ML20009B414
| ML20009B414 | |
| Person / Time | |
|---|---|
| Issue date: | 05/11/1980 |
| From: | Budnitz R NRC OFFICE OF NUCLEAR REGULATORY RESEARCH (RES) |
| To: | Harold Denton Office of Nuclear Reactor Regulation |
| References | |
| RIL-089, RIL-89, NUDOCS 8107150400 | |
| Download: ML20009B414 (5) | |
Text
l e
Y
\\ L UNITED STATES
((n p
NUCLEAR REGULATORY COMMisslON n
WASHINGTON, D. C. 20555
(
- ^j AMY 111990
%....+/
MEMORANDUM FOR: Harold R. Denton, Director Office of Nuclear Reactor Regulation FROM:
Robert J. Budnitz, Director Office of Nuclear Regulatory Research
SUBJECT:
RESEARCH INFORMATION LETTER NO. 89,
" STRUCTURAL AND MECHANICAL COMP 0NENT TEST TECHNIQUES" INTRODUCTION Section 3.9.2 of the Standard Review Plan, NUREG-75/087, entitled " Dynamic Testing and Analysis of Systems, Components and Equipment," gives NRC requirements for dynamic qualification testing.
These requirements include seismic qualification tests, usually limited to tests of prototype equipment conducted on shake tables, and in situ flow-induced vibration tests of piping and reactor internals.
Ordinarily, significant measurement and analysis of the test results does not occur, except for reactor internals. This is in contras+. with Japanese practice in which an in situ preoperational confirmatory seismic test is performed on each nuclear power plant using large shakers mounted directly within the containment. These large shaker tests yield confirmatory model information on virtually all structures and piping systems within the containment, including the containment building. The information so obtained is subjected to considerable scrutiny and analysis during the seismic safety assessment.
This RIL transmits the results of studies which explore and investigate the feasibility, costs, benefits, reliability, limitations and potential plant degradation associated with confirmatory in situ dynamic testing utilizing various means of exciting vibrations in safety-related structures and mechanical equipment.
Initially, the focus was on seismic testing; nonetheless -the same experimentally verified analytical models could equally be useful for calculating response for other dynamic situations, such as LOCA blowdown and different missile impact situations. This RIL treats only the various testing methods.
A forthcoming RIL will deal with the computer codes used to interpret data collected from these testing methods and the compatibility between test techniques and competer codes.
DISCUSSION The test methods evaluated in this RIL are intended to be used with in situ, full-scale and fully built nuclear power stations.
Approximately 30 such nuclear power plants have, so far, been tested worldwide since 1965 using confirmatory dynamic test techniques and plans of different scopes. Several of these tests were conducted in the USA. Potential objectives for confirmatory dynamic testing are:
8107150400 800511 PDR RES 8107150400 PDR
m Harold R. Denton For new plants, dynamic testing may be used for confirmation of design models employed in safety evaluations, and experimental assessment of the assumptions in computer codes.
For older operating plants, dynamic testing may be used for assessment of dynamic parameters, particularly damping.
For plants subjected to severe environmental or accidental events, dynamic testing may be used for damage assessment and requalification subsequent to the event.
For improved safety at all plants, dynamic testing may be used to detect structural degradation and construction errors.
For decommissioned plants, dynamic testing may be used to establish fragility levels of safety-related structures and mechanical equipment. The lack of good fragility data is a serious impediment to quantitative risk assessment.
As indicated in RESULTS, not all these objectives are presently achievable.
The confirmatory dynamic test methods evaluated include snap-back experiments, pulse generators, rockets, buried explosives, ambient vibrations and shakers. No attempt to indicate how confirmatory dynamic testing can be integrat' d into NRC licensing practice was made, although further effort e
is planned in this direction.
APPROACH RES required that its contractor, Lawrence Livermore Laboratory, engage consultants possessing significant experience in confirmatory dynamic testing of nuclear power plants.
The contractors were then asked to critique the reports prepared by their competitors.
Lawrence Livermore Laboratory then reviewed and commented on the work of their contractors and presented RES with their recommendations.
The results reported below are based on NRC's evaluation of the work of Lawrence Livermore Laboratory and its contractors.
RESULTS The practical choice for conducting confirmatory dynamic tests at nuclear power plants calls for the use of sinusoidal vibrators, especially eccentric mass units.
Linear hydraulic, reciprocal hydraulic and electro-dynamic shakers may also be used.
Hydraulic units are less portable while electrodynamic units are typically of much smaller capacity.
There is no known case where a nuclear power plant was dynamically tested using buried explosives and, subsequently, produced electricity.
Although buried explosives (more closely than other methods, but still with some deficiencies) model earthquake motions better than other techniques, they present problems in controlling damage, particularly
As-Harold R. Denton for facilities in the vicinity of the power station not designed for intense seismic motions.
Buried explosives also give rise to various environmental and safety concerns, do not offer great opportunity for repeating tests and lead to difficulties in handling data and extracting from these data high-quality estimates of dynamic parameters; Nonetheless, it is recognized that the evaluation of dynamic parameters is dependent on the test technique; thus, buried explosives, which most closely simulate earthquakes, may give more representative dynamic parameters appropriate for seismic safety evaluations.
Ambient vibrations, primarily due to the need to extrapolate through several orders of magnitude in response accelerations and the need to make statistical assumptions about the sourc1 of vibrations, usually lead to poor estimations of dynamic parameters.
However, costs are low for ambient tests.
Rocket and pulsor techniques are re-tly developed methods in which experience is still being gained. At the present time, with respect to nuclear power plants, it appears that relatively low levels of excitation can be achieved using rockets and pulsors.
Moreover, rockets may not be readily used in enclosed facilities and present safety problems during handling, installation and testing.
Snap-back tests are useful in studying local response, but cannot be employed for exciting containments or large equipment within containments due to the difficulty in achieving appropriate force levels.
In terms of reliability of the test data obtained, repeatability, control of plant degradation during testing, maintaining moderate cost, minimizing safety and environmental concerns, and imposing fewer constraints on utility of results, sinusoidal shakers are the preferred excitation source.
Snap-back and pulsor devices (not including rockets) may also be used, but are less attractive.
Buried explosives, ambient vibrations er' rockets can be used, but are the least desirable.
Of the potential objectives listed under DISCUSSION of this RIL, two may not be achieved by confirmatory dynamic testing.
Studies indicate that dynamic parameters change substantially even when no damage occurs in structures and equipment; thus, damage assessment by monitoring changing dynamic parameters does not appear feasible.
In addition, dynamic testing may not reveal structural degradation or construci. ion errors of importance in many situations, but could assist in identifying very gross degradations and errors.
Not as a stated objective, but as an accrued side benefit, confinnatory dynamic testing can uncover previously unrecognized phenomenology or sensitivities which affect safety evaluations.
Examples of this include energy transfer between steam generators undergoing vibratory motions, initially decreasing damping with increasing excitation level, as observed at-Diablo Canyon, and the great sensitivity of equipment eigenfrequencies to the torque applied to tighten supporting bolts, i
e Harold R. Denton Confirmatory dynamic testing, except when the objective is to establish fragility data, will be at low excitation levels in order to forestall damage, regardless of the test technique selected.
A limitation of low-level testing is that it may not yield accurate information for design basis level situat':ns. Usually, however, bounding values of dynamic parameters can be of considerable value.
Costs of confimatory dynamic testing vary depending on the compre-hensiveness sought. A reasonably thorough in situ test of a large comercial nuclear power plant including all planning and preparations, data reductions and associated modeling and analyses would cost on the order of 0.3 to 1.0 million current dollars.
For the lower cost estimate, a single test technique may be applied to a limited number of structures and systems. For the upper cost estimate, a few different techniques, i
useful for checking experimental results, may be applied to a larger number of structures and systems.
CONCLUSIONS i
1.
In situ confirmatory dynamic tests of nuclear power plants to l
assess design models and dynamic parameters are feasible, have been proven over the last 15 years, and have become the practice in Japan regarding seismic design adequacy.
Confirmatory dynamic testing can also be beneficial with respect to other severe environmental and accident situations.
2.,
The use of shakers appears to be the preferred and the best established test technique for the routine acquisition of confirmatory dynamic test data.
Shakers may be used for buildings, equipment, and components, but other methods may be used.
3.
There is not enough evidence to believe that confirmatory dynamic testing can contribute at this time to damage assessment or requali-fication,for example, in post-0BE inspection.
4.
Confirmatory dynamic testing can reveal phenomenology and sensitivities presently ignored or not understood, but which may be significant in seismic or other environmental and accidental safety evaluations.
5.
Confimatory dynamic testing can detect only very gross and major structural degradation and construction errors.
This could, however, be of great value despite the limitation to large effects.
The three contractor reports obtained through Lawrence Livermore Laboratory which document these findings and give greater detail are published as NUREG reports.
Harold R. Denton RECOMMENDATIONS Based on technical and economic considerations, this research indicates that it is feasible to use in situ dynamic seismic testing to confirm and resolve issues regarding seismic vulnerability at new and already operating plants. Good confidence exists that the results of such testing will assist in licensing decision making.
However, in situ dynamic testing cannot be a major contributor to damage assessment and identification of construction and installation errors at this time.
Robert J. Budnitz, Director Office of Nuclear Regulatory Research