ML20072D308
| ML20072D308 | |
| Person / Time | |
|---|---|
| Site: | Calvert Cliffs  |
| Issue date: | 02/01/1983 |
| From: | BALTIMORE GAS & ELECTRIC CO. |
| To: | |
| Shared Package | |
| ML20072D305 | List: |
| References | |
| NUDOCS 8303210192 | |
| Download: ML20072D308 (18) | |
Text
F
]
,L .
.c, .
Enclosura 3 CALVERT CLIFFS NUCLEAR POWER PLANT UNIT NO.1 SAFETY EVALUATION FOR PROPOSED TECHNICAL SPECIFICATION REVISION TO
- CONTAINMENT STRUCTURE POST-TENSIONING SYSTEM LIMITING CONDITION FOR OPERATION l
AND TENDON SURVEILLANCE REQUIREMENTS BALTIMORE GAS AND ELECTRIC COMPANY FEBRUARY 1,1983 8303210192 930307 gDRADOCK 05000317 PDR
l
-e
- 6
- 9 Q NTENTS SECTION
- 1.0 . PURPOSE
' 2.0 ' BACKGROUND 3.0 COMPARISON OF EXISTING AND PROPOSED SPECIFICATIONS 4.0 SAFETY EVALUATION
5.0 CONCLUSION
1.0 PURPOSE Pursuant to the requirements of 10 CFR 50.59, this report pro. ides a safety evaluation in support of the proposed revision to Technical Specification 3/4.6.1.6, Containment Structural Integrity.
2.0 BACKGROUND
Post-tensioning of the Unit I containment building was completed in 1972. The Unit 1 integrated leak-rate and initial structural integrity tests were completed two years later in 1974 prior to plant operation. In accordance with the surveillance program reequired by Technical Specification 3/4.6.1.6, the first, third, and fifth year tendon surveillances were performed in 1975,1977, and 1979, respectively. The intent of these periodic surveillances, as described in the bases for the technical specification, is to ensure that the structural integrity of the containment will be maintained comparable to the original design standards for the life of the facility.
The results of the first, third, and fif th year surveillances indicated that the containment post-tensioning system was performing adequately; however, some of the litt-off forces that were measured during the third (1977) surveillance fell below the limits established by th( technical specification. Although there was no evidence that these lower measurements indicated an overall trend of degraded performance by the post-tensioning system, it was felt that a confirmatory reanalysis of the containment design was needed to verify conformance with the original design standards. An engineering evaluation was then performed by Bechtel Power Corporation which included a reanalysis of the post-tensioning system utilizing updated material properties data for the concrete used in the Unit I containment. The results of this evaluation confirmed that the existing post-l tensioning system will continue to satisfactorily perform its design function j throughout the licensihJ lifetime of the plant.
I
During the engineering evaluation described above, the following additional facts became clear:
- 1. More realistic data were presently available for estimating time-dependent losses in the tendons than were available for use when the existing technical specification was written.
- 2. The ratio of total prestress force to total design pressure force for the original design was higher than that required by the design load combinations for all three groups of tendons (hoop, vertical and dome).
- 3. The existing technical specification has the following shortcomings:
. The specification does not include time-dependent prestress force limits.
. It does not require normalizing of measured lift-off forces.
. It does not account for the expected scatter of tendon lift-off force measurements.
. It does not incorporate acceptance criteria for the average lif t-off force of each group of tendons.
}
. It ' does not ' recognize the necessity to perform an engineering evaluation of surveillance results. This evaluation is required to determine whether an apparent reduction in individual or multiple tendon performance is indicative of an overall degradation of tendon performance within one of the three families of tendons, or:
was caused by environmental conditions not properly accounted for in the test procedure; was the result of measurment errors introduced by a specific characteristic or deficiency inherent to the test method employed; or was an isolated occurrence not significantly affecting the structuralintegrity of the containment.
. The action statement is not appropriate for surveillance inspections conducted on end anchorages, adjacent concrete surfaces, or the liner plate in that it does not recognize that fact that these inspections are conducted while the plant is shut down.
The proposed changes to Technical Specification 3/4.6.1.6. are intended to eliminate these short-comings and in so doing, provide for more thorough and meaningful surveillance practices at Calvert Cliffs.
l
1 3.0 COMPARISON OF EXISTING AND PROPOSED SPECIFICATIONS The proposed revision only affects the action statement of the Limiting Condition for Operation (Specification 3.6.1.6), and the surveillance requirements for Containment Tendons, (Specification 4.6.1.6.1). The minimum required number of surveillance tendons and the required surveillance schedule are not affected. No changes are being proposed to the requirements for inspections of end anchorages, adjacent concrete surfaces, or the liner plate (Specifications 4.6.1.6.2 and 4.6.1.6.3).
3.1 Limiting Conditions for Operation The action statement in the existing technical specification is a single paragraph which requires, when the structuralintegrity of the containment does not conform to a level consistent with the acceptance criteria specified in the surveillance requirements, that structural integrity be restored within 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> or that the plant be shut down.
In the proposed revision to the specification, the action statement is divided into two parts. The first defines the action that would be applicable during tendon surveillances; the second defines the action that would be applicable during surveillances of end anchorages, adjacent concrete surfaces, and the liner plate. This segregation of the action statements recognizes an important distinction between the two basic types of surveillances governed by this specification. Inspections and tests of tendons can be performed during plant operation. On the other hand, inspections of end anchorages, adjacent concrete surfaces, and the liner plate must be performed when the plant is shut down.
1 l
The proposed action statement relative to tendon surveillance is similar to the existing specification, except that the 24 hour2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> limit for restoring structural integrity has been ' deleted nnd replaced with a regi!rement to perform an engineering evaluation demonstrating the continued ability of the containment structure to perform its design function. If continued containment integrity cannot be assured by engineering analysis'within 90 days, the plant must be shut down.
The proposed action statement relative to the surveillance of end anchorages, adjacent concrete surfaces, and the liner plate requires restoration of structural i
l Integrity or an engineering evaluation to verify structural integrity prior to l
returning th'e plant to operation.
3.2 Tendon Lift-Off Testing Existing l
l The existing specification requires that individual tendon lift-off forces lie between a maximum and a minimum bound. These bounds are constant values, expressed in pounds per tendon wire, which are applicable to all l three groups of tendons (dome, vertical, hoop). The lower bound is l
l approximately equal to the minimum effective prestress after 40 years l based on the estimated prestress losses assumed for the original design, and l
l Is about 57% of the ultimate strength of the installed post-tensioning l
wires. The upper bound is approximately equal to the prestress at lift-off during the initial installation, or abou'; 74% of the ultimate strength of the wires. If the lift-off force of any tendon lies outside the predicted bounds, testing of the two adjacent tendons is required. If more than one tendon
out of the original sample population lie outside the predicted bounds, it is considered to be evidence of abnormal degradation of the containment structure.
The existing specification requires that lif t-off testing include an unloading cycle in which each of the surveillance tendons is detensioned to determine if any wires are broken or damaged.
Proposed Revision Figures 4.6-1, 4.6-2 and 4.6-3 of the revised specification are graphs that define the variation in prestress force (in kips per tendon) with time (in years af ter initial prestressing) for hoop, vertical and dome tendons. Each graph has time dependent plots for expected prestress, lower limit of the expected prestress range, and lower bound prestress for individual tendons. The required average prestress force is plotted as a constant 1
value that lies between the expected prestress and lower bound prestress l plots. The graphs do not include an upper bound,'since the purpose of lif t-off testing is to verify that the tendons can perform their intended function, and higher than expected prestress is advantageous.
The " expected" prestress plot shows the variation of prestress with time based on the "best estimate" of time dependent losses. For the original design, shrinkage and creep losses were extrapolated from 28 day test results available at that time. When the 1977 engineering evaluation was performed, test data were available for shrinkage and creep of concrete cylinders made from the same ingredients and mix design as the containment structure and loaded at 180 days and 365 days. This test data was correlated to the age after placement at which the dome and cylinder wall were prestressed and provided a more reliable estimate of the losses.
This same test program provided elastic moduli for the concrete for use in determining more reliable estimates of the elastic losses. Also, test data was available from the wire manufacturer which confirmed the estimates of wire relaxation losses. The combined net effect of using the revised losses is slightly higher expected prestress forces at 40 years than the corresponding value of the existing specification.
The " lower limit expected range" of prestress shown on the graphs is a plot parallel to and about five percent lower than the expected prestress plot.
The lower limit was determined by considering variations in material properties and measurement tolerances. The five percent difference is consistent with the scatter between mean and minimum lift-off values experienced during past surveillances at Calvert Cliffs and other plants.
The lower limit of expected range results in a prestress level (ratio of total prestress load to toal design pressure load) of at least 1.20 after 40 years of service.
The " required average" prestress plot is the average prestress force that was verified to be acceptable for all of the FSAR load combinations by reanalysis of the containment structure af ter the 1977 surveillance. This required average prestress provides a prestress level equal to or greater than 1.20 in all cases. It is a conservative limit established for surveillance use, and is higher than the minimum prestress level required to meet all design criteria at the end of the life of the plant.
t l
The " lower bound individual" prestress shown on the graphs is the lowest plot parallel to the expected prestress plot. The difference between the lower bound individual plot and the required average prestress plot
-Increases with time, with a maximum difference of about ten percent at 40 years. The lower bound individual plot is provided to account for anticipated maximum scatter of individual test results.
The revised specification requires that the normalized lift-off force of each tendon equals or exceeds the lower limit of expected range for the time of the tests. If the normalized lift-off force of any tendon lies between the lower limit expected range and lower bound individual limit, testing of the adjacent tendons is required. If the normalized lift-off force of any of the adjacent tendons lies between these same limits, it is considered to be evidence of possible abnormal degradation of the containment structure.
The revised specification has two additional acceptance criteria for lift-off tests. It also is considered to be evidence of possible abnormal degradation of the containment structure, if:
a) .the normalized lif t-off force of any tendon lies below the lower bound individual limit; or b) the average of the normalized lif t-off forces for any group (hoop, vertical or dome) lies below the required average pre-stress limit. ,
= .
3.3 Wire and Filler Grease VisualInspection The requirements for visual inspection of wire and filler grease are essentially the same in the existing and revised specifications. However, the revised specification is more prescriptive in that it identifies the presence of abnormai corrosion, cracks or other damage to a wire as evidence of possible abnormal
. degradation of the containment structure.
3.4 Wire Testing The requirements for tension testing of tendon wires are the same in the existing and revised specifications. The existing specification identifies the failure of any one of the wire samples to meet the minimum tensile strength test as evidence of abnormal degradation of the containment structure. The revised specification identifies such a failure as evidence of possible abnormal degradation of the containment structure.
4.0 SAFETY EVALUATION The proposed revisions to the existing technical specifications are logical, definitive, specific, and a significant improvement over the existing specifications.
4.1 Limiting Condition for Operation In lieu of an unspecified time frame to determine abnormal degradation, as it exists in the present specification, the revised specification includes a schedule for specific action in the event that possible abnormal degradation is identified; i.e.,
9_
Lm-
Existing Proposed Assure continued containment integrity Not specified 90 days by engineering analysis, after evidence of possible abnormal degradation of containment tendons is detected, or initiate plant shutdown.
The revised specifications recognize that the acceptance criteria of the containment tendon surveillance requirements are in reality preliminary warning signals that need to be investigated. None of the occurrences identified as evidence of possible abnormal degradation is in itself an indication that requires emergency action. Instead, further testing and/or engineering analysis are indicated on a case by case basis. It is noteworthy that the wording "possible abnormal degradation" used in the revised surveillance requirements is consistent with the wording of Regulatory Guide 1.35.
l The inherent safety of the containment structure design, relative to the l
) occurrences that the revised tendon surveillance requirements consider to be possible abnormal degradation of the containment structure, is discussed in the following paragraphs. This inherent safety assures that reasonable periods of time are available for engineering analysis supplemented by additional testing, as appropriate, following the detection of such evidence.
i 1
~
i l
4.2 Tendon Lift-Off Testing The prestress limits specified in the existing and revised specifications, expressed as a percentage of the ultimate tensile strength of the tendons, are as follows:
One Year After 40 Years After Initial Prestress Initial Prestress Existing Specification Limits for Measured Lift-Off Forces Maximun Prestress 74 % 74 %
(Hoop, Vertical & Dome Tendons)
Minimum Prestress $7% 57%
(Hoop, Vertical & Dome Tendons)
Proposed Specification Limits for Normalized Lift-Off Forces Expected Prestress, Hoop 63% 60 %
Vertical 64 % 61%
Dome 62 % 58 %
Lower Limit Expected Range, Hoop 60 % 60 %
Vertical 61% 57 %
Dome 58 % 54 %
Required Average Prestress, Hoop 51 % 51%
Vertical 59 % 59 %
Dome 52% 52%
Lower Bound Individual, Hoop 49 % 45 %
Vertical 56% 53 %
Dome 51 % 47%
The use of normalized lif t-off forces is required by the revised specification.
Normalizing of the measured lift-off forces corrects for variations in the elastic losses, resulting from progressive shortening of the concrete as a function of the sequence in which the tendons were post-tensioned, so that the corrected value is indicative of the average prestress for the containment.
u f a The revised specification has added the requirement that the average of the l normalized ilft-off forces for any group equals or exceeds the specified required average prestress force. This is an important improvement, since the design of the containment structure is based on the average prestress force in each group of tendons. The containment design criteria requires that the contalment remain in membrane compression for the initial structural integrity test and other working stress design load cases. For combined factored load cases, membrane concrete cracking is permitted. This cracking allows the tendons to be retensioned from the effective value until force equilibrium is satisfied. In all cases, a total of three hoop, three vertical, and three dome tendons are allocated to surveillance use and are considered to be totally ineffective.
The ratio of total effective prestress force to totalinternal pressure force is a useful gage of membrane compression. For the Calvert Cliffs design, the ratio of the minimum prestress force given in the existing specification to the design pressure force is 1.30, which is higher than the ratio required by the design load combinations. Reanalysis of the contalment structure design performed during the 1977 engineering evaluation verified that the reductions of prestress to the
" required average" prestress values, when combined with all required design loads, did not significantly affect the stress levels in the containment structure.
Lift-off test requirements of the revised specification consider the following occurrences to be evidence of possible abnormal degradation of the contaiment structure:
12 -
a)- The normalized lift-off force of more than one tendon in any group (hoop, vertical or dome) lie between the expected lower limit and the lower bound individual limit.
. b) The normalized lift-off force of any tendon lies below the lower bound individual limit.
c) The average of the normalized lift-off forces for any group (hoop, vertical or dome) lies below the required average prestress limit.
There is no immediate hazard associated with continued operation of the plant, when any of the occurrences consMered to be evidence of possible abnormal degradation is detected. The containmed structure would continue to serve its design function if the ratio of total required aveage prestress force to total design pressure force were reduced. As an example, if the ratio were reduced to 1.10 for the hoop tendons and the prestress levels of the vertical and dome groups were reduced proportionally, the post tensioning system would require a total of only 772 tendons (414 hoop,178 vertical and 180 dome) with the specified required average prestress. This suggests that a total of 90 tendons could have zero prestress. Or to state this another way, the required average prestress in this example could be reduced from 51% of the ultimate tensile strength to 46% for the hoop tendons, from 59% to 53% for the vertical l tendons, and from 52% to 47% for the dome tendons with no violation of design function.
1
{ s. e:
x Even if the initial' results 'of surveillance tests suggest acceleratid loss of prestress - below _the design pressure, containment integrity would be maintained.- Pressurization of the containment from a postulated accident causes the containment to grow outward and upward. If the average prestrer,s level !s below the accident pressure such that membrane compression is lost, further expansion of the contalment will cause the tendons to elongate. This eloagation results in restressing of the tendons, restoring the pressure-retaining capability of the containment and creating an equilibrium condition in the ;
structure. This phenomenon utilizes the accident pressure to increase the level of prestress and has been employed to demonstrate by analysis that the ultimate capacity of prestressed containment structures exceed their design capacity by a factor of 2 or more regardless of the initial prestress level. Two such analyses are presented in References (1) and (2), which are technical papers that were presented at the Workshop on Containment Integrity sponsored by the U.S.
Nuclear Regulatory Comission in June,1982. Although no such analysis was conducted specifically for the Calvert Cliffs containment, the similarities in design concept with those that have undergone a-detailed evaluation further enhance the conclusion that significant margins exist in the post-tensioning system beyond those identified in the design. These margins will prevent loss of containment function should lower than expected prestress levels occur.
The warning signals identified as "possible abnormal degradation" may require a significant review effort, including engineering evaluation, before resolution is firialized. Post-tensioned containments represent a sophisticated design concept requiring detailed analysis. In addition to the complexity of the design, l
l
.(1) Internal Pressure Capacity of Prestress Concrete Containment, P. Shunmugavel and T. Johnson.
(2) Analyses of Zion Containment, A. Walser.
, - _ _ -_ _ _ _ _ _ . -_ _ -_. __ . _ . ~ _ . _ _ - , - - _ _ - .
7 the number of components are substantial. There are a total of 871 tendons (465 hoop, 202 vertical and 204 dome) installed in the containment structure.
Of these, a total'of 9 tendons (3 of each group) are allocated to surveillance utilization and therefore are not considered to participate in the post-tensioning function, leaving 362 tendons (462 hoop,199 vertical and 201 dome) to contribute to the effective prestress of the structure. Recognizing that significant margins exist in the design of the structure, it is appropriate that a period of up to 90 days be provided to allow a timely but thorough and intelligent conclusion with regard to any early warning signals.
4.3 Wire and Filler Grease Inspection The revised tendon surveillance requirements consider the presence of abnormal corrosion, cracks or other damage to a wire to be evidence of possible abnormal degradation of the containment structure. There are a total of 77,533 wires (41,562 hoop,17,890 vertical _and 18,081 dome) installed in the Calvert Cliffs containment structure, not counting those in the nine tendons allocated to surveillance use. The detection of an abnormal condition in one of three samples in a total population of 77,533 wires does require additional investigation, but cannot realistically be considered a significant hazard.
I' o
4.4 - Wire Testinst The revised specification also considers the failure of any one of the wire samples to meet the minimum tensile strength test to be evidence of possible abnormal degradation of the containment structure. Again, the occurrence of one failure out of nine test specimens requires additional investigation, but does not constitute a reduction of safety margin. The same rationale discussed in i l
Section 4.3.6 with regard to the required number of tendons also applies to the required number of wires.
5.0 CONCLUSION
The preceding safety evaluation demonstrates that substantial margins exist above and beyond the conditions identified as " evidence of possible abnormal degradation." These margins provide assurance that the time limits specified for meaningful evaluation of occurrences considered to be evidence of possible abnormal degradation are commensurate with the continued ability of the containment structure to perform its design function.
The proposed revisions to Technical Specification 3/4.6.1.6 will not increase the consequences or probability of occurrence of accidents described in the FSAR, nor will they create the potential for any new accidents or reduce the margin of safety as defined in the bases for this technical specification. Thus, the proposed change does not constitute an unreviewed safety question. Surveillances performed in
- accordance with the revised specifications will, in fact, provide greater assurance that the post-tensioning system is performing its function and that the containment structural integrity is being maintained.
m