ML20212D491
| ML20212D491 | |
| Person / Time | |
|---|---|
| Site: | Vogtle  |
| Issue date: | 02/24/1987 |
| From: | Bailey J GEORGIA POWER CO. |
| To: | NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM) |
| References | |
| GN-1349, NUDOCS 8703040130 | |
| Download: ML20212D491 (6) | |
Text
~
.. t-e Georgd Power Company
- Rout)2 Box 299A -
Wrn &sboro, Georgia 30830 Teleg me 404 554 9961 '
404 724 8114 Southem Company Services. Inc.
Post Ot' ice Box 2625 Birmingham, Alabama 35202 Telephone 205 870-6011 Vogtle Proj.ect February 24, 1987 U. S. Nuclear Regulatory Commission File: X7BC35 Attn: Document Control Desk Log:
GN-1349 Washington, D.C.
20555 NRC DOCKET NUMBER 50-424 OPERATING LICENSE NPF-61 V0GTLE ELECTRIC GENERATING PIANT - UNIT 1 DIGITAL METAL IMPACT MONITORING SYSTEM (DMIMS) STATUS Gentlemen:
In response to FSAR Q492.1, VEGP committed'to provide a final DMIMS baseline report prior to power operation. In order to prepare the report it is necessary to collect data at the 100% power level. Attached is a description of the system, status, and procedures associated with the DMIMS. The attached Unit 1 status report will be supplemented with the final baseline report which is presently scheduled for submittal 90 days after reaching the 100% power level.
Should your staff require any additional information, please do not hesitate l
to contact me.
l.
Sincerely, t
.&. 6 J. A. Bailey Project Licensing Manager JAB /KWK/caa Enclosure xc:
J. P. O'Reilly G. Bockhold, Jr.
R. E. Conway NRC Regional Administrator L. T. Gucwa NRC Resident Inspector j
R. A. Thomas B. Jones, Esquire J. E. Joiner, Esquire D. Feig B. W. Churchill, Esquire R. W. McManus M. A. Miller (2)
Vogtle Project File l
t t 8703040130 870224 PDR ADOCK 05000424 P
PDR L
c._
Digital Metal Impset Monitoring System (DMIMS) Status Issee_
The NRC requires the. licensee-to provide a final baseline report prior to exceeding 5% power which contains the following:
A.-
An evaluation of the Loose Parts Monitoring System for conformance to Regulatory Guide 1.133.-
B.
A description of the system. hardware, operation, and implementation of the loose parts' detection program, including plans for startup testing, acquisition of baseline data, and alarm setting.
C.
A description and evaluation of diagnostic procedures used to confirm the presence,of a loose part.
VEGP Response:
In response to FSAR Q492.1, VEGP committed to provide a final DMIMS baseline-report prior to power operation (Amendment 13 - January 1985). Upon evaluation of this commitment _it has been determined that it is necessary to collect data at 100% power. The desirable alarm settings which will be provided in the baseline report are based on the smallest peak impact value generated by the 0.5 ft.-lb. simulated impacts and evaluation of the background noise level. The evaluation of the background noise level results in alarm settings that will be as sensitive as possible without causing false
- alarms.- The acoustical properties of the NSSS change with respect to temperature, pressure, and steam, and steam flow. The background noise levels are different for each monitored location _and each sensor will respond differently to the same excitation. The final baseline report takes into account the individual response that each sensor will have to a given
. excitation.
Also,- this report considers the crest factor of the background j
- noise for that particular sensor and location in the normal mode of operation
~
to determine the alarm setpoint. VEGP believes the 100%-power recording of baseline data provides the necessary information required to determine alarm settings 'that are' as close to the background noise level as possible without causing false alarms..
l..
A.
Conformance to Regulatory Guide 1.133 i_
j VEGP FSAR Section 1.9 contains VEGP's conformance to Regulatory Guide 1.133.
i -
B.
DMIMS System Design and Hardware The DMIMS consists of 12 active instrumentation channels, each comprising a piezoelectric accelerometer (sensor), signal conditioning, and diagnostic equipment. Two redundant sensors are fastened mechanically on the RCS at the upper head region (Reactor Pressure Vessel), lower head region (Reactor Pressure Vessel) and each steam generator (Reactor Coolant Inlet Region). Sensor mounting details are provided in Table 492.1-1.
FSAR Section 4.4.6.4, Q492.1 and GN-429
~ (Proprietary), Foster to Denton, dated 10/12/84 provide details of the DMIMS design.
F'. :
.s DMIMS STATUS Page 2
' System Operation:
VEGP has developed operational procedures which provide instructions for plant personnel for the operation and implementation of the DMIMS. The following procedures implement the DMIMS program.
- Procedure 13902-1, " Digital Metal Impact Monitoring System" provides instruction for the operation of the DMIMS. Instructions include:
Aligning the DMIMS for operation Selecting an alternate channel Obtaining event reports DMIMS-self-test i
Power lost and restored
- Procedure 17063-1, " Annunciator Response Procedures for ALB 63 on Process control Panel" provides guidance for specific operator response to given
. alarms.
- Procedure 24679-1, " Digital Metal Impact Monitoring System Channel Calibration" provides instruction for an Analog Channel Operational Test and a Channe1' Calibration of the DMIMS. The performance of this procedure satisfies analog channel operational test and channel calibration surveillance requirements at VEGP.
- Procedure 14000-1, " Operations Shif t and Daily Surveillance Logs" provides guidance for daily surveillance to detect the presence or possibility of a loose part.
Plant personnel may actuate the data acquisition system to obtain data for further evaluation.
Preoperational Testing:
A preoperational test was performed as described in FSAR paragraph 14.2.8.1.95.
This test was performed to establish initial calibration l
operability of the DMIMS and initial alarm setpoints. The initial setpoints l
were established at:
- Signal Conditioner - 10g (full scale range)
- Alarm Setpoint - 1.2g
- Threshold - 0.lg These setpoints will be optimized after the incorporation of the 100% power test data.
l
9
- (
BMDB STATUS
(-
Page 3'
'StahtopTesting:
During the Initial? Test Program, startup testing of the DMIMS wac performed to:
determine final calibration of the DMIMS, demonstrate system sensitivity and operate DMIMS in conjunction with plant startup testing.
Startup testing will continue through the 100% power test data.
Data Acquisitions c
The data collection will.be in accordance with the latest revision of the Westinghouse DMIMS Baseline. Signature Acquisitica Procedure, NSID-EIS-82-10.
Data collected will be recorded on a 14 channel FM magnetic tape recorder set
- for a minimum bandwidth of 0-20 kHz, per IRIG specifications.
All data from the tests will be evaluated using spectral analysis and transient recording equipment to obtain information on the frequency and time domain characteristics of background signature and hiaulated impact data.
All data analyzed will be produced from a series of simulated impacts made at each location.at a specific distance away'from each accelerometer. Impacts made on the steam generator's primary and secondary sides are to be generated l
at a-distance of approximately 30 inches away from each accelerometer.
Impacts made on the Reactor Vessel Top are to be generated midway between the lifting lugs. Impacts made on the Reactor Vessel Bottom are to be generated 30 inches away from each thimble tube which has an accelerometer mounted on i
it. - Since the distance from the impact site to the accelerometers varied with each location, the signal amplitudes of each impact generated will vary l'
accordingly. Therefore, all factors, including impact site, distance and mass will be taken into account when choosing the final alara setpoint for the DMIMS.
Evaluation and Diagnostic Procedures GPC han developed controls which direct the operator to notify NSS System Engineering upon detection of a loose part or suspected loose part. NSS System Engineering vill evaluate for immediate action and notify Westinghouse to analyze data acquired. The Westinghouse diagnostic procedure will include the following:
1.
Acoustic Signal Analysis j-A. Observe impact shape in the time domain to qualitatively evaluate l
signal path and distance to accelerometers. By examining the impact's rise and decay times, and amplitude, its relative proximity to accelerometers may be qualitatively estimated.
B. Compare frequency spectra with known impacts (baseline data). The ratio of particular frequency components present in a metal impact are indicative of the characteristic length or mass of the impact source. The baseline data may be used to identify the particular characteristic ratios.
L
6 DMIMB STATUS Page 4
.C.
Evaluate time delays between channels to determine the direction'of the sound path.. This, in conjunction with impact shape, is used to' evaluate the source of the impact.
D. Evaluate the repetition rate of the signal. If the repetition rate is high, it is more likely. that a loose part does exist as opposed to the impact being noise-generated from a normal plant transient.
II.
- Significant Issues Affecting Determination of Required Corrective Action A. Mass of Objects: The loose part mass may be determined by comparison of the impact to baseline data. In general, it has been noted that objects with a larger mass have grater potential to damage some plant components.- - If the loose part mass is known, assumptions may be made about its impact velocity and a more accurate judgment may be made concerning the plant safety.
B. Magnitude of Impact: As with object mass, impact magnitude is also a good indicator of the potential damage which may be inflicted by a loose part.
If the impacts are of relatively large magnitude a greater potential for damage exists. This may be determined by observation of the impact signal in the time domain.
C. Location of Impact Site:
Certain areas in the plant are more susceptible to loose part damage than others and have greater potential risk concerning plant safety.
Examples of these areas are the fuel region, the tube sheet, and the secondary side steam generator U-tubes. Location of the impact site may be made by noting which accelerometer exhibits the highest amplitude for a given impact and by determining the time delays between accelerometers which detected the same impact. It is noted that some impact sites may have safety implications and some sites.may result in only commercial considerations.
D. Mobility: Mobility is determined by noting whether time delays between accelerometers remain constant, whether the impact signal is detected in different vessels, and noting changes-in relative amplitude. The damage potential of a highly mobile loose part is uncertain due to the possibility of the part becoming lodged in, or migrating to, a sensitive location. If a part has limited mobility, a more accurate assessment of plant safety effects may be made.
However, for such a loose part the damage may be concentrated in a single area.
E. Repetition Rate: Repetition rate may be determined by timing the rate of occurrence of the impacts. Impacts with a higher repetition rate, of course, have greater potential to cause damage.
[T '
... 4 DIGMB STATUS Fase 5 F. Retrieval Expense vs. Potential Damage: -Once a reasonable assessment of potential damage has been made the retrieval expense should then be estimated.
Consideration must be given to.the type of -retrieval, probability of. retrieval, and plant scheduling.
Certain types of retrieval are more expensive than others and should be weighed differently. Parts in a reactor vessel, for example, are more
~
expensive to retrieve than in the primary side of a steam generator.
A.certain amount of uncertainty exists when a determination of impact location is made. Because of this, the possibility exists that once a retrieval has been attempted the loose part will.not be found in the expected location. This could occur as a result of the' part moving after the last impact was detected or inaccuracies in the data analysis. Factors to be considered are the location of the part, whether it lies in an accessible area, the number of impacts available for analysis, and the number of accelerometers which -
detected the impacts and are usable for analysis.
If the plant is scheduled for an outage in the near future it may be
-more cost-effective to wait until the scheduled outage before attempting the retrieval oper *. ion. This, of course, must be weighed against the potential damage waich may occur in this time period.
l l
l 1005V
--