ML20044B611
| ML20044B611 | |
| Person / Time | |
|---|---|
| Site: | Nine Mile Point  |
| Issue date: | 02/09/1993 |
| From: | Office of Nuclear Reactor Regulation |
| To: | |
| Shared Package | |
| ML20044B612 | List: |
| References | |
| NUDOCS 9302170159 | |
| Download: ML20044B611 (10) | |
Text
-
i j
OiCLOSUi!E 2
>S Rf Cut 1
C
,Yn UNITED STATES i
5
/E NUCLEAR REGULATORY COMMISSION 1
o WASHINGTON, D.C. 20666 l
l l
SAFETY EVALUATION BY THE OFFICE OF NUCLEAR REACTOR REGULATION TOPICAL REPORT INSTRUMENTATION SETPOINT METHODOLOGY
]
GENERAL ELECTRIC COMPA_NJ NEDC - 31336
1.0 INTRODUCTION
This report provides the basis for acceptance of and limits on the acceptance of the topical report submitted by General Electric (GE) describing the basis
)
for instrumentation (trip) setpoint selection for instruments that are not operated in a harsh environment. The criteria for approving the topical l
report are General Design Criterion 20, 10 CFR Part 50.36 and Part 50.46.
l l
1.1 DESIGN FEATURES AND PARAMETERS Since 1976, considerable interest has been expressed about how setpoints are selected in nuclear power plants. Of particular cancern is the adequacy of the setpoints with regard to assumptions made in the accident simulations.
The result of this interest was publication of proprietary topical reports by the nuclear steam suppliers, revisions of industry standards for setpoint calculations and revision to Regulatory Guide 1.105 " Instrument Setpoints for Safety Related Systems." This regulatory guide endorses ISA-S67.04-1982 l
"Setpoints for Nuclear Safety Related Instrumentation Used in Nuclear Power Plants."
l The topical report that is the subject of this safety evaluation report (SER) was originally published in October 1986 and was subject to minor revisions in-early 1992.
However, the work involved in the develo ment of this topical is l
a continuation of efforts that hav ntp to previous staff SERs gg@M (gb C
-\\
u
. (e.g. " Staff Report on Setpoint Methodology for General Electric Supplied Pro-tection System Instrumentation" dated May 15, 1984).
2.0 EVALUATION I
i 2.1 General Comments The current topical report is an important document that is critical to the understanding of how GE selects setpoints.
However, in the April 8, 1992' responses to the staff's questions on NEDC-31336, GE states that the examples used in the topical report "are not to be used by any plant to validate their.
t individual setpoints. That is, each plant must provide their own plant unique analysis for the setpoints....
The examples are used to show the safety margins and typical channel' errors that might be expected.
Since plants have different instruments, environments, seismic and other requirements only examples have been provided."
The remainder of this evaluation is devoted to a review of the individual sec-tions of the topical report. Section 5 " References" was not reviewed.
2.2 Section 1 " Instrument Setooint Methodoloov" This section starts with a general description of how setpoints are established. The general description is followed by a set of definitions of terms that are used in the methodology. Next, the methods identify the relationship between the different setpoints and the required data that form the terms of the calculations.
Finally, the combining of terms to generate and test the setpoints is discussed.
The definitions used in the topical are in general agreement with recognized industry standards such as ISA-S67.04-1982.
Table I provides a quick cross reference between the definitions in a top down sequence.
,-yr
,o
-o--a
i
.I i
.l
- TABLE.I NEDC-31336 ISA-S67.04-1982 RG 1.105 Licensing Safety Limit Safety Limit Safety Limit Analytical Limit Not Defined Not Defined i
Allowable Value Not Defined Not Defined (Tech Spec Limit) i Nominal Trip Setpoint Setpoint Upper Limit 4
Setpoint l
Steady State Operating Not Defined
- Not Defined j
Although the analytical limit and allowable value are not defined in ISA-567.04-1982, they are described.
The. descriptions _of these setpoints are similar to the definitions used in the topical report.
However, the definitions used by GE, in some instances, distribute the error terms differently. For example, a term used by GE in calculating the margin between the Licensing Safety Limit and the Analytical Limit is described as part of the margin between the Safety Limit and the Allowable value in ISA-S67.04-1982. The following definitions used by GE are acceptable to the staff.
l
" Analytical Limit (AL): The value of the sensed process variable established as part of the safety analysis prior to or at the point which a desired action is to be initiated to prevent the safety process variable from reaching the associated licensing safety limit."
i
" Allowable Value (AV) (Technical Specification Limit):
The limiting value of the sensed process variable at which.the trip setpoint may be found during instrument surveillance.
Usually prescribed as a license condition."
Although the sequence of the terms used by GE in the development of the various setpoints may be different, the*following data are required and are consistent with the data needs of ISA-567.04-1982.
These mandatory data terms are:
-w
j i Analytical Limit Channel Instrument Accuracy Channel Calibration Accuracy Channel Instrument Drift Process Measurement Accuracy Primary Element Accuracy Most of these error allowances are defined to include subterms.
The use of these terms is typical of other vendor methods and consistent with industry standards.
Therefore, they are acceptable to the staff.
The numerical methods for using accuracy and drift terms to calculate the lower order setpoints (as identified in Table I above) involve formulations that are consistent with the definitions used.
Independent, random, and normally distributed variables are combined by the square root of the sum of the squares.
Non-random (biased) and dependent variables are summed algebraically. The methods are designed to result in a 95 per cent probability of providing a channel trip before the process variable reaches the analytical limit, considering drift, assuming a one-sided normal distribution and a 95 per cent confidence level. With the exception of using a single sided test, these methods are consistent with industry practice and are, therefore, acceptable to the staff. The use of a single-sided test to define the probability that a trip will occur between two limits (e.g. Nominal Trip Setpoint and Allowable Value) is acceptable to the staff within the context of this topical report. The probability of a false trip (early trip) is a separate calculation when determining whether a technical specification setpoint will be satisfied. However, the use of a single-sided test for instrument channels that provide trips or permissives for increasing and decreasing variables (e.g. reactor level) is not supported by analysis and is not consistent with the general approach reflected in the current industry practice.
The staff, therefore, finds this set point methodology for increasing and decreasing variable unac'ce'ptable.
Similarly, the use of a single sided test for safety-related indicators and recorders is unacceptable.
4l
. Section 1 also described the requirements for determining drift data and the effect of such environmental factors as harsh environments. Where GE differs from the usual industry practices in these matters, GE has submitted justification for their positions. The staff agrees with the positions i
presented and notes that the methodology does not consider the application of harsh environments to equipment located inside containments for plants that vent the safety relief valves directly into the containment.
In these cases, an instrument-by-instrument review of the harsh environment would be necessary to support the GE position.
2.3 Section 2 " Instrument Accuracy and Drift Methodoloov" Usually, the accuracy terms and subterms are given as point values. The data for Rosemount and Gould transmitters and trip units have been provided by the manufacturers as curves.
Section 2 of the topical report deals with these curves over the expected ranges of these terms in GE plants.
For the most part, the staff found that this material was a more detailed discussion of some of the material presented in Section 1.
However, in our review of the setpoint changes proposed for Fermi 2 (a separate issue from the review of the topical report), the staff discovered that plant-specific conditions or procedures could result in changes to the equations presented in Section 2.
For example, different equations are used to calculate Rosemount trip unit accuracy for a BWR-6 than are used for a BWR-4 or BWR-5. At Fermi 2 (a BWR-4) the BWR-6 equation was used because of the use of Power Generation Control Complex cabinets.
The material presented in Section 2 is acceptable to the staff.
2.4 Section 3 " Instrument Setooint Descriptions" This section presents the results of ca'lculations (but not the calculations per se) using typical plant data for many of the trip functions in a BWR.
. ___. _ ~__
i 6-Although the methods used by GE are generally recognized and are responsive to l
the staff's previously stated concerns, the specific studies included in the topical are only applicable to a few of the instruments provided by or i
specified by GE.
Therefore, the numerical results presented in this topical-
)
have not been demonstrated to be generally applicable across the product lines. Based on our reviews of operational experience, we do not agree with 4
GE as to the accuracies involved in the setpoints for-safety / relief valve mechanical trips.
We would expect a plant-specific analysis for such -
1 setpoints.
i 2.5 Section 4 "NRC Open items"
)
5 Section 4 is dedicated to the resolution of the open items identified in a l
previous "NRC Staff Report on Setpoint Methodology for General electric I
Supplied Protection System Instrumentation."
-l Three of these open items are of particular concern because of the interrelationships with other issues or th'e repetitive nature of' the' I
i underlying concern expressed by the staff or the unique nature of the BWR.
+* -
These three issues are evaluated below. However, as a general finding, the topical report is responsive to staff concerns expressed in previous SERs on J
this subject matter.
t A.
Section 4.3 " Allowable Values" I
1 Originally, the staff had recommended that the difference between the Allowable Value and the nominal Trip Setpoint only include the q
drift allowance for that portion of the loop that was checked during the monthly setpoint verification.
The staff was concerned that the i
use of a single technical specification value and a single nominal trip setpoint could lead to a situation where trip unit drift was excessivebutcorrectiveact{onwasnottakenbecausethedrift 4
allowance included the transmitter drift.-
,g L 2 J...
,,,,-y
,._....,3.,w,,.,.,-,r 77,,.
y
l l
i
- GE responded to this, in part_by describing how drift is determined.
They also provided a discussion of the probability that a severely drifting trip unit would be ignored.
Although the probabilistic discussion'may be viewed as argumen-tative, the fact remains that the NRC does not reouire separate Allowable Values for trip units and transmitters in the same loop of any other vendor.
Therefore, the staff accepts the GE response.
B.
Section 4.4 "Expandino Manufacturers Performance Soecifications" The NRC staff requested more information on'the degree to which manufacturer data was verified in the field.
GE responded with a description of how drift data for the instruments presented in Section 2 of the topical report was verified in the field.
The staff finds that the methods presented provide a reasoned approach to determining whether drift is random and estimating the mean value.
However, we also note that Section 1 of the topical states that " Methodology used in establishing drift allowances is.
strongly dependent on the specific application... adequacy.must be.
j demonstrated empirically and.in the field."
l Although GE has adequately demonstrated that drift is a random and normally distributed variable for the instruments studied in this topical, we do not accept the assumption that drift is' random and normally distributed for all instruments.- Therefore, we' agree with GE in requiring that each instrument be demonstrated to have random drift by empirical and field data.
9-i
.s
.k.
4 j-
)
1 1 i
4 C.
Section 4.5 "APRM Validation Calculations" i
j The staff recommended that the values used in.the accuracy calcu-lations for the APRM system be supported by field data.
Section 4.5 l
describes the source and treatment of the manufacturer field data, j
l The discussions appear to be well reasoned and. complete..However,.
j we note that the Fermi 2 Power Uprate effort disclosed additional information and used different equations for trip unit accuracy.
The staff considers this an additional acceptable example of how f
field data and experience can and will modify the'.information i
i presented in Sections 2 and.3 of the topical report ~.
i' j
3.
CONCLUSION i
1 NEDC-31336 is an important reference for understanding how.GE selects instrumentation setpoints.
It demonstrates.that the methods used by GE comply with industry standards and are responsive to previously stated staff concerns.
However, because the topical report is limited to the development of a few l
sample calculations, it is not to be used by any plant as the sole basis for f
individual, plant specific setpoints. That is, each plant must provide its own plant unique analysis for the setpoints. The examples given in the j
topical report are used by GE only'to show the safety margins and typical channel errors that might be expected. Since p'lants have different i
instruments, environments, seismic and other requirements only examples have been provided by GE.
Plants that have safety relief valves that vent directly into containment must substantiate the GE position on harsh tenvironments.
I i
I i
..-....+,n.-
....i
i
. Where instruments are used that are different from those presented in Section 2 of NEDC-31336, the licensee must demonstrate that drift is, or is not, random and normally distributed and can be quantified.
i Except for the selection of safety / relief valve setpoints, the general methods used by GE are acceptable. The methods used to select mechanical valve lift and reset points are not within the purview of HICB.
The use of a single sided test for instrument channels that provide trips for increasing and decreasing variables (e.g. reactor level) is unacceptable.
Similarly, because this topical report is only intended to be used for setpoints of equipment operated in a non-harsh environment, the use of a single sided test for safety-related indicators and recorders is unacceptable.
1 Principal Contributor:
R. Scholl 504-2886 l
l e-l
1
. {
4.
REFERENCES i
1.
NRC letter 8405240338 dated May 15,.1984 " Transmittal of NRC. Staff Report on Setpoint Methodology for General Electric Supplied Pro-tection System Instrumentation" 2.
NRC letter dated March 11,1992 " Request for Additional Information
- Licensing Topical Report NEDC-31336" t
3.
GE letter Serial MFN-085-92/DFR-92-021 dated April 8,1992 "GE Response to NRC Questions Concerning Instrument Setpoint Methodology" 4.
Regulatory Guide 1.105 Rev. 2 " Instrument Setpoints for Safety Related Systems" 5.
ISA-567.04-1982 "Setpoints;for Nuclear Safety Related Instrumen-tation Used in Nuclear Power Plants" i
a a....---
i I
.)
1 j
Mr. B. Ralph Sylvia June 23, 1993 commitment, we request that NMPC provide the staff with the detailed technical assessment within 6 months of the date of receipt of this letter. The staff's i
j safety evaluation for NEDC-31336 was transmitted as an enclosure to a letter from Bruce A. Boger (NRC) to David J. Robare (GE) dated February 9, 1993. A j
copy of that letter and the safety evaluation are enclosed for your use, a
i Sincerely, 4
Original signed by:
John E. Menning, Project Manager Project Directorate I-1 i
Division of Reactor Projects - I/II Office of Nuclear Reactor Regulation i
Enclosures:
1.
Letter 2.
Safety Evaluation cc w/ enclosures:
See next page DISTRIBUTION:
Docket File NRC & Local PDRs PDI-1 Reading SVarga JCalvo RACapra CVogan JMenning EMarinos, 8/H/3 OGC ACRS (10)
CCowgill, RGN-1 0FFICE PDI-1:LA PDI-1:PM PDI-1:D NAME CVogan 4 JMenningN71 RACapra M
/7/M93 L/n/93 6 /23/93
/ /
/ /
DATE OFFICIAL RECORD COPY FILENAME: G:\\NMP2\\NM2SE.LTR
.r
- - - -- w w,
w
,,,-,