not exceed the value assumed in these analyses.

The licensee found this problem while performing startup.test ST 23.6, one purpose of which was to determine if maximum feedwater pump runout flow was within the value assumed in the FSAR.

When it was determined that the settings were improper, adjustments were made.

This LER was reviewed in NRC Inspection Reports 50-387/83-05 and 83-06.

We consider this to be a

potential problem which was corrected as a result of proper testing.

J.l SEA Concern In February 1983, while the reactor was shut down, the Residual Heat Removal System shut down twice within an hour as a result of spurious tripping of the alternate power supply breakers.

This occurred during preventive maintenance on a related motor.'n engineering modification was scheduled for completion on July 1 to help prevent recurrence.

J.2 NRC View LER 83-30 discussed the Residual Heat Removal (RHR).System shutting down twice within one hour as a result of spurious trips of the Division 1 Reactor Protection System (RPS).

Technical Specifications require that two shutdown cooling loops be capable of operation and that one of those loops be in operation when the plant is in Cold Shutdown.

Technical Specifications also state that, with neither RHR shutdown cooling loop in operation, the licensee must establish reactor coolant circulation and monitor reactor coolant temperatures and pressures within one hour.

After each RHR trip, the licensee reestablished RHR shutdown cooling within the one hour time limit. It should also be noted that, as described in the LER, very little decay heat was present in the core and negligible increases in temperatures were observed while the RHR loops were inoperable.

There was no hazard to public health or safety from this event.

Even if there had been significant decay heat and neither RHR loop had been capable of operation, there are methods of injecting water into the core to keep it cool.

Also, removing the heat generated by the reactor via the turbine bypass valves to the main condenser is an acceptable means of decay heat removal that could be used in similar situations.

Enclosure 10 K.1 SEA Concern K.2 The residual heat removal system removes heat generated by radio-active material in the reactor fuel.

The heat is generated even with the control rods in the full-in position; thus, the term "cold shutdown,"

though widely used, is actually a misnomer.

Even when the fission process is stopped, enough radioactive waste material is contained in the fuel cells to generate core damaging heat.

NRC View L.l Because the word "cold" usually connotes a temperature below that of the human body, the phrase "Cold Shutdown" can be misleading to non-technical individuals.

"Cold Shutdown" is a functional plant condition defined in facility Technical Specifications.

Specifi-cally, "Cold Shutdown" refers to a reactor condition when reactor coolant water (primary system) temperature and pressure are such that the water will not boil (i.e. less than 200 degrees Fahrenheit at or near atmospheric pressure).

t1aintaining the reactor coolant system in such a condition provides an enhanced safety situation when work is performed on the primary or its supporting system.

This condition has both nuclear safety and personnel safety consi'derations:

the reduced system pressure and the low temperature lessens the likeli-

, hood of pipe breaks or steam leaks.

Residual Heat Removal (RHR) is required during Cold Shutdown to continue to remove decay heat when shutdown evolutions are conducted.

RHR prevents decay heat from raising plant temperatures above the Cold Shutdown condition.

SEA Concern In February 1983, while the plant was shut down, an operator "making rounds" discovered a broken valve in the RHR System.

The valve was severely vibrating, had lost packing (seals),

the valve position indicator had vibrated to an extent that it fell off, and the welds on the bracket securing the valve had broken.

The LER stated that the damage was caused by excess flow through the valve.

An inspec-tion of the alternate RHR Sy'tem also revealed cracks in the pipe hangers but repairs were not made until the first system was again operable.

The operating procedures for the RHR System were revised, but no mention was made of more frequent inspection of system com-ponents.

Also, defective welds were the subject of an investigation early during the reactor test

program, and no.correlation was made between those welds and the RHR bracket weld failures.

L.2 NRC View LER 83-34 describes problems with the RHR system due to operating the system outside its design parameters.

The RHR "B" low pre'ssure coolant injection (LPCI) injection throttle valve was originally designed to maintain flow above 10,000 gallons per minute (gpm).

Early in plant life, this resulted in too large a cooldown rate while

Enclosure 1

on RHR shutdown cooling because there was very little decay heat in the core.

To avoid excessive cooldown rates,'he licensee began throttling this valve to flows less than 10,000 gpm.

This resulted in heavy vibration in the RHR piping near the valve, broken welds on a

RHR pipe hanger, packing in the valve vibrating out of the valve, and in a disabled valve position indicator.

The licensee declared the RHR loop inoperable and repairs were made to the valve and the hanger welds.

Subsequently, the other RHR loop was inspected.

Cracks were found on hanger welds in that system and these welds were also repaired.

After the repairs, the licensee revised the operating procedures to better assure that the valves would not be throttled below a 10,000 gpm flow rate.

NRC review of this event is documented in NRC Inspection Report 50-387/83-05.

We are satisified that there is no connection between this RHR vibration induced weld failure problem and the weld problem identified earlier in the startup testing phase in 1982.

This problem of throttling below design flow was the subject of NRC Information Notice 83-55, dated August 22,

1983, and addressed to all power reactor facilities.

NRC review noted a procedural discrepancy which permitted operation of the system outside the operating range of the throttle valve.

Nonetheless, PP&L complied with the Technical Specifications during discovery and repair of the problems noted.

Normal operator checks, which are conducted on each eight hour shift, identified this problem.

Nore frequent checks are unnecessary because this problem was detected and corrected without violating NRC requirements for system operability.

M. 1 SEA Concern On November 24, 1982 the reactor was operated with only two coolant leak detectors available to quantify coolant leakage in the reactor drywell.

The Technical Specifications require at least three leak detection systems to be operable at any time.

The operator continued reactor operation in violation of the Technical Specifications based upon a presumptive misinterpretation of the license requirements.

Two weeks after the utility proposed corrective action, a spill of 5,000 to 10,000 gallons of radioactive water was reported by a plant spokesperson to the news media.

Two weeks after that spill, a smaller spill was reported to the media.

Event reports apparently were not filed with the NRC as a result of these spills, nor from a 270 gallon premature release that occurred on December 21, 1982.

The spills reinforce the utility's poor judgement in operating the plant with less than the required number of leak detection systems avail-able.

N.2 NRC Yiew LER 82-60 documented the licensee's discovery of a discrepancy in their Technical Specifications (TS's).

As originally written, the

Enclosure 1

12 TS's stated that two drywell floor sump level monitoring system

channels, one primary containment atmosphere particulate radio-activity monitoring channel, and one primary containment atmosphere gaseous radioactive monitoring system channel must each be operable.

The gaseous/particulate radioactive monitoring system can be aligned to sample either the drywell or the wetwell and the TS did not stipulate aligning it only to the drywell.

On November 24, 1982, the licensee aligned the monitoring system to the drywell because its review concluded that the system would more accurately detect a coolant leak if it was so aligned.

The licensee submitted a TS change request to clarify how those systems should be aligned and restricted operation to assure that one gaseous and one particulate monitoring system were always monitoring the drywell.

The licensee did not violate the TS and issued LER 82-60 to document its review and the.bases for the proposed TS change request.

On September 2, 1983, the NRC issued Amendment 15 to the facility operating license to incorporate the requested change into the TS's.

These monitors could not have prevented or mitigated the spills which SEA noted as having occurred later.

The two spills referred to in the SEA report occurred outside the primary contai'nment, the first in an RHR pump room and the second in the radwaste building.

Neither

.,was of sufficient radioactive quantity to require reporting in accordance with the operating license.

Nonetheless-,

the NRC views such unplanned spills and releases of any magnitude as undesirable, regardless of whether release limits are exceeded, and insists upon measures to prevent recurrence.

Both spills were reviewed by the NRC as documented in NRC Inspection Report 50-387/83-03.

The SEA Report also discusses the 270 gallon inadvertent release which occurred on December 21, 1982.

The licensee did report this via the Emergency Notification System (ENS) in accordance with 10 CFR 50.72 require-ments.

This release and the licensee's corrective actions are documented in NRC Inspection Report 50-387/82-45.

Neither the spills nor the premature release caused a release of radioactivity to the environment in excess of NRC-regulations.

Because radioactive spills and releases are of public interest even when below the threshold for formal reporting to the NRC, licensees often inform the public of such occurrences.

The NRC encourages openness by licensees on matters of this nature.

N. 1 SEA Concern Although the spilled radioactive water was reported contained within the plant, cleanup of the leakage involved exposure of workers to radioactive material.

Further, although the radioactive water was reportedly treated, the plant is allowed to release radioa'ctive water into the Susquehanna River as long as the radioactivity is within prescribed limits.

Responsible public health officials believe that the most prudent public health policy is to reduce exposure to carcinogens (such as radioactive water) to th'e lowest leve'1 possible.

Enclosure 1

13 N.2 NRC View Unnecessary radioactive releases should be minimized.

As part of this philosophy the NRC does have an approach called As Low As Reasonably Achievable (ALARA).

Some releases are necessary and are allowed after appropriate calculations and evaluation are conducted.

'he ALARA philosophy is also applied to occupational radiation exposure.

There are regulatory limits established for maximum personnel exposures, including workers, as well as maximum levels for liquid and airborne releases.

The ALARA philosophy includes addi-tional measures that further reduce exposures and releases to levels that are as low as reasonably achievable.

Inspection Reports 50-387/83-03 and 82-45 describe the attention given by the NRC to the events identified by SEA.

0.1 SEA Concern SEA has identified several instances where problems have been en-countered with the rod control system.

The problems include (1) loss of indicator lights, (2) stuck control rods, (3) inability to with-draw control rods, and (4) incomplete testing procedures.

These problems were encountered between November 22,

1982, and February 20, 1983.

Although none of these problems prevented insertion of the

rods, an action that shuts down the reactor, the concern appears to be lack of aggressiveness on the part of the licensee to develop a

reliable system.

0.2 NRC View As indicated above, insertion of the control rods into the reactor core is the means by which prompt reactor shutdown is achieved.

None of the control 'rod-related events resulted in any inability to shut down the plant or in operation outside Technical Specification requirements.

LER 82-58 describes the November 22, 1982 problem with control rod 46-23.

The full-in position indication was lost for this rod.

When Instrumentation and Control (IKC) technicians attempted to determine the cause, the-problem disappeared and could not be duplicated.

The LER noted that this had previously occurred, as reported in LER 82-55.

In LER 82-55, the licensee stated that the position indica-tion mechanism"for another rod had been replaced on November 14,

1982, when its full-in position did not actuate.

The licensee took the actions required by the Technical Specifications in both in-stances.

LER 83-15 describes a startup on January 21,

1983, when control rod 34-52 would not withdraw using normal control rod drive pressure (the SEA report stated the date as December 21, 1982).

The control

Enclosure 1

14 rod was declared inoperable.

The licensee's Technical Specifications allow plant operation to continue, under certain conditions, with as many, as eight (8) control rods inoperable.

The licensee freed the rod by increasing the drive water pressure to the rod in accordance with an approved procedure.

This action was acceptable.

LER 83-17 describes the January 26, 1983 event concerning the rod worth minimizer (RWM).

The position indication system for control rod 46-23 was providing intermittent spurious signals.

The rod was verified to be in the correct position and the RWM was bypassed.

The licensee's Technical Specifications allow the RWM to be bypassed if a second licensed operator or other technically qualified member of the unit verifies all control rod movement and compliance with the prescribed control rod pattern.

The licensee did take this required action.

Also, as stated in the LER (the SEA report is in error), the licensee repaired the position probe on control rod 46-23 during a

subsequent outage.

This action was acceptable.

LER 83-39 describes the February 25, 1983 event (SEA lists the date as February 22, 1983) during which an inadequacy in plant startup procedures resulted in a channel functional test for the Rod Block Monitor and Recirculation Flow Unit being missed. 'echnical Speci-fications require a channel functional test of'hese instruments

. monthly, and within 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> prior to startup if not performed within the previous seven days.

The licensee was only performing the checks monthly.

Review of the monthly checks revealed that the instruments were always operable.

Such inadequacies in surveillance procedures were discussed in a NRC special inspection (Report 50-387/83-20) conducted on July 26-29,

1983, and during a Management Meeting held with the licensee on August 30, 1983.

The NRC report of that meeting was forwarded to the licensee on October 27, 1983.

0 0

y Ir,~

I c4

'P I'

ll-%

yl

FROM'.

Son. ARen Specter ACTION CONTROL COMP L DEADLINE DATE CONTROL NO 1356l DATE OF DOCUMENT TO:

INTERIM REPLY FINAL REPLY FILE LOCATION l]~ g+

PREPARE FOR SIGNATURE OF:

CHAIRMAN EXECUTIVE DIRECTOR OTHER:

DESCRIPTION g LETTER 0 MEMO 0 REPORT 0 OTHER Encloses ltd fm Carild P.. M~Btz, Susquehanna Enyfrarv'antal Advocates coacernfng utAfty'e.hand) fng of'qa$ ptmt rm'lf'unctfons of the reactor's safety systm at Susquehanna SPECIAL INSTRUCTIONS OR REMARKS ASSIGNED TO DATE INFORMATION ROUTING GHGM OoVoung Benton Stal'fo

'He't4mes UJnder trod SEC/M 2M6 NRC FORM 232 (640)

EXECUTIVE DIRECTOR FOR OPERATIONS PRINCIPAL CORRESPONDENCE CONTROL

U ~,

I II 14 I

C

~a C

1 Kl" I 4 1

83 2266 NRC SECRETARIAT Logging Date 9/19/83

.CI Commissioner Exec. Dir.loper.

CI Cong. Liaison

-Q Public Affairs OIA Date O

Gen. Counsel 0 Solicitor O Secretary 0

Inspector & Auditor Pollc Evggluetlon Incoming:

From:

To:

lic evhnt Sen Arlen S ecter Const Ref Gerald R. Schultz us ue anna Environ voca es Date req ur er snves sgatlon x o an sng o

equipmen ma unc son Cl Prepare reply for signature of:

0 Chairman CI Commissioner 0

EDO, GC, CL, SOL, PA, SECY, IA, PE O

Signature block omitted 0

Return origirial of incoming with response

)C3 For direct reply'uspense:

Sept 27 Q

For appropriate action O

For information RF, OCA to Ack NRC FORM 62A g" oVI For thecommission:

b'I

'Ie C

amP

'Send three (3) copies of reply to Secy Correspondence and Records Branch Rec'tl Off, EQ5 I:~

Oa~e.. 2

. <. P~P g ~ on~A ACTION SLIP c

r I.c'

p 1

t