ML20153D624
| ML20153D624 | |
| Person / Time | |
|---|---|
| Issue date: | 06/22/1998 |
| From: | Boger B, Sheron B NRC (Affiliation Not Assigned) |
| To: | Collins S NRC (Affiliation Not Assigned) |
| Shared Package | |
| ML20153D328 | List: |
| References | |
| NUDOCS 9809250120 | |
| Download: ML20153D624 (7) | |
Text
-
I
$3 staog g g k UNITED STATES a e < NUCLEAR REGULATORY COMMISSION WASHINGTON, D.C. 3000H001
..... June 22, 1998 l
MEMORANDUM TO: Samuel J. Collins, Director Office of Nuclear Reactor Regulation FROM: Brian W. Sheron, Acting Associate Direc For Technical Review, NRR
/ '
Bruce A. Boger, Acting Associate Director For Projects. NRR !
SUBJECT:
FOLLOW UP ACTIONS TO NRR AD HOC REVIEW PANEL ON l DIFFERING PROFESSIONAL VIEW ON TECHNICAL l SPECIFICATION TRIP SETPOINTS AND ALLOWABLE VALUES FOR
! INSTRUMENTATION -
In a memorandum dated March 17,1997, a differing professional view (DPV) was expressed regarding Technical Specifications (TS) trip setpoints (TSPs) and Allowable Values (AVs) for instrumentation. A panel reviewed the concerns addressed in the DPV and provided a report which you forwarded to us in a memorandum dated October 31, 1997. In that memorandum, you requested that we provide the actions proposed to address the panel's recommendations and the schedule for completing those actions.
We hava completed a review of the panel's recommended follow up actions regarding the DPV. In addition, we have met with the Office of the General Counsel (OGC), and they
, have indicated that the wording of applicable regulations and rulemaking history support the view that the TSPs corresponds to the limiting safety system settings (LSSS) which are i required to be in the technical specifications (TS) as recommended by the DPV panel.
However, we have also met with members of the NRR technical staff responsible for instrumentation and for TS. Based on this integrated interface, we believe an attemate 4 interpretation exists. Furthermore, the issue is of neg!igible safety significance and implementation of the recommendations would require extensive staff and industry resources. Moreover, we believe that current TS satisfy thevaquirements of 10 CFR 50.36 4
for the reasons discussed below.
We believe that the DPV panel and OGC, in developing their opinion applied more emphasis to the word " setting" in 10 CFR 50.36 than to the description of the LSSS as limitina safety system settings. We further believe that the regulatory record is not clear and does not specifically identify either TSPs or AVs as the LSSS. The DPV Panel acknowledges that 10 CFR 50.36 could reasonably be read as allowing the use of either the TSPs or the AVs as the LSSS. Therefore, rather than trying to interpret en unclear regulatory record, we have focused on the precedent set by the staff in implementing this regulation. This history indicates that the staff intended AVs to be acceptable for the LSSS in TS.
, Moreover, the current staff position as implemented in the improved Standard TS (iSTS) is 1
9909250120 980915 PDR TOPRP EMVWEST i \
C PDR 950?Nb/NQ- -- _ - - _ . - -.
o S. Collins 2 June 22, 1998 1
! to represent the LSSS as AVs in a single-column format with the exception of the I i Westinghouse ISTS which includes both AVs and TSPs in a dual-column format. OGC
! agrees that the regulatory record is not totally clear. The lack of clarity is further evidenced by the extensive discussions held by the DPV panel as well as discussions subsequent to the panel decision. Given the lack of a safety issue and the legal ambiguity, I we believe that the appropriate decision should be influenced by operational issues
. assor'ated with the TS and the impact on both industry and staff resources of any change.
] More discussion of the NRR staff view is nresented in the attachment.
j There are three importr,:d TS operationalissues with which the staff is concerned relative 3
to the DPV panel's recommended follow up sa:tions: (1) licensees with only TSPs in thair TS have experienced issues in implementing setpoint methodologies and in complying with channel functional test and channel calibration TS requirements, (2) licensees with the ;
single column AV format determine instrument channel operability and satisfy applicable '
safety limits based on AVs not TSPs, and (3) the assignment of the LSSS exclusively to TSPs is inconsistent with setpoint methodologies that support the ISTS, the previous STS,
- current regulatory guidance, and previous industry standards. The current ISA Standard,
- 567.041994, "Setpoints for Nuclear Safety Related Instrumentation," which was
- referenced by the panel but has not been endorsed by the staff, allows the LSSS to be j either AVs, TSPs, or both; this standerd also allows the LSSS to be controlled by plant
! procedures.
4 An obvious approach to implement the panel's recommendations would be to list TSPs in a
! single-column format in the instrumentation function tables in the TS. We believe this is j not appropriate because it would neither ensure control of setpoints, margins, and d
uncertainties sufficient to satisfy setpoint methodologies, nor would it provide for f operability determination consistent with approved setpoint methodologies, national
{ consensus standards, and 10 CFR 50.36. As a minimum, AVs provide TS operability J
criterie; this approach is consistent with the i3TS. An option that would both meet this operability requirement ano address the DPV recommendations would be a two-column TS format containing both TSPs and AVs. This option would control operability criteria not just by plant procedures but by the TS as required by 10 CFR 50 36. TS Surveillance Requirements for channel functional tests and channel calibratioc wou!d assure that the operability criteria were satisfied. 79 ensure that instrumentation settings were left within the TSP tolerance, the TS could explicitly require resetting an instrument within the TSP tolerance in the event the as found satting was outside the tr,ierance.
While the option of requiring a two-column format with both AVs and TSPs in TS might be viewed as optimal because it would address both TS operability and the DPV panel recommendations, it would represent a backfit for plants that have relocated TSPs from the TS. This option would require significant staff and industry resources to implement; it I couN not be justified based on safety; and it would not be required to meet the j requirements of 10 CFR 50.36 for the LSSS. Thus, the only compelling argument to !
pursue this option would be a compliance backfit based solely on an interpretation of 10 i CFR 50.36 that TSPs are required by regulation to be included in the TS. l l
1 4
r
S. Collins 3 June 22, 1998 As discussed above and further elaborated upon in the attachment, we do not believe that the regulations or the supporting regulatory history clearly support the OGC view or the differing professional view. We believe that an equally valid case can be made that AVs satisfy the requirement that the TS include LSSS. Therefore, we do mt believe that it is an appropriate use of chher NRC or industry resources to implement the DPV panel's recommendations, in summary, the staff believes the current approach to implementing the 10 CFR 50.36 requirement for LSSS by means of AVs is acceptable and can continue to be applied to the TS conversions to the ISTS. However, the staff does and will allow a two-column format with both TSPs and AVs as an option to licensees whc prefer this approach.
Attachment:
As stated
l j .-
- 4-
~
As stated in 10CF.R 50.36,"Each applicant for a license authorizing operation of a production or utilization facility shall include in his application proposed technical specifications in accordance j
with the requirements of this section." Additionally,10 CFR 50.36 requires that a," summary 3 i statement of the bases or reasons for such specification, other than those covering i l administrative controls, shall also be included in the application, t-f shall not become part of i
the technica!' specifications. The technical specification will be ufrmd from the analysis and evaluation included in the safety analysis report and may include, .cch additional technical i
specifications as the comrnission finds appropriate."
i j Among the items to be included in the technica: specifications (TS) according to 10CFR 50.36
- are safety limits, limiting safety system settings (LSSS), limiting conditions for operation (LCO)
} and surveillance requirements.
Safety limits for nuclear reactors are limits upon important process variables that are found l j
necessary to reasonably protect the integrity of certain of the physical barriers that guard !
against the uncontrolled rclease of radioactivity. Further, should a safety limit be exceeded l
- the reactor must be shut down and the licensee must notify the commission, review the j matter, and record the results of the review including the cause of the condition and the
! corrective action taken to preclude recurrence.
The LSSS are settings for automatic protective devices related to chose variables having significant safety functions. Two points about the LSSS are made, (1) where an LSSS is l specified for a variable on which a safety limit has been placed, the setting must be so
- chosen that the automatic protective action will correct the abnormal situation before a i
safety limit is exceeded, and (2) if during operation, it is determined that the automatic j safety system does not function as required, the licensee shall take appropriate action.
i .
An LCO !s the lowest functional capability or performance level of equipme it required for j ' safe operation of the facility. When an LCO is not met, the licensee must shut down the
- reactor or follow any remedial action permitted by the technical specifications until the j condition can be met.
Surveillance requirements are requirements relating to test, calibration, or inspection to i
assure that the necessary quality of systems and components is maintained, that the facility operation will be within safety limits, and that the limiting conditions of operability will be met.
As can be seen from the above, the requirement for an " Allowable Value" (AV) is not specifically required by 10 CFR 50.36 and it would appear, therefore, not to be necessary to include an AV in the TS. In fact, the definition for an AV is not specifically stated in TS whether they are 4
" custom", standard, or the more recent improved standard technical specifications (iSTS).
However,10 CFR 50.36 also does not specifically define or require a " trip setpoint" (TSP)
The purpose of this discussion is to show that although the term " Allowable Value" is not specifically definad in 10 CFR50.36 or in TS, by choosing AVs to specify LCOs for controlling ;
channel operability and as LSSS for preserving applicable safety limits, the requirements of 10 CFR 50.36 are met. The discussion will also show that although TSPs play ar 'itegral part of Attachment
\"
2 l
l operating within the LSSS, it is more reasonable when writing TS to select AVs to specify LCOs rather than TSPs. What is important is thct TS provide appropriate limits consistent with 4
staff approval of licensee's various setpoint programs.
1 i
Although the inclusion of LSSS in the TS is a requirement of 10 CFR 50.36, the relationship of i LSSS to safety limits, LCOs, AVs, TSPs, setpoint methodologies, standards, and regulatory
] guides continues to be the subject of debate within the nuclear power industry. The assignment of the LSSS to a single TS variable and associated setpoint methodology is difficult to resolve i
for a variety of reasons. Among these reasons is a lack of national consensus setpoint i
' standards during the early years of the nuclear power industry and the various unique and not entirely compatible setpoint methodologies originally provided by NSSS suppliers and j contractors. These variations, together with various staff requirements and interpretations res"Ited in various TS instrumentation requirements, TS formats, LSSS interpretations, j opt ibility determinations, and at times excessive LER generation.
t
' Recently, OGC was asked to provide an opinion as to whether the term, " Limiting Safety System Settings" as used in 10CFR 50.36(c)(1)(ii) refers only to manually adjusted TSPs or whether LSSS can be interpreted as embracing AVs which are derived from the TSP and analytical limit instrument channel uncertainty evaluation. The OGC discussion of the LSSS,
" Meaning of Limiting Safety System Settings in CFR 50.36," dated March 16,1998, referenced the guidance document supporting the original 10 CFR 50.36 rule. Although the evaluation by OGC concluded that the LSSS are the TSPs, it is the staff's opinion that the TSPs were not the LSSS as originally implemented. The staff's opinion is based on the implementation of the
" custom" TS, previous STS (circa 1980), ISTS (1992), and reviews of setpoint standards development up to the current industry standard revision. The OGC evaluation stressed that LSSS were " settings"; however, the staff considers the word " limiting" to be just as significant.
As pointed out in the OGC evaluation, the original wording for the LSSS was " Maximum Safety System Setting." This was subsequently changed to " Limiting Safety System Settings" for the final rule. The reason given for the change was that safety limits, safety limit settings, and conditions of operation can be limited at either upper or lower bounds and therefore the word
"!imiting" was more appropriate. In other words, LSSS are the " limiting" safety system settings or the rnaximum/ minimum settings that safety system settings may have and continue to satisfy the safety limits. The guidance document supporting the initial issuance of 10 CFR 50.36 also states that,"Ihe licensee is free to coerate with anv safety system settina within the limitina Y.abat "
Based on OGCs review of the 10CFR 50.36(c)(1)(ii) and original 50.36 rule making, it was their opinion that contrary to the ISTS single-column AV format that assigns the LSSS to AVs, the LSSS are represented by the TSPs since they are considered the instrument " settings". The staff considered the reasoning behind OGC's opinion and agrees that the inclusion of the TSPs in the technical specification could have some benefit. The staff does not believe, however, that a single setpoint column TS with only the TSPs can be implemented without significant confusbn and impact on licensees because such an approach would not include necessary limits on parameters. To successfully implement the DPV panet/OGC format, the staff concludes that a two column approach with AVs and TSPs would be necessary. Furthermore, Attachment
~ '
i l c l 3
, the staff believes that a single column TS utilizing AVs can be successfully implemented and i still meet the requirements of 50.36 as discussed below.
The DPV panel stated, "The panel found these concems primarily administrative in nature (i.e.,
- related to compliance with 50.36) and do not adversely affect the safe operation of reactors that are subject to TS requirements that do not include TSPs." In addition ". . . the panel found that the issues identified duQ these discussions and deliberations do not adversely affect the safe operation of reactors due to TS requirements associated with those issues." Given this general conclusion, the question arises
- if the panel recommendation is to be implemented as stated, is
- determined that backfitting would be required if the agency adopted the position that the LSSS I~
be defined as the TSPs and only the TSPs. Although the safety significance was determined to
- be minimal by the panel, the adoption of the panel's recommendation in conjunction with OGCs i opinion would require a compliance backfit for those licensees that previously assigned the
! Operating experience show6J a need for guidance on the selection of instrument setpoints (pressure, differential pressure, flow, level, temperature, power, radiation, time delay) used to
) initiate automatic protective actions and alarms. The lack of a national consensus standard on setpoint methodology led to the development of Regulatory Guide (RG) 1.105 Rev.1. This RG was an attempt by the staff to describe an acceptable method to ensure that instrument setpoints in systems important to safety were initially within and remained within specified limits.
The RG notes that the most frequent abnormal occurrence reports submitted between January 1972 and June 1973 involved drift of protective instrument setpoints outside TS limits. The single most prevalent reascn for a parameter being outside TS limits was the fa. i that the TS al! owed insufficient or no margin between the setpoint and the " technical specification limit" to account for instrument inaccuracies, vibration, or minor calibration variation.
RG 1.105, Rev.1 defines various terms including " technical specification limif' stating that it was the limit prescribed as a license condition on an important process variable for safe operation. However, there is no clear discussion as to the relationship of LSSS to technical I specification limits. There is no discussion about LSSS being defined as TSPs. And there is no discussion of the concept of an AV. However, the RG does expiain,"the setpoints should be established with sufficient margin between the technical specification limit for the process variable and the nominal trip setpoint to allow for (a) the inaccuracy of the instrument (b) )
uncertainties in the calibration, and (c) instrument drift that could occur during the interval i between calibrations." Based on a review of ' custom
- TS formats and the discussions and 1 resulting positions taken in the RG, the techn/ cal specification limit as referenced by the RG l was to be the limit prescribed as a license condition. [ltalics added to quotations for emphasis.)
Regulatory position 6 of RG 1.105, Rev.1 states that the assumptions -the minimum margin with respect to the LSSS, drift rate, and the relationship of drift rate to testing intervaMsed in selecting the setpoint values are to be documented. The actual assignment of LSSS to particular variables is not stated. However, the implication that the TSPs' relationships to the LSSS be defined and that sufficient margin be provided between the setpoints and the technical Attachment
. l
- 1
- 4 e
j specification limits implies that the TSPs were not required to be the LSSS. Althouoh
" technical soecification limit"is not a 10 CFR50.36 (3rm it can be seen from RG 1.105. Rev.1 j and the " custom" TS that the LSSS were not soecifically intended to be or assioned to the TSPs. Based on the above. the LSSS and the TSPs were not considered identical and the l TSPs by themselves were not considered the LSSS.
4
! The staff believes that the DPV panel and OGC applied more emphasis to the word " setting"
, than to the description of the LSSS as limtting safety system settings. The development of i 10 CFR50.36 and " custom" TS does not support the opinion that the TSPs chosen by the l licensee must be included in the TS. The limiting values are the values to be listed in the TS. 1 However, in the initial STS, the staff provided a means to document the TSPs. The reason for 2
this was to more fully monitor licensees' drift assumptions over surveillance intervals to better j control TSP values. Subsequently, with the development of the ISTS single column AV format,
) the staff again documented the AVs as the " limiting" settings which if exceeded the licensee i
must take appropriate action. The TSPs would be set by the licensee consistent with the j assumptions used to develop channel uncertainty allowances. Obviously the licensee should I establish the TSPs conservatively with respect to the AVs such that safety and operability limits
! would not be exceeded except on rare occasions. In other words the TSPs must be set such
! that they are bounded by the limiting values (i.e., AVs) consistent with the setpoint calculations.
i One relaxation associated with the iSTS format is that instrument performance criteria may not be as well controlled within the TS. In developing ISTS, instrument performance was recognized as a programmatic objective appropriately left to the licensee and not required to be in the TS. From a TS point of view, as long as the trip function criteria are met (the channel trips within the AVs/LSSS) the channelis operable. The performance parameters of the channel with respect to the setpoint calculation and uncertainty assumptions are documented during survei"ance (per IEEE 338 and RG 1.118) or inspection and are evaluated by the licensee.
The staff's position is that the TSPs are not the limiting values with respect to TS and that a single column format specifying the AVs as the LSSS is acceptable to control parameter values such that operability is maintained and safety limits are not ce exceeded. The staff believes this position is supported by the 10 CFR50.36 rulemaking documentation and a review of
" custom" TS developed per the guidance of 10 CFR50.36 shortly after it was issued. The staff's opinion is that a single-column format, with only the TSPs, based on current setpoint methodology would require revisions to TS, setpoint methodologies, Regulatory Guides, industry guidance and standards with a potentialloss of operating margin and with no improvement in safety. The DPV panel concurred that the safety significance of not including the TSPs in the TS was minimal. It is tha staff's opinion, therefore, that the ISTS single-column AV format is an acceptable and preferable implementation of the requirements of 10 CFR50.36.
Attachment
,