IR 05000409/1985019

Insp Rept 50-409/85-19 on 851023-25.No Violations or Deviations Noted.Major Areas Inspected:Followup on Alert Declared by Licensee & Actions Re Confirmatory Action Ltr on Flux Level Alarm Failing to Scram on 851023

ML20136H394
Person / Time
Site:	La Crosse File:Dairyland Power Cooperative icon.png
Issue date:	11/19/1985
From:	Boyd D, Eng P, Ring M, Villalva I, Joel Wiebe NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION III)
To:
Shared Package
ML20136H356	List: IR 05000409/1985019 ML20136H351
References
50-409-85-19, CAL-85-14, CAL-RIII-85-14, NUDOCS 8511250097
Download: ML20136H394 (11)

Text

.

-6

< .

U.S. NUCLEAR REGULATORY C0FNISSION

REGION III

Report No. 50-409/85019(DRS/DRP)

Docket No. 50-409 License No. DPR-45 Licenree: Dairyland Power Cooperative 2615 East Avenue South La Crosse, WI 54601 Facility Name: Lacrosse Boiling Water Reactor Inspection At: Lacrosse Site, Genoa, WI Inspection Conducte : October 23 through 25, 1985 (M. Ring @ lb'

Inspectors:

U"t*

MN. Eng (a -

Ul"gb_

,

Date 4 4 G j%

I. Villa D_a//-/

te 9- 8f J. Wiebe //-/f- f5

- Date Approved By: D. ief . //- / 7' M Reactor Projects Section 2D Date In_spe_ction n Summary Inspection on October 23 24 and 25,1985 (Report No. 50-409/85019 DRS/DRP Treas Inspectedi 3peciaT, i~e,a_m Wspe~c~tTon to followupbyon an alert

~

iTielicensee and to verify licensee actions related to Confirmatory Action Letter (CAL-RIII-85-14) regarding what appeared to be a failure to scram on the morning of October 23, 1985. The inspection involved 95 inspector-hours onsite by four inspectors including 36 ' inspector-hours onsite during off-shifts.

Results: The special inspection team determined that in the October 23, 1985, alert event The Nuclear Instrument Channel 6 (NI-6) was functioning acceptably and that there was no failure of the Reactor Protection System to produce a

- reactor scram. No violations or' deviations were identified, however, 3 unresolved items and 1 open item were identified requiring additional NRC inspection.

8511250097 851119 PDR ADOCK 05000409 G PDR L

-

.

..

'

-

.

DETAILS 1. Persons Contacted J. Parkyn, Plant Superintendent

• L. Kelly, Acting Plant Superintendent,. Assistant to Operations Supervisor / Training Supervisor

• B. Wery, QA Supervisor

• L. Goodman, Operations Engineer E. Moore, Reactor 3perator P. Moon, Shift-Supervisor

• Denotes those personnel.present at the exit interview.

Additional station technical and administrative personnel were contacted by the inspectors during the course of the inspection.

2. Synopsis of Event / Formation of Inspection Team Shortly;after 5:00 a.m. on October 23, 1985, during routine startup operations, the La Crosse Boiling Water Reactor (LACBWR) reactor operator observed an annunciation of the D panel " Channel 6 Flux Level (Hi)" alarm without the expected automatic scram normally associated with this alarm. The reactor was shutdown by manual insertion of the control rods and an alert was declared in accordance with the LACBWR Emergency Plan.

Following notification of the event, NRC Region III issued a Preliminary Notification (PN) of Unusual Occurrence,'PNO-111-85-92, describing the event and a Confirmatory Action Letter (CAL), CAL-RIII-85-14, to Dairyland Power regarding actions to be taken following the event. A special inspection team consisting of M. A. Ring, Test Programs Section Chief and Team Leader, P. L. Eng, Operational Programs Inspector and Inservice Testing Specialist, I. Villalva, Senior Resident Inspector (SRI)

assigned to the LACBWR site, and J. S. Wiebe, Senior Resident Inspector, formerly assigned to LACBWR, now assigned to the Duane Arnold site, was formed to follow up on the alert and to verify the licensee's actions relative to the CAL.

3. Confirmatory Action Letter (CAL) Verification Confirmatory Action Letter, CAL-RIII-85-14, from J. G. Keppler, NRC Regional Administrator, to J. W. Taylor, General Manager, Dairyland Power Cooperative, was issued on October 23, 1985, describing the NRC's understanding of actions to be taken by Dairyland Power regarding the alert event. -The letter was divided into eight action areas. Licensee actions and inspector observations regarding each area are described in the following paragraphs.

a. . Item 1 of the CAL: " Conduct a thorough review to determine if a trip signal was actually received by the Reactor Protection System (RPS)."

.

.

.

(1) This area (review of the event) of the CAL necessarily received most of the licensee's attention as well as that of the inspection team and dictates the course of action for several of the other CAL items. Efforts in this area were hampered by a lack of retrievable data regarding the event. This is primarily due to two factors. First, the LACBWR facility was not built with a Sequence-of-Events Recorder which most operating nuclear power plants have. Hence, a record of the alarm sequence or trip signals and actuation times was not available for review.

Secondly, the strip chart recorders available for nuclear instrument (NI) channels 5 and 6 had not been turned on prior to the event. Since NI channel 6 produced the event and NI channel 5 did not, one of the major areas of concern (CAL Item 4) was the difference in the observed actions of the two wide range, pcwer range channels (NI-5 and NI-6). By not having the strip chart recorders energized, the licensee and the inspection team were deprived of potentially illuminating information regarding the behavior of the two NI channels. Consequently, analysis of the event was primarily performed utilizing interviews of operations personnel (conducted by the SRI shortly after the event) and a reconstruction of the event performed by utilizing test signals to each of the power range NI channels. The inspection team reviewed the licensee's operating procedures as well as other industry standards and guidelines regarding operation of the NI-5 and 6 recorders and found that no specific requirements existed to ensure the NI channel recorders were energized. The LACBWR Operating Manual, Volume IV,

" Instrumentation, Control, and Electrical Distribution,"

Paragraph 4.5.1.(9) under Nuclear Instrument System Startup vaguely stated, " Energize recorders on Panel D-3 when required by turning the Instrument Power switch in recorder to ON". The recorders for NI-5 and 6 are located in Panel D-3. Even though the operating procedures did not specifically require operation of the recorders, good engineering practice would dictate operation of the recorders whenever the channels are operating and particularly during startup when either of these two channels (NI-5 and 6) has an increased probability (1 out of 2 logic versus 2 out of 4 logic at power) of being required to produce an automatic protective action. This issue was discussed with licensee management who committed to the revising their operating procedures to add specific requirements for energization of the NI recorders. Pending completion of procedure changes and subsequent NRC review, the topic of NI recorder energization is being followed as an unresolved item (409/85019-01).

(2) In order to conduct a thorough review of the circumstances the alert, an understanding of the LACBWR Nuclear surrounding Instrument (NI) system is necessary.The Nuclear Instrument System utilizes eight channels consisting of source range, channels 1 and 2, intermediate range, channels 3 and 4, and power range, channels 5, 6, 7, and 8. Power range channels NI-5 and 6 are wide range instruments capable of covering a range from 10-6 percent to 150 percent of full power.

.

.

.

.

Detectors for.all eight channels (ex-core type) are located in horizontal-instrument tubes which penetrate the biological shield at approximately the core centerline and run tangent to the inner face of the biological shield. The detectors are deployed at four 90-degree intervals around the reactor with NI channels 5 and 6 on directly opposite sides (east and west respectively) of the core. Both wide range channels NI-5 and 6 consist of a compensated ion chamber, a dual high-voltage power supply, and a picoameter. The meters are marked in two major lir, ear scales: one is from 0 to 60 percent, and the other is from 0 to 150 percent. Actual full-scale value of the picoameter is dependent on the setting of the range switch which selects the multiplier for the scale and provides a full scale range of 60 x 10 8 percent to 150 percent in 18 steps. In the upscale direction, NI-5 and 6 produce a warning alarm on the "C" panel at 110 percent, when set on the 0 to 150 percent scale. This alarm is labeled " Channel 5 (or 6) Flux Level (Hi)" and is produced by either channel detecting high flux levels. During a startup. NI-5 and 6 operate in a 1-of-2 logic while at power they operate in a 2-of-4 logic in conjunction with NI-7 and 8.

At 115 percent, NI-5 and 6 produce another alarm (also labeled

" Channel 5 (or 6) Flux Level (Hi)") on the "D" panel and a reactor scram which is indicated by a "D" panel alarm labeled

"All Rod Scram." The D panel has the capability of identifying the first-out annunciator indicating in red the scram signal which was received first. The C panel does not have this capability and alarms indicate in white.

(3) From interviews with the reactor operator and shift supervisor who were on duty at the time of the. event and from available logs, the following information concerning the alert was established. At approximately 5:00 a.m. on October 23, 1985, the plant had been in a normal startup with all control rods partially withdrawn. The 0500 log reading shows the control rod pattern to be almost symmetrical with rods 2-5 and 14-29 at 72 inches withdrawn, rods 1 and 10-13 at 16.8 inches withdrawn, L rods 7-9 at 17 inches withdrawn and rod 6 at 17.2 inches withdrawn.

Even though the rods were being withdrawn to establish operating control rod pattern and were responsible for the increasing flux, because of the relatively symmetrical pattern, the rod position was judged by the team to have had little effect on the event

! other than producing the increasing flux signal and contributing to the difference between channels 5 and 6. Shortly after i 5:00 a.m., during this routine increase in power, the operator

'

was alerted by initiation of the "D" panel high flux alarm (nominally set at 115%) on NI Channel 6. The alarm panel indicated high flux was a first-out signal (i.e. in red). The

<

"C" panel alarm (nominally set at 110%) had also occurred but it is not known if there was sufficient time between alarms for the. operator to have taken any action. Subsequent testing the the NI-6 alarm actuations showed both the C and D panel alarms to be occurring almost simultaneously. Since the D panel high

.

'

l

.  !

l flux alarm and the reactor scram are both nominally set at 115%, i the operator immediately looked for indications of a reactor scram. Failing to see the "All Rod Scram" light, rod position indicating scram, or the rod display which shows accumulator status indicating a scram, the operator shut down the reactor by manually inserting rods. The operator indicated he had observed NI channel 5 to have read approximately 100% and NI channel 6 to have read approximately 112% at.the time of the alarm with the NI 5 and 6 channels on the 150 x 1 5% scale.

The reactor shutdown by inserting rods proceeded normally without any further complications. The licensee's response (Attachment 1) to this CAL item described a slow increase in power wit:i a possible noise spike of a few percent as being responsible for the alarm activation. Review of the Intermediate Range Channels 3 and 4 recorders shows no evidence of a spike.

Since NI-5 and 6 recorders were not energized, further definitive evidence is not available. Upon review of the LACBWR Emergency Plan, the on shift operations personnel concluded that the Er:ergency Plan classifies an event requiring operation of shutdown systems with failure to scram automatically, but with successful manual insertion of sufficient control rods for subcriticality, as an alert. An alert was therefore declared at 5:39 a.m. on October 23, 1985. It appears that there may have been some confusion among shift operations personnel as to the proper classification of the event. This issue was not examined by the inspection team but will be followed by the SRI.

(4) In order to understand the operation of the NI-6 trip and alarm circuits, the inspectors utilized General Electric drawing No.

114B5426, " Trip Unit," attachment 2 to this report. In this drawing, K1 is the relay which leads to the D panel alarm labeled, " Channel 6 Flux Level (Hi)," and sometimes referred to as the trip alarm. The K2 relay feeds the actual trip or scram

. circuit and subsequently the "All Rod Scram" annunciator. The 110% C panel alarm is fed from a separate circuit. From this drawing, it can clearly be seen that any adjustments to the relay. set points are performed upstream (left hand side of the drawing) of both relays and would therefore be expected to affect both the trip and the trip alarm equally. The following is a general description of the NI-6 trip circuit:

The trip circuit consists of an oscillator circuit which controls a switching transistor. The NI-6 output (i.e., input to the trip unit) is compared to a reference value in the trip circuit. If the NI-6 output is less than the reference value the oscillator is allowed to oscillate. The oscillation turns on the switching transistor which energizes both the trip alarm relay (K1) and the trip relay (K2). When the NI-6 output is greater than the reference value the os:illator is turned off.

This turns off the switching transistor and'deenergizes both the

~ trip alarm relay and the trip relay causing a trip alarm and RPS trip signal. If the oscillator circuit should fail and

_ _ _ .

__ ___ __

.

.

.

stop oscillating, the switching transistor would turn off and deenergize.the trip alarm relay and the trip relay causing a trip alarm and RPS trip signal.

-With the inspection team present, the licensee performed testing of both the'NI-6 and NI-5 channels in an attempt to simulate the events which resulted in the alert while closely monitoring the behavior of the channels. This was done by injecting a flux. test signal via the installed test device in the the NI-channel and slowly approaching each of the setpoints while monitoring both the raw flux input and NI channel output on a strip chart recorder. Obse= vers monitored the response of each of the alarm lights, indicated percent power, and the NI channel drawer lights for trip and for trip alarm.

A review of strip charts made during testing shows that as the NI 6 output approaches the reference value, the oscillator apparently oscillates intermittently. This causes the switching transistor-to turn on and off rapidly. The voltage provided to the trip alarm relay and the trip relay also is turned on and off rapidly (alternating current). The response of the two relays and the circuits controlled by these two relays to the rapidly changing voltage is different. The difference may be a result of different relay-construction, different circuits being supplied, or a combination of these two effects. The trip alarm relay supplies a solid state annunciator system which locks the alarm in to provide first out indication. Therefore, when the rapid switching of the switching transistor caused the trip alarm relay to " chatter" the solid state annunciator system would lock in the first out alarm, giving indication of a trip signal.

The trip relay, however, feeds anothei -hanical relay.

Depending on the magnitude of,the oscillo, the chatter of the trip relay (clearly visible on the drawer lights during testing) as a result of the rapid switching of the switching transistor may not cause the mechanical relay being fed to deenergize until the effective " root mean square" value of voltage drops below a certain value.

During a simulation of a slow approach of NI-6 output to the reference value using the test signal, it was noted that the trip alarm annunciated at a lower level than the level at which'the trip relay would completely deenergize. Between the level at which the trip alarm would actuate and the trip relay would completely deenergize, the trip relay would clearly

" chatter." The trip relay was tested in this manner several times and completely deenergized at a setpoint consistent with Technical . Specification requirements during every test.

In the case of NI-6, it was also observed that the 110% C panel i alarm occurred almost simultaneously with the 115% D panel trip i alarm.

- - _ _ _ _ _ _ _ _ _ - _ _ _ _

-_

.

.,-

. .

~ The. apparent malfunction noted would always cause the trip-

. circuit to deenergize the trip alarm relay and the trip relay

~ atia level.less than~or equal.to the trip setpoint. .The

~

inspectors'could not postulate a further degradation of_the-circuit _that_would cause a failure to trip or a trip greater than the setpoint and still be consistent with the observed responses of the instrument during and following the event.

With respect to Item 1 of the CAL, the inspection team concluded that the date to fully. document whether a trip signal was actually received _at the time of the event is unavailable.

However, the inspection team believes: that the phenomena of the trip alarm signal occurring at a slightly lower flux level than the trip as oescribed abcve was, in fact, what occurred; that-no trip signal was actually recieved by the Reactor Protection. System (RPS); that no ATWS (Anticipated Transient Without Scram) condition. existed and if a trip signal had been received.by the plant, it would have scrammed.

b. Item 2 of the CAL: "I'f such a signal was received, determine why the control _ rods failed to automatically insert."

.As' described in the. licensee's CAL response (Attachment 1) and Item 3.a. of'this report, the licensee concluded that a trip signal;was not received. The inspection team agreed with this conclusion.

c. Item' 3 of -the CAL: "If such-a trip signal was not received, determine'why the annunciator was activated."

As described in Item'3.a.(4) of this report, the annunciator (clarm)

was_found during testing to be actuating approximately 2% lower than the trip on NI-6. It is believed that the flux signal.seen on NI-6

was sufficient to actuate.the alarm at the lowered actuation point but not sufficient to actuate the trip'at the correct setpoint. This

' separation in actuation points was demonstrated via test several times. In allicases, the trip relay completely deenergized at a setpoint consistent with; Technical Specification requirements.

d. Item 4 of the CAL: " Determine what caused the signal that nuclear instrumentation' observed and why nuclear instrument channel number 5 was not affected."~

.The signals being described by NI-5 and NI-6 are believed to both have been. valid signals proportional to the increasing flux being developed _in~the reactor as power was being increased from the source range through the. intermediate range into the power range. At the time of the event, power was in the 1 5% range which represents the low end of the intermediate range. It is not unusual for there to be a small' difference between detectors NI-5 and NI-6 at this level.in a startup; The 0500 log readings showed NI-5 to be at 56% and NI-6 to

be at 62%. The operator indicated that at the time of the event he observed NI-5 to indicate about 100% and NI-6 to indicate about 112%.

-

.

.

.

The conclusion reached by the inspection team was that NI-5 was unaffected because it was reading 5-10% lower than NI-6, but that

.NI-6 did, in fact, respond to a. valid flux signal with the alarm function actuating about 2% lower than its expected band.

e. Item 5 of the CAL: " Maintain all affected equipment related to the Reactor Protection System in such a manner that it can easily be

.kept or placed in the "as found" condition. Therefore, take no action which would destroy or cause to be lost, (other than necessary to protect the health and safety of the public) any evidence which would be needed to investigate or reconstruct this event."

-

As described in the licensee's response to the CAL (Attachment 1),

this portion of the CAL was not strictly followed. Shortly following the event, licensee personnel felt the alarm and trip setpoints on NI-6 may have been set too low and performed an adjustment to ensure they were set within the proper band. While this action is contrary to the CAL wording, it was performed shortly after the _ event when there may have been some confusion as to the specific requirements placed on_the licensee by the CAL and before the licensee had received the actual CAL (approximately 7:30 - 8:00 a.m. on October 23,1985). However, as described in Paragraph 3.a.(4) of this report, any adjustments to_the setpoint are performed upstream of both relays K1 and K2 and would affect both thel alarm-and the trip equally. For this reason and because NI-6 could be returned to its as found condition, the adjustment performed by the_ licensee

.was judged by the inspection team to not'have impeded the investigation or reconstruction of.the event.

f. Item 6~ of the CAL: " Review operator actions taken immediately following his recognition of the failure to trip and determine if these actions were in accordance with your procedures and policies (specifically, determine why the-control rods were manually inserted rather than by inserting'a manual trip signal)."

(1) The licensee's response to this item states that the operator's

~

indication of power level was approximately 110% which is below the nominal 11S% scram point and that the ATWS procedure allows either a mantal scram or control rod insertion. The operator did, in fact, reduce power by control rod insertion. The-licensee's ATWS procedure describes symptoms of an ATWS event and dictates certain actions in order to enable operators to recognize such an event and properly carry out corrective action. Of the listed symptoms, only one (4.12.3(1)) was applicable to the event in question. Symptom 4.12.3(1) states,

" Reactor scram initiated without the expected decrease in reactor power due to the failure of control rod insertion."

The inspection team believes (as described in Paragraph 3) that this symptom-in fact, did not occur in that power level was below the scram setpoint, no "All Rods Scram" was received, and no actual trip signal was initiated. However, because the trip alarm (Flux Level (Hi)) and the trip are set at the

.

.

.

same trip point, the trip alarm was received and the trip was not; the operator had indication that he might have an ATWS condition. Because the power increase had been relatively slow as indicated by intermediate range recorder traces, the operator

~

could tell by looking at the nuclear instrumentation that there was not an immediate safety concern due to some kind of power excursion hence he chose to insert rods rather than scram. The immediate actions required by the ATWS procedure are as follows:

" Attempt to manually insert control rods by:

(a) Depressing manual scram pushbutton.

(b) Depressing all rod insert pushbutton."

The licensee indicated that their training is such that the above words would provide the operator with a choice of either action. Because of the specific wording and its related training at LACBWR, the inspection team believes that it is debatable whether or not the operator should have considered the plant in an ATWS condition under symptom (1) and whether or not he should have scrammed under the immediate actions.

However, the inspection team has concerns regarding the procedure wording and training on how to implement the procedure such that an operator may not initiate a scram _in an event where the reactor protection system calls for a scram and one is not received. The licensee is reviewing this area for possible procedure and/or training changes. This subject will be treated as an unresolved item pending further inspection (409/85019-02).

(2) While the CAL only addressed operator actions following the event, there was also a question regarding operator actions

,

prior to the event. The licensee's startup procedure step i

2.1.2(4) states, "Each time the neutron flux indication on the wide range Channels Nos. 5 and 6 approaches 75 percent on any 150 range scale (or 30 percent on any 60 range scale), upscale the switches to the next range." The operator did not do this.

! Had the instruments been upscaled at the proper point, the set point for both the alarm and trip would not have been reached and the event would not have occurred. The malfunction in the oscillator circuit would still have been present and may have manifested itself in a different fashion at some later point.

Failure to follow the startup procedure is considered a violation of 10 CFR 50, Appendix B, Criterion V; however, no

notice of violation is being issued in this case because the

'

violation was identified to the inspection team by the licensee and appears to meet the tests of 10 CFR 2, Appendix C, IV.A.

It should be noted that failing to upscale is actually a conservative action in that it simply causes the reactor to trip sooner. However, failing to upscale also may precipitate on unnecessary challenge to a safety system. The licensee has

.. -

.

~

-

-

.

!

- , -

'

.

committed to review the circumstances surrounding the event, including the need for upscaling, with each of the operating shifts. This action was partially complete at the time of the exit and will be followed as.an unresolved item pending completion of the corrective action (409/85019-03)

g. Item 7_of the CAL: "A formal report of your findings and. conclusions will be submitted to the Region III office within 30 days."

The licensee's report was received in the Region III office on-October 24, 1985,- and is included as Attachment 1 to this report.

h. Item 8 of the CAL: "We further understand that startup will not occur. until authorization to restart is obtained from the Regional Administrator or his designee."

The inspection team followed the licensee's investigative and testing actions previously discussed in this report. The inspection team discussed its findings with cognizant Regional personnel and with NRR representatives,-R. Dudley and G. Holohan. The Regional Administrator's designee, D. Boyd, authorized the plant to resume operation at approximately 2:30 a.m. on October 24, 1985.

No additional violations or deviations were identified; however, three areas require further review and are considered unresolved items.

4. Additional Observations a. In order to' preclude recurrence of this event,- the licensee replaced th'e NI-6 drawer with a spare drawer. This replacement drawer was then carefully tested in the same manner as the suspect NI-6 drawer had been tested. During this testing the inspection team observed that it was possible to make NI-5 and the replacement NI-6 behave in a manner similar to the suspect NI-6 by very slow application of'the test signal. However, the. magnitude of the difference with which the alarm relay would precede the trip relay was less than .5% (barely discernible) as opposed to approximately 2% for the suspect drawer. .In the opinion of the inspection team, this does not constitute a problem for continued operation. The licensee then reperformed normal surveillance testing for NI-5, 6,

-7 and 8 to verify proper operation of all four channels. In the opinion'of the inspection team, normal surveillance testing, as demonstrated by the licensee's personnel, would not be likely to discover the alarm relay actuating slightly early unless a very careful, slow approach to the setpoint was used (i.e., unless the technician was looking for this problem) or a malfunction of greater magnitude occurred, b. To further confirm the proper operability of the nuclear

instrumentation high flux scram capability, the inspection team

! requested the. licensee to demonstrate a successful scram utilizing

! the NI-6 trip signal. The licensee agreed and performed this action

! by withdrawing a signal rod.in accordance with approved plant l

1 L.

_ _ ___________________

.

.

.

procedures for rod scram timing. The rod was then scrammed by turning the NI-6 range switch all the way downscale until the background was sufficient to reach 115% on the 150% scale and trip the reactor. All systems, including rod position indication, "All Rods Scram" alarm, both " Channel 6 Flux Level (Hi)" alarms, the drawer lights, and accumulator pressure indication, functioned correctly in this demonstration.

c. Because of the quarantine placed on affected equipment by CAL Item 5, the. licensee was unable to disassemble the removed NI-6 drawer to troubleshoot and determine if any individual components had been degraded in any manner prior to the exit. This effort is continuing and will be followed as an open item (409/85019-04).

No violations or deviations were identified; however, one area requires further review and is considered an open item.

5. Open Items Open items are matters which have been discussed with the licensee which will be reviewed further by the inspectors, and which involve some action on the part of the NRC or licensee or both. An open item disclosed during the inspection is discussed in Paragraph 4.c.

6.~ Unresolved Items Unresolved items are matters about which more information is required in order to ascertain whether they are acceptable items; items of noncompliance or deviations. Unresolved items disclosed in the

,

inspection are discussed in Paragraphs 3.a.(1), 3.f.(1), and 3.f.(2).

l l

7. Exit Interview

The inspection team met with the licensee representatives denoted in

!

Paragraph 1 on October 25, 1985. The inspection team summarized the scope of the inspection and the findings. The licensee acknowledged the

,

statements made by the inspection team with respect to the open and

,

unresolved items and the violation described in Paragraph 3.f.(2). The

!

inspection team also discussed the likely content of the inspection report with regard to documents or processes reviewed by the team during the inspection. The licensee did not identify any such documents /

processes as proprietary.

_ _ - _ _ _ _ _ _ _ .-

_ . - - ___ . _ _ _ _ _ . . _- _ . . .-- - _. . _ __

'

OCT 24 '85 11:15 P02 PM* "8

,

.. ' OCT 24 '85 11820 1110 P02

.

ATTACHMENT 1 i

!

l B I COOPERATNE P.o sox en . reis ust av souts u oosst usemsw we (000) 706 4000 l

October 23, 1986 In reply, please j refer to I.AC-11211

D0C E T NO. 50-409 Mr. James Regional Achinistrator U. 8. Nuclear Regulatory Commission Region III 799 Roosevelt Road-Glen Ellyn, IL 60137 SUBJECT: DAIRYLAND POWER COOPERATIVE LA CROSSI BOILING WATER REACTOR (LACBWR)

PROVISIONAL OPERATING LIC2NSE NO. DFB-45 manama! TO CONFIMATORY ACTION LRTTER Reference: (1) Letter Emppler to Taylor, dated October 23, 1985

Dear Mr. Esppler:

You sent us a Confirmatory Action Letter regarding an apparent improper response of the reactor protective system (Reference 1). We have reviewed the items listed and our response follows.

'

M 11mf1 Clasanct a thoraangh misw to Atarnise if a trip aismal actually man moeived by the nonctor Protection Spute (ME).

RfC RESPONSI

!

A thorough review has been conducted. It has been concluded that a scram  !

signal was not received. An alaru signal was received which gave one visual indiestion of a possible trip condition. i

i M IfEV 2 1/ seat a alamm1 aus receive 4 hearafar adtr the emotimJ sed Almiled to autamuticelly insert.

DPC B5P0153 N/A v"W g ; C

,, ,- ,, s

,

,

I OCT 24 '85 11815 P83 N4 Ne 'P

. ,

OCT 24 '85 11821 1110 P03

'

f ATTACHMENT 1 Mr. James G. Reppler October 23, 1985

,

! U. S. Nuclear Regulatory Commission I.AC-ll211 page 2

! Region III

i h M 11St 3 i

If such a trip algael non not noeined, determine edly the annunciater ame j

actiewted.

DPC RESPONSE Due to an apparent failure of an oscillator circuit on the trip stage of NI-6 that should have ensured that both the alors and trip relays actuated at the same time; the annunciator canne in approx. 'Fi power before the trip signal actuated. A very slow increase in power to approx.1104 indicated on the 150 x 10-st scale, in conjunction with a possible noise spike of a few percent activated the annunciator but not the trip. We were able to duplicate the separation between the annunciator and the trip releys many times, with the

"same results. This has been observed by the NBC Besident Inspector and the personnel sent fram Negion III.

M11W4 Betermine anbet conned the signal that nuclear inutnmentation sheerred and entr nuclear imetriaaset channel nam 6er 5 aan not aWiscted.

DPC RESPONSE At this power level (10-st) there was a spread of 64-10% between NI-5 and NI-6. The 0500 readings on 10/23/86 abowed NI-6 to be at 664 and NI-6 to be 82% on the 150 x 10-** scale. This spread would allow NI-6 to reach an actuation condition without NI-5 reaching an octuation point.

M5 Akistain all affected equismaet minted to the naector ypotection craten is auch a naamar that it can anelly be knpt er placed in.the *an fened" cuadifiae. Sarafare, tate no action atlet meeJd duetswy er comes te he last, (other than nacaesary to protect the health and anfety of the public)

any evidance adtiet amald he needed to investigste er nacenetruet this event.

DPC RESPONSE Although some minor adjustment was done on NI-6, it can be placed in the as-found condition. No further action will be taken on the dresser, pending NRC concurrence of the course of action deessed necessary.

M 11EV 6 Arrieur eqparater actione tahan Jamediately following" his reansmitian af the f>llure to trip and deteraise if thane actione ante in accernemoon erith your i J and polician (specifian117, determine a6y the centrol roer naare manualir lacerted rather than by ineecting a amenal trip aisual.)

OCT 24 '85 lit 16 P84 ## "

, ,

' OCT 24 '85 11:22 1110 PO4

.

, ATTACHMENT 1 Mr. James G. Emppler October 23, 1985 U. S. Nuetear Regulatory Commission LAC-11211 Region III page 3 DPC MSPON$E The operator's indication of an apparent trip condition was annunciator D5-4, his power level indication was at 1104, which is well below the nominal scran point of 115%. Be immediately reduced power to sub-critical by inserting #1 control rod. The ATWS procedure allows either a manual scram or control rod insertion. (Ref: IACBWR Op. Manual Vol. I, Sec. 4.12.) This was accomplished.

M.221tf f .

t A forum 1 report of posur fladians and casc[uskaan edil be sa6mittad to the narico III office withis .N dow.

BPC MSPONSE This letter constitutes the requested response.

M EMV 0 hir fkarther nederstand that startap will mot occur sutil metherisatlan to rautart la obtained than the nnglamal khalaistrator er his daniginee.

DPC M SPONSE The investigation has been discussed with and testing perforised under the cognizance of NRC representatives. Restart will await authorization by the Regional Administrator or his designee.

Very truly yours, DAIRYLAND F0WER COOPERATIVE I James W. Taylor, General Manager j FL:JDP:IMK:dh cc - Document Control Desk U. 5. Nuclear Regulatory Casamission Washington, D. C. 205h5 NRC Resident Inspector J. Stang, NRC Project Manager M ET C 118 30

_ ._ _

-_-

_ _ _ _ _ _ _ _ _ _ _ _ _ . _ _ _ _ _ _ _ _ _ . . _ . _ _ . _ .

) > . . '

'f;,2:s, :n. s. : s

.

.

.

.

,

ATTACHMENT 2 rJ ,,

.

.C .'J.,;;; il'f r . .. ,

, O .#

.'ii.#

.

[

g , - L ~ < (n . .

-

.

so go a e - .nsa.ee

.

.'

? ' j."

f

? & y r .' n..,

r . , . . ,

y Ns G Q;::

. =

,N <e. yo  : ts p ,, =

-

• t-

<i

@ w2 ,

g)  ; T u _.n;; .

p a O

0, <

-t- -

T ,

r - v :. :. ,

. .,

,

g g f\ [, k k I '. P. .

~

,E J .; C[h'

N  :.,x. -te..,-...g..y. .' x? .- ..u.+ . . - -

q,l U o e 3J y

,

. .ll Wl ==8*

.

g g

ga

,

- ' .

a

% N .

& 4.

4 "> E ,

.

~ *

h (ek

====.9 t. f l'

' d y , ,n 4

a a.

> g

.,

,

% g al .,g 4' 3 d,r %'. ?

j _

7. . ,<

" g% ., )- g I

! A .,

,/ Y

] { ! 4

."~i-d' r e w t 4+y '- w,. . .' -3 Y

=

D"w 2nW o 7 m8 4 e 5 .- Q.oy

-

N N *

-

..

"I o ~ 4'#

t' j ' -* C ( s , -

- i

>

O

"'

Q 3 u0 2 3 G e s

ts

. .

rat ed O.,

e " . 3 ... v.-: ,- w

a,, .,

. ,

,

t. -m w.'*.o e . , <

5 ,. . . .

E u b2

,,

e en

' s'

.

ta.

r - .

O_

j Q

-

l

.

i,

- . .

o.4' o-

-

' * ^ #

--

-r, s c m ~wm:

' ' -

'

f. g f ,L. Jnh,;l dh4 y y t 7 .- 4(4 {. .a c{ M m ik w a ,. g i N,

.

y [NT y 'y at

@

h t

i

$;~6;-D

'" ' 4 Q .TL;N

.- ', A.4). 3 I

an

&s v Wi"

,.

V-- gl .Mi9;r a

!3 k. -.t. a

N*

n (r ES- ep@p N

Ai f M L,p

-

e Yblb.1:

n.w mz rl . _1 , pe wo,

'

og 7 ,. ..n S

-

p

-

,.

[ o .,, t afe yV4 ,

m

{ml.f Y{mhy.l. % .

.*N

.

g

.. TC y

c'

> V 5f::n ?%.: a y _.-l . .

,

, , e

,

, y s.;.:.. c = ,n . . + . . . . ..

. , . .

ar e es}d

.

y I & Wy 4. . 21. v .f,.t v -

.

-

.

t .n g

-

u'H , Opus 4,-..v. ,y ,u a f

,t

..

. ., - c

' 6. -!. 2.f .

,

O r- Mss Qip

-

s '-

-

y- . A y4 .

f it ' _ e . . . , .

) 1 0 4 0 % NW .: .

n., .w ..,.. %.. s - + . -. s.

':

7 2 a .s .

i.,_^_ .l ._

'

. "

' ')

p D

f.i

,

e s o n i 3 9yv'e~.

o 'So.

, 2 s v.

l 1 .

,,

ce er,m

-

ra 44 47- pno61

.,sL 0 _ g i

'

l,

'

-

.. -

s. ;l* ; U "A

! * cm 4 g 3-

.

gy

.

. v r- g , *

,

, 1-1Y l ,nC

,h ===

i N001 l g g V  !

j "{ pq ..!!'"

l .c;* i . .' , , , .

1, ; w .n g- N

.

O g h a p ' , e , " . @; s ';.' .

v : c' e. . o +< g;~ 4 ',~ y;; g-. e ( : .h u. =n.

.a .

d >p w y o ,J e; , . .c.. s -

-c

,

.e ei ,i yQ l. - -

.

> " " " " " " " '1g - ei g

'

b. ,, J 2 , f.' -[' . ,, ' ,l

ta 4 OI ,, ,

I [. ,

• O

us

',e s:

d' 5 -

h '

, .

K Iigg / 3 2- 4 g

b  :

,

yW ' u.,. 1 :,o n

. -

.

,,,

-

.

g.g .,

ga . ,

,

.r y .wg....v .,.

'

a..

, ' * ' -  :

~

W e '

y[' 5 . ' * - %M'

qf,+, g .,9 g, -

t.: r m . .,

ej$

.

g ,, , 2, o' '. +- : '

g W EO

~l f.

a

o,, a'8sa. $ A

'

5 : A; geg I 5<4

-

j r$

a f. * ~ s l < #..'.f.x

.

.

. 'R . '. . -

4 '

!

.f. (? ,'

s,g. . . . . os .. .

.. . . . .

- -

-

,

q d _'.-(.

'

+ + +

.y 4 I i f ,, , w *,' _y.

y r: ,

- b

,

,

ti $' _h_ 'l'*: ]. l