ML20138G800
| ML20138G800 | |
| Person / Time | |
|---|---|
| Issue date: | 11/04/1985 |
| From: | Minogue R NRC OFFICE OF NUCLEAR REGULATORY RESEARCH (RES) |
| To: | Harold Denton Office of Nuclear Reactor Regulation |
| References | |
| RIL-142, RIL-142-S01, RIL-142-S1, NUDOCS 8512160405 | |
| Download: ML20138G800 (5) | |
Text
--
~ _ . _
, T45 Filg N':.i a :t Ole 1.'o. *
/
R h. .-
CSugie w d *-t g -
NOV 4 1985 =ca Request lio.
. m. .
. me. egr5cg3 '
HEt:dRA'i' J U.:o F' R: harold R. Dantaa, Director h,
Jin iw or huck tr Reacur Ruualdtian- e
-~
FbR.
Fai0.:: Rdr rt J.1;i% gut, Diractar ' ' 7,fkg ditica of .uclesr Ragunatorj h,esearcn .-- - __
SUSJECT: SUPP EGEi T TO RESEA,iCi. If,F06.ATION LETTER, l'O.142 "F.ELIAILITY MLY51S OF STIFF VERSUS FLEXIikE PIPIhS" Your staff requestad tnat we amplify the su'aject IUecarch Infarmation Letter to batter satisfy your needs. In ciccordance, the following m teristi is orfared.
- 1. The purpase of tha inv:stigation wa; to d2vulep a technical buis far flexible piping desigas which would iuprave piping rcliability and mininire the use of pipa supports (that is, rigid restraints) and snubbers. Aq adjunct to this purpose is the identification of sit-uction whare fledble piping d:sigi. is desircble. This study treats the renoval of pipe supports and snubbers in existing operatianal piping and ia piping baing designed or redesign 3d.
E. Tin statemnt in the originc1 research inferation letter that, "The
. piping raliability is either i., proved or affected varj little by the increased piping flexibilitj" is explained as fallows: high energ/
. piping exp2rienccs greater reliability when flexibility is incrassed; low energy pipiag suffers a very slight decrease in reliabilitj uhan flexibility is 1.icreased. This result is tru2 within a wide raaga of dssumptions PCrtaining to earthquake frequeaCy, thermal trcnsient frequency and failure rates for snubbars. '
- 3. The principal reccim ndation of the study is that flexible piping design :.hould be encouragad based on benitrits achieved in terms of improved and more effective plant maiatenance, reduced radiation ex-pasures to workers and mare favcrable construction and maintenan:e economics. In addition, safety and reliability actually increases for high energy piping and decreeses insignificantly for Icw energy piping when flexibility is increased.
- 4. Thus, there are clear adecutd0cs to tne general adaptian of flexible piping designs. tienetheless, in achieving aorc flexible piping de-signs through the reduction in pipa supports and snubbers, certain cautions most continue to b2 observed as listed below:
- a. Any attempt to provide flexibility by removing or omitting pipe supports or snubbars near component nozzles should be done only 8512160405 851104 PDR RES emc s > PDR ounesanss>
oats >
n:c ronu s S n0-80) NRCM OMO OFFICIAL RECORD COPY l
P V
Harold R. Denton 2 NOV 4 1985 when a careful evaluation of the resulting nozzles loads is undertaken, and it is demonstrated that these loads are within acceptable bounds.
- b. The removal of pipe supports and snubbers to produce more flexi-ble piping will lead to greater pipe displacements, and it should be determined that these displacements can be tolerated without impact to neighboring structures, components and sys-tems.
- c. In existing operational piping, the removal of pipe supports and snubbers will in general increase loads in those pipe supports and snubbers which remain. Those pipe supports and snubbers which are not removed should be checked to determine whether they have sufficient margin, or alternatively whether they should be strengthened. To a certain degree, careful selection of pipe supports and snubbers to be removed could eliminate any need to strengthen remainina pipe supports and snubbers.
On the other hand, acceleration environments for the qualifica-tion of on-line components were not determined to be a critical concern. However, as with the piping itself, it should be de-termined that component displacements can be tolerated without impact to neighboring structures, components and systems.
A copy of the original Research Information Letter 142 is attached for easy reference.
Original Signed by:
ROMAggggg Robert B. Minogue, Director Office of Nuclear Regulatory Research
Enclosure:
As stated Record Note: This Supplement was requested by NRR, and was reviewed and approved by NRR.
See previous concurrences MSEB:DET J0'Brien:sp MSEB:DET:C JRichardson J DDIRES
[/GArlotto Dross a h
1C v1 1 10/ 15'85 10/15 /85 10//(f/85 10/ '/85 N/4 /85
'(.
i Harold R. Denton 2 4 1985 when a careful evaluation of the resulting nozzles loads is undertaken, and it is demonstrated that these loads are within acceptable bounds.
- b. The removal of pipe supports and snubbers to produce more flexi-ble piping will lead to greater pipe displacements, and it should be determined that these displacements can be tolerated without impact to neighboring structures, components and sys-tems,
- c. In existing operational piping, the removal of pipe supports and snubbers will in general increase loads in those pipe supports and snubbers which remain. Those pipe supports and snubbers which are not removed should be checked to determine whether they have sufficient margin, or alternatively whether they should be strengthened. To a certain degree, careful selection of pipe supports and snubbers to be removed could eliminate any need to strengthen remaining pipe supports and snubbers.
On the other hand, acceleration environments for the qualifica-tion of on-line components was not determined to be a critical concern. However, as with the piping itself, it should be de-termined that component displacements can be tolerated without impact to neighboring structures, components and systems.
A copy of the original Research Information Letter 142 is attached for easy reference.
Robert B. Minogue, Director Office of Nuclear Regulatory Research
Enclosure:
As stated Record Note: This Supplement was requested by NRR, and was reviewed and approved by NRR.
I l
M dT MSEB:0ET:C 0:DET 00:RES D:RES l
J y p JRichardso
/ -
GArlotto Dross RBMinogue 10/t9/85 10/;3/85 10/ /85 10/ /85 10/ /85
1
- I
, , p ugg d' k UNITED STATES -
)
[ g(A*g NUCLEAR REGULATORY COMMISSION 3 WASHING TON, D. C. 20555
- %, ..V... J l NOV4 1985 MEMORANDUM FOR
- Harold R. Denton, Director Office of Nuclear Reactor Regulation FROM: Robert B. Minogue, Director Office of Nuclear Regulatory Research
SUBJECT:
SUPPLEMENT TO RESEARCH INFORMATION LETTER NO. 142
" RELIABILITY ANALYSIS OF STIFF VERSUS FLEXIBLE PIPING" Your staff requested that we amplify the subject Research Information Letter to better satisfy your needs. In accordance, the following material is offered.
- 1. The purpose of the investigation was to develop a technical basis for flexible piping designs which would improve piping reliability and minimize the use of pipe supports (that is, rigid restraints) and snubbers. An adjunct to this purpose is the identification of sit-uations where flexible piping design is desirable. This study treats the removal of pipe supports and snubbers in existing operational piping and in piping being designed or redesigned.
- 2. The statement in the original research information letter that, "The piping reliability is either improved or affected very little by the increased piping flexibility" is explained as follows: high energy piping experiences greater reliability when flexibility is increased;
' low energy piping suffers a very slight decrease in reliability when flexibility is increased. This result is true within a wide range of assumptions pertaining to earthquake frequency, thermal transient frequency and failure rates for snubbers.
- 3. The principal recommendation of the study is that flexible piping i
design should be encouraged based on benefits achieved in terms of l '
improved and more effective plant maintenance, reduced radiation ex-t posures to workers and more favorable construction and maintenance i economics. In addition, safety and reliability actually increases for high energy piping and decreases insignificantly for low energy piping when flexibility is increased.
- 4. Thus, there are clear advantages to the general adoption of flexible piping designs. Nonetheless, in achieving more flexible piping de-signs through the reduction in pipe supports and snubbers, certain
! cautions must continue to be observed as listed below:
1
- a. Any attempt to provide flexibility by removing or omitting pipe supports or snubbers near component nozzles should be done only
- -. =
r- . .
Harold R. Denton 2 NOV4 1985 when a careful evaluation of the resulting nozzles loads is undertaken, and it is demonstrated that these loads are within acceptable bounds,
- b. The removal of pipe supports and snubbers to produce more flext-ble piping will lead to greater pipe displacements, and it should be determined that these displacements can be tolerated without impact to neighboring structures, components and sys-tems.
- c. In existing operational piping, the removal of pipe supports and snubbers will in general increase loads in those pipe supports and snubbers which remain. Those pipe supports and snubbers which are not removed should be checked to determine whether they have sufficient margin, or alternatively whether they should be strengthened. To a certain degree, careful selection of pipe supports and snubbers to be removed could eliminate any need to strengthen remaining pipe supports and snubbers.
On the other hand, acceleration environments for the qualifica-tion of on-line components were not determined to be a critical concern. However, as with the piping itself, it should be de-termined that component displacements can be tolerated without impact to neighboring structures, components and systems.
A copy of the original Research Information Letter 142 is attached for easy reference.
,krN ' $ .Y Robert B. Minogue, Director Office of Nuclear Regulatory Research
Enclosure:
As stated I
6 s,m *8 UNITED STATES
.\ NUCLEAR REGULATORY COMMisslON
[, ,j wassincrow. o.c. runs AUG 11 W
, k...../
MEMORANDUM FOR: Harold R. Denton, Director Office of Nuclear Reactor Regulation FROM: Robert B. Minogue, Director Office of Nuclear Regulatory Research
SUBJECT:
RESEARCH INFORMATION LETTER, N0. 142 " RELIABILITY ANALYSIS OF STIFF VERSUS FLEXIBLE PIPING" Introduction Current design criteria for piping evolved under the presumption that higher seismic margins necessarily improve plant reliability. Conserva-tive design against earthquake loads has relied increasingly on rigid supports, snubbers, and other types of seismic restraints to stiffen
/
i piping systems. The resultant decrease in flexibility, however, is likely to cause higher nonnal operating stresses because of the restraint of
! thermal expansion. Furthermore, because of the large uncertainty inherent in predicting seismic effects (compared to that in predicting thermal effects), seismic loads dominate the design even though seismic loads occur very infrequently. As a result, stiffening a piping system to improve its resistance to seismic loads may actually decrease its overall reliability during normal cperation.
In order to quantify piping reliability, pipe failure probabilities were computed by Lawrence Liversore National Laboratory. Pipe failure was assumed to be caused by fatigue crack growth at pipe weld joints. Two types of failure modes were considered initially, i.e., a through wall crack (leak) and a complete pipe severance (break). Pipe support and on-line component failure medes were subsequently introduced. Studies were undertaken to evaluate changes in reliability of pipe supports and on-line components as a function of piping flexibility based on available fragility data. Pipe failure probability was estimated by applying a
- Monte Carlo method with a stratified-sampling scheme to simulate the life histories of the piping system.
1 Selected piping systems with related design data were collected from real
- nuclear power plant designs. Flexible piping designs were created from the existing designs by removing rigid supports and/or snubbers. Piping stresses for various designs were calculated for the reliability assessment.
This assessment recognizes the characteristic difference between regular pipe supports and snubbers. While removal of regular pipe supports changes both seismic stress and thennal expansion stress in a piping l
,. , . _ - _ - - - . . _ , , - - . . . _ _ . , , , _ . . , ~ - . m.% ,y ,..,_y_,,m,,, _ . , . . _ . - _ , , , , , - _ _ , , , _ . _ - .
Harold R. Denton 2 g, 1
a system, removal of snubbers affects only seismic stress. However, a piping system including snubbers may not exhibit the desired reliability because snubbers are known to have a high failure rate. The possibility of snubber malfunction is incorporated in this assessment. ,
l The malfunction of pipe whip restraints was also treated. The situation where the pipe comes in contact with a restraint device during nonnal i operation due to an imperfect installation was investigated. Actual 1 stresses caused by the malfunction were calculated and an assessment of the safety impact on the piping was undertaken by perfoming a reliability analysis with and without the malfunction. The intent was to confim that the ralfunction of pipe whip restraints introduces higher themal stresses and reduces the overall piping reliability.
Results l
Based on this research, the change in piping reliability was determined to be largely insensitive to the change in piping flexibility for the cases studied. The piping reliability is either improved or affected very little by the increased piping flexibility as a result of removing rigid pipe supports and/cr snubbers.
small; values of the order of 9 10,P1pe failure probabilities are general per reactor year are reported for pipe rupture and 10'6 per reactor years are reported for pipe leak. It is concluded that flexible piping design is desirable based on reliability considerations, and that flexible piping design also offers many other benefits.
Although it has been demonstrated that piping systems can be made more J
reliable by adopting flexible piping designs, increase in piping flexibility usually results in greater pipe displacerents. Displacement criteria or requirements to confine piping displacements may be needed.
' Since pipe supports and on-line corponents (such as pumps and valves) are l important parts of a piping system, the effects of increased piping flexibility on the reliability of supports and components were investigated. Results indicated that the support reliability usually exhibits a moderate decrease as the piping flexibility increases. It is felt that the supports in a flexible piping design need to be either upgraded or subjected to further investigation.
For large components, such as steam generators, pressurizers, and large pumps, the global effect concerning the component support failure due to increased nozzle loads was evaluated. It was found that in general the global effect is rather insignificant. However, the local effect at the vicinity of the nozzle may need to be further investigated. Without such an investigation, it is suggested that removing pipe supports which are close to nozzles should be done with extreme care.
i
I I
N rold R. Denton f J. " i s y
.~
For self-supporting on-lir.e valves, it was discv.o tu tim t. .a .o've acceleration may or may not increase with piping flexibility. Neverthe-less, valves usually have sufficient design margins to accomodate the higher acceleration and are able to maintain functionality. The problem in this case is the increased valve displacement usually associated with a flexible piping design. Specific design consideration may be needed in order to limit the valve displacement.
Malfunctioning pipe whip restraints were found to increase thermal stresses by a factor of 15 in the worst case; this malfunction increases pipe rupture probabilities by one order of magnitude and pipe leak probabilities by two orders of magnitude approximately.
Evaluation This research provides strong technical support for the NRC Piping Review Comittee's non-mandatory recomendations aimed at reducing the number of snubbers at nuclear power plants. Flexibile piping provides easier access for plant maintenance, reduces radiation exposures during maintenance, reduces thennel stresses during plant operation and positively affects construction and maintenance economics. Moreover, in general, piping reliability is improved.
Additional infonnation on this work may be obtained from John O'Brien (x37854) of the Mechanical / Structural Engineering Branch, DET.
l kkd h Tit- M Robert B. Minogue, Director Office of Nuclear Regulatory Research 0
_ -_m.. , _ . _ _ . . . . _ . . _ , . . _ _ . . _ . . _ , , _ . _ - . . . , , . - _ _ _ , _
f
" , , . ,1. : i:. in f..:; .,y bi5
,,~
r
Reference:
NUREG/CR-4263, "Reliebility Analysis of Stiff Versus Flexible Piping; final Project Report" May 1985, Prepared by S. C. Lu and C. K. Chou, Lawrence Livermore National Laboratory. !
I b
l l
l 1
-__ _ _ . . _ _ _ _ . . . _ , . . - . . _ - _ . . . , - . . _ .