IR 05000250/1989025
| ML17345A709 | |
| Person / Time | |
|---|---|
| Site: | Turkey Point  |
| Issue date: | 05/22/1989 |
| From: | Casto C, Payne C, Peebles T NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION II) |
| To: | |
| Shared Package | |
| ML17345A708 | List: |
| References | |
| 50-250-89-25, 50-251-89-25, NUDOCS 8906060261 | |
| Download: ML17345A709 (14) | |
Text
~
~
Report Nos.: 50-250/89-25 and 50-251/89-25 Licensee:
Florida Power and Light Company P. 0. Box 14000 Juno Beach, FL 33408-0420 Docket Nos.:
50-250 and 50-251 Facility Name:
Turkey Point 3 and
Iicense Nos.:
DPR-31 and DPR-41 Date Signed Inspection Conducted:
April 6-8 and April 24 - May 4, 1989
,~/)
Inspectors:
C. Casto, Team Leader Team Members:
G. Weale B. Haa ensen"--'--
.r
.,
)
ayne, Lice Examiner Date'Signed Approved by:
T.
eebl s, Chief Operations Branch Division of Reactor Safety Date Signed Scope:
This was a special announced inspection of crew training and evaluations on the Turkey Point simulator in accordance with Confirmation of Action Letter 89-01, dated March 29, 1989.
Both the facility and Region II inspectors evaluated the ability of four newly formed crews of licensed operators to function as a team in order to allow continued operation of either Unit 3 or Unit 4.
The inspection was divided into two site visits: April 6-8, 1989 and April 24-26, 1989.
Results:
During the first site visit, it was noted that corrective action had been taken on some deficiencies identified during the requalification examination of March 1989.
In particular, a licensed operator, in training for a Senior Reactor Operator license, had been assigned a collateral responsibility in the control room to serve as communicator for the Plant Supervisor - Nuclear during plant events which require entering the Emergency Plan Implementing Procedures.
890602 02h1 890522 PDR ADQCK 05000250
PNU
This action allows the Plant Supervisor - Nuclear to better manage the event in progress and to increase his ability to monitor rapid changes in plant condition.
Additionally, it was noted that the position of Shift Technical Advisor had been upgraded with improved direction, procedures and training.
I However, weaknesses were identified with two out of four crews in their ability to properly classify events per the Emergency Plan Implementing Procedures.
The facility determined that terminology and format used in the Emergency Classification Table of these procedures were confusing to the operators.
As a result, improvements were made to the table and training provided to each operator.
(See Section 2 of the report details.)
A second site visit followed which included observing a portion of the above additional training and observing all associated crew evaluations on the simulator with emphasis on event classification.
Overall, the crews performed in a satisfactory manner.
(See Section 3 of the report details.)
No violations or deviations were identifie REPORT DETAILS 1.
Persons Contacted Licensee Employees-J.
Odom, Site Vice President-J. Cross, Plant Manager
="L. Lagarde, Plant Emergency Preparedness
.-gT. Finn, Plant Training Supervisor-7$G. Hollinger, Operator Training Supervisor L. Goebel, Operator Training B. Schimkus, Operator Training P. Finegan, Operator Training NRC Representatives-C.
G.
B.
//C.
"R.
Casto, Chief, OLS
Weale, Sonalysts Haagensen, Sonalysts Payne, Senior Operator Examiner Butcher, Senior Resident Inspector Attended exit interview April 8, 1989 Attended exit interview April 26, 1989 NOTE:
Acronyms and initialisms used throughout this report are listed in the last section.
2.
April 6-8, 1989 Site Visit The inspectors observed simulator evaluations of all four shift crews scheduled to be used to operate either Unit 3 or Unit 4 when the commit-ments of CAL 89-01 were fulfilled.
Two out of four crews exhibited weakness in their ability to properly classify events in accordance with the EPIP's.
Details are provided below. It was noted that a weakness identified during the March 1989 requalification examination had been addressed by the facility.
In particular, the PSN was unnecessarily burdened'n his Emergency Coordinator role by also acting as communicator.
In response, the facility assigned an RO, in training for an SRO license, a collateral control room responsibility of serving as communicator for the PSN during plant events which require entering the EPIP's.
This action allows the PSN to better manage the event in progress and to increase his ability to monitor rapid changes in plant conditio Shift Crew 4 (1)
Scenario Number 114 - Pressurizer Steam Space Ieak A pressurizer PORV fails open due to PT-445 failing high.
The crew recognized and responded to this event by attempting to close the PORV block valve.
The valve began to travel shut when the breaker for the block valve tripped causing the block valve to stop traveling.
The actual leak rate at the time was 65lbm/sec (450 gpm) of steam.
The crew completed a leak rate check by calculating the difference in PRT level change over time.
The crew determined the leak rate to be approximately 107 gpm while acknowledging. the rate to be only a rough estimate given that the leak was steam and not water.
The crew continued to mitigate the event by performing the actions of EOP, E-l.
At the time, SI flow was greater than 400 gpm and two charging pumps were providing makeup flow to the RCS.
However, neither the pressurizer nor the PRT were full at that point, so the high flow rates of these systems in themselves did not indicate a measure of the leak size.
Eventually, the pressurizer went solid and the PRT filled causing the rupture disk to blow (as. designed).
When this happened, the flow out the PORV could be assumed to be water.
Yet, with indications of SI flow remaining high and actual flow of approximately 450 gpm out the PORV, the PSN considered the leak rate to be 107 gpm.
The PSN classified the event at time T+8 minutes as an ALERT based on section 3 of the Emergency Classification Table.
The PSN did explain to the crew that he was waiting for containment pressure to reach 4 psig to reclassify the event as a Site Area Emergency.
This indicated that the PSN considered all other criteria for a SAE under section 3 to have been met.
The SAE Emergency Action Ievel under section 3 states in full:
1)
RCS water inventory balance indicates leakage greater than" charging pump capacity (210 gpm)
as indicated by:
a)
Pressurizer low pressure trip or RCS pressure decreasing uncontrollably.
AND b)
High containment pressure (4 psig) or increasing containment sump level or increasing containment radiation level.
AND c)
Pressurizer level decreasing rapidly."
After the PRT ruptured, the containment sump level was increasing (therefore satisfying item b) and the PSN was knowledgeable of this fact.
However, he failed to declare an SAE.
The root cause of this problem appeared to be the format
and wording of the EAL.
This EAL was designed for an RCS break into containment and assumes the operator will not be able to discern the actual leak rate.
So it gives him various key plant parameters to evaluate as an aid in reaching a
classification.
Early in the event, the PSN recognized he met each of the criteria above for an SAE except for item b).
However, later in the event when item b) was eventually met, item c) was no longer met in the PSN's opinion, because the pressurizer had gone solid and level was no longer "decreasing rapidly".
The PSN was further misled because he felt he knew the leak rate to be '107 gpm which was only half the general criteria of an SAE and therefore an ALERT classification was all that was necessary.
The operators must understand that though a pressurizer steam space break dumps into the PRT, thus delaying or at least misleading them in a proper classification of the event, it is still a LOCA and the containment boundary is being challenged once the PRT is ruptured.
Also, they must understand the basis behind each of the classification criteria and use this basis in conjunction with the wording of the criteria to evaluate the classification of the event.
(2)
Scenario 102 - Steam Generator Tube Rupture The crew identified the SGTR and manually tripped the reactor.
An Sl and Phase A isolation were also manually initiated.
They progressed through EOP, E-0 and into E-3 (SGTR).
The crew called for a chemistry sample and had the HP personnel rope off the turbine deck areas around the condenser air ejectors.
This was a leak path to the environment until the S/G was isolated.
The Chemistry Department completed a leak rate determination ( 5-10 minutes)
and informed the PSN that the leak rate was 450 gpm.
Several times during the event, the PSN called chemistry to verify this value.
The crew elected to man the TSC for support; however, they did not make outside notifications.
Based upon the leak rate determined by chemistry, the PSN classified the event as an UE despite the criteria given in section 4 of the classification table for an ALERT.
This criteria states in part:
" 2)
A rapid failure of steam generator tubes has occurred (leak of >500 gpm) as indicated by:
a)
b)
Valid alarm on PRMS R-15 or R-19 AND No significant increase in containment sump level AND One of the following:
(1)
rapidly decreasing RCS pressure, OR
(2)
reactor trip on low pressurizer pressure, OR (3)
safeguards initiation on low pressurizer pressure, OR (4)
one steam generator level increasing rapidly.
(See Section 3)"
The operators did have valid radiation alarms, no containment sump level increase and all of the following:
RGS pressure decrease Reactor trip (manual)
on low pressure SI (manual)
on low pressure One steam generator level increasing Also, section 4 has the operator loop back to section 3, Primary Depressurization to verify that the event should not be classified by that section at a higher levels In this situation, those criteria were not met.
However, the PSN should have realized that if a leak which is >50 gpm but less than charging pump capacity in conjunction with high process radiation readings is classified as an ALERT in section 3, then surely a 450 gpm leak with all other valid indications is also an ALERT condition.
Again, the PSN-was confused and relied on a number from the classification table (500 gpm)
compared to a leak rate based on an inaccurate chemistry sample to finely discriminate classification levels.
This action to obtain a
secondary chemistry sample is called for by the ONOP in order to determine which S/G is ruptured, not by the EPIP to classify the event.
b.
Shift Crew 2 (1)
Scenario 114 - Pressurizer Steam Space Teak This is the same scenario that shift crew 4 had trouble with and this crew performed with nearly the same results.
The PSN classified the event initially as an ALERT.
He did so apparently based on the fact that pressurizer level was not decreasing vice containment pressure had not yet exceeded
psig (see section 2.b.(1) above).
The PSN had the operators monitor for PRT rupture; however, when informed that the PRT had in fact ruptured, the PSN failed to reclassify the event.
The scenario guide showed that the licensee expected the conditions of the event to cause an escalation in emergency classification to a SAE after the PRT ruptured and thus provided a challenge to the containmen At the termination point of the scenario, the pressurizer was solid, the PRT was ruptured, the actual leak rate was 64.5 ibm/sec, RCS pressure was 1030 psig and SI flow was 429 gpm.
Also, two charging pumps were providing makeup to the RCS but the crew had secured one SI pump.
(2)
Scenario 102 - Steam Generator Tube Rupture During this event, the actual S/G rupture size was increased to greater than 500 gpm.
Also, no chemistry results were given to the PSN.
The PSN properly classified the event as an ALERT.
However, it was determined after the scenario by the NRC observers that the PSN based his classification on a combination of section 3 and section 4 ALERT criteria rather than the one correct section 4 criterion.
The PSN concluded that section 4, ALERT criterion 2)
(>500 gpm)
could not be determined since no chemistry results were provided.
He did believe that ALERT criterion 1) of section
was met by receipt of valid radiation alarms (he disregarded the coincident loss of power requirement also stated in the criterion) and that ALERT criterion 1) of section 3 was partially met in that the leak was
>50 gpm and there were radiation alarms.
The PSN informed the communicator of the ALERT based on sections 3 and 4, then informed the state of Florida that the ALERT was based on on section 4.
This was the correct classification but for an incorrect reason.
The problems discussed above further highlight weaknesses identified in the March 1989 requalification examinations.
During these examinations, one SRO classified an RCS leak which was much greater than charging pump makeup flow as an UE.
Later in the event he properly reclassified it as
.
an ALERT.
A second SRO also classified a
Neither he nor his crew corrected this error.
Another SRO during a loss of AC power, classified the event as an SAE though initially he did not classify the event as an ALERT.
The EPIP section for this event had a note which'tated that as soon as a loss of both 4 KV buses was experienced, the event should be classified as an ALERT.
The SRO read this note but still initially failed to properly classify the event.
Similarly, a third SRO did not initially classify the same event as an ALERT.
However, he was assisting the BOP operator during this time and was constructively engaged in controlling the plant.
Over 20 minutes passed in this manner before he finally entered the EPIP's, and then he immediately declared an SAE.
This event highlighted the need for a communicator.
Since the PSN felt that he could not be wrapped up in the EPIP's during the event, it went without being classified longer than permitted by plant procedure i,
3.
April 24-26, 1989 Site Visit Following identification of problems during the April 6-8, 1989 site visit, the facility determined that the Emergency Classification Table possessed deficiencies which significantly contributed to the operator's poor performance.
The sections of the table were felt to be unnecessarily obscure and confusing.
Additionally, the format of the (able was believed to be poorly designed from a human factors point of view.
As a result, a major revision to the table was made to correct these weaknesses.
After most of the changes were made to the table, a lesson plan and student handouts were developed by the Training Department to inform the shift crew operators.
This training was split up by shift crew into two classroom portions (Part A and B) and two simulator portions (Part A and
').
The second half of the Part B simulator portion was used to evaluate the PSN's ability to classify events and to give the entire crew a written examination on using the classification table.
The following training sessions were observed by the inspector:
Shift 2 Classroom Part B
Shift 2 Simulator Part A
Shift 1 Simulator Part B
Shift, 2 Simulator Part B
Shift 3 Simulator Part B
Shift 4 Simulator Part B
It was noted by the inspector that the Emergency Classification Table was in a state of change and had not been released in final form.
As a result, the operators were being taught information which was potentially inaccurate and subject to change.
The NRC is concerned that the facility was attempting to move too quickly in trying to make procedure corrections while concurrently providing operator training on these changes.
Teaching the operators procedures that are being changed as they are being taught is not a recommended teaching technique.
Operators that have to.unlearn a recent change because it was not correct are receiving negative training and could lead to operator confusion in the plant. It was also noted, as a strength, that feedback obtained from the operators during the training sessions was used to further improve the classification table.
The facility professed a policy of having the operators classify an event by evaluating the EAL's from the highest classification down (i.e., from General Emergency down to Unusual Event).
The purpose of this policy was to ensure the facility achieved the highest classification possible consistent with the facts of the event in a prompt manner while also ensuring the event was not underclassified.
One of the improvements
incorporated by the facility was to rearrange the Emergency Classification Table so that the operators could evaluate the table from the highest classification down and still read the table from left to right. It was noted by the inspector however, that the lesson material was not taught in a manner consistent with the above policy and the changes made to the table.
An overhead transparency of the revised table would be shown on a screen.
The instructor would then review the entire table with the class and emphasize the changes made as well as the basis for each table requirement.
However, instead of starting at the left side of the table with General Emergency and working his way across the table as the operator was expected to do in real life, the instructor would start at the right side of the table with Unusual Event and work his way to the left up to a General Emergency.
The technique provided negative training to the operators, the impact of which was later observed during the simulator assessments.
a
~
Shift Crew 3 (1)
Scenario 114 - Pressurizer Steam Space Leak During this scenario the APSN was acting in the PSN position and was responsible for classifying the event.
The PSN acted in the position of the APSN.
The operator correctly classified the event as an SAE due to section 2 of the Emergency Classification Table, however, his technique in using the table was not consistent with facility policy in that he read the table from right to left (Unusual Event to General Emergency).
Also, he overlooked several sections of the.table which were located on the bottom of the same page of another EAL.
This error did not result in a misclassification by the operator in this situation but highlighted the need to assess whether placing only one EAL per page would be appropriate.
b.
Shift Crew 2 (1)
Scenario 67 - Small Break Loss of Coolant Accident This was a fairly rapidly developing s'cenario.
The crew had indication of approximately
gpm flow mismatch between charging and letdown.
Additional charging pumps were started and letdown was isolated, however, leakage still indicated high.
The Reactor Operator then reported a flow mismatch of greater than 100 gpm.
A reactor trip with manual SI and Phase A initiation followed shortly thereafter.
From there the event proceeded through the EOP's in a normal progression.
In accordance with his training, the PSN initially classified this event based on the plant conditions at the time the EPIP's were first entered.
As a result, the PSN declared an UE based on section 2 of the table and tasked the communicator with making the required notifications (at the time, conditions were such that an SAE was in effect per section 2).
Two minutes later, the K'SN upgraded the event to an ALERT due to high radiation levels in the containment.
One minute later he changed it to
an ALERT due to a LOCA (section 2).
One minute later still, the PSN upgraded the event to its proper classification of SAE.
While the communicator did not have time to report any of the intermediate classifications, the PSN's method of classifying the event could lead to misunderstanding of the situation as it existed.
The more appropriate action to take in this event would be to evaluate plant conditions at the time of classification and classify the event at the highest level consistent with those conditions.
(2)
Scenario 58 - Steam Break Outside Containment During this scenario the APSN was acting in the PSN position and was responsible for classifying the event.
The PSN acted in the position of the APSN.
The operator correctly classified the event as an UE due to section 4 and 1 of the Emergency Classification Table, however, his technique in using the table was not consistent with facility policy in that he frequently read the table from right to left (Unusual Event to General Emergency).
Also, he overlooked four sections of the table when he reviewed only through the section 18 EAL.
This error did not result in a misclassification by the operator in this situation but highlighted the need for operator training to better emphasize reviewing the entire table during initial and follow;up classification activities.
c ~
Written Examinations All four written examinations that were administered to the shift crews were reviewed by the inspector.
With the exception of several typographical errors, the examination appeared to be well written and adequately covered a range of event and classification situations.
It was noted that while only the PSN and APSN were evaluated on the simulator at classifying events, both the SRO's and RO's took the written examination.
This was done to ensure minimum competency of the RO's in the event. they were required to temporarily act as Emergency Coordinator.
4.
Action on Previous Inspection Findings (92701)
(Closed)'FI 250,251/OL-91-01, STA Performance During Requal Exams This item was generated to follow up a licensee commitment to improve STA performance while assisting the operating during plant events.
During requalification examinations during March 1989 (see Exam Report 50-250/OL-89-01),
the STA's performances were judged to be less than satisfactory.
The licensee has since upgraded the training of these personnel and developed operating aids for the performance of their duties.
During the site visits of April 1989, it was noted that STA performance had significantly improved and that STA input to the operating crew was both timely and accurat,
Exit Interview An exit interview was conducted on April 8 and April 26 following each site visit.
The inspectors discussed in detail the results described above.
During a telephone conference call on May 4, 1989 between NRC and FPL management, it was determined that the Emergency Classification Table had been changed back to its initial format of evaluating event classifications from UE up to General Emergency as read from left to right.
This change was verified with a copy of the table obtained from the facility. It was also noted that only one EAL was listed per page.
Additionally, it was determined that the final version of the modified table was near approval.
In accordance with their SAT program, the facility will train the operators on changes made since their upgrade training was completed.
Proprietary information is not contained in this report.
Dissenting comments were not received from the licensee.
6.
APSN-BOP-CAL-EAL-EOP-EPIP-HP "
IFI-KV-IOCA-NRC "
ONOP-PORV "
PRMS-PRT-PSN-PT-RCS-RO-SAE-SAT-S/G "
SGTR-SI-SRO-STA-TSC-UE-Alternating Current Assistant Plant Supervisor - Nuclear Balance of Plant Confirmation of Action Letter Emergency Action Level Emergency Operating Procedure Emergency Plan Implementing Procedures Health Physics Inspector Follow-up Item Kilovolts Loss of Coolant Accident Nuclear Regulatory Commission Off Normal Operating Procedure Power Operated Relief Valve Process Radiation Monitoring System Pressurizer Relief Tank Plant Supervisor - Nuclear Pressure Transmitter Reactor Coolant System Reactor Operator Site Area Emergency Systematic Approach to Training Steam Generator Steam Generator Tube Rupture Safety Injection Senior Reactor Operator Shift Technical Advisor Technical Support Center Unusual Event