LIC-90-0564, Forwards Updated, Position Paper Re Seismic Capability of Welded Steel Piping. Documentation Demonstrates No Generic Thermal or Seismic Safety Issue Exists for Welded Steel Small Diameter Piping at Facility

Forwards Updated, Position Paper Re Seismic Capability of Welded Steel Piping. Documentation Demonstrates No Generic Thermal or Seismic Safety Issue Exists for Welded Steel Small Diameter Piping at Facility

ML20058K342
Person / Time
Site:	Fort Calhoun
Issue date:	06/29/1990
From:	Gates W OMAHA PUBLIC POWER DISTRICT
To:	NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM)
References
LIC-90-0564, LIC-90-564, NUDOCS 9007050015
Download: ML20058K342 (9)

Text

~

y l l

, .. t l 1..

L;

June 29, 1990 Omaha Public Power District LIC-90-0564 1623 Harney Omaha, Nebraska 68102 2247 402/536 4000 U. S. Nuclear Regulatory Commission Attnr Document Control Desk Mail Station P1-137 Washington, DC 20555 l

REFERENCES:

- 1. Docket 50-285

2. Letter from OPPD (K. J. Morris) to NRC (Document Control ,

Desk) dated December 2, 1988 (LIC-88 0506) l

SUBJECT:

Update of Plans Regarding Small Bore Pipe Support Spacing Reference 2 provided a description of Omaha Public Power Districts (OPPD)'

actions planned with respect to safety-related small bore piping. _These actions included 1) documenting as-built small-bore pipe support locations, and

2) review of as-built pipe support locations against the new nomograph / Alternate Seismic Criteria. OPPD is extending the completion date of these two items to December 31, 1994 and December 31, 1995, respectively.

OPPD's current position on piping and restraint qualification issues related to small diameter piping is presented in the attached documentation. This documentation demonstrates that no generic thermal or seismic safety issue ,

exists for welded steel small diameter piping at FCS, OPPD is developing.a generic program for resolving outstanding piping issues to present to the NRC.

This program will identify priorities, schedules and resources based on the ,

relative safety significance and benefit of resolving various issues. l 1

These new commitment dates are consistent witn the emerging generic piping program schedules and will allow the interim development and approval of criteria and procedures which would be appropriate for performing qualification reviews, if warranted, of small diameter piping systems.

If you should have any questions, please contact me.

Sincerely, l l

N. ..

L W. G. Gates Division Manager Nuclear Operations WGG/sel Ok kObh $bOhib$sr_;

P PDC c: LeBoeuf, Lamb, Leiby & MacRae A. Bournia, NRC Project Manager R. D. Martin, NRC Regional Administrator, Region IV P. H. Harrell, NRC Senior Resident Inspector 0[

10 1

• b h!b$ ^4* M r.(Py l ( { a%()(jf))p ( ) (j gy(g l

4 ,

OPPD POSITION PAPER REGARDING THE SEISMIC CAPABILITY OF WELDED STEEL PIPING R.E. Lewis, P.E.

May 3, 1990 I. PURPOSE <

To document the following:

i

1) That existing seismic design features (of CQE piping at Ft.

Calhoun) and data obtained from various walkdowns, provide a high level of confidence that the piping will perform adequately for-the design basis earthquake events.

2) That a significant level of conservatism exists in the design basis seismic qualification process for Ft. Calhoun.

3) That welded steel piping systems are inherently rugged seismically.

4) That various technical and documentation deficiencies, which have been identified in our piping seismic analyses over the '

years, are insignificant in comparison to the overall seismic capacity of the systems.

5) That no significant seismic safety concern exists, for welded steel piping systems, at Ft. Calhoun.  ;

II. EXISTING' CONDITIONS The CQE piping at Ft. Calhoun was originally designed for seismic '

-loads, using methods less rigorous than current practice but which were deemed appropriate at the-time. Since then, "as-built" walkdowns, complete reanalysis and substantial modifications were p(using updated erformed to methods) "as-built reconcile and design discrepancies" found during resolution of NRC IEB 79-02 &

79-14. In addition, we have performed verification walkdowns for the District's Design Basis Reconstitution program, and continued'to evaluate.and resolve various seismic issues in combination with other work requiring reanalysis of piping. Therefore a significant margin of seismic capacity has been incorporated into the CQE piping systems as they exist today.-

However, miscellaneous documentation and technical' issues, with regard to demonstrating the seismic qualification of CQE piping at Ft.'Calhoun, have surfaced over the years. Some of the past piping design inputs, assumptions and modeling practices are undocumented and/or inconsistently applied. The basis for many of the " apparent assumptions" are unknown and not supported by historical industry practice.

by current The associated standards calculations and practices. and documentation Calculations are deficient which reconcile these-deficiencies, often indicate that various piping and/or -

restraint limits.

components are outside the Seismic class I design basis

g 9 g I

Calculations and criteria, prepared for.the District's Alternate Appendix F Criteria documents (Ref. 9 & 10), demonstrate that significant conservatisms exist in the Design Basis seismic inputs when compared to those derived with current practice methods and

. criteria. Many of the technical deficiencies noted above, are insignificant in comperison to these excess conservatisms.

• *

• y 4 9 ,

e III. EXPECTED PERFORMANCE UITH EXISTING CONDITTONS >

The current and past seismic analysis practices, as they apply to nuclear piping design and qualification, are now known to significantly overpredict seismic failures (Ref. 7: p.136, p.158, p.164; Ref. 8: p.67, p.217, p.220, p.233). The over-conservatisms for piping seismic anslysis are three fold: 1) analysis methods 2) seismic input data and 3) code acceptance criteria.

Analysis methods have been refined over the years, but still '

typically calculate seismic responses based on the assumption of-linear elastic behaviour, use of response spectra methods, conservative Regulatory Guide 1.92 modal combinations (Ref. 7: p.158, p.167) and Regulatory Guide 1.61 damping values or lower in lieu of higher empirical values. pipe stresses calculated by the response spectra method can be 2 to 8 times that resulting from a more exact time history method (Ref. 15, p.6). Actual behaviour of seismically excited piping systems displays significant energy dissipation and is non-linear in nature Measured responses are(Ref. 7: p.134; Ref. 8: p.205, p.229, p.233).

usually much lower than calculated (Ref. 7:

p.168; Ref. 8: p.67, p.220, p.229).

In addition, the seismic input Floor Response Spectra (FRS) contain artificial conservatisms in the form of Regulatory Guide 1.122 peak broadening, smoothing and enveloping (Ref. 7: p.134, p.158). The seismic structural analysis process, which is used to derive the piping FRS, contains similar conservatisms to the piping analysis process and therefore compounds the conservatisms. Calculations, performed for the District's Alternate Appendix F Criteria (Ref. 10),

demonstrate that significant conservatisms exist in the Design Basis seismic inputs when methods and criteria. compared to those derived with current practice The piping code criteria, against which calculated seismic inertia stresses are compared, is known to be unrealistic and overly conservative (Ref. 7: p.164, p.168; Ref. 8: p.67, p.229, p.233).

Failures modes due to seismic loading have been shown, throuch testing and industry research programs (Ref. 13 & 14), to be caused ~

by fatigue and ratcheting instead of plastic collapse which is the basis for the piping code criteria. Numerous stress reversal cycles (f ar more than would result from actual earthquakes) are needed to induce failure. The codes, however, treat inertia as a primary stress and limit the stress amplitudes without regard to the limited number of cycles involved.

Therefore the cumulative conservatisms inherent in the seismic qualification process are significant and lead to over-prediction of seismic failures. However, relaxation of these excess conservatisms is not currently permited in our licensing basis.

The actual earthquake experiences of others, have proven tian and again that typical power piping systems are inherently rugged seismically. A significant amount of piping performance experience data has been collected from surveys of at least 29 different >

earthquakes around the world and from testing recently performed for/by recognized- industry organizations. This data demonstrates that the rare instances of seismically induced piping failures can be attributed to a limited number of root causes as follows (Ref. 1&

2):

Seismic Anchor Movement (SAM)- insufficient flexibility in piping restrained by_different structures or equipment that move rela-tive to each other, can result in breach of pressure boundary and/or restricted flow.

Spatial Interaction (SPI)- inadequate clearance between vulnerable components, like control valves, and other structures or equipment can cause loss of function and/or restricted flow.

Corrosion - excessive thining of the pipe wall, due to corrosion, can result in breach of pressure boundary and/or structural-fail-ures.

Non-Welded Joints - screwed fittings are vulnerable to breach of pressure boundary failure due to the stress concentrations and reduced cross section present at the thread roots and bolted flange joints might develope leaks.

s Welded steel piping systems, designed to standard industry piping codes (even when not designed seismically), exhibit adequate perfor--

mance in earthquakes (with recorded peak ground accelerations as high as 0.99) with few exceptions (Ref. 1). Piping restraint failures, on these same lines, are not uncommon and provide pressure boundary protection by acting as mechanical fuses to provide additional flexi-bility where needed to accomodate SAM movements (Ref. 2). It can be concluded from the above that good seismic design results from provid-ig the necessary flexibility to accomodate SAM, providing sufficient clearances to prevent Spl, providing allowance for errosion/ corrosion and avoiding the use of non-welded joints. performing rigorous dynam-ic analysis (which is known to significantly over predict the seismic behavior of piping) and applying arbitrary code stress limits (which are known to be unrealistic and over conservative for seismic loads; Ref. 3,

p. 2-17) are not required to achieve adequate margin against l aeismic induced piping failures and may in fact result in designs which are less reliable than those for which no seismic design is employed (Ref. 2: R. Cloud, p. 81 & Sect. 5.1.3, p. 84).

l' l

1

.Recent testing through programs such as the "EPRI/NRC Piping and Fitting Dynamic Reliability. Program" (Ref. 13) have provided-the industry with considerable insight into the credible failure mechanisms relative to seismically' induced response. This data shows that (Ref. 1) :

" Sound and uncracked piping is " Fail-Proof" to any conceivable seismic inertia loads in LWR-type plants (more than 10 times current ASME SSE-allowable load levels are required to induce pipe failuro)."

"There is a natural mechanism to relieve high seismic elastic calculated pipe stresses, and the mechanism is plastic energy dissipation which behaves like increased damping."

"The piping failure mode under super seismic type loading is fatigue and ratcheting, and not static collapse as assumed by the current ASME Code rules."  ;

"The ratchet / fatigue failuro condition has no sharply defined stress level at which pipe functionality is lost as in the case of a static collapse mode."

" Current ASME Code Allowable Stress Criteria with linear dynamic analysis is a very conservative and unrealistic measure of pipe failure caused by excessive seismic type loads. (The fatigue i criteria of the Code is OK) ."  ;

"The research explains why it is not necessary to provide horizon- i tal support to piping subjected to seismic loading to protect the pipe from stress collapse. The main benefit from horizontal sup-port is to prevent excessive pipe displacement."

" Fatigue and ratcheting damage due to one or two large earth-quakes will generally cause insignificant damage to=any reason-ably designed piping system (with or without nuclear standards)."

In summary, seismic induced stresses in velded steel piping systems is not a credible safety concern provided that no SAM, SPI, errosion/

corrosion or non-welded j oint issues exist. In addition, it has been-recommended that explicit earthquake design requirements be eliminated, for piping subjected to a SSE-ZPGA.less than 0.2g, since no such observed piping (Ref.

low magnitude system failures have been noted for earthquakes of 2, Sect. 5.2.2, p.

85).

-e e 1

With respect to the Fort Calhoun CQE piping systems, the following is

-known:

1)~The equipment to which CQE piping is attached and the piping i itself, was designed to be seismically rigid, and eas anchored to i resist seismic loads. Therefore, relative movement between l equipment and piping located in the same building structures, at the same elevation, will be insignificant with respect to inducing SAM loads on piping. Furthermore, the Containment and Auxiliary building structures are founded on a common mat and designed to resist seismic loads. Relative SAM movements between and within these structures is known to be insignificant. Seismic movements between the Auxiliary and Turbine building structures is significant. The CQE Main Steam and Feedwater piping routed between these structures has been evaluated for the resultant SAM loads. No other'CQE piping is presently routed between these two-structures. Therefore, no generic SAM concerns exist for CQE piping at Ft. Calhoun.

2) The most vulnerable piping components, from the standpoint of SPI concerns, are valve actuators. The seismic design of CQE piping systems, at Ft. Calhoun, typically provided three way seismic restraints on the piping at the location of control 1 valves. In addition, out-of-plane restraints were provided for eccentric valve operator masses. Seismic displacement of the valves into other structures or equipment, is thereby prevented.

In general, the original seismic design of CQE piping provided sufficient restraint to acheive rigid- response with respect to.

the building structures. In addition, modifications were made for  ;

IEB 79-02 & 79-14'which.provided additional restraints to limit

. pipe stresses and support loads to within. acceptable limits.

Therefore, interaction with other piping components and adjacent valves is unlikely. In addition, we have observed that other structures and/or equipment located near CQE equipment were designed and constructed to prevent SPI. Field inspections of CQE piping, perf ormed for the District 's IEB 79-14 program, evaluated clearances between piping components and surrounding structures-and equipment to_ ensure'that thermal expansion would not be j inhibited. Clearances were also observed'during walkdowns performed for the Design Basis Reconstitution project. Clearances '

were-found to be adequate. Further review of SPI, with respect to control valves, will be performed for resolution of NRC USI A-46.

For the CQE Main Steam and Feedwater piping, one instance of-

, potential SPI was noted between a MS Isolation Valve and an L adjacent FW seismic restraint structure. The potential SPI results from the significant displacements of the MS line l predicted to occur in the event that all lateral seismic L restraints would fail (this has been conservatively assumed for l l purposes of demonstrating operability under worst case l conditions). A modification was performed to eliminate that SPI L concern (Ref. 11 & 12). In general, no generic spi concerns exist l for CQE piping systems at Ft. Calhoun. A program is to be I

' implemented in the near future for USI A-46 which will provide additional documentation or corrective actions to support this conclusion.

l l

1 i

l

3) The use of corrosion resistant materials, chemical treatment and current monitoring and replacement programs for controling errosion/ corrosion are judged to eliminate these effects from being a concern with respect to seismic induced failures.

4) All known joints in CQE piping (excluding instrumentation lines) at Ft. Calhoun are either welded or flanged. The flanged joints were specified to use high strength bolts so that complete failure due to seismic loads is unlikely although some leakage may result. The welded joints recieved inspections as required by the appropriate codes to ensure quality and critical welds are periodically re-inspected thru the District's ongoing ISI program..Therefore, no generic concern with regard to non-welded or poorly welded joints exists at Ft. Calhoun.

5) The Fort Calhoun site SSE-ZPGA was determined to be 0.12g, from seismological studies, and conservatively established as 0.17g for design basis. Since these values are less than 0.2g, explicit design requirements for seismic loads would seem unwarranted.

IV. CONCLUSION We conclude that: 1) a significant margin of seismic capacity exists in "as-built" design features of CQE piping at Ft. Calhoun the design basis seismic qualification process significantly 2) that over-predicts seismic responses 3) that welded steel piping is inherently rugged seismically 4) that outstanding technical and documentation deficiencies, regarding the seismic qualification of CQE piping at Ft. Calhoun, are insignificant in comparison to the demonstrated superior seismic performance of welded steel piping in actual earthquakes 5) and that no generic seismic safety concern exists with regard to CQE piping at Ft. Calhoun.

1 l

. . ,4 REFERENCE LIST

1) Procedure for Seismic Evaluation and Design of Small Bore Piping (NCIG-14), Sept. 29, 1989, EPRI NP-6628, Project Q101-16,17

2) Report of the U.S. Nuclear Regulatory Commission Piping Review Committee, USNRC, NUREG-1061, Vol. 2 ADDENDUM

3) Report of the U.S. Nuclear Regulatory Commission Piping Review Committee, USNRC, NUREG-1061, Vol. 2

4) Regulatory Analysis for Resolution of Unresolved Safety Issue A-46, Seismic Qualification of Equipment in operating Plants, USNRC, NUREG-1211

5) Task Actions Plans for Unresolved Safety Issues Related to Nuclear Power Plants, USNRC, NUREG-0649, Rev. 1

6) Generic Implementation Procedure (GIP) for Seismic Verification of Nuclear Plant Equipment, SQUG, Dec. 1988, Rev. 1

7) "Recent Advances in Seismic Design of Piping and Components",

Proceedings of the 1985 Pressure Vessels and Piping Conference, PVP-Vol. 98-3

8) " Seismic Engineering- Recent Advances in Design, Analysis, Testing and Qualification Methods", The 1987 Pressure Vessels-and Piping Conference, PVP-Vol. 127

9) " Alternate Seismic Criteria & Methodologies for Fort Calhoun Station", Volumes I & II, July 1988.

' 10) " Generation of In-Structure Response Spectra for Fort Calhoun Unit-1", Vol. I, II & III, Jan. 1989 t

l

11) MR-FC-89-45, "Feedwater Supports in Room 81 & Turbine Building"

12) MR-FC-90-19, " Main Steam Supports in Room 81 & Turbine Building" L 13) EPRI Report No. RP1543-15, " Piping and Fitting Dynamic l

Reliability Program", Dec. 1989

14) EPRI Report No. NP-3746, " Dynamic Response of Pressurized Z-Bend u Piping Systems Tested Beyond Elastic Limits and with Support Failures", Dec. 1984

15) NUREG/CR-3718, UCID-19722, " Reliability Analysis of Stiff Versus Flexible Piping - Status Report", April 1984 l

-a-