ML19309A858

Forwards Documents Re Investigation Into Extended Outages at Facility.Documents Include Trial Staff Statement Before PA Public Util Commission,Consumer Advocate Statement & CN Dunn Testimony

ML19309A858
Person / Time
Site:	Beaver Valley
Issue date:	03/28/1980
From:	Levin J PENNSYLVANIA, COMMONWEALTH OF
To:	Russell W Office of Nuclear Reactor Regulation
References
I-79070314, NUDOCS 8004010426
Download: ML19309A858 (54)

Text

s * .* f,

[

COMMONWEALTH OF PENNSYLVANIA PENNSYLVANIA PUBLIC UTILITY COMMISSION P. O. BOX 326S HARRISBURG, Pa.17120 March 28, 1980 IN REPLY PLE ASE REFER TO OUR FILE Mr. William T. Russell 50- W Chief, Technical Support Office of Nuclear Reactor Regulation Nuclear Regulatory Commission Washington, DC 20555 Re: Investigation into the extended outages of the Beaver Valley 1 Nuclear Generation Station Docket No. I-79070314

Dear Bill:

I am enclosing quite a lot of information with this letter on the Beaver Valley investigation. You should find enclosed the following:

1. Trial Staff Statement of Issues (submitted at a prior hearing)

2. Consumer Advocate Statement

3. Testimony of Clifford N. Dunn Statement No. I submitted for hearings scheduled April 1 and 2 I have omitted the exhibits attached to Dunn's testimony since it mostly consists of NRC IE reports and the like. If you require copies, give me a call.

~

I will write further concerning the issues in a few days.

Very truly yours,

} .

(AA.^m d ohn A. Levin \ L Assistant Counsel 9, l . t (717) 783-2804 f JAL/j em 8 0 04010 4'24d -

g, ,. .

i BEFORE THE PENNSYLVANIA PUBLIC UTILITY COMMISSION Investigation Upon the Commission's:

Own Motion'Into the Extended  :

Outages of the Beaver Valley 1  :

Nuclear Generating Station  : I-79070314 i

REPLY OF OFFICE OF CONSUMER ADVOCATE TO STATEMENT CF ISSUES OF ~

PENNSYLVANIA POWER COMPANY AND DUQUESNE LIGIIT COMPANY I

I In a Prehearing Memorandum dated February 5, 1980, and a letter to the Administrative Law Judge of MarEh 5, 1980, the Office of Consumer Advocate set forth the legal and factual issues which we believe to be relevant to the present proceeding. The present e

memorandum is intended to serve as a reply to the Statement of Issues filed by the Duquesne Light Company and Pennsylvania Power Company.

Initially, we would note our agreement with the Companies' statement at page 2, ttat the primary issues in .this investigation involve "the facts surrounding the outage [s]" at the ' Beaver Valley nuclear generating plant which began on March 9, 1979 and November 30, 1979.

Where our positions differ is in the scope of the investigation into these facts and the ultimate purpose and goals of this investigation. Contrary to the Companies' view, this is not a mere I , fact-finding mission in which the Commission is powerless to act. Under l

the Commission's order of November 26, 1979, the Companies were <lirected to explain why their net energy clauses should not be adjusted to remove

t

• the costs of replacement energy incurred during the outages at Beaver Valley and why their base rates should not be adjusted to remove the costs associated with Beaver Valley. The purpose of these hearings is to review those explanations and determine whether the possible adjustments suggested by the Commission's order are appropriate.

The Companies' claim that these adjustments can only be made prospectively and in the context of a full rate hearing is without I

merit. Certainly, with regard to rates collected via the net energy clause, the Commission has the authority to order the refund of charges for unreasonably incurred expecses, without the need for a full rate case. To hold otherwise would mean that utilities could impose higher rates automatically through their fuel adjustments, but that the Commission would be powerless to order refunds of improperly collected rates without a full rate case. As the Commission has no opportunity to determine the reasonableness of tpese rates before they are put into effect, it must be permitted to review and correct them promptly on a retrospective basis.

This same issue was raised in the New York Public Service Commission's January 16, 1979 opinion regarding the operation of the fuct adjustment clause during an outage at the Consolidated Edison Indian Point No. 2 nuclear plant (Case 27123, Opinion 79-1, discussed more fully in the Consumer Advocate's February 5, 1980 Prehearing tiemo rand um) . In that case, the New York Commission approved a refund of energy costs which were imprudently incurred during a 1976 refueling outage of the nuclear plant and which had been passed through to the company's customers via a fuel adjustment clause. Though delaying the implementation of its refund order until the termination of the utility's then pending rate case, the New York commission rejected the 2

1 3 ,

e argument that its order of a fuel adj ustment rate refund constituted impermissible- retroactive ratemaking. As noted by the Commi m on:

The purpose of deciding whether the outage reasonably could have been shortened is to determine whether some of the fuel adjustment charges, resulting from the outage, were unreasonable. The Judge found that some charges were unreasonable, in that sense, and should be disbursed to present customers as a credit through the fuel adjustment clause; the Company excepts, on the ground that such a decision would be retroactive ratemaking. We deny the exception.

The fuel adjustment clause is a general formula which, when properly applied, allows a utility to recover from customers its reasonably necessary expenditures for fuel and purchased power if the expenditures exceed a particular level. This type of formula rate, like any other rate, is acceptable only if it is viewed as being subject to the Company's obligation to charge for no more than just and reasonable costs, and our authority.to insure that consumers pay no more than those costs. But the charges under a formula rate, unlike thore resulting from a fixed rate, cannot be tested for reasonableness prospectively. Of course, we are fully aware of the general rule in this State against retroactive ratemaking,*i.e., ratemaking based on determinations that rates charged in the past were unreasonable. But none of these cases announcing that principle invola.ed formula rates. They all dealt with fixed rates, i.e., rates that established a fixed charge per unit of consumption and whose reasonableness

• could be assessed prospectively.

Contrary to Con-Edison's claims, those cases do not protect utilities from having to repay amounts charged in excess of the approved tariffs. Thus, a utility would be responsible for adjusting its )

previous collection from customers under a fuel j I

adj ustment clause, if the costs the company incurred were adjusted after the collection occurred under the fuel adjustment clause. Similarly, an adjustment is l

warranted where, as here, it is demonstrated that the Company incurred fuel charges because of a clear lack of reasonable care. With respect to fixed rates, the l statutory scheme contemplates a " regulatory lag" l

during the period in which the public is notified of proposed new rates, and the proposed rates are held in abeyance pending a public inquiry into the reasonableness of the underlying costs which the new

! rates purportedly reflect. A formula rate, in'-

contrast, is a mechanism for passing costs on without this kind of regulatory lag; it therefore must be l l

l 3

i n construed as applicable only to prudently incurred expenses. Only in this way can a formula rate be imposed consistently with the requirement that utility rates be just and reasonable.

Id_., at 34-35 (footnote omitted). See also Magee Carpet Co. v. Pa. PUC, 174 Pa. Super. 438, 402 A.2d 229 (1954), where the Superior Court upheld Cie Commission's refund of fuel adjustment rates which had been found to be excessive.

Indeed, the refund mechanism is the only reasonable method of reviewing a utility's automatic fuel adj ustment charges. See 66 Pa.

C.S.A. $1307(d) and (e). If the utilities want to receive the benefit of automatic pass-through of fuel costs--without the need~ for a full-blown rate hearing--they should not be heard to complain that such I

a hearing is needed before any of those rates efn be refunded.

As to a possible changn in base rates arising from the removal of costs associated with Beaver Valley, a more thorough analysis of e

other expenses and revenues of the Companies as a whole may be necessary. This need not be an insurmountable task, however, as demonstrated by the Commission in its treatment of base rate issues in the af termath of the accident at Three Mile Island. . At that time, the Commission acted quickly to remove the costs associated with TMI-2 from the base rates of both Metropolitan Edison Company and Pennsylvania Electric Company. P'JO v. Metropolitan Edison Company, 29 PUR 4th 502 (1979). The Commission's response was facilitated by the agreement of the parties to the use of test years, capital structure, and rate of return from each company's most recent rate case. The records of those previous cases were incorporated into the record of the THI-2 case and a full-blown rate proceeding was not required. ..

4

In the case at bar, the two companies have either recently filed, or are reportedly in the process of preparing to file, requests for rate increases. Indeed, in susper. ling Penn Power's current rate request at R-79121020, the Commission directed the presiding Administrative Law Judge to make specific recommendations concerning "the issues arising out of the commission's current inve s tiga tions . . .

involving... outages at Beaver Valley I which are relevant to setting rates for Penn Power." A similar direction may be expected in the anticipated. upcoming .Duquesne rate case. The timing ~f o these 4

proceedings thus gives rise to the possibility that any rate base findings in the present investigation can be incorporated into the pending rate investigations or vice-versa. Thus even if a change is to be made in the Companies' base rates, there is no need to engage in a full-blown rate case as part of these proceedings.

e t

t

• 9 5

i

~

CONCLUSION The scope of the present investigation includes the authority to investigate and make findings of fact regarding the outages at Beaver Valley No. I and to determine whether and in what amounts adjustments to the base rates and/or net energy clauses of the Companies are appropriate. The Commission has the power to order, at the conclusion i'

I of this investigation, permanent prospective adjustments to the net i

energy clauses and base rates of the Companies' involved, and may also order a refund of those charges which already have been improperly imposed on consumers through the Companies' net energy clauses.

i Respectfully submitted,

- E (d)Q:, I Irwin A. Popowsky V

. e Assistant Consumer Advocat Ashley Schannauer Legal Assistant DATE: March 21, 1980 .

I i

.l l . ~-

I 6

' l

. 1 l

l 1

i BEFORE THE PENNSYLVANIA PUBLIC UTILITY COMMISSION In Re: Investigation upon  : I-79070314 (Duquesne)

Commission Motion into the  : I-79070318 (Penn Power)

Extended Outages of the Beaver  ! P-79040173 (Trial Staff Petition)

Valley Cenerating Station  : P-79040174 (Pittsburgh Petition)

Recommended Questions and Issues To be Addressed Pursuant to Order of Investigation i

By its Order entered July 27, 1979 at I-79070314 the Com-I mission instituted a formal investigation to determine'if any, or to what extent, adjustments should be made to Duqu,esr.e Light Company's (Duquesne) base rates and net energy clause to reflect the impact of the extended outage of the Beaver Valley 1 nuclear generating station if that unit were not in (ull commercial operation by October 15, 1979. That same Order consolidated the Commission's investigation with previously instituted investigations dceketed at P-79040173 and P-7904D174 as the issues to be raised in those proceedings were considered to be closely related to the Commission's investigation.

By Order entered August 3, 1979 at I-79070318 the Commission joined Pennsylvania Power Company (Penn Power) in its investigation of the Beaver Vai',ey 1 outages since Penn Power is part owner of that generating stat!on.

By furth'r Order of November 28, 1979, the Commission reaf firmed its intent to examine the prolonged outages of Beaver Valley. The Commission stated:

. ;+;

"[The] news that (Beaver Valley] will again be shut down before year end for an extended period reinforces our view that a conclusion as to the fully operational status of Beaver Valley I must be based upon a hearing record and not upon pleading alone. The continuing impact of the extended past and apparent future outages at Beaver Valley 1 on the base rates and net energy clauses of both Duquesne and Penn Power merits investigation as to whether, prospectively, the unit is in full commercial operation. . ..

- On August 17, 1979, Duquesne filed motions to dismiss Commission Trial Staff's (Staff) petition at P-79040173 and the petition of the City of Pittsburgh (City) at P-79040174.

~

We have reviewed Duquesne's motions and an' '

answer filed by Staff of August 23, 1979. We

' conclude that the investigation instituted

pursuant to Staff and City's petitions must go forward in order to resolve the factual disputes embodied in the p1'eadings. Duquesne's motions to dismiss will therefore be denied "

Additionally, by the terms of its order of November 26, 1979, the Commission directed Duquesne Light and Penn Power to quantify the present and projected costs of the prolonged Beaver

' Valley shutdown in base rates and energy costs through October, 1980.

l Accordingly, the Trial Staff here states recommended Questions and issues to be addressed in this consolidated proceeding:

1) What is the legal definition and regulatory

i. effect of " commercial opera' ion", if any?

r .f 2) .What factual circumstances must exist before

'I an electric generating station may be de-l s' termined to be in " commercial operation"?

l 1.

3) What legal and factual circumstances.must exist l for an electric generating station to remain in " commercial operation"?

4). What is the operating history of Beaver Vs11ey Unit I (" Beaver Valley")?

s e  %,

o. .

~*

5) What were the factual and technological causes of all Beaver Valley shutdowns after it first achieved criticality?

6) What utility management decisions or actions caused, lengthened or shortened those shutdowns?

What were the bases for such decisions or actions?

7) What consultant, contractor, agent or employee decisions or actions, caused, lengthened or shortened those shutdowns? What were the bases for such decisions or actions?

i j 8) What governmental or regulatory decisir n , orders ,

or' actions caused, lengthened or shortened those shutdowns? What were the bases for such decisions, orders or actions?

9) What costs were and are projected to be associated with each shutdown?

l

10) How did Duquesne Light and Penn Powe~r treat such costs for Pennsylvania ratemaking purposes?

11) How did Duquesne Light and Penn Power carry their respective loads during each shutdown?

Did that affect base rates or energy clause

- collections?

12) What contracts, bonds, agreements, indemnities

.i or law exists with respect to allocation of liability for any direct or incidental damages f' suffered (or anticipated to be suffered) by Duquesne or Penn Power as a result of shutdowns at Beaver Valley.

13) What rate base and what operation and maintenance expenses associated with Beaver Valley 1 are presently reflected in the base rates of Duquesne Light and Penn Power?

!. 14) What is likely to be the operating history of Beaver Valley 1 for the next few years?

4

& W

e ,-

15) Should an adjustment be made to the base rates and/or energy clauses of Duquesne Light and Penn Power as a result of the Beaver Valley chutdowns?

Respectfully submitted.

0 .

/1

} ,/

h'?t'V J !),Wmp '

John A. Levin ds'sistant Counsel v.

Gregg C. Sayre Assistant Counsel .

Pennsylvania Public Utility Commission Dated: 4 ab flA O F f

'I o

1' l

e I e t

l l

4 - - . - , , , . . - - , . n.

Duquesne Statement No. 1 I-79070314 P-79040173 P-79040174 Witness: C. N. Dunn Date:

INVESTIGATION UPON COMMISSION's i

OWN MOTION into the Extended Outage of the Beaver Valley No. 1 Nuclear Generating Station Order entered July 27, 1979 DIRECT TESTIMONY of CLIFFORD N. DUNN l

Q-1 Please state your name, business address and position with j D' quesne Light Company.

A-1 Clifford N. Dunn; 435 Sixth Avenue, Pittsburgh; Vice President, Operations.

Q-2 How long have you occupied that position?

l Since October 1, 1975.

i A-2 t

Q-3 What are your duties?

A-3 As a member of the management group, my functions include:

participating in the preparation and administration of the l operating budget dnd the capital improvement budget; responsible for operation and maintenance of the physical properties; advise Chairman of the Board, President and other officers with respect l

l to these functions .

l Q-4 Please summarize what formal education you have had, and your l

experience in the electric utility field.

j A-4 In June, 1949, I graduated from Carnegie Institute of Technology with a Bachelor of Science degree in Chemistry and was hired by the Duquesne Light Company. Between 1949 and 1957 I worked in the Chemical Section, Power S tations Department. From 1957 to 1962 I was assigned to the Shippingport Power Station with a variety of assignments, including station chemistry, testing and training.

In 1962, I was promoted from Shippingport to the Central Office of the Power Stations Department as Superintendent of Personnel and Records with administrative responsibility for all aspects

2.

of the Department's safety, training and personnel programs,

! including union relations. In 1964 I was promoted to the position of Technical Assistant to the Vice President of Opera-tions. My functions were primarily administrative in nature.

l They covered such areas as (a) capital and operating budget l

expenditures, (b) personnel. practices, (c) union grievance procedures, and (d) technical problems with principal emphasis

, on the ue related to power generation. In 1965,- I re-entered l Carnegie Institute in the night school program and graduated in May, 1971 with a Bacnelor of Science degree in Administration l

l and Management Science. On August 1, 1972 I was promoted to the position of General Superintendent of the Power Stations Department which title was changed to Manager, Power Production, on December 19, 1973. As head of the Power Stations Department i

l I was responsible for the overall supervision of the operation-and maintenance of the Company's generating stations. On October 1, 1975 I was promoted to my present position.

! Q-5 Have you testified previously before this Commission?

A-5 Yes. I testified before this Commi:sion in the following Duquesne Light Company rate cases:

C-18808 C-19276 RID-89 -

RID-373 R-79010740 ,

l I have also testified in other administrative investigations i

before the PUC. .

3.

Q-6 What was the Beaver Valley issue as stated in the Commission's Order entered July 27, 1979 which instituted this investigation?

A-6 The issue was the extended outage of the Beaver Valley No. 1 nuclear generating unit which started March 9, 1979 and had not ended at the time the Order was entered.

Q-7 What is the background for the Beaver Valley outage associated with this investigation?

A-7 On October 27, 1978, Duquesne notified the Nuclear Regulatory Commission (NRC) by letter (confirming an oral report on l . . . . . . . . .

. October 26, 1978) that during a' review of stress calculations

\f 31

, , . ,1 for the safety injection piping inside containment, Stone &

, ., u Webster had discovered some discrepancies. The Stone & Webster memorandum to Mr. Werling, Beaver Valley Station Superintendent, dated October 23, 1978 on this matter, is presented in Exhibit A .

The Duquesne Light Company letter to the NRC of* October 27, 1978 is presented in Exhibit B. The initial report indicated that

/' the discrepancies were due to a hand calculation. The recal-(

f culated stress levels exceeded allowable stress. However, they did not exceed yield stress (deformation with or without breaking) 1 except in one case out of six flow paths and thus no loss of O

j \

, N

, ,n:t , ^.

! -tr*i e

_c, I

1

. ..., i;, . , , r pg .  !. A 'j... . .

4. \

- Iy ., I gr.; b i

t, in safety function would have occurred even under postulated accident conditions. These recalculated values were determined j using NUPIPE, a computer code acceptable to the NRC in 1978 s' On December 6, 1978, Duquesne filed a report which contained a brief history of the events. A copy of the report is presented in Exhibit C. The report noted that in 1974, two of the six safety injection lines were coded into a computer code called PSTRESS-Shock II, which, according to the report, was an< acceptable ; l '

technique at that time and was the basis for all computerized .

Category I pipe stress analysis calculations. The " error" was reported as being a mis-application of the simplified analysis technique (chart method) employed during the original design.

In reviewing the data developed to support the adequacy of these two pipe support modifications, significant discrepancies were observed between the data from the original computer program (PSTRESS-Shock II) used to analyze earthquake loadings and a f

currently acceptable computer program (NUPIPE). On March 8, 1979, representatives from Duquesne, Stone & Webster, and the NRC met in Washington to discuss these differences. As a result of these discussions, it was determined that the difference arose in a sub-routine of the original program which used an algebraic summation of the predicted loads rather than.a sub-routine which summed the absolute value or the square root of the sum of the 7

squares. _ , . > .

5

w(f e , . ,

9 5.

On March 9, 1979, Duquesne brought Beaver Valley No. 1 Unit to a hot shutdown condition. Based on preliminary information developed between March 10 and 12, 1979, it appeared that certain piping conditions might be subject to overstress in the event of a design basis earthquake. Based upon this analysis Duquesne on March 13, 1979 initiated action to bring Beaver Valley to a cold shutdown which was accomplished on March 14, 1979.

In addition, on March 13, 1979 the NRC issued an Order which required Duquesne to show cause: (1) why the piping system for j all potentially affected safety systems should not be re-analyzed for design basis earthquake events using an appropriate piping analysis computer program which does not combine loads alge-braically; (2) why any modifications indicated by such re-analysis should not be made; and (3) why facility operation should not be suspended pending such re-analysis. The Order also required that Unit No. 1 be shut down immediately and remain shut down pending further order of that Commission. Similar orders were sent to l

the operators of four other units. A copy of the March 13, 1979 l

! Show Cause Order is presented in Exhibit D.

i On March 31, 1979, Duquesne filed a Response'to the Order to Show Cause agreeing to re-analyze the seismic load for affected facility piping systems u' sing an appropriate piping analysis computer code and to make any necessary modifications and re- l questing, absent a sufficient threat to the public health and

6.

safety, that upon completion of the re-analysis of and any necessary modifications to the affected piping systems required to assure safe shutdown capability and the capability of affected piping systems associated with the Engineering Safety Features and the Emergency Core Cooling System, that the facility be per-mitted to resume operation pending completion of re-ana,1ysis of the balance of the affected piping systems and any necessary modifications of the remaining affected piping systems. A copy of the March 31, 1979 letter is presented in Exhibit E.

By letter dated June 11, 1979, the Company informed the NRC that the pipe stress re-anal i had been completed and, with the installation of two hydraulic snubbers, no pipe is overstressed i

in the event of the occurrence of the Design Basis Earthquake.

! The Company informed the NRC that it had initiated the engineering and installation activities necessary to instal 1 two (2) additional I

hydraulic snubbers and to modify approximately fifteen (15) existing supports. Based on re-analysis of over 900 supports, using the currently acceptable techniques, it was actually only necessary to add three (3) new supports and modify eight (8) existing supports in order to meet safety criteria and permit the station to return to service. The Company expressed its intent to submit a report during the week of June 18, 1979 in support of its request for l,

station startup.

l

. . j 7.

P on June 19, 1979, Duquesne submitted copies of the Report on the re-analysis of Safety-Related Piping Systems for Beaver Valley No. 1 Unit. This report is rather voluminous. Because of its size and the limited number of copies, it is not made an exhibit in this proceeding but the report is available for inspection at the offices of Thomas & Thomas, 212 Locust Street, Harrisburg, Pa.,

as well as my office in Pittsburgh. It is also available as a public document.on' file with the NRC. The report provided a summary of the results for the re-analysis of safety-related piping that was per-formed. Noting that within the next two weeks it would complete modifications necessary to assure that the modified design would not result in seismic loads on piping systems which will impair the functional capability of engineered safety features during an earthquake, the Company requested that the NRC modify its March 13, 1979 Order to permit interim operation of the unit until the time for refueling. During the interim operation, the company agreed to continue its re-analysis and make any necessary additional modi-i fications during the refueling outage.

l l The NRC requested additional information which was provided l

in the form of three revisions dated July 11, July 18 and July 27, 1979. These revisions are available at the same l

locations indicated earlier. In addition, NRC inspectors made frequent visits to the station to review the modification effort.

l 8.

On August 7, 1979, Duquesne sent a letter to the NRC stating the modification work was completed and the station ready for operation. On August 8, 1979, the NRC lifted the Show Cause Order. A copy of this document is presented in Exhibit F.

In addition, the Company made maximum use of the outage time to do whatever work needed to be done with the unit shut down.

A list of major work activities is shown in Table IV.

Q-8 When did Beaver Valley return to operation? -

A-8 Station heatup started immediately after the NRC Show Cause

! Order was lifted. The uni.t was synchronized on August 17, 1979.

Q-9 What was Beaver Valley's ava.i' ability factor subsequent to August 17, 19797 A-9 The Beaver Valley No. 1 Unit was available 94.5% of the time I

as shown below:

Hours _

Availability On line or reserve Period  %

Aug.18 thru 31 326 336 97.0 September 631 720 87.6 October 728 745 97.7 l

November 697 720 96.8 Total 2382 2521 94.5 I  ;- ) .

.I O

I

9.

On November 30, 1979 the Beaver Valley unit was taken out of service for refueling, scheduled maintenance, safety inspections and safety modifications required by our operating license and/or the NRC.

Q-10 What is the status of the other nuclear power stations which were shut down by the NRC Show Cause Order of March 13, 1979?

A-10 Five stations were shut down on or about March 14, 1979. The Maine Yankee station returned to operation June 5, 1979. Beaver Valley was the second to return to operation on August 17, 1979.

CMie Fitzpatrick No. 1 Unit returned on September 7, 1979 and the Surry No. 1 Unit returned on October 24, 1979. The fifth unit, Surry No. 2, was shut down on February 4, 1979 for refueling and steam generator replacement. It is not expected to return to operation until late spring, 1980.

Q-11 The first paragraph in the PUC Order required Duquesne to file a statement explaining why its rate base should not be adjusted to remove the cost associated with the Beaver Valley No. 1 Unit, i

if that unit is not back in full commercial operation by October 15, 1979. Did Duquesne file a response?

A-11 Yes. Duquesne's response is contained in Exhibit G.

I Q-12. Will you summarize the reasons why the Commission should not take action to reduce Duquesne's rate base due to the current outage of Beaver Valley No.17

=

• 10.

A-12 First, the Commission's concern was expressed on page 2, paragraph 2, of the Order and I quote, "The rate payer cannot be expected to pay, both the base rates associated with this plant and the increased energy costs related to its outage, for an indefinite period." The unit returned to service on August 17, 1979; therefore, the period was act indefinite.

In fact, it returned to service much eTrlier than the October 15, 1979 date set by the Commission.

Second, outages of base load generating units, both fossil and nuclear, do occur for a variety of reasons and can be of very

. short or very long duration. Outages are a normal and expected I

t risk in utility operation. The Beaver Valley outage should not be treated any differently than any other outage. Customers realize the cost benefit of large units when these large units operate and should realize the cost penalties (or regular busi-

', > 1 i'd .-..,,n.

s -rj'f. ' "i 1 n, ness risks) of extended outages. -

=,

Third, the Beaver Valley shutdown was the result of the initial action taken by Duquesne and subsequent action taken by the Nuclear Regulatory Commission (NRC) in the interest of public l

safety. Duquesne cannot insure against outages resulting from precautionary actions deemed necessary to insure public safety.

The utilities and the NRC have very active programs to identify problems which occur at one or more reactors and to take the necessary action to assure the problems are made known to other

11.

potentially affected reactor operators and that appropriate steps are taken to resolve these problems. Such steps can and do involve the shutdown of reactors or extending an existing outage. These shutdowns can be for substantial periods of time.

To alter Duquesne's base rates would be to penalize Duquesne for precautionary action taken in the public's interest and would require Duquesne to become an absolute guarantor in its electric operation even during a period of transition in standards applicable to nuclear. stations such as now exist out of the Three. Mile Island nuclear accident.

Fourth, to reduce Duquesne's rates would result in further deterioration of the financial viability of Duquesne and other electric companies and would inevitably affect the electric industry's future cost of capital with accompanying adverse effects on electric ratepayers. Duquesne just settled the proceeding at R-79010740 in order to halt the deterioration of its financial position, and, as a result of inflation and changes in operations, is forced to file a new rate proceeding in the near future. To take any action which_would seriously jeopardize Duquesne's current financial position would be contrary to the public's interest.

Q-13 The second paragraph in the PUC Order requested Duquesne to

! file a statement explaining why its Net Energy Clause should I

not be adjusted to remove those costs of replacement energy

[

l

12.

associated with the outage of the Beaver Valley No. 1 Unit, if that unit is not back in full commercial operation by October 15, 1979. What is your response to that statement?

A-13 The question is now moot as far as the outage which started in March, 1979 since the unit was returned to operation (synchronized) on August 17, 1979 and supplied energy to the public in a very reliable manner. Moreover, all the reasons cited in Answer 12 would apply to this answer.

Finally, the data for this period indicates that the ratepayer did not bear increasing energy costs. Initially, it must be noted that if the unit had not been shut down on March 9, 1979 it would have been shut down for refueling and modification work starting in mid-May, 1979 and continuing through the summer. Thus, any costs experienced subsequent to May, 1979 would have occurred even if the Show cause Order had not been issued. Moreover, during the period between March 9 and June 30, 1979, Duquesne was a net seller of energy when the energy bought under the CAPCO Buy / Sell arrangement established January, 1979 (before the outage) is properly considered Duquesne's capacity.

The purchase power expense included in the NEC for the months of March through June, 1979 was a credit of $751,000 and a declining NEC rate as shown below: .

l l

l l

13.

Billing Month - 1979 Billing Rate - $/KWH January .006608 February .006531 March .004184 April .004400 May .003544 June .002804 July .002470 August .002383 The net energy cost data for each month subsequent to May, 1979 and which affect the billing months subsequent to July, 1979 are those which would have occurred since Beaver Valley would have shut down in mid-May, 1979 for refueling and modification work if it had not been shut down in mid-March, 1979 by the NRC Show cause. order.

For the months of July through November, 1979, purchase power expense included in the NEC calculation was a $183,000 credit while the NEC rate remained essentially constant as shown below:

Billina Month - 1979 Billing Rate - $/KWH l

l September 0.003100 October 0.003111 November 0.002706 December 0.002943 January, 1980 0.002711

! February 0.002986 March 0.003373 I

. Q-14 Is the Beaver Valley shutdown in any way similar to the Three l

Mile Island situation?

14.

,i.,

A-14 No. The reason for the shutdowns are completely different.

In addition, the Beaver Valley outage ended well before the

./

October 15, 1979"date set by the Commission while the outage time for either of the Three Mile Island units is still undetermined.

Q-15 During the course of the Beaver Valley outage from March 9 to August 17, 1979, did Duquesne Light company make any estimates of its increased cost to replace the generation, lost at Beaver Valley?

A-15 Yes. Daily estimates were made using actual after-the-fact conditions but assuming Beaver Valley would have generated 8400 MNH each day of the 162-day period. The 8400,MWH are ,

equivalent to operating.at a 91% capacity factor. The estimated increased generating and purchase power cost amounted to about

$19,880,000 for the 162 days or $123,000 per day.

Q-16 What was the financial effect on the average residential customer?

A-16 The financial impact can be estimated as follows:

Increased cost reflected in the Net Energy Clause $123,000/ day Residential Sales March-August, 1979 1,313,400 MWH Total Sales March-August, 1979 6,642,200 MWH Percent Residential Sales of Total Sales- 19.77 Residential Customers as of 6/30/79 493,233 S123,000/dav x 0.1977 = $0.049 per day 493,233 residential customers or $1.48 per month

15.

The effect is de minimus at best.

Q-17 What effect would a $19,880,000 expense have on Duquesne's financial condition?

A-17 A S19,880,000 added expense would equate to a loss of S0.32/ share using a 0.5167 tax rate and 31,750,000 shares of stock. Such a loss would have a serious effect on Duquesne's already anemic current earnings. Such a loss would mitigate the beneficial effect of the $58,000,000 increase in annual. revenue which the Commission found " . . . to be in the public interest under the present circumstances" at R-79010740.

Q-18 The third paragraph of the Commission's Order asked Duquesne Light Company to submit a detailed analysis of the cost associated with Beaver Valley No. 1 currently in their rate base as per settlement approved at R-79010740. What is your response? .

i A-18 Inasmuch as Duquesne's base rates are the result of a settlement where no determination of original cost, rate of return, revenues, operations and maintenance expenses, depreciation, taxes, cost i of capital, etc., was made, it is impossible to determine which costs associated with Beaver Valley are included in the settlement rates.

Q-19 The fourth paragraph of the Commission's Ord.er required Duquesne to project the anticipated energy costs, with and without Beaver

16.

Valley No. 1 in service, for each month through October, 1980.

What is your response?

A-19 In response to the Commission's Order entered November 26, 1979 regarding this investigation, Duquesne did supply a projection of costa related to the Net Energy Clause for the year 1980.

That projection is contained in Exhibit G and Table I to this testimony. Such projections are very difficult to make with any degree of accuracy since the total cost .is affected by changes in such factors as (a) fuel costs, (b) customer usage (which determines the daily peaks and daily energy production),

(c) forced and scheduled outages, (d) the need and cost for emergency energy by Duquesne and other utilities, and (e) economy purchases and/or sales with other utilities. The need for pro-jections for the months of October, November and December, 1979 are now moot since actual data are now available for those months and any projection into 1980 now would be different than that which would have been made within 30 days of July 27, 1979.

Q-20 In your opinion, will the Beaver Valley station be a reliable generating unit in the future?

A-20 Yes. It has demonstrated reliable operation in the past. During the coal strike Beaver Valley No.1 was largely responsible for the City of Pittsburgh having power. Without Beaver Valley No. 1, customer outages could have occurred. Since its return to service t

l-

.=

17.

on August 17, 1979 it has operated in a reliable manner. The major problems which caused outages in 1976-78 should not occur again. They are shown in Table II. This table was developed for my testimony in R-79010740. It is the same data except I deleted the outage time in 1977 for the Low Head Safety Injection Pump and related problems. This was done after my personal review of the many other events which occurred during that period. It was really a period of maintenance activities. 'The safety in-jection pump and related problems were only in an early stage of evaluation. From an equipment operating standpoint, I am very confident that shutdowns due to past equipment problems should not recur but I cannot be as optimistic for the future as to shutdowns due to NRC actions taken in the interest of public safety. Public opinion and the political atmosphere resulting from the Three Mile Island accident foretell an era of even greater public safety requirements. With greater emphasis on public safety, nuclear power station availability could be adversely affected, at least in the near term. If the public interest demands ever increasing levels of safety, then the

public should pay the costs as they do with the many other products they buy.

18.

The performance of nuclear reactors decreased significantly in 1979 as shown by the following data:

1979 1978 1977 Capacity Factor 60.5 67.7 64.4 Availability Factor 67.9 74.8 73.4 Forced Outage Rate 14.3 9.5 7.6 Number of Reactors 67 70 66 1

! The data for 1979 deleted TMI-2 after April 1, 1979. -

Q-21 Did the Presidential Committee (Kemeny Commission) which investi-gated the Three Mile Island Accident make any recommendation on i

,I the subject of rate-making agencies and the cost for safety-related changes at nuclear power reactors?

A-21 Yes. Their report makes the following recommendation (page 69) :

" Utility rate-making agencies should recognize that im-plementation of new safety measures can be inhibited by delay or failure to include the costs of such measures in the utility rate base. The Commission, therefore, recommends that state rate-making agencies give ex-plicit attention to the safety implications of rate-  !

making when they consider costs based on ' safety-related' changes."

That recommendation has a direct application to the question

raised in this investigation. It applies to the outages of i l l March-August, 1979 as well as the outage which started December 1, l l

i

19.

1979. When a utility, either on its own initiative or on an order of a regulatory agency, shuts a nuclear power station down in order to make safety inspections and/or safety modifi-cations, the rate-making agency should recognize the need and cost for this work and not inhibit the utility's ability to j

pay for such cost by failure to recognize such cost in utility rates.

l l

I l

I l

l i

l 1

! 20.

Q-22 You testified that Beaver Valley was shut down on November 30, 1979 for refueling, scheduled maintenance, safety inspections and safety modifications. What are some of the major work tasks to be completed?

A-22 The outage started December 1, 1979. The work will consist of many mandated NRC inspections and modifications, several Duquesne

~

Light Company requested plant modifications in addition to the actual refueling of the reactor and normal plant maintenance.

The crit'ical path identified for the outaga is attributable to NRC mandated inspections, tests, modifications and station im-provements. The major tasks to be accomplished Que te safety requirements established by the NRC are:

• Repair and modification of the hydraulic snubbers.

' Automatic lowhead safety injection system pumps re-circulation modification. ,

Additional alarms on the emergency diest.1 generators.

Chlorine detectors for the control room.

Modifications to the refueling water storage tank.

Auxiliary feedwater pump recirculation pipe modifications.

Security systems improvements Modifications to increase the net positive section head of the Quench Recirculation Spray Pumps.

( Pipe modifications and as built drawings as required by I

i NRC IE Bulletin No. 79-14.

l l .

21.

Base plate and bolt tests and inspections as required by NRC IE Bulletin No. 79-02.

Improvement of solenoid operated valves as required by NRC IE Bulletin No.79-01A.

Inspections of all stainless stee: piping which could be used for transferring borated water as required by NRC IE Bulletin No. 79-17.

Safety improvements deemed necessary from the Le,ssons Learned from the TMI accident as required by NUREG's 0578.

In addition, both low pressure turbine spindles will be re-I placed with those built for Unit No. 2. This is a prudent

- management decision.

Q-23 How long might the current Beaver Valley outage last?

A-23 Current schedules show a mid-July, 1980 return-to-power date.

This date may vary for many reasons, including, availability of craft personnel, a better definition of work involved than that

presently known, space and equipment limitations, material deliveries', documentation of all work on safety-related systems as required by our Quality Assurance Program, NRC inspections l

l of documentation, etc.

Q-24 Is this the " first" refueling or. cage for Bea'ver Valley?

'A-24 Yes. ,

Q-25 What has industry's experience been as to outage time for the so-called "first refueling" as is the situation at Beaver Valley?

O

1

• l

. *. 1 22.

A-25 Based on a study of 39 reactors which refueled for the first time between January, 1974 to June, 1979, the following statistical data were deve' loped:

Mean refueling time 58.6 days l

Standard Deviation 16.9 days l

Range 27 - 111 days This was the time associated with t1 refueling activities only. Each reactor has its own uniq 9 fueling . situation '

which leads to the wide range in time shown above. "*r example, j if the refueling work is not the critical work, then the refueling-l work itself may be accomplished over a longer time. In the case of Beaver Valley, an eight to ten week outage period would be reasonable for the normal refueling / maintenance / routine inspection /

testing activities.

Q-26 Has a statistical study been made as to the total time for the

" first refueling" outage?

A-26 Yes. Using the same sample, the following statistical data were developed:

Mean outage time 86.1 days Standard Deviation 28.3 days Range 42 - 156 days

l 23.

l Q-27 Why are the first refueling outages so long?

A-27 Because this outage is the natural time to make changes to the station equipment based on its initial pe;iod of operation. It is also the period for warranty inspection of the turbine. Many inspections and other functions are being done for the first time; therefore there is a greater time factor.

Q-28 Is such data valid today?

A-28 It can be used as a base to which additional appropriate con-sideration must be given for the increased safety inspections and safety modifications- required by the NRC, some of which are identi-fled in Answer 22. Some of these inspec'lons should be once-in-a-station lifetime events. The same cannot be said for modifi-cation work since safety-related equipment is subject to the ever changing requirements imposed by the NRC in the interest of public or employee safety. In addition, the ti,me to perform the refueling activities must be reviewed in the context of the total outage schedule since the manpower allocated to this activity may be reduced if it is not the activity on the so-called critical path, i.e., the job which determines the outage time.

.Q-29 Were there other reactors shut down at the time you prepared this testimony due to one or more NRC Bulletins you mentioned previously?

A-29 Based on information available from NRC publications, the Farley

No. 1 Unit (860 Mw) of the Alabama Power Company was shut down on March 8, 1979 for " refueling" and the inspection requirements

24.

of NRC Bulletin 79-02. .It returned to operation on November 5, 1979. Thus the outage time was 241 days. The Oconee No. 3 Unit (860 MW) of the Duke Power Company was s'.iut down for refueling on April 28, 1979. At the same time they reported making an investigation and modification of possible safety problems re-lated to the Three Mile Island No. 2 (TMI-2) accident. The re-l -

fueling started May 5 and was completed June 30. From July 1 the station was on a forced shutdown for the inspection re-quirements of NRC Bulletins 79-02 and 79-14. It returned to service October 30, 1979. Thus the total outage time was 184 days. The Salem No. 1 Unit (1079 Mw) of the Public Service Electric and Gas Company of New Jersey was shut down on April 3, 1979 for refueling and modification to seismic hangers and anchor bolts (NRC Bulletins 79-02 and 79-07). It returned to service on December 27, 1979. Thus the outage time was 266 days. There are other reactors shut down for refueling and, while not so reported, could be performing similar inspections. Thus the problems of meeting the requirements of NRC bulletins, particularly 79-02 and 79-14, are not unique to Beaver Valley.

Attached as Exhibit H are copies of the bulletins associated with i

the current outage and their revisions as well as the Duquesne responses where responses have been submitted as of March 17, 1980.

l The information contained in the exhibit is self-explanatory.

i 1

l 25.

Q-30 What is the projected outage time for those reactors shut down for refueling as of December 31, 1979?

A-30 The projected outage times are about two to four months. How realistic some of these estimates are remains to be seen. i 1

Q-31 Why will the Beaver Valley outage which started December 1, 1979 last into mid-1980 to do this inspection work?

i A-31 The Beaver Valley situation is a unique combination of events.

It is this first " refueling" outage which normal,1y would be longer than any future refueling outage. The sheer volume of work in terms of " normal first refueling" outages plus recent NRC Bulletin requirements, plus the NRC requirements to implement changes as a result of the lessons learned from the Three Mile Island accident, plus other work requirements with limited qualified manpower, etc., all lead to an extended outage.

Q-32 Has Duquesne Light Company taken an extraordingry effort to cope with the work load?

l A-32 Yes. A special team has been defie.ed to manace the outage. An organizational chart for this group is shown in Table III. The Engineering and Construction Division and Power Stations Department are giving this Beaver Valley work top priority over other work requirements. The goal is to complete the work as scheduled, or earlier, if possible. In spite of this all-out effort, the goal may not be attainable due to factors which are r.at known or cannot

' be reasonably predicted at this time. The positions of Outage

?.

I 26.

Manager and Outage Coordinator are filled by Duquesne Light Company employees. The Outage Coordinator is a special position devoted entirely to coordinating the efforts of engineering, construction and operation using computer oriented scheduling techniques. The Outage Manager has the authority to cross departmental organization lines as necessary to reduce outage time and solve problems. The position is responsible to the Vice President of Engineering & Construction and Vice President of Operations for the outage efforts.

Q-33 Why did the company elect to shut down on November 30, 1979?

A-33 Refueling could not be accomplished during the show-cause outage because the first fuel reload, planned and analyzed by Westinghouse and approved by Duquesne Light could not be implemented. The level of remaining reactivity of the existing fuel wac still too 1.x F. and required reduction by further burn-up to,within the limits used in the reload safety analysis. Any change of the re-fueling plan would have required re-analysis by Westinghouse and a complete safety review; this re-review was estimated to require a six month effort. Attached as Exhibit I is a graph showing the status of fuel burn-up as of the show cause outage period and the projected burn-up based on various generating unit capacity factors. It was necessary to achieve a burn-up between 400 to 427_ equivalent full power days (EFPD) to satisfy the reload safety

27.

analysis. During this end-of-life operation, the unit must be operated at lower capacities due to the low levels of boron concentration in the reactor coolant system at the end of core life. The goal was to operate up to (but not exceed) the 427 EFPD figure. This would result in the maximum use of energy from the fuel to be removed. By operating to the maximum burn-up, additional time was available for the engineering and material procurement associated with the work to be done to satisfy NRC requirements of NUREG 0578. The station was shut down on November 30, 1979 with a burn-up of 42 6 EFPD.

Q-34 Returning to Table II, a long single outage for the period was noted as due to the ' transformer f ailure. Would you explain briefly what happened?

A-34 Yes. The transformer was manufactured by General Electric. It is connected 21.5 kV " delta" on the generator side and 345 kv

" star" on the high side with grounded neutral. The nameplate rating is 945 MVA at 55 rise. The transformer failed on July 28 1978 while the station was operating at approximately 770 Mw. There were no abnormal conditions or events in the station or on the transmission system prior to the failure.

The failure appears to have been a primary to secondary failure

.on the B phase. An inspection of the transformer in the shop indicated the faul egan in the third or fourth layer as I probably strand-to-strand or turn-to-turn conductor insulation

28.

f ailure which burned both to ground and through the -insulation tube to the inner low voltage winding. The inspection failed to produce any conclusive evidence as to the reason the trans-former failed. The warranty c. the transformer had expired at the time of the failure. Duquesne expects to recover the costs to repair the transformer under its existing fire insurance coverage. There is a $375,000 - $500,000 deductible provision in the insurance policy.

_l Q-35 What actions did Duquesne take immediately following the trans-former failure on July 28, 1978 to expedite its repair or re-I placement and to make full use of the outage time to do other

,necessary work?

A-35 A meeting was held on July 31, 1978 to assess what can and must be done and who will do it. A copy of the minutes of that meeting are presented in Exhibit J. The exhibit shows clearly that a very concise, immediate action plan was developed and implemented within three days of the f ailure.

Q-36 What specific actions were taken to locate a replacement transformer?

A-36 Working .through the vendors and other utility contacts between August 1-3, 1978 we learned:

a. A transformer was located at Millstone which might be available but must be returned in a short time period.

29.

b. A transformer might be available from Midland No. 2 Unit but would result in a 20-25% capacity reduction on Beaver Valley No. 1 if used.

c. A 975 MVA transformer was available from the Midland No.1 Unit due to delays in the startup vf that unit.

d. Two 600 MVA transformers might be available at Canonsburg, Pa. since they were built as spares for AEP. .

e. A new transformer would require 13-14 months to manu-j facture.

1

(

f. Repairs to the damaged transformer would extend the outage to about March, 1979 at the earliest.

On or about August 7, 1978 a decision was made to negotiate a purchase or lease arrangement for the transformer from the Midland No. 1 Unit of Consumers Power Company since this trans-former best suited our technical and time -requirements. A Purchase Agreement was signed August 24, 1978. The transformer wrs shippe by rail (4 miles to the river) on August 22, 1978 and then by barge from Mt. Vernon, Indiana to Beaver Valley.

It arrived at Beaver Valley on September 9, 1978. Meanwhile the failed transformer was disassembled and shipped to Pittsfield, Massachusetts by railroad car on August 29, 1978.

.--e - , - -

, 1

=,

30. I Q-37 What work schedule did Duquesne develop to make effective use of this outage time and to reduce the work schedule for the refueling outage' scheduled for the fall of 1978?

A-37 The original outage work schedule was issued on August 14, 1978.

This schedule, handwritten in bar-chart form, contained 135 general categories and specific work activities. General categories were estimated since specific work in so'me categories had not yet been identified. The original outage schedule began on July 28, 1978.and was to be completed on November 27, 1978, approximately 17 weeks (123 days) total scheduled duration.

The final (actual) schecile was a computerized network of 1,427 detailed work activities developed from the original 135 summarized activities. The actual start of this final schedule was also July 28, 1978. Completion was the unit gynchronization date, December 19, 1978, approximately 20 weeks (144 days) total duration. The transformer work itself was completed on November 21, 1978 when the transformer was energized from the transmission system. Comparing this total duration to that of the original schedule, the original estimated outage duration was exceeded only by 17 percent.

The critical path which created the November 27 completion date l

on the original schedule was divided between two major modifications:

_ _ _ _ _ _ _ _ _ - _ _ _ _ . _ _ y

31.

a. The Westinghouse Turbine / Generator Overhaul followed by the LA Coniensate Pump Shaft Seal Replacement; and

b. DCP 220 - The Main Unit Transformer Replacement.

The final critical path for the actual schedule was DCP 234, (MOV-S I-867A, B ) BIT Inlet Valve replacements. This job was not included on the original schedule, but was added on November 3, 1978. The work involved the replacement of two valves which had a history of cracked valve seats. They were r.eplaced with valves of different design. All work, excluding heat trace checkout, was completed on December'15, 1978, permitting unit synchronization to occur on December 19, 1978.

Following are the general work categories scheduled during the Main Unit Transformer Outage:

Westinahouse Reliability and Availability Procram Items The following WRAPS items were planned to be totally l completed during the outage as Design Change Packages (DCP) :

l l 1. DCP 196 - Improve Feedwater Flow Measurement.

1 l

2. DCP 199 - Upgrade the Fuel Transfer ' System.

3. Underwater TV camera.

DCP 196 and the Underwater TV Camera were completed during the outage. DCP 199 was installed but remains to be tested at the beginning of the Cycle ;-JT Refueling Outage.

The following WRAPS items were planned to be partially completed, as noted, during the outage:

32.

1. DCP 190 - Steam Line Break Protection (cabinet installation only).

2. DCP 200 - Automatic Steam Generator Level Control (cabinet installation only).

3. DCP 191 - Turbine Lo-Reactor Trip Delete (conduits and cable only)

4. DCP 195 - RCP Seal Maintenance System (cubicle grating modification only). ,

5. DCP 197 - Transit Time Flowmeter (conduits and l

cable only).

l Only DCP 195 was performed as planned during the outage.

DCPs 190 and 200 were performed, but not completed as planned 4

due to a delay in resolving the cabinet mounting methods.

DCPs 191 and 197 were deleted from the outage because engineering and material resources were not available. ,

In-Service Insoections Seventeen In-service Inspection packages were originally scheduled to be performed during the outage and all packages were completed.

l l

Maior Modifications Forty-five major modifications were originally planned to be performed, eithe. totally or partially. During the

33.

course of the outage, three major modifications were added to make the combined total of 48. of these 48, 40 were scheduled to be completed totally. Four were scheduled to be partials, and four were listed for consideration.

The three major modifications that were added during the outage were: DCP 223 - Diesel Generator Field Flash Modifica-tion; DCP 232 - Adding a Relief Valve to the Regenerative Heat Exchanger; and DCP 234 - (MOV-SI-867A and MOV-SI-867B)

BIT Inlet Valve replacements.

DCP 223 was added to the outage in November. It was originally estimated to take four weeks to accomplish but three weeks' time was saved by utilizing existing cable versus the routing of new conduits and cables. DCP 223 was completed on December 2, 1978.

DCP 232 was also added to the outage in November. In-stallation was scheduled to occur in parallel with project engineering's seismic analysis of the installed relief valve.

Installation began on November 24, 1978. Installation was approximately 90 percent complete on December 1, 1978 when project engineering determined that insufficient time was available to complete seismic design of the relief valve in timo for the scheduled plant startup. At th.at point, all installed equipment was disassembled and removed from contain-t L ' ment.

1

34.

DCP 234 was authorized to begin on November 3, 1978.

Within one week af ter work began on DCP 234, it became the critical path job. Due to several engineering scope changes and material delays, .this job took six weeks to complete, ending on December 15, 1978.

The following tab'le displays the results of these 48 major modifications:

Actual Results Originally Performed Maior Modifications Scheduled Completed Partially Deleted Scheduled to complete 40 15 11 14 Scheduled as partials 4 -- 1 3 Listed for consideration 4 0 0 4 Totals 48 15 12 21 Insufficient engineering and material resources were the reasons for deleting 21 major modifications and performing 12 as only partial installations. .

Periodic Eauipment Testing Sixty-seven regular frequency periodics were identified, l

scheduled, and completed by the Maintenance Department during the outage. Sixty-three belonged to the DLCo. Instrumentation group, and four belonged to. the DLCo. Electricians group.

,i

. Maintenance Work

There were 427 corrective action maintenance work requests scheduled during the outage. Of these 427 MNHs, 423 were

completed. Four were deferred to the future refueling outage, j'

, i

.. i 35.

Station Testing During the outage, 22 station or system tests were scheduled to be performed. Nineteen of these were completed.

Three were deferred to refueling.

' At the completion of the Main Unit Transformer Outage, 1

39 startup activities were performed. Their occurrences were between November 18 (fill and vent condensate system) to the final unit synchronization date of December 19, 1978.

Q-38 Did Duquesne take every reasonable action to make effective use of the authge time to complete necessary work and to minimize

. the outage time?

A-38 Yes.

Q-39 Did Duquesne consider any legal actions against General Electric to recover damages?

A-39 Yes. A legal opinion was solicited from outside counsel. The legal conclusions were (1) Duquesne was barred from a warranty claim since the 30-month warranty had expired; (2) there was no evidence to prove negligence; and (3) the chances of winning a l

case based on strict liability were slight. Therefore no legal 4

actions were instituted.

Q-40 -I again direct your attention to Table II. What were the activi-ties associated with the outage for the Low Head Safety Injection and Recirculation Spray Pumps?

9

~

36.

A-40 One of the major activities was the installation of a crosstie j 4

between the Low Head Safety Injection Pumps and the Recirculation Spray System. This modification was performed to provide increased reliability of the Low Head Safety Injection (LHSI) system due to possible bearing wear and vibration related problems on the LHSI pumps. The solution involved the installation of 6-inch lines from the discharge of the outside recirculation spray pumps to the upstream side of valves on the discharge. of the LHSI pumps.

This included installation of pipe, tees, elbows, fitting, gate valves and check valves. Also installed were the necessary supports and hangers.

A second major activity was a design change to improve the net positive suction head for the Recirculation Spray and Low Head Safety Injection Pumps. This modification was performed because in the event of a major loss-of-coolant accident (LOCA) , the vapor pressure of the water in the containment sump, supplying the recirculation spray (RS) pumps and the Low Head Safety Injection (LHSI) pumps during recirculation phase, may be closer to the containment pres'sure than original analyses had indicated. This situation results in a calculated available net. positive suction head (NPSH) at the RS pumps for a short period of time and at the LHSI pumps during the initial recirculation phase, which is less I

than the value originally specified. The solution to this problem was to divert water from the spray header to the pump suction of i

t i

i ..

37.

l the Outside Recirculation Spray Pump by the installation of i

4-inch lines to distribution sump. The piping was routed from the 10-inch quench spray line in containment to the sump from the Recirculation Spray Pumps. Associated hangers and rein-forcing pads were also installed.

A third major activity involved the overhaul of the outside Recirculation Spray Pumps. Specific tasks were:

1. Both Recirculation Spray Pumps were removed, disassembled and inspected for running clearances.

2. Seismic ring clearances were inspected on both recircu-lation spray pump cans.

3. Pump cans were inspected and plumbed.

4. Each recirculation spray pump discharge bowl and shaft columns were machined for parallelism, then optically aligned. .

5. Each recirculation spray pump shaft flange and pump can flange were machined to provide parallel seating surfaces.

6. Pumps were then reassembled and installed, and aligned.

l The fourth major activity was the overhaul of the Low Head Safety Injection Pumps. This work was similar.in scope to the six specific tasks for the Recirculation Spray Pumps plus the instal-lation of bearings of a new design in each pump shaf t column.

This problem was first identified at the North Anna Nuclear Station

\

l

. 38.

of the Virginia Electric Po'ver Company. Since the Beaver Valley pumps were similar to those at North Anna, the NRC required Duquesne to perform these activities so as to achieve a higher level of equipment reliability and public safety.

Q-41 What is being done during the current outage to make the permanent modifications to these systems?

A-41 Permanent modifications to ensure adequate NPSH for the RS and LHSI pumps during all postulated modes of operation will be made during the ongoing refueling outage. The planned modifications include dhe following:

- Available NPSH for RS pumps (RS-P-1A, 1B, 2A and 2B) will be increased by injecting cold water from the quench spray system to the containment sump.

Each quench spray pump will be equipped with a larger impeller installed in its existing casing. (To provide, the additional flow required for the cold water injection into the RS pumps suction.)

Required NPSH for the LHSI pumps (SI-P-1A and 1B) will be de-creased by installing cavitating venturies in the discharge line of each pump and a restrictive orifice . in the common dis-charge line to limit pump runout flow.

- Performance of the QA and RS header nozzles will be improved by replacing 2/3 of the existing spray nozzles with a different type nozzle, and plugging the remaining nozzles.

s.

39.

- The existing gravity feed chemical addition system will be replaced by a redundant positive displacement pumped in-

~

jection system.

- Modification to Reactor Water Storage Tank to increase the amount of water available.

This modification work is one of two major tasks which appeared to be the critical path when the outage started. This is the best judgment as of March, 1980. However, the work associated with 1

the completion of the Three Mile Island Lessons Learned changes (NUREG 0578) may become the critical path.

Q-42 There have been recent stories in the news media about cracks in a number of Westinghouse turbines and that Beaver Valley is

! one, Would you explain?

! A-42 I first became aware of the problem in the fall of 1979 when Westinghouse invited Duquesne to send a represeptative to a meeting on October 30, 1979 where Westinghouse would discuss the i

situation. Westinghouse indicated in the meeting that in the I

last three months they had discovered keyway cracks in turbine discs of modern design. They reported finding four discs with cracks after inspecting 69 discs. This inspection involved seven spindles of which three had cracked discs. They attributed the cracks to stress corrosion. j l

The NRC became aware of the problem on November 17, 1979 and,  ;

l after further investigation transmitted this information to the l .

40.

industry via IE Information Notice 79-37 dated December 28, .379.

t i On February 25, 1980, the NRC sent a letter to Duquesne asking i

for specific information on the turbine disc problem. The information was requested by March 18, 1980.

, . Q-43 Are these cracks in the Beaver Valley turbine disc?

A-43 Yes. The turbine discs were inspected by Westinghouse personnel i

using testing techniques and equipment developed specifically for this task. Their inspection work was overseen by Duquesne j personnel. On January 5, 1980, Westinghouse completed the testing and evaluation work. The'y reported finding two discs with cracks, l ane on each low pressure spindle. There are ten discs per spindle.

Q-44 What is Duquesne planning to do?

A-44 We immediately arranged to ship the two Beaver Valley No. 2 Low Pressure Spindles from Philadelphia, wtsre they were stored, to l

Beaver Valley. This was done since the repair time for the

! original spindles was estimated to be eighteen months at best.

I I The new spindles are at Beaver Valley now. They are being in-spected with no cracks found so far.

Q-45 Is Duquesne considering legal action against Westinghouse?

l A-45 Yes.

Q-46 Thic current outage represents a period of major work activities.

t What time and effort is Duquesne devoting to. the integration of these activities into an overall master schedule and to monitor and control the secuence of events?

, e \

s.

41.

A-46 Duquesne is devoting a major effort to the scheduling and review function. Again, I refer to Table III and the position of l I

Scheduling Supervisor. His . function is to obtain information from engineering, construction and station scheduling personnel which is necessary to develop the overall master outage schedules.

Exhibit K contains some typical, detailed, level three work schedules. The first page shows the detail activities associated

, with DCP-130 which is in progress both in the station yard and I the auxiliary feed pump room. The second page shows the detail i

activities associ.ated with specific segments of DCP-189; it also shows dates when this specific schedule was reviewed and revised. The third page is another segment of the work on DCP-189.

The fourth page is another example of detailed work sheets used to follow progress on the task of inspecting and overhauling the 36 snubbers in containment. All this information serves as input to a computer data base which in turn is used in a computer program to produce various reports used by the outage management team to monitor and plan the hundreds of tasks. One such computer generated report is also presented on pages 5 through 12 of Exhibit K. It is a simplified bar chart used to evaluate work now in progress and work scheduled in the near future. The bar

chart shows all major tasks grouped by the major categories of work, i.e., major modifications (DCP'c-Design Change Packages),

1 Inservice Inspections, Three Mile Island Lessons Learned items, l i

- a e

4 42.

maintenance, periodic testing, station testing and plant operations. Work schedules are reviewed and planned at (a) daily meetings of those responsible for the physical work at the station, (b) a weekly meeting of those engineering, con-struction and operation personnel who are charged with the tasks of knowing the status of each work task at a given point in time and keeping the outage time to that originally established and (c) monthly meetings between top level supervision assigned to outages and the Vice President-Engineering and Construction, and the Vice President-Operations. It is important to recognize that the specific work schedules shown in Exhibit K are subject to changes in time based on any new information which is developed in the daily and weekly review meetings. The detail schedules are not cast in concrete at any time. They must change to reflect (a) change in work scope, (b) availability of engineering in-formation, drawings, materials, (c) work interferences, (d) manpower priorities, (e) availability of station systems, etc.

The all objective is to meet the end date of July 22, 1980.

i Another computer generated report is shown on page 13 of Exhibit K.

This is a very detailed work schedule in terms of specific activities over a short time period. Work schedules are even reviewed at one further level, i.e., at the foreman / worker level.

4 4 e

e A

43.

All this detail is deemed necessary to provide reasonable assurance that the workmanship, including documentation, meets the standards established by our quality assurance program which in turn enforces standards for public safety.

Q-47 The events you have described in this testimony cover many activities and involve several highly technical fields. Are

~

you technically qualified as to the details involved in all these activities.

A-47 My knowledge on these subjects is derived from a combination of reports, correspondence, personal involvement, discussions, etc. during the course of each event and agairs reviewed prior to preparing this testimony. I was involved to the depth necessary by virtue of my position.

Q-48 Does this conclude your testimony?

A-46 Yes. ,