ML17081A538

2-Columbia-2017-02 Draft Outline Comments

ML17081A538
Person / Time
Site:	Columbia
Issue date:	03/09/2017
From:	Vincent Gaddy Operations Branch IV
To:	Energy Northwest
References
Download: ML17081A538 (10)

Text

DRAFT OUTLINE COMMENTS Facility: CGS First Exam Date: 2/27/17 Written Exam Outline Comment Resolution RO Question 4 K/A replacement does not appear Developed a question using the justified. original K/A.

Knowledge of the operational implications of the following concepts as they apply to MAIN TURBINE GENERATOR TRIP: Pressure effects on reactor level. (RO-3.5)

Reason for rejection: System does not operate like the K/A statement implies. From COLUMBIA GENERATING STATION SYSTEM DESCRIPTION, Volume 4, Chapter 2, NUCLEAR BOILER INSTRUMENTATION (NBI): Changes in reactor pressure influence the reference line and 1 variable line hydrostatic pressures in the same direction and magnitude; this effectively eliminates changes in level indication due to changes in reactor pressure K/A is asking for effects on actual RPV level due to pressure increase, not just indicated level. I.E., level initially decreases due to void collapse from pressure increase, then increases due to SRV/BPV opening, EOC-RPT and Feedwater.

Should be able to write a question that matches the original K/A.

RO Question 29, Incorrect method for replacing a K/A Resampled. New suggested used. If unable to write a question to K/A A1.05, must K/A: 203000.A1.03.

2 randomly select from other K/As within A1 - there are 6 others above 2.5.

SRO Question 94, The 10CFR 55.41/.43/.45 Developed a question using the references included with NUREG 1123 K/A original K/A.

statements are not absolutes. 2.1.9 does not list a 55.43 reference, but that doesnt mean you cant write 3

an SRO question to one. Directing personnel activities is an inherent SRO function. Re-evaluate ability to write an SRO question to 2.1.9. Question has an SRO importance rating of 4.5.

SRO Question 100: Same explanation as Q-94 Developed a question using the 4 above, regarding generic K/A 2.4.31. SRO original K/A.

importance rating of 4.1. OBDI 202 - IOLE Process Rev 2

Administrative JPM Outline (Date)

Comment Resolution Number the SRO Admin JPMs A-5 thru A-9 for Renamed SRO Admin JPMs distinction from RO JPMs. from A-1 through A-5 to A-5 1

through A-9 on SRO Admin JPM Outline (Rev 1).

Ensure RO A-2 is not identical to the JPM run in Changed JPM description on 2015, 2009. RO Admin JPM Outline (Rev 1) from Determine the Main Turbine Load Change Recommendation when raising Main Turbine load from 5% to 90% to Determine the Main Turbine Load Change Recommendation when raising Main Turbine load from 15% to 2 85%. Updated JPM to reflect changes.

Note to NRC: Did not see this JPM on 2009 exam per ADAMs document ML091030041 (Columbia - CGS - 2009-03 Final Operating Test). JPM uses different values from that on 2015 exam. OBDI 202 - IOLE Process Rev 2

Control Room / In-Plant System JPM Outline Comment Resolution For quantitative criteria at bottom of form, include the Attached actual numbers to 1 actual number for the attribute in parentheses, e.g.: quantitative criteria.

(A)lternate path 4-6 (5)

An in-plant Safety Function 6 EDG JPM has been on Replaced P-1 (Alt Path) with Alt the last 3 exams, and 4 of last 5 exams. Replace with Path In-Plant JPM (Restart of 2 a non-DG JPM for reduced predictability, such as RPS Motor Generator) (Both swapping inverter power supplies, etc. Safety Function 6)

S-1: (RRC) RRC-P-1A START IN MODE 1 . How is The only difference was the this different from 2015s? Not listed as P. Already circumstance that required at upper limit of Previous 2 exam JPMs without this securing the pump. In 2015, (3). the pump continued to increase in speed without operator demand which required securing the pump (pressing the Stop PB). On the current outline, the pump fails to come up to the required speed which, again, requires securing the 3 pump (by pressing the Stop PB). The same action is required in both cases but directed from different procedures because of the way the pump flow control system failed in each case. In hindsight, maybe not enough of a difference.

Replaced with another Safety Function 1 JPM (see below).

Overall a high degree of repetition from previous

• Replaced P-1 (Alt Path) exams. 9 of 11 direct from bank. 3 are from previous with Alt Path In-Plant JPM 2 exams (upper limit), possibly a 4th,, plus a 5th from 3 (Restart of RPS Motor exams ago. 4 exams in a row with DG-related in- Generator) (Both Safety plant jpm. High risk of predictability. Function 6)

• Replaced S-1 (Alt Path)

JPM (Safety Function 1) with non Alt Path Sim 4

Safety Function 6 JPM (TRANSFER BUS SM-3 FROM TR-S TO TR-N)

• Changed S-3 to Alt Path version to make-up for S-1 change to non Alt Path

• Reclassified S-3 from Bank to New JPM based on type OBDI 202 - IOLE Process Rev 2

of Alt Path which currently does not exist in our bank

• Removed Safety Function 6 JPM P-1 from SRO-U CR/IP Systems Outline and replaced it with Safety Function 1 JPM P-2 (this was done to ensure SRO-U JPMs met requirement for having all different Safety Functions). OBDI 202 - IOLE Process Rev 2

Simulator Scenario Outline Comments Note when reading resolutions: Scenario 1 is now Scenario 4 (Spare) and Scenario 4 is now Scenario 1 based on Scenario 4 NRC Comment 3.

Scenario 1 Comment Resolution Scenario 1: The first creditable I/C malfunction

• Deleted Normal evolution.

doesnt come until Event 4. An efficient scenario construct typically contains either 1 Reactivity event

• Deleted non-creditable OR 1 Normal event to start the scenario, and then event requiring just a Tech moves on to a Component malfunction in event 2. It Spec call and replaced with also would be preferable if one of the other existing Component malfunction I/C malfunctions could be made into a TS call, rather (RCIC coupling failure) than having to use event 3 as a stand-alone TS call requiring Tech Spec call.

with no creditable operator actions. Scenarios should To balance out the run 60 to 90 minutes, but if youre spending the first scenario, the ATC must be 1 30 minutes doing normal stuff it takes away from directed to carry out the evaluating safety-related competencies. Lets action of tripping the RCIC discuss. turbine while BOP monitors reactor parameters. RCIC controls are just to the left of the normal area of responsibility for the ATC.

• First creditable I/C malfunction starts at Event 2.

Scenario 1 Event 5: Is there a way to set up initial Unfortunately no. Two IRMs conditions such that a TS or TRM entry IS required would have to be inoperable in for IRM failed upscale? If so you can get rid of Event the same trip system for an 3 Rod Accum Trouble Alarm which is TS only. LCO Action statement to apply.

Having an inoperable IRM as part of the initial conditions requires the IRM Bypass switch to already be positioned to bypass the inoperable IRM.

Switch only allows one IRM in 2 given trip system to be bypassed at a time. There would be no Verifiable actions for the ATC when second IRM fails during the scenario.

Added new event (see Comment 1 resolution above) to address this. OBDI 202 - IOLE Process Rev 2

The ATC only has 1 I/C malfunction prior to the Major Added new event (see event, and only 1 after Major for a total of 2 - not a Comment 1 resolution above) 3 balanced scenario. to address this.

Scenario 1: Under target quantitative attributes, Yes.

EOPs entered/requiring substantive actions is intended to mean those EOPs entered post-trip in Counted PPM 5.3.1 entry 4 response to the Major Event. So event 4 entry into based on secondary PPM 5.3.1 would not count (but maybe youre containment flooding (Major) counting the entry per event 7?).

Scenario 1: Target Quantitative Attributes: should Updated wording.

5 say EOP-based Critical Tasks.

Scenario 1: Target Quantitative Attributes: In order to Deleted Failure of FDR-V-607 count as a creditable I/C malfunction, it has to have a from Target Quantitative verifiable action in response to it. Failure of FDR-V- Attributes table.

6 607 doesnt appear to have any verifiable actions (and is not credited as a component malfunction on D-1), so it should not count as a Malfunction after EOP entry in Quantitative Attributes..

Scenario 1 Event 7: Make the SW-V-29 failure to Made SW-V-29 failure its own auto open a separate event. event.

7 With scenario outline changes, what would have been Event 8 is listed as Event 6.

Scenario 1 Event 7: From training dept. experience, Some risk of scramming solely what is the risk of crew tripping the reactor directly in based on the OBE, especially response to OBE? Could cause loss of beans. when Suppression Pool leak develops. ABN-EARTHQUAKE directs controlled plant shutdown but also gives option to scram reactor if the plant has been damaged or indications exist that may affect the safe operation of the plant.

8 Removed minimum seismic event and OBE (since in ABN for both) to minimize possibility for scramming early based on ABN caveat allowing scram.

With only a single piping failure in the scenario, a seismic event is not required as stated in NUREG-1021 (Rev 10):

Mechanistic component failures are well-documented events that occur each year OBDI 202 - IOLE Process Rev 2

and often in multiple numbers.

However, non- mechanistic failures (e.g., pipe breaks) generally occur singularly; therefore, unless there is a connective precursor, such as a seismic event, it would not be realistic or credible to have several piping systems fail during any one scenario.

Scenario 1 Event 8: Merge the two parts into a single Merged and annotated as statement, and credit it as a Component malfunction. described.

9 Reactor mode switch fails to scram reactor, requiring With scenario outline changes use of manual scram pushbuttons to scram reactor now listed as Event 7.

prior to wetwell level lowering to 19 feet 2 inches.

Scenario 1 CT #2: What is the bounding threshold for Changed CT to read When successfully opening 7 SRVs? When would they fail wetwell level cannot be the CT? maintained above 19 feet 2 inches, initiate emergency 10 depressurization by opening seven (7) Safety Relief Valves (ADS preferred) within 10 minutes of wetwell level lowering to 19 feet 2 inches.

Scenario 2 Comment Resolution Scenario 2 Event 7: Failure of RWCU-V-4 to close is Split out from Event 7 as Event a separate creditable event. Split it out from event 7. 8. Previous Event 8 now Event 9.

1 ATC creditable event moved from Event 7 to Event 8.

Scenario 2 Target Quantitative Attributes: The SLC- Combined both under P-1A shear and P-1B flow blockage count as a single Reduced SLC injection 2 malfunction, as they will be under a single event, for capability and counted as 1 of Malfunctions after EOP entry in Quantitative the 3 total in Quantitative Attributes. Attributes table.

Scenario 2 CT-2: Explain better what exactly Changed CT to read Maintain maintain plant parameters entails. What are they RPV level above -186 inches.

physically doing / parameters controlling? Short excursions below -186 inches does not constitute 3 failure of CT provided level restored and maintained above -186 inches within 10 minutes of going below -186 inches. OBDI 202 - IOLE Process Rev 2

Scenario 3 Comment Resolution Event 5 is not a separate creditable event, it is the Merged previous Event 5 into required response to Event 4 which allows Event 4 to Event 4.

be called a Major. Merge as part of Event 4.

Note that Event 4 above 1 became Event 5 when another creditable event was inserted into the scenario (as Event 3).

See resolution for Comment 5 below.

Event 3 can be a Component malfunction or a Deleted Event 3 reactivity Reactivity manipulation for a single operator, but not manipulation to allow credit for both per NUREG 1021 ES-301: ATC Component malfunction.

With the exception of the SRO TS evaluations, each Reactivity manipulation will still 2 event should only be counted once per applicant; for occur but will not count for example, a power change can be counted as a credit.

normal evolution or as a reactivity manipulation and, similarly, a component failure that immediately results With scenario outline changes, in a major transient counts as one or the other, but Event 3 is now listed as Event not both. 4.

There are 2 creditable malfunctions after EOP entry, Changed Target Quantitative not 3. Attributes to reflect 1 creditable malfunction after EOP entry.

Note: The 3 listed creditable malfunctions (as previously listed) was the Steam leak in Main Steam Tunnel; MS-V-28A through D fail to auto close (relay fault); MS-V-22A will not close (mechanical). If we only 3 count the MSIV failures as 1 creditable malfunction (see NRC comment 4 below) and no longer count the MSLB as a creditable malfunction (based on NUREG reference in NRC Comment 2 above)., this leaves only 1 remaining Am I missing something? OBDI 202 - IOLE Process Rev 2

The MSLB in event 6 is a separate Major event. The Separated MSLB into separate failure of MS-V-22A, MS-V-28A-D is a separate Major event.

component malfunction.

4 The failure of MS-V-22A and MS-V-28A-D put under separate event as Component malfunction.

Scenario 3 has only one creditable I/C malfunction for Inserted new event into the BOP, and only 2 for the ATC. Not an efficient scenario as Event 3 to give distribution of events. BOP another creditable event opportunity.

5 Staggered ATC and BOP creditable events for better distribution of events.

CT#2 does not appear to have any bounding criteria. Changed CT #2 to read With Before a 3rd area exceeds MSOT perhaps? 2015 a primary system exam? discharging into secondary containment and area temperature exceeding maximum safe operating 6 level in more than one area, initiate Emergency Depressurization (ED) by opening seven (7) Safety Relief Valves (ADS preferred) within 10 minutes of second MSOT being exceeded.

CT#1 - Safety Significance - If secondary Replaced word approaches 7

containment temp exceeds its MSOT with exceeds.

Scenario 4 Comment Resolution Event 8 counts as a single malfunction after EOP Changed Target Quantitative 1 entry as it drives a single response. Attributes table to reflect 2 malfunctions (vice 5).

CT#3: Is there a required time limit, bounding criteria Changed CT to read After by which RPV level must be restored about TAF? ED, and within 10 minutes of RPV pressure lowering to 200 psig, restore and 2

maintain RPV water level above TAF (-161 inches) using Low Pressure ECCS systems.

This appears to be a more robust scenario than Will run as Scenario 1 while 3 Scenarios 1 or 3. May consider running this scenario. making original Scenario 1 the Spare (Scenario 4). OBDI 202 - IOLE Process Rev 2

GENERAL

1. Submit common forms in a single file. E.G., all 301-1s in a single file, 301-2s, etc.

(Will submit forms as described on subsequent submittals.)

2. Only 4 examiners are assigned to the team now. Consequently, while running scenarios there will only be 1 examiner free. So, all 3 in-plant JPMs will not be able to be administered on the same day, as originally scheduled. Most likely schedule solution is to run 1 in-plant JPM in parallel with scenarios Tuesday, and run the remaining 2 in-plant JPMs in parallel with simulator JPMs on Thursday. Alternate approaches are welcome. (Will change schedule as described.)

3. Confirm if class size is still 12. (Class size remains 12.)

4. Reformat schedule so that it can be printed on a standard 8.5 x 11 sheet.

(Will change schedule as described.)

OTHER FACILITY CHANGES MADE TO OUTLINE

1. Changed Examiner-directed Trigger numbers in each scenario outline such that they are numbered sequentially from first event to the last event. OBDI 202 - IOLE Process Rev 2