ML20039B764
| ML20039B764 | |
| Person / Time | |
|---|---|
| Site: | Midland |
| Issue date: | 12/10/1981 |
| From: | CONSUMERS ENERGY CO. (FORMERLY CONSUMERS POWER CO.) |
| To: | |
| Shared Package | |
| ML20039B761 | List: |
| References | |
| NUDOCS 8112230532 | |
| Download: ML20039B764 (63) | |
Text
,.
.~
1
+
e-
+
a i-i i
7
)
i i
i i
T' ANALYSIS OF BURIED SAFETY-GRADE PIPING FOR
^
. MIDLAND PLANT UNITS 1 AND 2 4-3.
i 1
r Submitted by:
Consumers Power-Company
-8112230532 811555 12/10/81 PDR ADOCK 05000329 A
fDR,
1.
b---
~. _ _ _. _.. -, _.
TABLE dF CONTEN'"S_
ANALYSIS OF BURIED. SAFETY-GRADE PIPING' FOR MIDLAND PLANT UNITS 1 AND 2
-Page I.
SCOPE 1
II. -
SUMMARY
2 A.
BACKGROUND 2
B.
FUNCTIONAL CAPABILITY - PROPOSED SOLUTION 3
C.
ANALYTICAL SOLUTION DIFFICULTIES 5
III. C2 TAILED DISCUSSION 7
A.
SERVICE WATER SYSTEM 7
1.
Profile and Ovality.
7 2.
Ovality / Buckling 9
~
3.
Future Settlement 15 a)
Predicted values 15:
b).
Monitoring program 17 4.
Seismic 18-a)
Seismic. analysis 18 b)
Variable soil properties 20 c)
Effect of pipe deformation 21 on scismic forces d)
Code requirements 24 ii 12/10/81
.Page 5.
Rebedding 25 ' -
a)--
Size verification of*8-inch lines 25 b)
Rebedding of;8-and110-inch-
.25 service water lines 6.
Verification 26 a)
Preservice 26 b)
Inservice 27 i
B.
DIESEL FUEL OIL LINES 28 1.
Profile
.28 -
2.
Future Settlement 29 C.
BORATED WATER STORAGE TANK LINES 30 1.
Rebedding 130 2.
Future Settlement 30 D.
MISCELLANEOUS GENERIC
'BJ ECTS - 1.
Anchor and Component Loads 31 2.
II Under I 32 3.
Overburden Loads 33 IV.
LIST OF REFERENCES IV-1 V.
LIST OF TABLES V-1 VI.
LIST OF FIGURES ~
VI-l VII.. APPENDIXES VII-l iii 12/10/81-
I.
SCOPE This document addresses the open iyggs identified in the draft Safety Evaluation Report (SER) that are related to Seismic Category I piping buried in the plant fill.
These items were identified in Sectian 3.7.3 (Page 3-7, Items 4 and 5 ),
and Section 3.9.3 (Page 3-13, Items 1 through 6) of the. draft SER.
The discussion of the items identified in Section 3.9.3 will provi a the information needed to resolve the items ide.tified in Section 3.7.3.
The Seismic Category I buried piping systems included in this document are:
a.
Service water system lines b.
Diesel fuel oil lines c.
Borated water storage tank lines A complete list of the included pipe line numbers is included in Table I-l and their locKtion is shown in Figure I-1.
The control room pressurization lines are also Seismic Category I lines buried in the plant fill.
They are not l
addressed in this document because they have recently been installed (May 1981) and therefore have not been 6
subjected to settlement.
1 12/10/81 1
II. -
SUMMARY
II.A BACKGROUND The NRC~ staff han expressed concerns for the adequacy of buried safety-related piping at the Midland nuclear
_ plant due to settlement.
These concerns were-orig expressed in the NRC Requests Regarding Plant Fill [gylly.
Questions 16 through 20. -These requests.will hereinafter ~
be referred to as "50.54( f) Question (s)...-.".- The' concerns have been discussed tur Consumers Power Company L and the NRC staf f throughout 1981; -in January, May, and October meetings, and in' numerous telephone conversations.
To resolve the NRC concerns, extensive measurements have been taken on piping location, elevation *,fand ovality, so that the current condition of the' piping is well-defined.
The as-built condition was, however, generallylless well-defined. -This has made it difficult to establish how much of the current profile was caused by settlement since installation and how muca of it is due to as-built condition.
Discussions have been continuing on methods for establishing the current stress condition in this piping.
As an alternative, a. " demonstration solution" has been proposed to establish that the -pipe has sufficient dimensional stability to. maintain its functional ~capabilit;'.
4 2
12/10/81
~,,, -,,.., - - -,
e---
no
II.B FUNCTIONAL CAPABILITY - PROPOSED SOLUTION In a telephone conference on August 25, 1981, the NRC concern for maintenance of functional capability was expressed.
The telephone conference was to provide the NRC staff response to the demonstration solution approach proposed by Consumers Power Company in an August 10, 1981, telephone conference.
The demonstration solution as proposed August 10, 1981, and changed and supplemented in the October 6,
- 1981, meeting consists of:
1.
Passing a device through the pipelines to a)
Establish that the pipe has not buckled, or b)
Manually obtain ovality measurementa in large lines 2.
Performing periodic hydrostatic testing, including leakage measurement, to ensure pipe integrity 3.
Performing periodic flow verification test-ing to ensure functional capability The program is to demonstrate whether the pipe has re-tained sufficient dimensional stability to maintain the system's{ggctionalcapability.
Standard Review Plan SRP 3.9.3 recognizes the validity of this approach and provides guidance.
This guidance includes, in part, the statement that, "Since the treatment of functional capability, including collapse and deflec-tion limits, is not adequately treated by the Code for all situations, such factors must be evaluated by designers and appropriate information developed.
(code requirements are discussed in more detail in Section III.A.4.d).
This guidance indicates that an alternative to determining stresses is to demonstrate that the areas of discontinuity retain sufficient dimensional stability.
Teledyne Engineering Services (TES) stated in a letter to Consumers Power Company,
" Retaining sufficient dimensional stability is, in fact, the only basic question to be answered and is directly related to assuring functional capability of the piping" (see Appendix A).
3 12/10/81
The program has demonstrated acceptable current dimen-sional stability by inspecting the pipe to determine cross-sectional shape-(ovality) which is directly related to stability.
These results are discussed in Section III.A.l.
Continued functional capability will be demonstrated by flow verification tests to be con-ducted during plant operations.
An additional check of functional capability will be provided by the inservice inspection (ISI) program (see Section III.A.6).
This type of testing will not explicitly show that no pipe deformation is occurring; rather, it demonstrates that deformation sufficient to reduce the flow below minimum requirements has not occurred.
To ensure that the system contains sufficient margin to prevent loss of functional capability, conservative acceptance limits have been established using code guidelines and standacdc for buried pipe in Amer [ggn Water Works Association (AWWA) Specification Mll We have also based our acceptance limits on the general piping standards for bending nuclear pipe according to ASME Section III codes.
Our confidence in these limits is supported by the results of various pipe experiments reported by E.C. Rodabaugh and S. E., Moore in NUREG/CR-0261(5).
The introduction to NUREG/CR-0261, under " Relevance to Functional Capability," states, "We do not have any test data in which large enough displacements were applied to produce significant reductions in flow area; e.g., 50% reduction of flow area.
We would guess that to produce such a condition in straight pipe by appli-cation of a moment load, a rotation of 30* or more over a length of about 2 pipe diameters would be necessary."
The NUREG discussion then states, "The moment to produce this ' kink' in the pipe might not be much greater than the ' limit moment'; the displacement would be far in excess of any normally-used criterion for defining a
' limit moment.'
It is importan t to note that exceeding the deflection corresponding to a limit moment does not necessarily mean that functional capability will be significantly impaired."
These conclusions indicate that it would take far more deflection than can con-ceivably occur in buried pipe due to settlement to significantly impair the pipe's functional capability.
The AWWA conclusions are discussed in Section III.A.2.
4 12/10/81
II.C ANALYTICAL SOLUTION DIFFICULTIES The difficulty with analytical solution is separating the as-built condition of the piping (i.e., the local installation discontinuities) from the deflections due to settlement.
The misalignments and discontinuities reflected in the field data are inherent in the fab-rication process.
Project quality records indicare that the pipin standards?67as fabricated and installed within acceptable (15/32 inch, local mismatch; io/32 inch,
+3 overall mismatch; 12 inches, overall locat n).
The calculated stresses based on field. deflection measurements cannot be relied upon because the measure-ments include installation discontinuities as well as soil settlement.
For example, allowable angular mis-matches of weld joints are magnified over a long length of pipe and can appear as " knees" along a straight line (see Figure II-1).
Assuming that these knees are due to soil settlement results in concentrating the curvature at the knees, thereby significantly overesti-mating the stress levels.
Deflections of this magnitude resulting from settlement would result in gross local deformations that would have been apparent during examination.
Using the calculated stresses, these deflections would produce ovality well beyond 8%.
The analytical solution using empirical data is further complicated by the measuring tolekance.
Measurement inaccuracies can cause apparent pipe oscillations to be overemphasized.
In 1979 profiling was done to approximately 11/4-inch accuracy, with measurements every 10 feet.
A parametric study over a 20-foot span using vorst case measurement errors, (1/2-inch deflec-tion) yielded a calculated elastic s' tress of 55 ksi.
This stress alone is greater than the allowable stress.
The latest reprofiling has been done to a tolerance of 11/16-inch, but the number of survey points has also been increased, thus decreasing the flexibility and increasing the sensitivity to the measurement toler-ances.
To develop a computer moder of the piping, a rigid restraint in the vertical plane forces the pipe into tne measured profile configuration at the survey locations along the pipelines.
This does not allow the pipe to flex according to its geometric and material properties.
These abrupt changes (knees) at the 5
12/10/81 l
survey locations concentrate the pipe curvature near these local discontinuities, resulting in artificially high local stresses.
Thus, fitup and installation differences (discontinuities), assumed to be settle-ment, will result in erroneous, very high calculated stresses.
Structural Mechanics AssociatesI ) performed calcula-tions by modeling the pipe as a beam on elastic foun-dation to determine the soil loading necessary to cause the measured deformations.
This study showed that soil ',adings as much as three times the conser-vative esL_; ate of the soil capacity would have been needed (see Figure II-2).
The limited information available about presettlement as-built conditions proves that we have an unrealistic calculational sol-ution.
The modeling technique was further refined to include nonlinear aspects of the pipe and soil para-meters.
The computer results would not converge on the measured pipe configuration.
This demonstrates again that, in certain locations, the ueasured pipe profiles could not occur due to soil settlement alone.
The problem of developing an accurate analytical model is complicated by the presence of the soil around the pipe and the scil / pipe interaction.
The soil character-istics such as friction and soil support mechanisms are very difficult to approximate.
As the pipe tries to deform (ovalize), pressure devblops between the pipe and the soil which counteracts the ovalization and maintains the pipe geometry and, thus, functional capability.
The basic analytical problem is how to separate the as-built condition of the piping from the deflections due purely to settlement.
We have concluded that the profile data cannot be used in a traditional flexi-biliti analysis unless an agreement can be reached on a method to accomplish this separation.
6 12/10/81
III. DETAILED DISCUSSION A.
SERVICE WATER SYSTEM
- 1..
Profile and Ovality In 1979, a profile of one line in each trench was done.
The profiling was done to approxi-
'ely + 1/4-inch accuracy with measurements uvury 10 feet.
In August 1981 new profile and ovality measure-ments were started in all service water system piping.
This was to obtain more accurate information and to profile the condition of all lines which had not been measured.
Reprofiling -and ovality measure-ments of the service water supply and return lines were completed in October 1981 (see results in Appendix B).
The 1981 profiles involved cleaning the interior surface and marking it at a minimum of 5-foot increments for measurement. - Measure-ments at some-locations,'particularly in elbows, were as close as 1.5 feet apart.
Measurements were also taken 2-1/2 inches on either side of pipe welds.
The tolerance on the measurements was estimated to be +1/16 inch.
(See Section II.C for discussion of the effect of these tolerances.)
To do the 1981 profiles, a unique apparatus was developed by Southwest Research Institute (SwRI).
The pipe elevation profile measure-ment system developed by SwRI for this effort is shown in Figure III-1.
The device uses a pressure transducer moved within the pipe and positioned on the pipe bottom (as determined using a bubble level on the transducer).
It measures the differential pressure between a reference water column and a water column ending at the transducer.
The system used in 1979 was similar, but involved a visual measurement rather than sensed differential pressure.
7 12/10/81
Ovality is measured at the same locations as elevation with another SwRI instrument-(See Figure III-2).
The. device uses rotating arms to obtain-both maximum and minimum diam-eters.
Their azimuth orientation is recorded with the-azimuth location of the longitudinal fabrication weld.
Fittings were measured using the same measurement arm; however, this required removing it from the rolling platform (dolly) which was used in streight pipe sections for accurate positioning.
The ' ovality measurements for both straight pipe and fittings have been plotted and are shown along with the profile data in Appen-dix B.
They generally were less than 2%.as compared to the manufacturing tolerances of 1%-for straight pipe (ASME SA155) and 1.76%
for fittings (ANSI B16.9).
Some piping fabricator catalogs (NAVCO, in particular) include a note that ovality may change due to handling.
They indicate that for pipe manufactured to a 1% tolerance, experience shows that 2% or more ovality is normal for pipe installed in a trench ready for backfill.
For the ovalities measuked at Midland, there is no way to determine how much is due to settlement, but in any-case the ovalities measured are within the range considered. normal for newly installed pipe.
8 12/10/81
III.A.2 Ovality / Buckling The bending stresses induced in the buried pipe by settlement are similar to Zabrication bending stresses because the support provided by the surrounding soil is similar to the radial support provided by a bending mandrel.
The acceptance criteria for ovality that we propose to use is 8% as stated in ASME Section III codes (NC-4223.2 and NC-3642.1) as the tolerance for installation and fabricated bends.
Most codes that discuss ovality relate it to the fabrication of bends.
Most of the codes limit the ovality in the bend area to be a maximum of 8% (ovality defined by (D
-D - )/D The bending / forming requiremeE6E in EENE S8c) tion III, ANSI B31.1, B31.3, and PFI ES-3 all.acorporate this limit.
Some of these codes imply that this limit is a
" good practice" tolerance rather than a limitation imposed because of material ductility considerations.
For example, ASME SA155 fabrication requires forming to a cylinder and joining with a full penetration weld.
This indicates the pipe material can take a permanent set in a manufacturing process substantially in excess of the 8% limit without sustaining damage.
Likewise*,
ASME SA106 requires flattening a section of pipe between parallel plates to a diameter approximately one third of the original diameter without any ev2dence of damage.
ANSI B31.1 earagraph 104.2.1.c and ASME III, NB-4223.2, provide for fla ttening greater than 8%.
It is evident from the above discussion that the codes indicate that considerable deformation can be sustained without damaging tne integrity of the pipe, and that restricting the ovality to 8%
is conservative when the actual ductility of the pipe is considered.
It should be noted that the existence of ovality does not in itself imply a structural failure of the pipe.
9 12/10/81
It should also be noted that the codes, and hence the. code considerations of bending and ovality, are based on an assumed failure where the moment carrying capability of the pipe is a maximum.
This presumes that after the instability point is reached, the conditions which caused the instability continue to prevail as in a load-controlled situation and that deformation will increase without limit.
Settlement, however, is a deflection-controlled condition where the settlement-induced secondary stresses may cause localized yielding, but are not self-driving to failure.
In the letter from TES (Appendix A), the appli-cability of the current ASME III Code require-ments were discussed in the following manner.
For the piping systems we are addressing here it is important to recognize that the entire buried pipe was subjected to soil settlement.
This is really a different situation than that addressed in current Section III criteria (NC-3611.2(f)) for non-repeated anchor movements.
Many of the reasons for this differ-ence have been discussed above and demonstrate the important variations between non-rapeated anchor motions (building settlement for a non-buried pipe) and general soil settlement.
NUREG/CR-0261(5) provides an experimental relationship between moment and ovality just before buckling.
The experimental results of Reference 12 of the NUREG defines flattening as the decrease in the diameter in the plane of the moment divided by the original diameter (D
-D This formula hasbeenverifiedbytele$k8n)/D.e 8cnference between J.F. Sorensen, author of Reference 12, and W.J. Cloutier of Consumers Power Company.
This definition of flattening is different from the definition of ovality used throughout this document, which is based on ASME (D
-D The difference restits in the flattE8fng c8b3r)/D.ising 10 12/10/81
half of the ovality.
The NUREG states that the flattening is a function of the diameter-thickness ratio (D/t), and is shown to be 4.5% for D/t = 100 for small-scale tests and 5.5% for large-scale tests.
This represents the flattening at the maximum load just before buckling.
The underground service water system piping D/t varies between 69 and 96.
Considering the calculation method of Reference 12, the ovality reported in the experiment would be 9 to 11%.
All analyses / experiments discussed thus far reflect analytical models or experimental conditions which conservatively neglect stabilizing influences present under actual site conditions.
These influences include the following.
a)
The assumption of an infinitely long pipe neglects the restraint provided by. adjacent cross sections undergoing a smaller degree of ovalization.
b)
The minimum specified yield stress values used in the analyses / experiments neglect the extra capacity indicated by the stress-strain data from the actual pipe material used at Midland.
i c)
The increase in the predicted buckling resistance of the pipe due to the service pressure was neglected.
d)
The confinement and cross-sectional support provided by compacted fill surrounding the pipe was neglected.
The cumulative conservatism represented by these four stabilizing influences is sufficient to raise our confidence about the appropriateness of the 8% acceptance criteria established to determine a pipe's worthiness as safety-grade piping.
Thecodemostdirectlyapp((gabletosteelpipe buried in fill is AWWA Mll In Chapter 8, Earth Loads on Steel Pipe, the following excerpts discuss the mechanism by which buried steel pipe support loads.
11 12/10/81
4 Although the maximum load-carrying capacity of flexible pipe depends to some extent on the wall thickness and its section modulus, the pipe, by deflecting, is able.to make full use of the load-carrying ability of the earth surrounding it.
As the pipe may change shape without failure, it transfers part of the vertical load into a horizontal or radial. thrust which is resisted by the passive pressure of the earth at its sides as these move outward.
When the wall itself is rigid, this movement may not occur.
It follows that.the rigid pipe must carry the whole load itself, whereas the flexible pipe divides the.
load with the earth enclosing it.
Therein lies the inherent difference between rigid and flexible behavior and the explanation of why the classical bending-moment formulas apply to the analysis of rigid pipe but not to the analysis of flexible pipe.
At this point, when deflection is mentioned, the engineer accustomed-to thinking in terms of flex ure or bending-moment formulas in rigid construction is likely to contend
~
that permanent deflection can occur only after the yield point has been passed and that, therefore, a pipe so stressed has failed structtrally and is dangerous.
The simplest rebuttal to this argument is to-recognize that the steel in a finished pipe has, in the manu-facturing process, been cold coiled,-
uncoiled, bent, curved, or twisted a number of times-and has been stressed beyond the yield point each time; yet, after all these operations have been a mpleted, the r
finished steel pipe is :1 sed for all manner of high-press 2re work without fear or hesitation.
12 12/10/81
If the engineer still is hesitant to restress a part of the finished pipe wall beyond the yield point by slightly deflecting it underground, let him consider what happens to the test speci-men by which the pipe strength is measured according to' specification.
Usually it is sliced as a ng from the end of a finished pipe, cut at one side, uncurled from the circle into a flat piece, and then.put in a tensile-testing machine which proceeds to show that after once more passing the elastic limit, the steel still possesses the specified strength.
In a way, the deflection underground is simply a finished forming operation.
Therefore, where steel pipe such as is here discussed is concerned, the word
" failure" must define a state of falling short of satisfactory performance and not a state in which localized stresses appear-to pass the yield point of the-material as judged by the results of bending-moment formula analysis.
These excerpts support the provision in SRP 3.9.3 that the pipe is acceptable hs long as it retains sufficient dimensional stability to ensure functional capability.
More specific to ovality tolerance, AWWA Mll, Sec -
tion 8.23.1 states, " Deflection of unlined pipe, or of pipe lined af ter installation, may safely reach 5 per cent of nominal diameter."
This deflection is nominally equivalent to 10% by the formula (D
-D used 1 / ASME and is based on failure"8f tggn)/Dcoa8ing, not any limitation of the pipe.
AWWA Mll, Chapter 8 also states, "Real collapse failure of steel pipe does not occur under earth loads until a condition is reached where the vertical diameter has been decreased about 20 per cent of the nominal diameter and the horizontal diameter has been increased a similar amount."
~
13 12/10/81
From the. foregoing guidance based on research, experimental results,'and years of experience, we feel that applying the 8% ovality criteria recommended by ASME is a very conversative acceptance criteria-for ovality due to settlement.
i 1
i l
l i
14 12/10/81 I
i i
l
\\
l--
I
III.A.3 Future Settlement a)
Predicted values The responses to 50.54(f) Questions (2) 4 and 27 contain a discucsion of the methods used to estimate future settlement.
The response to Question 27 includes the-following description of the twc. settlement components (Figure 27-1 is-attached as Figure III-3):
The distinction between
[ primary consolidation and seco]ndary compression settlement is made on the basis of the physical processes which control the time rate of setclement.
In primary consolidation settlement, the time rate of settlement.is controlled by the rate at which water can be expelled from the voids.
In the case of secondary compression settlement, the speed of settlement is controlled.
largely by the rate at which the soil skeleton-itself yields and compresses.
The transition time between these two processes is conveniently identified as that time when excess pore water pressure becomes essentially zero.
This time, denoted as t
is shown in Figure 27-1.
100 It has been observed in many laboratory and field measurements that the relationship between the magnitude of secondary compression and time is approximately a straight lino on a semilogarithmic plot after the primary 15 12/10/81
consolidation har been completed, as shown in-Figure 27-1.
Thus, the
- settlement AH can be expressed approximately as:
AH = -C log t /t a
2 y
where t and t are two specifia time heriods on the extrapolated secondary compression line and C
is the settlement per a
log cycle of time during secondary compression.
The response to Question 27 contains a much more extensive discussion of cettlement, prediccion method, and the basis for conservatism and acc.tracy than is presented here in these excerpts.
Supplemental Figures 27-51 through 27-198 show settlement vs log time plots for the diesel generator building.
They show that the fill is in the secondary compression settlement phase.
l In March 1980 a preliminary settlement estimate was provided for calculhting future pipe stresses.
The estimated settlement envelope was determined based on measured time-settlement data from Borros anchors buried in the plant fill.
Thil estimate resulted in a settlement envelope of 0 to 3 incnes for the 40-year plant life to be used in analyzing the piping buried in the plant fill.
This estimate did not include settlement that occurred prior to March 5, 1980, nor has settlement since March 5, 1980, been used to adjust the predicted value of future settlement.
i 16 12/10/81
III.A.3.b)
Monitoring program The service water system (SWS) future settle-ment shall be monitored at the terminal ends, before the first anchor point _ of each pipe as it enters the buildings.
The first pipe anchor inside the building'is the most rigid-anchor in the system, compared to the soil bedding outside the building.
Therefore, it is most susceptible to high stress.
The settlement to be monitored will be the differential settlement between the pipe anchor and the underground piping.
Thi.s settlement limit shall be. established fr*m the amount of settlement (Ay) the piping can tolerate before it reaches the ASME III code criteria for nonrepeated anchor movementi (3S,).
A representative cantilevered igngth of piping shall be used to calculate this limit.
The settlement limit must be corrected for any settlement which has already occurred and has induced anchor stress because settle-ment occurring before the piping was fitted to the pipe anchor does not cause pipe stress at this location.
The technical specifica,tions shall require a report to the NRC when the settlement reaches 75% of the maximum allowable settlement limit.
Upon reaching the 75% settlement point, the monitoring frequency shall be monthly, and engineering evaluations will begin.
The monitoring frequency for the monitoring points shall be similar to the monitoring frequency established and implemented for monitoring of structure settlement points throughout the plant.
The monitoring points shall be surveyed at 90-day intervals for the first 5 years of operation, and on a yearly basis for the remainder of operating life.
The anchor points as well as'a point on the piping as it enters the wall penetration shall be monitored.
If the differential settlement between the anchor and the desig-nated piping point reaches the 75% reporting limit, we would decrease the official moni-toring interval to 30 days, to better assess the settlement rate and severity.
17 12/10/81
III.A.4 Seismic a)
Seismic analysis The seismic analysis performed on the buried pipe uses the. theory and techn[gye
_ presented in Section 6.0 of BC-TOP-4A to calculate-forces and stresses at connections, tees, and bends in buried pipe.
These stresses are summarized in Table III-1.
The analysis is based on the equations for-beams on elastic foundation..The soil-subgrade modulus for each_ case is calculated
. based on the soil and pipe properties.
The method considers the ef fect of soil strain on the subgrade modulus and uses the vari-ation of shear modulus :with shear strain for sand as: developed by Seed and Idriss (Reference 8, Figure 2-5).
The analysis calculates forces and the corresponding stresses due to seismic movements of the surrounding soil and connecting st-uctures.
It is a static analysis based on the maximum soil strain which is, in turn,-based on the. magnitude-of the earthquake and the propagation velocities of the various seismic waves._
The flexibility and stress intensification-factors for welded elbows or pipe bends and welded tees are determined in accordance with ASME Boiler and Pressure Vessel Code,Section III, Division 1, Subsection NC, H
Table NC-3673.2(b)-1.
For each case with a bend, elbow, or tee, the analysis considers earthquake motion in two directions (i.e., parallel to each leg).
For each case with a connection to a building or component, the analysis consi-ders earthquake motion in three directions.
In the case of a bend, the transverse leg is assumed to deform-as a beam on an elastic
~
foundation due to the' axial force in the r
longitudinal leg (parallel to the earth-quake motion).
The displacement of the 1
18 12/10/81-
' bend is defined by the overall spring 4
constant-at the bend.
The_ spring constant of.the bend depends on the stiffness of the longitudinal and transverse legs as well'as the degree of fixity at the bend and at the far ends of the legs.
Tees and connections are analyzed in a similar manner.
t l
F l
4 l
t
(-
19 12/10/81.
III.A.4.b)
Variable soil properties The analysis. considers the following soil properties:
Poisson's ratio Unit weight
- Coefficient of friction (soil / structure)
Shear modulus Shear wave velocity compression wave velocity Surface wave velocity Maximum particle. velocity Maximum particle acceleration Maximum soil strain The soil subgrade modulus is calculated for each csse, based on the soil and pipe pro-perties.
The values used for these soil properties were those determined from the investigation work at the jobsite.
The sol.',
modulus of elasticity was varied +50%.
.The maximum particle' acceleration was increased 50% above the SSE value as a margin for.the site-specific response spectra.
20 12/10/81
III.A.4.c)
Effect of_ pipe deformation on seismic forces The pipe deformation will affect the seismic forces in two ways..The pipe settlement will cause'the idealized straight pipe to bend in the vertical direction, and the bending will cause the pipe cross-section to deform from the idealized circular cross-section to an ovalled cross-section.
To analyze the difference in the seismic loads on the idealized straight pipe and the actual settled pipe with a curved profile, a series of analyses was done on 26-inch dia-meter pipe configurations which varied from straight to a 5-degree bend.
The bend ra-dius was varied.from 1 pipe diameter to 100 pipe diameters (2,600 inches).
Figures III-4 and III-5 were used to, deter-mine what degree of bend and radius are re-presentative of the existing condition of pipe at the Midland jobsite.
Figure III-4 shows the measured profile for line 20"-2HCD-169 and focuses on the segment of greatest bend.
Figure III-5 shows the method used to deter-mine the bend radius and degree of bend.
This method establishes that the cross-section of this profiled line woul'd be a sag with ap-proximately 4 degrees of bend and a radius of 1,800 inches (90 pipe diameters).
The analyses showed that, for a straight pipe, the axial stress was approximately 5.5 ksi.
For a 5-degree bend, the axial stresa de-creased to 5.3 ksi and 5.4 ksi for 1-and 100-pipe diameter bends respectively.
For a straight pipe the bending stress is zero.
For a 5-degree bend the bending stress was 5.5 ksi and 1.3 ksi for 1-and 100-pipe dia-meter bends respectively.
With the configuration established in Fig-ures III-4 and III-5 (4-degree bend, 90 dia-meter bend radius) considered representative of existing conditions, both the axial stress (5.4 ksi) and the bending stress (1.1 ksi) were found to vary insignificantly from the straight pipe analysis.
21 12/10/81
k The seismic forces transverse to the axis of the pipe are so small that to distinguish between the forces on the theoretical cross-section and the forces on the ovalled cross-section is beyond the sensitivity of the methods used in the seismic analysis.
An indication of the effect of deformation on the transverse seismic forces can be obtained from determining the change in ovality resulting from seismic strains.
Assuming 2.5% ovality to be conservatively representative of pipe at the Midland job-site, the deformation resulting from a 2.5%
ovality'in a 36-inch diameter pipe is:
% ovality = 100 x (D
-Dmin)/D max g
2.5.= 100 x (D,,x - Dmin)/36 D
-Dmin = 0.9 inches max min = 0.45 inches
- D
-D
=D D
max g
g The additional deformation due to seismic strain transverse to the pipe axis is:
i Maximum soil strain = 0.000185 in/in Assume soil strain results directly in equal strain in the pipe.
Therefore, the seismic strain induced in the pipe _(Esse), is:
6 x 0.000185 = 0.00666 in E 0.007 in.
=
ESSE Assume this strain reduces the minimum diameter and increases the maximum di-ametir by the same amount.
D
-Dmin
- I
-Dg) max max Dmin)
= 2(D g
22 12/10/81
Adding the seismic strain results in:
D,,x - Dmin *
( o ~' min)
- 6SSE
=-2(0.45 + 0.007)
= 0.914
% ovality = 100 x (D,,, - Dmin)/Dn
= 100 x (0.914)/36
= 2.539%
Thus, the effect of the seismic loads on an ovalled pipe would be to increase the ovality from 2.5% to 2.539%, which is still within allowable limits.
The preceding discussions indicate that the seismic analysis of the deformed piping, considering the deformation of the piping, would result'in axial.and bending stresses virtually unchanged from those for a straight pipe, Land an increase in ovality from 2.5 to 2.539%.
i 4
1 23 12/10/81
III.A.4.d)
Code requirements There has been discussion with the staff on the treatment of seismic stresses and settlement -
stresses.
The staff's concern is that if our settlement stress calculations do not meet the 3S limit as specified for single-anchor point mo9ements in ASME Code Section III,.these stresses must be combined with the primary stresses in Equation.10 of Paragraph NC 3652.2..The stress effect of any single nonrepeated anchor movement is compared to a separate allowable (3Sc) in Equation 12 of Paragraph NC 3652.3.
Our position, based on settlement stress calculations, is that most of the piping is not overstressed above the code allowable (3S and in the' local areas where analysis indic)a;tes an apparent overstress, it is mainly_due to the analytical difficulties in-treating the profile data.
These difficulties were discussed in Sec-tion II.C.
Furthe rmore, if we do combine settlement stress with seismic stress, it would not be clear from an ASME code view-point which code allowable to use.for com-parison'with the calculated stress.
24 12/10/81
III.A.5 Rebedding a)
Size verification of 8-inch lines On October 28, 1981, diameter verification pigging operations-were conducted on four 8-inch diameter piplines.
The specific lines were 8"-lHBC-310, 8"-lHBC-311, 8"-2HBC-81, and 8"-2HBC-82.
The results indicated that each pipeline was greater than 7.781 inches in diameter and was not obstructed.
This indicates that none of the pipes has been flattened due to bending or heavy loads and they currently meet the 8% acceptance criteria for ovality.
I The pigging operation was conducted in accordance with Appendix C and-provided a go, no-go test to check ovality.
The re-sults are described in Appendix C.
b)
Rebedding of 8-and 10-inch service water lines Lines 8"-lHBC-81, 8"-lHBC-82, and 10"-OHBC-28 were.previously rebedded.
Service water lines 8"-2HBC-311, 8"-2HBC-310, and 10"-0HBC-27, near the east side'of the diesel generator building, which have not previously been rebedded, vill be rebedded to conform to a straight unstressed condition.
These lines are identified in the detail section of Drawing SK-C-745 (shown as Figure I-1).
25 12/10/81
III.A.6 Verification As discussed in Section II.B and briefly mentioned in Section III.A.2.b, it is necessary to demonstrate that the pipe has sufficient dimensional stability to maintain its functional capability.
This~will be accomplished by a program of preservice end inservice checks, tests, and inspections.
a)
Preservice current dimensional stability has been established by inapecting the pipe to determine cross-sectional shape (ovality).
Section III.A.1 discusses the equipment and technique for determining ovality, and provides the results of these surveys.
A construction hydrostatic test (ASME III.
NC-6221, NC-6129) will be done as follows:
o Test pressure of 1.25 x system design pressure o
Hold interval of 1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br /> to test inacces-sible weld joints o
Monitoring test pump leakage to estab-lish future l'eakage criteria 26 12/10/81
III.A.6.b)
Inservice Inservice inspection will be performed in accor-dance with ASME Section XI as committed in Midland Preservice and First Ten Year Interval Inspection Plan - for NDE and System Pressure Tests - Volume II.
The ISI program consists of inservice tests and hydrostatic tests to ensure pressure boundary integrity.
The inservice tests are described in Figure III-6 and the hydrostatic tests are des-cribed in Figure III-7.
The leakage acceptance criteria for these tests are shown in Figure III-8.
The ISI will be done with one unit at power during the test.
Tha remaining SWS train will supply cooling water to both uni?.a by crossover piping in the auxiliary building and turbine building.
Rapid restoration of the tested SWS train is possible because normal isolation valves will be used during these tests.
The flow verification tests to be conducted during plant operations are outlined in Fig-ure III-9,,and Tables III-2 and III-3 show mini-mum required flows and the number and location of flow measurement elements.
The requirements are proposed for inclusion in the technical specifications.
The monitoring program performs a trending evaluation of the test data to detect any decrease in flow, although acceptance cri-teria are met.
l l
l 27 12/10/81 n
III. RESOLUTION B.
DIESEL FUEL OIL LINES 1.
Profile The diesel fuel oil lines were installed in June 1980 after the diesel generator building surcharge program was completed.
The as-built elevations of those lines were surveyed approximately every 20 feet.
These elevations are shown in Drawing MPY-138Q (Figure III-10).
This drawing also shows piping support details.
It is mounted on Unistrut sections embedded in concrete at intervals along the pipe length.
The piping was then covered by approximately 2 feet of compacted soil.
t t
I l
[
l t
j 28 12/10/81
III.R.2 Future Settlement 50.54(f) Questions (2) 17 and 20 discuss the stresses induced in buried pipe due to settlement.
Both responses (Table 17-1, Note 6 and Page 20-2, third paragraph) indicate that the fuel oil lines are of such small diameter (1-1/2" and 2") that they have enough flexibility to withstand the predicted settle-ment without exceeding allowable stresses or affecting their structural integrity.
To substantiate this judgment, an analysis was done to evaluate stresses in the following diesel fuel oil lines due to predicted future settlement:
1-1/2"-lHBC-3, 1-1/2"-2HBC-3, 1-1/2"-lHBC-4, 1-1/2"-2HBC-4, 2"-lHBC-497, 2"-2HBC-497, 2"-lHBC-498, 2"-2HBC-498 This analysis assumed:
l.
Three inches of settlement was proportioned over a 40-foot pipe span with the 3 inches occurring at midspan.
2.
Simplified beam equations were used for
' buried piping continuously supported by soil.
The analysis indicated the highest stress value, including stress intensification factors, was 18 ksi in a 2-inch diameter line.
This is well within the allowable stress of 45 ksi (3S ) for these lines and substantiates the claim of fle.kibility made in the responses to 50.54(f) Questions 17 and 20.
29 12/10/81'
d' III. RESOLUTION i
C.
BORATED WATER STORAGE TANK LINES 1.
Rebedding i,
In the October 6 and 7, 1981, meeting.with the NRC staff, Consumers Power Company committed to.rebed the 18-inch BWST.line from the tank valve pit to the tank farm dike.
These pipelines are identified as 18"-lHBC-1, 18"-lHBC-2, 18"-2HBC-1, and 18"-2HBC-2.
This commitment'was made be-cause this-piping is in the area to be sur-charged as part of the remedial fixes on the tank' foundations.
The measured profile data taken in 1979 on_ pipelines 18"-2HBC-1 and 18"-lHBC-2 show maximum deflections of 1.92 inches and 0.96 inch, respectively.
These measurements are within the construc -
l tion tolerance of +2 inches for installation of piping and it may be assumed that soils settlement has not adversely affected this piping.
2.
Future Settlement Borated water storage tank lines have been cut loose at the valve pitito isolate them from the settlement caused by.the surcharge of the valve pit.
The existing program which monitors the settle-ment of the BWST and the auxiliary building,will l
provide data on the future settlement of these lines.
These monitoring points will indicate whether.the piping is overstressed due to settle-l ment.
1 l
l c
I 30 12/10/81 i
III. RESOLUTION D.
MISCELLANEOUS GENERIC SUBJECTS There are several subjects pertinent to most of the onried pipe.
Rather than discuss each subject several times as it relates to each piping system, this section will discuss each subject, including how it affects each piping system.
The subjects considered generic to all buried pipe and which are discussed in this section are:
o Anchor and component loads o
Effects of rupture of nonsafety-related piping on safety-related piping, components, and structures (referred to herein as "II under I")
o Overburden loads 1.
Anchor and Component Loads The loads induced into anchors and compo-nents by settlement of the underground piping are being analyzed to determine acceptable settlement used limits.
These limits will be in conjunction with the monitoring program discussed in Section III.A.3.6.
The settlement limit shall be established from the amount of settlement (Ay) the piping can tolerate before it reaches the ASME Code Section III criteria for nonrepeated anchor movements (3S ).
The limit will be the lesser of the seEtlement which causes the limiting stress or the settlement which causes contact with the penetration through the building.
The settlement limit will be corrected for any settlement which has_already occurred and has induced pipe stress at the anchor point.
31 12/10/81
III.D.2' II Under I In the draft SER(1) the NRC expressed a' concern for the effects of the rupture of nonsafety-related piping on safety-related piping, components, and structures.
This concern is referred to herein as "II under I."
This concern is a classic-II/I question, brought up during a discussion of underground piping settlement at Midland, but not peculiar to the Midland soils issue and not unique to Midland.
Pipe break encompasses not only whip _and jet impingerent, but also the related hazards of steam and liquid flooding, excess pressure, differential pressure, and temperature.
Of the foregoing effects of a pipe break, liquid flooding is the single item requiring evaluation for buried piping.
Analysir. of flooding is treated on a case-by-case and individual system basis.
The possible result of flooding would be a washout / loss of support.
A review was done to identify where non-Seismic Category I pipe passes beneath a Seismic Category I
-pipe or structure.
.A break in the non-Seismic Category I pipe was. assumed to cause a washout extending to the surface, th'us causing a loss of support for any Seismic Category I system.above it.
The unsupported length was determined using a side slope of 45 degrees, the vertical separa-tion, and the angle of crossing of the two systems.
The review indicated that for all non-Seismic Category I pipes passing beneath a Seismic Cate-gory I pipe, the maximum stress induced in the overlying Seismic Category I pipe was approxi-l mately 3 ksi for line 1-1/2"-2HBC-498.
[
The effect of a non-Seismic Category I pipe break on structures is' considered to be encompassed by
-Question 49{ggussedintheResponseto50.54(f) the break d Part c2.
The pertinent portion of this response is included as Appendix D.
I r
I 32 12/10/81-
II.D.3 Overburden-Loade This.section discasses the effects of overburden loads such as soil uead weight, heavy equipment, etc Question {jedolping.
on-the by
~
The Response to 50.54(f)
.ddressed this question.
The Response to Question 34 is attached as Appendix E.
The Response to Question 34 refers to the effect, at a depth of 6 feet, of a Cooper E-80 railroad load.
A review of the depth of cover (distance below ground surface) of all Seismic Category I lines' indicated that 6 feet is the approximate depth of cover on all lines except the diesel fuel oil lines.
The results, indicated in Appendix E, concluded that the 26-inch and 36-inch buried Seismic Category I pipes are adequate to withstand external loads, and stresses in pipes smaller than 26-inch diameter will be relatively low and are not critical.
Thedieselfueloillineshaveayjnimumcoverof approximately 2.2 feet.
AWWA Mll
- includes a graph (see Figure III-ll) showing the relationship between load (expressed as height of cover) and the diameter of steel pipe.
This graph shows that for diameters less than 20 inches, the amount of load needed to cause a 1K deflection increases almost infinitely.
According to this graph, a 1-1/2 inch to 2 inch diameter pipe would be virtually uncrushable when buried in the fill.
33 12/10/81
IV.
' LIST OF REFERENCES Referenced 1.
Safety Evaluation Report (draft), Sections 1, 32 3.6.2, 3.7.3, and 3.9, transmitted by R.L.
Tedesco's September 23, 1981, letter 2.
NRC Requests Regarding Plant Fill 2, 15, 29, (referred to as 50.54(f) Questions) 32, 33 3.
Standard Review Plan SRP 3.9.3, ASME Code-3 Class 1, 2, and 3 Components, Component Suppcrts, and Core Support Structures, Rev. 1, July 1981 4.
AWWA Manual Mll, Steel Pipe Design and 4,-11, 33 Installation, American Water Works Asso-ciation, 1964 5.
NUREG/CR-0261,-Evaluation of the Plastic 4, 10 Characteristics of Piping Products in Relation to ASME Code Criteria, July 1978 6.
Specification 7220-M-214, Piping System 5
Erection Fitup Control Requirements 7.
Evaluation of Pipe Behavior Due to Soil 6
Settlement for a Typical Buried Line for the Midland Nuclear Power Plant, Struc-tural Mechanics Associates,- May 1981 8.
BC-TOP-4A, Seismic Analyses of Structures 18 and Equipment for Nuclear Power Plants, Rev.- 4 9.
ICS Civil Engineer's Handbook, page 136; Fig. III-5 edited by Archibald DeGroot, International Textbook Company, 1956 l
IV 12/10/81
}
a
V.
LIST OF TABLES Referenced I-l Seismic Category I Lines to be Addressed 1
III-l Stress Summary for Buried SW Piping 18 III-2 Minimum Required Flows 27 III-3 Flow Measurement 27 v-l 12/10/81
t-
. SEISMIC CATEGORY I LINES TO BE ADDRESSED A.
Service Water System.(SWS) 8"-lHBC-310 26"-OHBC-53 8"-2HBC-81 26"-OHBC-54 8"-lHBC-81 26"-OHBC-55 8"-2HBC-310 26"-OHBC-56 8"-lHBC-311 26"-OHBC-15 8"-2HBC-82 26"-OHBC-16 8"-lHBC 26"-OHBC-19 8"-2HBC-311 26"-OHBC-20 10"-OHBC-27 36"-OHBC-15 10"-OHBC-28 36"-OHBC-16 36"-OHBC-19 36"-OHBC-20 B.-
Diesel Fuel Oil Lines (Fuel Oil) 1-1/2"-lHBC-3 2"-lHBC-497 1-1/2"-lHBC-4 2"-lHBC-498 1-1/2"-2HBC-3 2"-2HBC-497 1-1/2"-2HBC-4 2"-2HBC-498 C.
Borated Water Storage Tank (BWST) 18"-1HBC-1 18"-lHBC-2 18"-2HBC-1 18"-2HBC-2 TABLE I-l 12/10/81
..., ~.
s' I
sus ott ovv b
)\\
l\\
h g iiI I
II I-
.i
=
.O T
A o y
a t
d ll
't 3
ss4 5
F-5 O
fk k
Y o' 0 $
'a, 3 i i 3
I 1
! s*
c n a-o e s ~l e
J b
h J
s hd h' N N
q Q
.1 <s.,N s
N s 4 <
4 t
3 + r m e,-
I $
q l
4
'O 'o 'o 'n D 'o so sQ so
- O
-o
- O ~
'o
' ' 0 "'
)
4
$* O O O O O O O O
O o o o o
o.
Q f
E O O o o o o o o o
o o n o
a
$ NAN $] $N
\\g* &
8
)
F 444+ + t te mmn o
rn g
,b M_ C'
'A
'g
'gQ
'O N
N b
E
{
bm-K-
4W O el d 0 g3 (A g
ad f n o' J a a e in' -e
_a m40 0 0
=
<S n -mmn
-N
}
-) 2
")
'o 'o 'o 'O
-o ' o 'o o
o o
'o. o
'o
'O s
o, o 8
f,3 oC 0 0 o o C 0 0
M - y o
e o O o C 0
0. o o
s s
-00 00 to 07 3r _, _, _ ' 2
_s s as a
W i 4 _6 N
~
y 7 N d gj d
~
~~
f
~
xf
.r-O y g cJ % o b d OAU o d4 U h 9
-g-g.g.
d s
n
~
Q t' h *m
~
m ohkU f
n-s f M @m@
M dO O o
m i
3 p Y
=
k 'o 'o 'o 'O oo 'o 'O
'o
'O 'y 'o
'Q o
eo 5'
fh $$$g O.,
I 1
g g tR V
a.u-p g c: _tr oo
< e I
y j
r3 M 'a'a W a W 'a 'a
'in b % 'm
'M g
m a s a a
a
+ +r-#e u \\3 0 N3 e
c 2
r-r r-
\\g V
i t e e t e
2 H N N N - ~
e 4
o w
O 2
7 p2 a
3 Y $ Y fh86=d' y6I I
5 Z
5 i $i h-3
=
y 3 u
- A
}
=o p
n y 3%
3 3
3 1
Q 2 2 h 3 2 h h 2 e
ei
._ A e e w 4 6 A e 4 e y
I h
dN5
! !, ! E s a a $ h 5 $ $ 5#. M 8 e s s, s M A c*
w
=
=
.1 1 1 g
8 ' 1 g==
2 M
h TABLE III-l
MINIMUM REQUIRED FLOWS Required Line Description Flow (gpm) 8"-1 HBC-310 DG 1 A Supply 1,600 8"-2HBC-81 DG 2A Supply 1,600 8"-1 HBC-81 DG 1B Supply 1,600 8"-2HBC-310 DG 2B Supply 1,600 8"-1 HBC-311 DG 1 A Return 1,600 8"-2HBC-82 DG 2A Return 1,600 8"-1 HBC-82 DG 1B Return 1,600 8"-2HBC-311 DG 2B Return 1,600 10"-OHBC-27 DG 18128 Supply 3,200 g
h 10"-OHBC-28 DG 1B128 Return 3,200 26"-OHBC-53 DG 1 Al2A+TB Supply 9,225 H
[
26"-OHBC-54 DG 1 A12A+TB Return 9,225 b
26"-OHBC-55 DG 1B12B+TB Supply 9,225 26"-OHBC-56 DG 1B12B+TB Return 9,225 26"-OHBC-15 Aux Bldg A Supply 15,894 26"-OHBC-16 Aux Bldg A Return 15,894 26"-OHBC-19 Aux Bldg B Supply 15,894 26"-OHBC-20 Aux Bldg B Return 15,894 36"-OHSC-15 A Supply 25,119 36"-OHBC-16 A Return 25,119 36"-OHBC-19 B Supply 25,119 l
36"-OHBC-20 B Return 25,119 Required flows are based on FSAR tables 9.2-1 and 9.2-2. Worst <:ase values for each line were d'Aermined fror;1 the six operation modes and the ESF mode in those tables. Turbine building flows are based on poten%* flow under accident conditions (Mode 6).
l MOLAND UNITS 1 AND 2 NRC PRESENTATION 10/2/81 G 1868-02
FLOW MEASUREMENT Line Description Flow Element Locat6on 8"-1H8C-310 DG 1 A Supply 1 FE 1841 Cooler Outlet 8" 2H8C41 DG 2A Supply 2FE 1851 Cooier Outlet i
8"-1 H8C-81 DG 18 Supply 1FE 1846 Cooler Outlet 4" 2H8C-310 DG 28 Supply 2FE 1855 Cooier Outlet 8"-1H8C 311 DG 1 A Return 1FE 1841 Cooler Outlet 8"-2H8C42 DG 2A Return 2FE 1851 Cocier Outlet 8"-1H8C-82 DG 18 Return 1FE 1846 Cooler Outlet l
8"-2H8C 111 00 78 Return 2FE 1855 Cooler Outlet a
10"-OH8C 2T DG 18I28 Supply IFE 1846 +
Cooler Outlet 2FE 1855 Cooler Outlet 10"4H8C-28 DG 1BI28 Return 1FE 1846 4 Cooler Cutset 2FE 1855 Cecier Outlet 26"4H8C 53 DG 1 Al2A + TB1 Supply IFE 1876 Supply Line asetering F.4 I
26"-OH8C-54 DG 1 Al2A +181 Return 1FE 1876 Supply Line heelering Pit 26"4H8C 55 DG 18I28 + T82 Supply 2FE 1876 Supply Line neotering Pit e-3 26"-OH8C-56 DG 18t2B + TB2 Return 2FE 1876 Supply Line testerine Pit h
26"4H8C-15 Aus Skig A Supply 0FE 1995A +
Aun 8ksg A Supply Line p
IFE 1914A +
Soester Pwnp Discharge trj 1FE 1990A +
Chiller Outlet 2FE 19904 Chiller Outlet 26"-OH8C-16 Aus 8kig A Return OFE 1995A +
Aus Skig A Supply Line H
1FE 1914A +
Soester Pump Discharge i
1FE 1990A +
Cheller Outlet ta 2FE 1990A Chiller Outlet i
26"4HBC-19 Aus SkSg 8 Supply 0FE 19958 Aus Skig 8 - Return Line 26"4HSC 20 Aus Bide 8 Return OFE 19958 Aus 8kig 8 Return Line 36" OH8C-15 A Supply IFE 1878 +
Supply Line nestering Pit 0FE 1995A +
Aus 88dg A Supply Line 1FE 1914 A +
Sooster Pump Discharge 2FE 1990A Chiller Outlet 2FE 1990A Chiller Outlet i
36"-0H8C 16 A Return 1FE 1876 +
Supply Line 48eteN4 Pit OFE 1995A +
Aun 8ksg A Supply Line 1FE 1914 A +
Sooster Pump Diecharge 1FE 1990A +
Cheller Outlet 2FE 1990A Chiller Outlet 36"4H8C 19 8 Supply 2FE 1876 +
Supply Line teetering Pit 5
0FE 19958 Aun 8ksg 8 - Return Line 36"4H8C-20 8 Return 2FE 1878 +
Supply Line Reetering Pit 0FE 19958 Aus Bldg 8 Return Line (Thas list conftes cepetuidey to snesswo Okms in thaned servece weler system gwpeg wehg instened instrumentation in some areas, addehene8 enessurement devices ero installed Ihat may be considered preferehte shorneuvas)
Me tal 2.et G 6 96s OJ
VI.
LIST OF FIGURES Referenced I-l' Plan of Buried Q-Listed Pipe Locations, 1, 25
.Bechtel Drawing SK-C-745' II-l Initial Discontinuities in Installed 5
Pipe, Figure 6_from. Southwest Research Institute' Report, " Structural Analysis of Buried Pipeline," October 16, 1981 II-2 Linear Elastic Analysis Results for 6
Upper Bound Soil Properties, Figure 4-2 from Structural Mechanics Associates report, Evaluation of Pipe Behavior Due to Soil Settlement for_a Typical Buried Line for the ' Midland Nuclear Power Plant, SMA 13701.02, May 1981 III-1 Schematic-Pipe Elevation Profile 7
Measurement System, Page 8 of South-west Research Institute Report No. 1, Pipe Profile Measurements at Midland, August 1981 III-2 Sketch-SwRI Out-of-Roundness Measurement, 8
Instrument III-3 Typical Laboratory Time-Settlement Behavior 15 Under Constant Pressure.
III-4 Marked-Up Copy of Figure II-2 21 III-5 Determination of Bend Radius and Degree 21 of Bend III-6 Inservice Tests - Leakage Tests 27 III-7 Hydrostatic Tests - Leakage Tests 27 III-8 Leakage Test Acceptance Criteria 27 III-9 Flow Verification 27 III-10 Top Line Elevations of Diesel Fuel Lines, 28 Bechtel Drawing 7220-FSK-MPY-138 III-ll Relationship Between Calculated Height of 33 Cover and Diameter of 1/4-Inch Steel Pipe VI-l 12/10/81
6
+
,..,_.....,.3 l
5
^
41 I 3
l 4
f.!
N
}l!I., l'{
\\ n h....\\ rl:'.
9 'I i
..i.e! ini
,1
!o
~
il.
- - M a>
C a
m2 3
n.
~~
/
as
'I *J it!
r i,
2 Y :4 i NI!N UN(;p.j; it il[ :3{'12 h _"_
il;I$
3 h
i 'k ! !!l U
.[!.i,l:al ij'i 5
C (i
.l' l ::
ll..f1!
t e
~
II
! l M =
I.,I
wa
. g
=..
a a a qqqq,spq t{
n I
s_ _ llL,
7
!g l
ili 1 !33
- ___k F3-l lit t'
3 3{
y' lt
' ll 1
=
=
st i
.:g g
Tl I
--A q;;_
',Ng < t 7
1
=
y s
Jl-
\\
1 1.
\\
ifl[ - g i
um g'
. ]tR l
),..
i
=
! @s
[;
I.
< A
[A it.
e a
e
' b
, 't l
'[,E I
l l
V C
-?
II 71' s
i d
Ll.!,i.i!
l' Aj Ij 3l a]' e g
1,i,
'41) r/ g!!:!'a:
c-2 i
-nt, gi 39 w
l.
a Je e.eur-r-
--I
- l.]
[
s i,1
~]
- g. aq........
y
}g
,i...
s
(
\\ \\
.N ji j Y
C\\
)f
!{
k I
(
- p' 8 -
I 4
'j, I
]j g \\ " ~ *I W
i -!. s ill 1
3(
l I
- - L, tb
.i 5
E i-21 I
cor u I
4 p
e d als I
i s x
n J, e !b" 3d1 i
's
=
3G e-i
' di 5
Ct *~ %
~
e N
9
{
9
\\.
\\-
v r--- - W i
!!gy-l
\\
,,4.n w
, i 54d fo
.M i
of it.
- l}l k
0
+
i l
2::,
I-l~~
FIGURE I-l
1 GR00 TID REFERENCE PLAtlE
= x
.. i l..i
..ntl
. l.
.g i 3 g g...j...
i i
.t
..I i
yi;
,I g
y y
m r
H j
p^
/
1 WELD OFFSET viEldKlHK
'110 KINK,5 110 0FFSET s
Tr 0FFSET +
OFFSET +
KINK KillK it:1TIAL DISCONTittulTIES lli Itl5TALLED PIPE
~.
DISTANCE FROM READOtff POINT (PT) 20 40 6.0 8.0 100 1}0 140 ljo
- 10. C O
(-116. 0) l 1 252 K/ft A
E
-2.0 215 K/ft**/'
g-(.
)
-47.0)
Tl'1,o)
J (10.0) 17.0)
O u
5 (4.0) s 9
29.0
(-27.0) (-18.
9 06 (1 2 0)
-. -4. 0...,
4 7*
/
(-22,. 0)
V
~~
(51.0)
(1 *o}
i 295 K/L't.
M, 1 -6.0 m
s y
e Indicates Pipe Position at Survey Points Conservative Soil Capacity Estimates c:
Pipe Displacement Profile Uplift = 10 K/Pt H
-- Soil Sottlement Profile Bearing = 75 K/Pt
- Pipo bending stress in kai at maasurement point (typicakf
~
- Soil spring forces (typical) e y
LINEAR ELASTIC ANALYSIS RESULTS FOR UPPER BOUND S0IL PROPERTIES r
PRESSURE TR ANSDUCER y REFERENCE WATER COLUMN WATER LEVEL N q ULTHASONIC TRANSDUCER
,,. MQ[
C
/
CONDUCTOR \\
WATEN COLUMN W
SE CA
'*c #
S
!g
/ MONITOHINSTRtJMENT WQTER f
COLUWA ADJUST,
./
LEVEL GAUGE l 10 p- _ __
d MAG 14ET I
PHESSURE gg
[AXCAOLE THAIJSDUCER H
( 90 Sill 0 A la plpe )
g N
k
...u,.,A.,,...-
H H
H I
P MAGNET N2 20000
. J'o I,
-2
~8
% ?. 4ll 1000s srr.O o-o
~ g ;-
I
--n REFEllEtaCE CALIBRATION ULOCK REFERENCE CAllHRATION OLOCK U
FOR llEIGili FORTsuCMHLSS SCllEM ATIC-PIPE ELEVATION PROFILE MEASUHMENT SYSTEM
SrbP (Milt.pm)
SuDER BLOCK LNAX p/A.)
SL/ DER BLOCK ARM 2
{
(ninpa)
AEM.1 (l.EAD AQ4) 4 al'n La_:1 f
e; 8
y
]Illilll(lllllillatill a
c! :
g
-, /'
s W
M S7OP (MIM DM.)
L D/A. SCALE ~
A2Iku7~N SCALE D
b
=
SKETCH-621 007=DF,000NDNESS MEASUREMENT ~
/MSTRUMEN7^
YT I
SN E
T N
I OAO L
nem o p m-
_h ro O o = <-
T z
no grAD I
L O
SNO C
Y T
R N
A E
M E
L T
T z
e ES 2?e &
l Ll l
I l
l l
I I
I j
l l
--_8 N,
Y O
I f
l i
l I
I j I s l I
I a l
I,I R
I
~
~
A S
D'E S
N R -
~
O P
~
C M -
E O
~
S C "M
dem P 8 nO> h
@ 2 ?*g a memz o m
O 3 Cnm U 7 Qn2 oa"b2g O
E Cphuo r.3 e 2 q w ~
A* h a
m ous {,o $,.O y, Do8
$"n~o
- $nWfg C5 1 n u5 MyacM2 2 Mrg mHOCE yH "
uNmO 2O Oa i n
4 DISTANCE FROM READOLTP POINT (PT) 20 40 60 go 100 1p 14,0 Igo 100 i
0
.0L l
l 2T2'K/ft m
h*2.0 215 K/ft**/$
\\
This area detailed
/
(
1)
-47.0) on Figure III-5 (10.0) 17.0)
@ l'3.oy i
o
~
(29.0
(-27.0) (-18.
O Q.@
(1 2 0) 9 -4.0.u V
2g
/
R
% (-21.0)
(1,o) 215W4t /
g (51.0) s/
i.:
,V h-6.0 l
[
m l
{
i O
CohservativeSoilCapacityEstimates e Indicates Pipe Ar#/ddo at Survey Points Pipe Displacement Pacofile Uplift = 10 K/Ft
-- Soil Sottlement Profile Bearing = 75 K/Pt
- Pipa bending stress in kai at measurement point (typic f""
j
- Soil spring forces (typical)
LINEAR ELASTIC ANALYSIS RESULTS FOR UPPER B0UND SOIL PROPERTIES O
oa7fnnin alsh c/ den / #cdiW cncl depte. o/dend DeAIsilof Area F /?'
to '
1 zoaL/ealn o
@urc E-Y L...q'.'
z" a.rnu.
k
- m f Go Aei an,.4 us a o
c'clor ~ /0N m
-, H&
.__c____,_
J.i. /a.
m
,e = c ' t y 5 V
BM L
= 0aof + f 8G) 0 i
swoo p (
~
8 1
A.ra o</A A~
/doo.f si.
l
~
d.leyeu o f c<.,tue 38.Z-20 s
~
A' m (D. 30 *) r[h o f / ro N (/ n us)
//t ca es, cyc.
of.o f 6 a,c. e) n // 4 a atc.
s to s=
d.1 =
J. 62.
- 4 d.3 MD
'f F/GURE 22T-f*
l' l
INSERVICE TESTS - LEAKAGE i
TESTS l
e EACH INSPECTION PERIOD:
3,7,10,13, 17.., YEARS s
e NOMINAL SYSTEM OPERATING PRESSURE:
57 i
PSIG a
e ISOLATE BURIED PIPING e PRESSURIZE WITH TEST PUMP e MAINTAIN PRESSURE 4 HOURS e MEASURE FLOW RC E NTATIO 0/2/81 G-1868 06
HYDROSTATIC TESTS - LEAKAGE l
TESTS l
e EACH INSPECTION INTERVAL:
ONCE EACH 10 YEARS i
i a
e 1.10 x DESIGN PRESSURE:
115.5 PSIG 8
i i8 i
n e ISOLATE BURIED PIPING l
l 7
e PRESSURIZE WITH TEST PUMP l
e MAINTAIN PRESSURE 4 HOURS e MEASURE FLOW RC PHESENTATIO 0/2/81 G-1868 07 P
4 l
l j
LEAKAGE TEST ACCEPTANCE i
CRITERIA l
e SMALL ENOUGH TO DETECT PRESSURE BOUNDARY FAILURE
- s i
8 e LARGE ENOUGH TO ACCOMMODATE i
ANTICIPATED BOUNDARY VALVE LEAKAGE Y
~
e 0-5 GPM e RESULTS IN INSIGNIFICANT FLOW LOSS e TO BE REVIEWED FOLLOWING PRESERVICE TESTS NC ESENTATION O/2/81 G 1868-08
l FLOW VERIFICATION e ENSURE ABILITY OF BURIED PIPING TO l
MAINTAIN FLOWS REQUIRED FOR SAFETY j
FUNCTIONS l
l e ESTABLISH PUMP AND SYSTEM LINEUPS TO l
OBTAIN KNOWN CONFIGURATION THAT s
PROVIDE REQUIRED FLOWS i
E o UTILIZE INSTALLED INSTRUMENTATION TO a
VERIFY REQUIRED FLOW IN EACH BURIED LINE e ONCE PER YEAR e TO BE INCLUDED IN TECHNICAL SPECIFICATIONS NRC SE TATION O/2/81 G 1868-01
l t
l l
0 m.
~
\\
lM l
l Y1
~
t D, 3;,2J30532 NO. OF PAGES l#-
'l REASON-i O PAGE ILLEGIBLE.
O HARD COPY RLED AT:
PDR CF OTHER O BtlicR COP ( REQUESTED ON
/
/
DMAGE TOO LARGE TO FILM.
O HARD. COP (RLED AT:
OTHER O RLMED ON ADERTURE CARD NO. MMOS2' # /
i
- - - ~ ~ - - -
18 14 12 i
1 l
glo u
i
\\
.} 8
\\
z s
s Y
4 p~ /
e 2
10 20 30 40 50 60 Mrs Dimeter-A Fig. 8.5.
Belationship Betw'een Calculated Height of cover and Diameter of 1. Inch Steel Pipe The relationship was computed from Eg
?
8.3 and Eg 8.4a for, a defection of I per cent of the pipe diameter: lagfactor, IJ:
K,0.10: and soil wei ht. I10 lb/cnft. For t
Curve A, e = 30 psilin.:for Curve B, er =
700 psi.
i i
FIGURE III-ll
_ _ = _
VII. APPENDIXES
. Referenced JA.
Teledyne Engineering Scrvices letter, 3, 10 D.F.
Landers to W.J.
Cloutier of Consumers Power Company, November 11, 1981 B.
Southwest Research Institute, four reports
- 7,'8-on pipe profile measurements.at Midland C.
Procedures and Drawing for Diameter 25 Verification Pigging Procedure D.
Partial Copy of Response to Question 49 of 32 Nhd Requests Regarding Plant Fill E.
Response to Question 34 of NRC Requests.
33
-Regarding Plant Fill U
l 6
I a
l VII 12/10/81=
s
- p APPENDIX A 1
i 3
Teledyne Engineering Services Letter to W.-Cloutier ot Cohsumers Power Company, November 11, 1981:
i l
I-4 f
l l
i i
l l
l l
l 12/10/81 i
i t.
k
'#rELEDYUE ENGNEMRING, SERVICES 303 BEAR HILL R$AD WALTHAM, MAS $AONUSETTS 0225.
(617)890-3350 TWX(710)324 7508 November 11, 1981 5171-4 Mr. William Cloutier Consumers Power Co.
1945 West Parnall Road Jackson, MI 49201
Subject:
Technical Discussion - Buried Pipe at Midland Nuclear Station
Dear Mr. Cloutier:
Attached please find the subject technical discussion which essen-tially sumarizes my thoughts on the buried pipe at Midland.
If you have any questions, please do not hesitate to contact me.
Very truly yours, TELEDYNE ENGINEERING SERVICES 92 4 Donald F. Landers Senior Vice President Engineering Operations DFL/mac Encloture Y ' N R h
.L w[p;2'"'
cc:
J. C. Tsacoyeanes (TES) se, &
l NOV131981 U.TLUi] P.10 JECT StlASE21T l
I ENGINEERS AND METALLURGISTS
YM ENGNEERNG SENICES Project 5171 Discussion Regarding Buried Pipe at Midland Nuclear Station 1.0 Introduction At the Midland nuclear generating station it has been determined that some of the buried piping systems have been subjected to loading associated with settlement of the soil around them.
Measurements of the amount of settlement have been made using various techniques..These measurements have been used as input to a piping flexibility analysis and stress results obtained. The real technical questions are associated with:
1.
The capability of the system to perform its intended function, 2.
The validity of the calculated stress results using the measure-ment data as input, and 3.
Code requirements.
2.0 Functional Capability The important question to be answered is whether the piping system is capable to perform its intended function over the life of the plant. Soil settlement is a loading condition that occurs over a long period of time and is not cyclic in nature.
The only concern therefore is whether significant deformation has occurred to produce collapse of the pipe or to significantly reduce flow area.
The USNRC has provided guidance to the industry in this area in the past with Mechanical Engineering Branch Position Paper MEB-6.1 and cur-rently with Preliminary Standard Review Plan PSRP-3.9.3. Since PSRP-3.9.3 is significantly more definitive it will be used to draw some guidance from. The baseline criteria of acceptcbility under PSRP-3.9.3 is to limit stresses for specified load conditions or load combinations to Service Limit B allowables. The basis for this is that calculated elastic stresses are limited to values which demonstrate the theoretical limit load is not reached.
However, recognizing the restrictiveness of this approach, PSRP-3.9.3 provides alternatives which allow a significant increase in stress if it can be demonstrated analytically or experimentally (or combinations thereof) that discontinuity areas retain sufficient dimen-sional stability so as not to impair the component functional capability.
Retaining sufficient dimensional stability is, in f act, the only basic question to be answered and is directly related to assuring functional capcbility of the piping. Consumers Power has inspected the pipe geometry
i
~
TM ENGNEERING SERVICES Project 5171 to determine cross-sectional shape (ovality) which is directly related to stability.
These inspections indicate ovality readings of less than 2%
generally with maximum values of 3%.
These values are well within the tolerances of manufactured pipe and Code ovality allowables and, in fact, could have been present when the pipe was received at the site.
Paragraph NC-4223.2 of Section III allows 8% ovality in pipe after bending. Ovality that exceeds 8% must be justified by the design calcula-tions.
In the Class 2 Piping Design article of Section III the only concern related to ovality is the effect on pipe bends. Paragraph NC-3642 requires that, for pipe bends, the 8% ovality requirements of NC-4223.2 must be met. Since the measured ovalities are well within the Code allow-ables for fabricated pipe, functional capability is demonstrated using the techniques permitted by PSRP-3.9.3.
3.0 Calculcted Stress The calculated stresses that were based on deflection measurements are difficult to rely on because the measurements can include things other than soil settlement.
For example, allowable angular mismatches at weld joints that occurred during fabrication are magnified over a long iength of, pipe and can appear as " knees" along a straight line.
Assuming these
" knees" are due to soil settlement can result in significantly overesti-mating the stress levels.
Obviously, deflections of this type resulting from settlement would result in local deformations that would.be apparent during the examination work that was performed and, using the calculated stresses, would produce ovality well beyond 8%.
This, of course, is not the case a.d therefore the calculated stresses should not be relied on to determine acceptability.
The problem is further magnified by the presence of the soil around the pipe and how to consider this in the calculations.
In areas where calculations indicate large deformations (ovality) will occur the presence of compacted soil will have an effect.
As the pipe tries to deform (avalize) a significant pressure is developed between the pipe and the soil which counteracts the ovalization.
If compacted soil is not present throughout the entire system (which is the calculation assumption) then the results would be reasonable.
4.0 Code Requirements For the piping systems we are addressing here it is important to recognize that the entire buried pipe was subjected to soil settlement.
This is really a different situation than that addressed in current Section III criteria (NC-3611.2(f)) for non-repeated anchor movements.
Many of the reasons for this difference have been discussed above and demonstrate the important variations between non-repeated anchor motions (building settlement for a non-buried pipe) and general soil settlement.
~ !
~
TN ENGMERNG SERVICES
~ Project 5171 The major reason for this discussion is really related to the develop-ment of the Code criteria related to allowable stress for non-repeated stresses. The need for a criteria was raised because design agents were being asked by regulatory authorities to include the effect of relative building settlements in the piping analysis.
In the development of the Code criteria the majority working group on piping reaction was "that it was not a matter of concern" since it was a single non-repeated anchor point motion and was deflection controlled. However, the industry needed a criteria in order to accommodate regulatory comments and the allowable value of 3.0 S was determined to be appropriate.
c
APPENDIX B
~
Four Reports From Southwest Research Institute for Pipe Profile Measurements at Midland
- o Report No. 1, dated August 1981 o
Addendum to Report No.1, dated November 1981 o
Report No. 2, dated November 1981 o
Report No. 3, dated November 1981
- Attached as a separate volume i
12/01/81
.