ML18044A603
| ML18044A603 | |
| Person / Time | |
|---|---|
| Site: | Palisades  |
| Issue date: | 03/03/1980 |
| From: | Frost S CONSUMERS ENERGY CO. (FORMERLY CONSUMERS POWER CO.) |
| To: | Ziemann D Office of Nuclear Reactor Regulation |
| References | |
| 800227-02, 800227-2, IEB-79-14, NUDOCS 8003060522 | |
| Download: ML18044A603 (4) | |
Text
March ~. 1980 Revised -
correct typing error on page 2 of Attachment General Offices: 21!2 West Michigan Avenue, Jackson, Michigan 49201
- Area Code 517 788-0550 February 27, 1980 Director, Nuclear Reactor Regulation Att Mr Dennis L Ziemann, Chief Operating Reactors Branch No 2 US Nuclear Regulatory Commission Washington, DC 20555 DOCKET 50-255 - LICENSE DPR PALISADES PLANT - ADDITIONAL INFORMATION - IE BULLETIN NO 79-14 Consumers Power Company transmitted responses to IE Bulletin Numbers 79-02 and 79-14 by letter dated February 14, 1980.
Attachment D to that letter --
entitled, "NRC Bulletin I/E 79-14 Inspection and Analysis Prograrn, 11 dated February 7, 1980 -- was a description of our 79-14 activities.
Sect:i.on 4 of the program presented the causes for piping stresses exceeding those in the 1967 Piping Code.
Consumers Power presented an interim criteria for allowable stress limits in piping and pipe supports which would be utilized until the completion of our evaluation and modification program.
During the review of the remaining unanalyzed stress systems as described in Section 7 of the program, an additional cause of calculated piping stresses exceeding allowables has been noted.
Therefore, our previous submittal is being supplemented to describe the cause of the deficiency and the corresponding corrective action.
The problem noted was that "stubbed-in" joints in some low temperature, low pressure lines resulted in code stress limits being exceeded for the OBE condition.
An example where this problem was noted was the supply and discharge of service waterlines to the containment air coolers.
Stress limits are not exceeded for the combined effects of deadweight, pressure and thermal loads.
The code calls for use of large stress intensification factors for bending stresses at the location of the "stubbed-in" connections.
These particular code rules have their bases in fatigue considerations.
The systems under evaluation are not subject to significant fatigue duty.
Temperature fluctuations are not significant.
Load fluctuations result in low levels of
~f'\\
~':>, \\'
aoosoeoS 2...'-
stress during normal operation.
Stresses imposed during a postulated seismic event would result in fewer than 300 cyc~es.
2 The attached Explanation of Stress Intensification Factors in Local Piping Components indicates that these design considerations in the code were imposed to assure pressure boundary integrity after thousands of fatigue cycles.
Consumers Power Company concludes that this additional analyses discrepancy is not significant since the actual load cycles imposed by normal operation and a postulated seismic event do not approach the fatigue conditions considered in the bases for the code.
Consumers Power will complete all necessary modifications within the schedule committed to by our February 14, 1980 letter.
Steven R Frost Palisades Licensing Engineer CC JGKeppler, USNRC Attachment Explanation of Stress Intensification Factors in Local Piping Components
ATTAClll1ENT EXPLANATION OF STRESS INTENSIFICATION FACTORS IN LOCAL PIPING COMPONENTS 1
A high calculated stress condition has been identified at branch connection points in two stress systems in the Palisades design review program.
For those cases of concern, the high stress levels in the calculations can be attributed to the fact that fabricated branch connections were used (with no added reinforcement in the form of pads or saddles), in lieu of forged welding tees, thus requiring the imposition of relatively severe stress intensifica-tion multipliers on the calculated bending stresses (M/Z) at those points.
Where this condition has been identified the indicated high stress levels produced by the imposition of a stress intensification facto'r (SIF) on a local component* is acceptable, on an interim basis, until such time as permanent correction measures can be achieved to satisfy all requirements of the FSAR and the code rules, provided:
- a.
The calculated stresses in a system, determined without the imposition of an SIF, will not exceed the allowable stress for the material at the points in question.
- b.
The total number of full cycles of strain that the system will experience, from original plant start-up until a permanent fix is achieved, will not exceed 2000 cycles.
This position is quite defensible if one examines both the genealogy and the true definition of the term "stress intensification factor" as it was introduced in the 1955 edition of ASA B31.1 Code for Pressure Piping, and as it is still applied today.
Markl 1 s*original definition in his paper "Piping-Flexibility Analyses" (ASME 53, A-51)i>* July 1953 was as applicable then as it is now.
"... the stress-intensification factor will be defined here as the ratio of the bending moment producing fatigue failure in a given number of cycles in a straight pipe of nominal dimensions, to that producing failure in the same number of cycles in the part under consideration."
By this definition it can be seen that if a bending moment (a cyclic stress) will produce a fatigue failure in a local component after a given number of cycles and if its SIF is, say 5, then a point in a straight pipe of the same nominal dimensions and material will tolerate 5 times the bending moment (as cyclic stress), for the same number of cycles, before fatigue failure will occur.
- An elbow, branch connection point, or another component point where the imposition of an SIF is required under present code rules.
- ~This paper resulted from the work of a special Task Force on Flexibility, appointed by the ASA B31.1 Sectional Committee in 1951.
-.;,-*... ~-
2 This SIF concept in the code means, in the context of code rules, that if the stresses in all local components in a pipig system are calculated using the appropriate SIFs and are limited to the same stress levels as the straight pipe in the system then it will be reasonably assured that all parts of a system will have approximately the same endurance strength under cyclic loading effects. It does not mean that if a component has an SIF of 5 (or any number calculated under the relevant code rules) and if the calculated thermal expansion stress (M/Z) is 20,000 psi... that the imposition of the SIF will mean that a stress level of 100,000 psi will exist in the component.
A bending stress at a component point in a system will never exceed its M/Z condition, or 20,000 psi in the example here.
In the case of the Palisades problem, stresses were calcula~ed at branch connection points without the imposition of SIFs.
The actual stresses at those points will be no higher than those calculated but their fatigue life will be considerably lower than the improperly calculated stresses imply.
Therefore, the "real life" consideration is not consi.dered overalarming since the predicted requirement for lifetime cycles is so very low; probably less than 500 full strain cycles in a 40-year plant lifetime.
This very consideration was proposed in Appendix 2 to ARC Markl's paper noted above for the rules to be adopted in the code.
The M W Kellogg Company (probably the most knowledgeable A&E office in the world at that time) proposed:
"For calculations made in conformity with paragraph (a) (of Para 620 of Chapter 3 of Section 6 of B31.1), stress-intensification and flexibil-ity factors may be omitted if the piping system is not subject to more than 2000 stress cycles during its expected life." The term "full strain cycles" instead of "stress cycles" would have been more appropriate, in retrospect.
In any event, the proposal was not adopted in spite of its validity.
The code rules adopted in 1955, based on Markl's paper, were largely adopted from experimental data outlined in an earlier paper by Markl:
"Fatigue Tests of Piping Components" (ASME 51, PET-21), May 1951.
In this paper, Markl's summarization leads to conclusions of piping components/straight pipe flexi-bility. The final conclusions of this work are that components other than straight pipe are as flexible as the pipe.
Further differences in flexibility and geometry can be directly correlated to endurance in cyclical fatigue loading.
Finally, the endurance limits of piping components can be catego-rized in terms of "stress intensification factors."
Thus, piping components will maintain their loading limits, in terms of the calculated stresses, to the endurance limits of the straight pipe.
The data produced in Markl's earlier paper show that the probable fatigue-life limits for branch connections will far exceed the conditions expected in the life of the Palisades Plant.
Therefore, it is justified to operate under the interim criteria until final modifications are completed to update the plant to implied FSAR requirements.