ML19308C012

From kanterella
Jump to navigation Jump to search
Discusses IE Insp of Ranch Seco on 780321-22,28-31 & 0417.No Noncompliance Noted.Major Area inspected:780320 Reactor Trip & Cooldown.Pressurizer Level Maintained by HPI Manual Operation.B&W Performed Analysis of Transient Effects
ML19308C012
Person / Time
Site: Rancho Seco, Crane
Issue date: 04/21/1978
From:
NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION V)
To: Mattimoe J
SACRAMENTO MUNICIPAL UTILITY DISTRICT
References
TASK-TF, TASK-TMR NUDOCS 8001170752
Download: ML19308C012 (7)


Text

. ~, _ - _, _.

g s,,

v c

(4, UNITED STATES kANcpQ hCO 4

[q NUCLEAR REGULATORY COMMisslON

![

REGION V g

O.,

4

$UlTE 202. WALNUT CREEK PLAZ A

,o 1990 N. CALIF 09NI A 80VLEVARD 9 *tes#

WALNUT CREEK, CALIFORNI A 94596 April 21,1978 Docket No. 50-312 l

Sacramento Municipal Utility District P. O. Box 15830 Sacramento, California 95813 Attention:

Mr. John J. Ma'ttimoe Assistant General Manager ar,.d Chief Engineer Gentlemen:

Subject:

flRC Inspection of Rancho Seco This refers to the" inspection conducted by Messrs.;P. Johnson, A. Johnson, H. Canter, and J. Woessner of this office on March 21-22 and 28-31, and April 17, 1978, of activities authorized by NRC License flo. DPR-54, and to the discussion of our findings held by. the inspei: tors with Mr.R. Redriguez and othcr :mbcr: of, you ' staff at tne conclusion of the inspection.

Areas examined during th.s inspection are described in the enclosed inspectica report.

Within enese areas, the inscection consisted of selective examinations of procedures and representative records, inter-views with personnel, and observations by the inspectors.

I o items of noncompliance with NRC requirements were identified w. thin N

the scope of this inspection.

In accordance with Section 2.790 of the NRC's " Rules of Practice,"

Part 2. Title 10, Code of Federal Regulations, a copy of this letter and the enclosed inspection report will be placed in the NRC's Public T

Occument Rocn.

If this report contains any informatica that you believe to be proprietary, it is necessary tnat you submit a written application to this. office, within 20 days of the. date of this letter, requesting that such information be witnheid from p'ublic disclosure.

The applica-tion must include a full stateme'nt cf the. reasons why it is claimed tnat

~

the information is proprietary.

The application should be prepared cc that any proprietary information iuentified is contained in an enclosure 8001170 7 M

(

~'

c DETAILS 1.

Persons Contacted

  • R. Rodriguez, Manager of Nuclear Operations
    • L. Schweiger, Quality Assurance Director
  1. P. Oubre, Plant Superintendent
    • J. McColligan, Engineering /QC Supervisor
  • R. Colombo, Technical Assistant
  • N. Brock, Instrument Control Supervisor J. Jewett, Senior Quality Assurance Engineer
  • G. Coward, Senior Power Plant Engineer
    • D. Blachly, Power Plant Engineer The inspectors also talked with and interviewed several other licensee employees, including members of the technical and engineering staff, technicians, shift supervisors, reactor operators, and maintenance personnel.
  1. Denotes those present at exit interview on March 22, 1978.
  • Denotes those present at exit interview on March 31, 1978.

2.

Reactor Trio and Cooldown of March 20, 1978 At 4:25 a.m. on March 20, 1978,- a reactor trip and turbine trip

~

occurred, followed by a. rapid cooldown' of the ' primary systam which exceeded cooldown rate limits permitted by the technical specifications.

The transient also involved reactor coolant system (RCS); conditions' in the restricted region of the pressure-temperature limits. curve.

provided in the technical specifications.

The event was precipitated by a loss of non-nuclear instrument (MMI) power caused by a mcmentary short circuit which occurred while changing an indicator bulb in a push-button switch on the reactor console.

A special inspection was conducted at the site on March 21 and 22 to examine circcmstances related to the event and to review the licensee's evaluation and corrective actions.

Discussions with plan,: personnel and examination of related logs, recorder charts, computer records, and other documents provided the folicwing f'ndings related to the event:,

Initial Ccnditions - Immediately pHor to the event, the plant a.

was operating at a steady state power level of 70%.

An operator had removed a push-button switch frcm its socket en the reactor control console to change a status indicating light bulb installed on the back side of the switch assembly.

b.

Secuence of Events (elacsed time)

Tim Event 0

(4:25:44 a.m.) A light bulb frcm a puchbutton assembly fell into a switch sccket on the reactor control

2-

~

console resulting in a short to

ground, The circuit breakers to two +24 VDC power supplies in tht: flill cabinet opened, causing loss of several indication and control para-meters, including steam generator and pressurizer levels and all RCS tem-peratures.

1-2 Sec.

Loss of the RCS hot leg temperature input to the integrated control system (ICS) caused termination of feedwater flow.

Reduced heat removal in the steam generators caused RCS temperature and pressure to increase.

5 Sec.

The reactor tripped on high RCS pressure, followed by a turbine trip.

1 Min.(approx)

The secondary sides of both steam generators were essentially emptied due to operation of condenser bypass valves, atmospheric dump valves, and auxiliary steam loads.

1-7 Min.

RCS temperature remained relatively stable at 580-595" F.

After a brief increase, RCS pressure decreased

. ressurizerj p

slowly to about 2000.psig.

level was. maintained (using ccmputer;

-indication).by manual operation of a' high; pressure. injection (HPI). pump; 7 Min.(approx)

"A" steam generator level centro 1*

initiated emergency feedwater injection (the turbine-driven auxiliary feedwater pump had started as required on loss of feedwater flow).

I 8

However, internal cnaracteristics of the buffered circuits were apparently such that the output frca the "A" steam generator level channel drifted slowly downward, causing emergency feedwater injection; the cutput from the "S" channel drifted slowly upward.

The time of seven minutes is based upon testing by the licensee on Mar.ch 24, in which this behavior was duolicated.

This time is also consistent with th cor.pa:ce tr, m y, which showed RCS cooldchn corr.encing abet.: eig m.m :. m af:ar :.ie trip.

L

~__....._.-q

- ~

8 Min.

RCS cooldown commenced as a result of emergency feedwater ficw to the "A" steam generator.

Manual operation of 1

a main feedwater p. ump may have con-tributed to the cooldown.

13 Min.

Decreasing RCS pressure.(and temperature) caused a safety. features. actuation at'

.1600 psig, aggravating the RCS cooldown rate. All safeguards pumps and the motor-driven auxiliary feedwater pump (AFP) started.

Full auxiliary feedwater flow to both steam generators was initiated.

15 Min.

The RCS reached a minimum. pressure.ofu 1475 psig; pressure then increased to and was maintained at approxth:ately 2000 psig bylmanu' lico~ntrol of.an~ HPI a

pi'p.

32 Min.(?)

Tne computer alarm typer showed "A" steam generator full (+500").

(flote:

time interval is in question because of uncertain "A" steam generator level indication).

48 Min.

The computer alarm typer showed "B" steam generator full (+600").

75 Ain.

After investigating inver.ter alarms, operating personnel located and closed two tripped 7.5 amp circuit breakers in the tii1I panel, restoring all lost indications and controls.

RCS temperature at this time read g

285 F.

Operating personnel secured both AFP's and ccmmenced RCS pressure reduction (using pressurizer spray).

i

105 Min.

RCS pressure and temperature were restored to the permissible operating region of the technical specifications pressure-temperature limits curve (Figure 3.1.2-2).

c.

Cause of Event - flon-nuclear instruments are housed in seven cabinets, wttn cabinet ilos. 5 through 7 sharing a cccmon pcwer supply.

In addition to the 120 VAC instrument power supplied 1

.*A J***'.

...d-~

'~ = E $$ [ 1.\\- -

,e.

-~_

4 "D"

"J" Inverter

. Inverter Auto i

Transfer N

A 7.5 A 120 7 AC to titi!

Breakers )

)

s Cabinets 5 - 7

+24

-24

+24 3

t'24 V DC to ti!{I f

Switching relays.

-24 status lights, &

j other !{til uses

~

frca the vital "0". inverter (with backup power frcm the "J" inverter), the inverters also serv'd two i 24 VDC power supplies.

The DC power supplies operate relays which,.in conjunction with related pushbutton switches, select the various instrument channels to be used for control and indication (see illustration below; K1 and X2 are fictitious relay designations used for explanation only).

Actuation of a pushbutton en the reacter control console picks up either relay K1 or X2 (an interlock prevents simultaneous actuation of both relays) and closes a contact to select, as shown below, either Channel A or B for Integrated Control System (ICS) input, panel indication, and recorder input.

In the example shcwn, loss of relay pcwer i

would cause both K1 and X2 contacts to open.

tio control or i

indication signal would then be available, even if AC instrument power were not interrupted.

Computer inputs are selected by separate (normally closed.) relay contacts in such a way that most are not affected by loss of the 124 VDC supplies.

T Channel A XI hat Panel Indication 11 4

---> Integrated Control System T

Channel 3

> ?.ecorder hat m

l

_. ~,,.

T.

T u_ __

1

~5-20, 1978, as a status indicating light At 4:25 a.m. on March bulb tehind one of the pushbutton switches was being replaced, the bulb inadvertently fell into the switch socket, causing a momentary short which opened the 7.5 amp breakers shown in the The resulting loss of DC' power to switching first figure above.

relays and other full functions caused the loss of several control and indication parameters, including pressurizer and Reacting to steam generator levels and all RCS temperatures.

a step decrease in RCS hot leg temperature, the ICS (sensing an apparent reduction in reactor heat output) terminated feedwater flow in an attempt to maintain outlet steam con-The feedwater flow termination caused a reduction in ditions.

heat removal by the steam generators, which in turn caused RCS The reactor tripped on pressure and temperature to increase.

Turbine high RCS pressure, and a turbine trip followed.

bypass valves, atmospheric dump valves, and auxiliary steam loads emptied the steam generaurs after about one minute.

The injection of emergency feedwater (see footnote on Page 2) initiated plant cooldown about eight minutes after the trip; manual operation of a main feedwater pump may also have con-tributed to the cooldown, but the relative timing and magnitude Auxiliary feedwater and safety could not be established.

features actuation five minutes later (at an RCS pressure of 1600 psig) significantly aggravated the temperature transient.

AnalysisofEvent-Regetor,coolantsystemtemperaturedecreased d.

'from approximately SgG,F at 0433 (based on computer trip memory printout);to 285 F at approximately 0540, when in-dications were re3tored.

This represents a cooldown rate of approximately 275 F per hogr, compared to the technical specifications limit of 100 F per hour.

RCS pressure and temperature were also in the restricted region of the pressure-temperature limits curve (Technical Specifications, Figure 3.1.2-2) for 30 minutes after and for an estimated 10-15 Sabcock minutes before temperature indication was restored.

and Wilcox_ (B&W). performed an analysis!(under ASME,Section III)tof the ef.fects of this transient on the reacter vessel, steam generators, control rod drives, fuel, and certain other

~

~

I plant components, and gave concurrence'on March 22,1978, to '

, plant startup: subject to certain recc=endations (see Cor-rective Actions, below).11RR reviewed the.B&W.analy'sision March 24,~ and also concurred in plant startup subject to

~

fulfillment of the B&W reccuendations.

Babcock & Wilcox re-quested additional inspection of steam generator tubes during the next refueling outage, and was planning the following analyses af ter resumption of plant operation:

(1) deter-mination of the effect of tne transient on the RCS accumulative usage factor and (2) an ASME,Section XI, analysis of selected primary components (to add additional conservatism to the Section III' evaluation).

w

~ ~ ~

.. ~.

~ ~ -..

Corrective Actions _ - The Management Safety Review Committee e.

(MSRC) reviewed the event on March 21, 1978.

The MSRC (1) appointed a committee of three individuals possessing con-siderable experience with the Rancho Seco facility to invastigate the event and recommend corrective actions *,' and (2) directed the Plant Review Committee (PRC) to review the circumstances of the event and, if warranted, recommend returning to power subject to compliance with B&W recommendations.

The PRC reviewed the event on March 22, 1978, and recommended resuming operation after issuance of a special procedure to implement B&W recommendations.

These recommendations included restrictions on rate of power increase during initial startup; performance of fiNI instrument checks; increased surveillance of icose parts monit] ring and primary and secondary radiochemistrw nr one week; previsions for restoring NilI in the event of a power loss; and a 75". l'imit on power level until B&W QA evaluation of their analyses is completed.

The plant was subsequently restarted on March 24, 1978.

f.

Followuo Review - Cocumentation of the increased surveillance prescribed by the licensee's procedure was examined during the week following startup, with no irregularities observed.

Review of the reactor power recorder chart showed that the rate of power increase between 62 and 72% had exceeded the 3%

per hour recommended by B&W.

The licensee had documented this discrepancy on a nonconformance report, and had discussed the matter with B&W.

The licensee stated that the rate-of-increase recommendation had been conservatively established with no specific basis, and that exceeding it was not considered by B&W or the licensee to represent a safety concern.

During the exit interview,' the licensee stated that followup review of the transient would consider the feasibility of changes to prevent a severe plant transient as the result of a single failure in the NNI system.

The licensee also stated that the Region V cffice would be informed wheti B&W gives concurrence for operation above 75% power.

Licensee Event Report 78-1, describing the March 20 occurrence, was submitted by the licensee on March 31, 1978.

tia deviations or nonccapliance items, other than those described in the licensee's report, were identified by the inspector during review of the LER or during the ensite review of the event.

  • The licensee's event report stated that the findings of this committee will be transmitted to the Region V office when available.

.4.P...

..-e.--w..,m.+e==Ee

>-~*e*-

=

~

  • * * " " ' " * ' - * * ' ' *