ML091830616
| ML091830616 | |
| Person / Time | |
|---|---|
| Site: | Palisades  |
| Issue date: | 06/10/2009 |
| From: | John Ellegood Division Reactor Projects III |
| To: | |
| Stacy Rosenberg | |
| References | |
| Download: ML091830616 (14) | |
Text
U.S. NUCLEAR REGULATORY COMMISSION NON-CONCURRENCE PROCESS SECTION C - TO BE COMPLETED BY DOCUMENT SPONSOR NRC FORM 757 (3-2009)
The TIA Program Manager and members of the Technical Specifications Branch have met in person and by phone with the non-concurring individual in order to come to a common understanding of the concerns expressed and potential means to address them. See attached documentation of the e-mail correspondence and summarization of concerns that formed the bases for discussion. The TIA has been revised to take the concerns into account to the extent possible.
PRINTED ON RECYCLED PAPER NAME Stacey Rosenberg TITLE Chief PHONE NO.
301-415-2357 ORGANIZATION NRR/DPR/PSPB ADAMS ACCESSION NO.
ML091340595 TITLE OF DOCUMENT TIA-reevaluation of Implementation of LCO 3.0.4a CONCURS NON-CONCURRING INDIVIDUAL (To be completed by document sponsor when process is complete, i.e., after document is signed):
NON-CONCURS WITHDRAWS NON-CONCURRENCE (i.e., discontinues process)
CONTINUED IN SECTION D Use ADAMS Template NRC-006 NRC FORM 757 NRC MD 10.158 (3-2009)
DATE 09/29/2009 SIGNATURE - DOCUMENT SPONSOR WANTS NCP FORM PUBLIC WANTS NCP FORM NON-PUBLIC DATE 09/29/2009 SIGNATURE - DOCUMENT SIGNER ACTIONS TAKEN TO ADDRESS NON-CONCURRENCE (This section should be revised, as necessary, to reflect the final outcome of the non-concurrence process, including a complete discussion of how individual concerns were addressed.)
/RA/ Stacey Rosenberg
/RA/ Thomas Blount
TIA 2009-005 as currently written J. Giessner Non-concurrence concerns J. Ellegood Non-concurrence concerns Proposed way ahead to address concerns Quotes GL 87-09 statement that The restriction on a change in operational modes or other conditions should apply only where the Action Requirements establish a specified time interval in which the LCO must be met or a shutdown of the facility would be required.
GL 87-09 never explicitly says that the action did or did not have to be done before the mode change occurs; it states conformance to the Action Requirements establishes an acceptable level of safety for unlimited continued operation of the facility.
The GL does not clearly establish that Mode transition may occur prior to completion of required actions and the letter provides no argument to establish an interpretation of the GL 87-09 verbiage. The GL does state "Conformance with the action requirements establishes an acceptable level of safety for unlimited continues operation of the facility." This statement implies that GL intended to require completion of the action requirement, since, if not completed, an acceptable level of safety would not be established.
Revise the TIA to address the lack of clarity here. The fundamental difference between the way this regulatory position was expressed in GL 87-09 and how it is now expressed is that in the GL, and the Palisades amendment to implement it, the limitation on mode changes was expressed in a negative sense, i.e. the restriction applies only where, while in the current version the limitation is phrased in a positive sense, i.e. shall only be made when.
An Action Requirement that is within the limits of its associated Completion Time is not an Action Requirement with an established time interval in which the LCO must be met or shutdown of the facility would be required, and therefore the regulatory position of GL 87-09 would allow for conduct of a mode change under those circumstances.
TIA 2009-005 as currently written J. Giessner Non-concurrence concerns J. Ellegood Non-concurrence concerns Proposed way ahead to address concerns Uses the regulatory position of GL 87-09 and the changes in the STS/ITS conversion as an aid to interpretation of the current TS and bases.
- 1. The central reason the TIA was reversed was review of GL 87-09 by the staff at NRR. The staff asserted that the TIA was not aligned with the GL and that key language from GL 87-09 was improperly translated during development of ISTS
[Improved Standard Technical Specifications]. The key section is that the GL solution to unnecessary restrictions on mode changes by Specification 3.0.4 also resolved the problem of inconsistent application of exceptions to Specification 3.0.4 which delayed startup under conditions in which conformance to the Action Requirements establishes an acceptable level of safety for unlimited continued operation of the facility. Application of LCO 3.0.4a
[improved or old format] STS was not intended to result in more restrictive requirements for individual specifications. TS 3.0.4a provided the allowance in one section of TS versus specific sections of TS allowing exceptions to TS 3.0.4.
- 2. Licensing basis is TS, its bases and interpretation, not GL 87-09. GL 87-09 can only be used to assist in determining the TS interpretation.
A generic letter supersedes an approved license document.
(That is, we have given greater weight to the regulatory position of GL 87-09 than to the current licensing basis for Palisades.)
Revise TIA to include Palisades amendment 130, which implemented GL 87-09, amendment 189 and its SER, and amendment 219 and its SER to show the traceability of the GL 87-09 regulatory position in the licensing documents for Palisades.
TIA 2009-005 as currently written J. Giessner Non-concurrence concerns J. Ellegood Non-concurrence concerns Proposed way ahead to address concerns To conform to the guidance in GL 87-09, the improved STS bases language should be interpreted to state that compliance with LCO 3.0.4a relies on both the Required Actions and associated Completion Times as they would apply to the discovery of inoperable equipment while operating in the applicable mode or other specified condition of an LCO.
- 1. The question the TIA tries to answer is this: does TS 3.0.4 a require the associated ACTION, that permits continued operation for Palisades plant (in TS 3.3.3 Action A.1) for an unlimited period of time, to be done before the mode transition occurs. The original TIAs conclusion was the associated ACTION (often called remedial action), does need to be completed before mode transition. The revised TIA reverses the position and determines that the action does NOT need to be done prior to the mode ascension.
- 2. [M]y strongest argument is this: if no action is taken before the mode transition, then the plant has not been afforded any additional safety DURING the mode transition. Since TS 3.0.4 was designed for the transition, the plant is in no different configuration until the action is taken. After the mode transition, taking the action serves no purpose other than require the plant to shutdown after a period of time if the action is not done (which would be required in a shutdown LCO anyway). This course of action would be NOT be aligned with the TS basis item which says Compliance with Required Actions that permit continued operation of the plant for an unlimited period of time in a MODE or other specified condition provides an acceptable level of safety for continued operation [my emphasis].
- 1. The letter states "key language in GL 97-09 was improperly translated." Implicit in this argument is the ISTS language does require completion of the required action prior to a mode change.
Since GL cannot over ride a license requirement, it follows that the current license does require completion of the required action prior to a mode change.
- 2. The letter also argues that since 3.0.4 uses future tense, no action need be taken prior to mode transition. However, the action is entered only when the LCO becomes applicable. Therefore, future tense would be used regardless of the intent of TS 3.0.4 since its use would be invoked prior to the planned mode change. In addition, the bases clearly states that "Compliance with the Required Actions that permit continued operation of the plant for an unlimited period of time in a MODE or other specified condition provides an acceptable level of safety for continued operation." It is important to note that in ISTS the Required Action is what must be done. The bases, therefore, clearly establishes that the acceptable level of safety is only achieved when the required action is complete.
Revise TIA to directly quote the wording of LCO 3.0.4a, When the associated ACTIONS to be entered permit continued operation in the MODE or other specified condition in the Applicability for an unlimited period of time; coupled with the definition of ACTIONS in Section 1.1 as that part of a Specification that prescribes Required Actions to be taken under designated Conditions within specified Completion Times.
The definition provided in the use and application section makes it clear that the translation from a two-column to three-column format did not actually change the coupling of Required Actions and their Completion Times that is implied by the improper translation of the bases. It also would serve to focus on the words of the LCO (ACTIONS) and their meanings rather than the vernacular use of the term remedial action.
This still relies on the understanding of the phrases to be entered and to be taken, which are in the future perfect tense, but the limitation on the time due to the phrase within specified Completion Times is consistent with that usage.
TIA 2009-005 as currently written J. Giessner Non-concurrence concerns J. Ellegood Non-concurrence concerns Proposed way ahead to address concerns I realize there is some lack of clarity in TS in the use of the future tense for to be entered, but one thing is certain: some remedial action needs to be taken. Mode changes have additional risk that must be managed.
Best addressed outside of TIA if the current licensing basis does not require the required action to be complete prior to a mode change.
Clearly in some cases at some other plants, the application of TS 3.0.4 (post change to ISTS) would have ended up being more restrictive. For example, I reviewed old TSs at Salem and DC Cook and found TS 3.0.4 exemptions in the specific instrumentation section of TS. The exemptions from TS 3.0.4 essentially said no action is required for the mode transition. But this is not true at Palisades. There were no TS 3.0.4 exemptions in safety instrumentation section of TS.
Therefore the use of TS 3.0.4 in this section would be less restrictive.
Need to discuss. The statements in GL 87-09 regarding the more or less restrictive effect after its implementation were brought up during earlier conversations and were included in earlier versions, but were removed from the TIA as they are not necessary to the analysis of the regulatory position. This concern may have already been addressed through earlier discussions.
TIA 2009-005 as currently written J. Giessner Non-concurrence concerns J. Ellegood Non-concurrence concerns Proposed way ahead to address concerns
[T]he staff opinion and the information provided in the TS basis and generic letters state that good practice should dictate that the plant should be started up with all equipment operable and that good judgment should be used. These statements, although honestly intended to provide a good framework, have no feasibility in enforcement space. Determining a standard that was not complied with is not objective; therefore, no performance deficiency would be found using the current oversight process. The licensee will follow the TSs, which they equate to sound judgment. So I take little consolation in thinking that Palisades should have adopted the good practice of either restoring systems or alternately established the basis for continued operation by placing the inoperable radiation monitoring channel in trip before mode transition [conclusion of the revised TIA]. The NRC would be hard pressed to enforce prudent operations.
As discussed in the non-concurrence, this is not suitable to address in the TIA, but will need to be addressed elsewhere.
It is noted that enforcement space with the current ROP, differs remarkably from SALP program for enforcement which existed at the time the GL 87-09 was issued. The SALP process was very subjective and including guidance directing licensees to establish good practice and not startup with inoperable equipment would be much more significant then than it is now, because the best practice could be enforced by assigning a poor score to reactor operations.
Since mode transition could present a possible challenge, some action should be taken to provide the plant additional safety or at least evaluate the remedial action as not being needed. Without requiring the remedial action, it may be important to risk assess the specific condition prior to mode ascension. Simply, require plants to do TS 3.0.4 b (risk based) and delete TS 3.0.4a from the ISTS.
As discussed in the non-concurrence, this is not suitable to address in the TIA, but will need to be addressed elsewhere.
TIA 2009-005 as currently written J. Giessner Non-concurrence concerns J. Ellegood Non-concurrence concerns Proposed way ahead to address concerns Includes statement that the mode change allowances of LCO 3.0.4a is premised on the reasonable expectation that the required actions will be taken within their completion times.
TIA proposes a standard to determine if mode change can be made as being a reasonable expectation that the required action can be completed in its specified completion time Revise TIA to explain that, while this is not a requirement of LCO 3.0.4a, the reasonable expectation that required actions will be taken within their completion times is necessary to comply with the LCO 3.0.2 limitation on intentionally entering an ACTION for operational convenience.
TIA 2008-002 is superseded by this TIA.
The TIA should state that it is reversing a previously established staff position.
Revise the TIA to address the fact that this TIA does not reverse a previously established staff position, but instead restores the applicable regulatory position to its state prior to the errors in TIA 2008-002. The applicable regulatory position had been established in GL 87-09, as implemented for specific licensees. TIA 2008-002 fundamentally modified the applicable regulatory position to include a requirement for completion of required actions prior to mode changes when using LCO 3.0.4a of the ITS; this change was outside of the scope of actions that can be taken in a TIA as it did not allow for proper vetting of the change to the regulatory position through the CRGR and taking into account other stakeholders input. The purposes of this TIA are to correct the problem of having modified a regulatory position using the mechanism of a TIA by cancelling that TIA (2008-002) and to explain that the previously established regulatory position of GL 87-09 is still applicable.
For the above reasons, in part, I do not concur on the letter Other concerns are implied by this that will need to be detailed in order to address.
1 Bowman, Eric From:
Ellegood, John Sent:
Tuesday, June 16, 2009 9:17 AM To:
Bowman, Eric
Subject:
RE: Non-concurrence on re-evaluation of a TIA As far as the letter content, no. During previous discussions (and earlier versions of the letter) some arguments were raised that I did not agree with.
Original Message-----
From: Bowman, Eric Sent: Tuesday, June 16, 2009 9:14 AM To: Ellegood, John
Subject:
RE: Non-concurrence on re-evaluation of a TIA
- John, Jjust want to check and see if you've got other concerns we will need to address because you closed the non-concurrence by stating that it was only in part due to what you've laid out on the form. I'm working at home today, so won't be at my office number; I'll be back in the office tomorrow.
Thanks for your help with this.
Eric
Original Message-----
From: Ellegood, John Sent: Tuesday, June 16, 2009 9:05 AM To: Rosenberg, Stacey Cc: Schulten, Carl; Bowman, Eric; Giessner, John; Taylor, Thomas; Betancourt, Diana
Subject:
Non-concurrence on re-evaluation of a TIA Attached is my non-concurrence on the reevaluation of the TIA on LCO 3.0.4a.
1 Bowman, Eric From:
Bowman, Eric Sent:
Thursday, June 18, 2009 12:47 PM To:
Ellegood, John; Giessner, John Cc:
Schulten, Carl
Subject:
RE: TIA 2008-002 It actually does serve the purpose of explaining the 7 day completion time. Also, because it's in the bases rather than the TS, it can't modify the meaning of things in the TS, but merely explain them This is particularly so for those that are in different sections as is the case here with the mode ascension allowance being in LCO 3.0.4 while this discussion is providing the bases for LCO 3.3.3's requirements for ESF channels/bistables.
From: Ellegood, John Sent: Thursday, June 18, 2009 12:41 PM To: Bowman, Eric; Giessner, John Cc: Schulten, Carl
Subject:
RE: TIA 2008-002 There would be no need for the 3.3.3 basis statement if 304a allows mode transition w/o completing the required actions.
From: Bowman, Eric Sent: Thursday, June 18, 2009 12:37 PM To: Ellegood, John; Giessner, John Cc: Schulten, Carl
Subject:
RE: TIA 2008-002 That interpretation would only apply if you accept the interpretation of LCO 3.0.4a that required actions must be complete. If you interpret LCO 3.0.4a as allowing mode transitions while in the allowed completion time for an action requirement there is no need to find an implied exception, which would leave us with a more logically established and harmonious framework for this. It might be worth pursuing the suggestion that action requirements should be entered prior to mode ascensions, but that's outside the scope of the TIA in my view.
From: Ellegood, John Sent: Thursday, June 18, 2009 12:32 PM To: Bowman, Eric; Giessner, John Cc: Schulten, Carl
Subject:
RE: TIA 2008-002 I'll point out that 3.3.3 that says startup is allowed to continue implies that you are already in Mode 2. Therefore you could go to Mode 1 w/o completing the required actions. Since is does not discuss a M-3 to M-2 transition, it implies that M2 to M1 is an exception-and if in M-3, the required action must be completed.
From: Bowman, Eric Sent: Thursday, June 18, 2009 12:24 PM To: Giessner, John; Ellegood, John Cc: Schulten, Carl
Subject:
RE: TIA 2008-002 The logic I would use in analyzing the licensing basis here is that the LCO 3.0.4a allows entry into a MODE when the ACTIONS to be entered permit continued operation in the MODE for an unlimited time, and ACTIONS are defined as that part of a TS that prescribes Required Actions to be taken under designated
2 Conditions within specified Completion Times. These provisions are both part of the TS and at face value would not seem to require completion of the Required Actions prior to entry into a MODE.
The bases for LCO 3.0.4a sheds some light on the proper interpretation of the LCO, but is not actually part of the TS. The bases is a bit ambiguous in that it cites the Required Actions as providing an acceptable level of safety, but does not address the other half of the question. The second half of the question would be whether the limitation on time allowed for operating subject to ACTIONS (or having Required Actions that have not been entered yet, but are within their specified Completion Times) also provides an acceptable level of safety.
This is addressed in the bases for LCO 3.3.3 Action A.1 in the statement that "[t]he 7 day Completion Time is based upon operating experience, which has demonstrated that a random failure of a second channel occurring during the 7 day period is a low probability event." The LCO 3.3.3 bases also uses the phrase "startup is allowed to continue," which is a lot more keyed to the allowance for mode ascension. That phrase is used in a number of LCO bases for non-shutdown LCOs in the NUREG-1432 STS.
While I agree that taking the action prior to mode ascension is the better because it provides a greater degree of protection, my primary focus has to be on the regulatory effect of the licensing bases. I haven't been able to find the pre-Amendment 130 version of LCO 3.0.4 and its bases, but am not sure that the existence or lack of exceptions to that version of LCO 3.0.4 would matter. The wording of Amendment 130 and its associated safety evaluation lead me to believe that we took actions that would allow mode ascension relying on the 7 day completion time for placing an ESF bistable in trip per LCO 3.3.3 Action A.1 because having an inoperable ESF instrument channel or bistable does not trigger an action requirement that would require restoring the inoperable instrument channel or bistable (meeting the LCO) or shutting down. It merely requires that the ESF bistable be placed in trip before the 7 day completion time expires.
I think the bottom line is that if we were to interpret the LCO 3.0.4a bases as requiring completion of Action Requirements prior to mode ascension, we would have a conflict with the LCO 3.3.3 Action A.1 bases. The LCO 3.0.4a bases does not literally require that interpretation, but is silent on the aspect of completion times for provision of an acceptable level of safety, and can be interpreted in a manner that is in harmony with the LCO 3.3.3 Action A.1 bases.
Eric From: Giessner, John Sent: Thursday, June 18, 2009 11:20 AM To: Bowman, Eric; Ellegood, John Cc: Schulten, Carl
Subject:
RE: TIA 2008-002 I did see this, but the key is not the LCO action time - the key is the mode ascension. Agree it is low probability, but so are other LCOs (which are S/D LCOs). The item in question is TS 3.0.4. If TS 3.0.4 was not applicable, the old TS would have said so. In the new TS my focus is on the action that gives you the assurance.
jack From: Bowman, Eric Sent: Wednesday, June 17, 2009 2:50 PM To: Giessner, John; Ellegood, John Cc: Schulten, Carl
Subject:
TIA 2008-002 Importance: High John/Jack,
3 I did a little further review of the Palisades TS in preparation for our phonecon on Tuesday and found the quote below in the bases for LCO 3.3.3 from Amendment 189, which I believe is the current revision to that LCO.
I've also attached the extracted pages from that amendment, which is available in ADAMS at ML993510369; the bases is at page 655 of the file. If we revise the TIA to cite these bases, would it satisfy your concerns with the document? I realize it still leaves the question open regarding enforcement of prudence and best practices, but can't really address that in the TIA.
If one ESF channel is inoperable, startup or power operation is allowed to continue, providing the inoperable channel actuation bistable is placed in trip within 7 days. The provision of four trip channels allows one channel to be inoperable in a non-trip condition up to the 7 day Completion Time allotted to place the channel in trip. Operating with one failed channel in a non-trip condition during operations, places the ESF Actuation Logic in a two-out-of-three coincidence logic.
If the failed channel cannot be restored to OPERABLE status in 7 days, the associated bistable is placed in a tripped condition. This places the function in a one-out-of-three configuration.
In this configuration, common cause failure of the dependent channel cannot prevent ESF actuation. The 7 day Completion Time is based upon operating experience, which has demonstrated that a random failure of a second channel occurring during the 7 day period is a low probability event.
Thanks!
Eric