ML20148F186
| ML20148F186 | |
| Person / Time | |
|---|---|
| Site: | Comanche Peak  |
| Issue date: | 07/15/1987 |
| From: | TEXAS UTILITIES ELECTRIC CO. (TU ELECTRIC) |
| To: | |
| Shared Package | |
| ML17303B208 | List: |
| References | |
| RTR-NUREG-0797, RTR-NUREG-797 DAP-EP-068, DAP-EP-68, NUDOCS 8803280086 | |
| Download: ML20148F186 (25) | |
Text
- -.
-- [ 1, 7 UM i
Y29
,e COMANCHE PEAK DESIGN ADEQUACY PROGRAM ENGINEERING EVALUATION COVER SHEET 4
TITLE Differential Snubber Lockuo of EVALUATION ID.NO.DAP-E P
,_qd3_
Dual Snubber Assemblies NO. OF SHTS.
19 DAP DISCIPLINE:
Pioina and Suecorts DAP ACTION PLAN NO.
IX SUPPORTING ENGINEEP,ING EVAll'ATIONS:
CONT.!D.NO. DAP-E-TITLE:
DAP-E-
_ TITLE:
2 DAP-E-TITLE:
DAP-E-TITLE:
SUPERSEDES ENGINEERING EVALJATION NO. DAP-E PURPOSE: Review and evaluate NUREG/CR-2175 ("Snubber Sensitivity Study")
relative to identified concerns about differential snubber lockup.
Provide supplemental information to the DSAP IX Engineering Evaluation for tht: external source issue on Axial / Rotational Restraints.
SCOPE:
Review of applicable portions of NUREG/CR-2175, external source docu-ment NRCT-19, and CPPP-7.
10NTENTS(SEESECTION4.0,DAP-8) 1.0 ABSTRACT 2.0 REVIEW PROCESS BASES FOR SAMPLE SELECTION (Not Applicable) 3.0 REFERENCE DOCUMENTAlION 4.0 ACCEPTANCE CRITERIA 5.0 EVALUATION
6.0 CONCLUSION
S 7.0 ATTACHMENTS (Att A/4 pgs, Att B/l pg, Att C/l pg)
REVIEWED APPROVED REV.NO.
REVISION ARIGINATOR DATE 3Y DATE BY DATE O
[ObEL 1/A/87 E L - ik!b7 G/~5 7)Wh'1 v-vvv ii <
i y Page 1 of 19 8803200086 % 3h 43 PDR ADOCK ppg E
1.0 ABSTRACT Pages 26 to 30 of external source document NRCT-19 (Reference 3.1), indicate an area of concern related to the effect of differential snubber lockup on dual snubber assemblies. Specific reference is made therein to a NUREG on snubber sensitivity which addresses this subject (Reference 3.2). This Engi-neering Evaluation presents a review and evaluation of the NUREG relative to the identified concern and relative to how SWEC treats this area procedurally within CPPP-7 (Reference 3.3).
2.0 REVIEW PROCESS 2.1 Applicable external source documents (References 3.1 and 3.9) were identified for review as input to this evaluation (refer to Section 5.1).
2.2 The subject NUREG (Reference 3.2) was identified for review along with a supporting NUREG (Reference 3.11) (refer to Section 5.2).
2.3 Applicable portions of SWEC procedure CPPP-7 (Reference 3.3) were identified for review (refer to Section 5.3).
2.4 Acceptance criteria for the evaluation was developed (refer to Section 4.0) based on ASME Code (Reference 3.4) requirements.
2.5 The evaluation was performed as described in Section 5.4.
2.6 Conclusions reached as a result of the evaluation are presented in Section 6.0.
1 Page 2 of 19 l
l l
O 3.0 REFERENCE DOCUMENTATION 3,1 Transcript of meeting between NRC staff and CYGNA, at Bethesda, Maryland, July 3, 1984 (Document ID No. NRCT-19).
3.2 "Snubber Sensitivity Study," NUREG/CR-2175, July, 1981.
i 3.3 "Design Criteria for Pipe Stress and Pipe Supports," SWEC procedure No. CPPP-7, Rev. 2.
3.4 ASME Boiler and Pressure Vessel Code,Section III,1974 Edition, through Winter, 1974 Addenda.
3.5 Transcript of NRC meeting, at Arlington, Texas, March 23, 1985 (Document 10 No. NRCT-17).
3.6 CPSES Final Safety Analysis Report up through Amendment 55.
3.7 "3.9.3 ASME Code Class 1, 2, and 3 Components, Component Supports, and Core Support Structures," USNRC Standard Review Plan, Rev. 1, July, 1981.
3.8 "Axial / Rotational Restraints," Comanche Peak Design Adequacy Program Engineering Evaluation No. DAP E P-012.
3.9 "CASE's Partial Answer to Applicants' Statement of Material Facts as to which there is no Genuine Issue Regarding Allegations Concerning Consideration of Force Distribution in Axial Restraints," Affidavit of Case Witness Mark Walsh before the Atomic Safety and Licensing Board, August 27, 1984 (Document ID No. CASE-15).
3.10 "Justification of Design Load for Struts / Snubbers and Lugs Used in Conjunction with Riser Clamp," SWEC Calculation No. GENX-042, Rev. O.
Page 3 of 19 i
7-
, ~ ~. -, -
t i
s 3.11 "Single vs Dual Snubber Installations," NUREG/CR-2032, Nay, 1981.
3.12 "Gaps," Comanche Peak Design Adequacy Program Engineering Evaluation No.
DAP-E- P-013.
4.0 ACCEPTANCE CRITERIA There are specific criteria upon which to base an evaluation of the Snubber Sensitivity Study ("SSS") (Reference 3.2). Applicable portions of the FSAR (3.98.3.4 of Reference 3.6) refer to Subsection NF of Section III of the ASME Boiler and Pressure Vessel Code ("Code") (Reference 3.4).Although limited (as described below), some guidance in this area is provided in the Code.
Paragraph (b) of NF-ll21 of the Code implies (by omission) that the rules of Subsection NF include consideration of tolerances.
Even though specific tolerances are recommended in the Code (see below for Appendix K), it's not readily apparent from other Code paragraphs how design consideration of tolerances is included. Applicable Code paragraphs are summarized below.
NF-3271.2(a)
The supports shall be designed to provide the required sup-porting effort and to permit the designed piping movement.
NF-3274(b)
Snubbers shall be carefully applied to ensure that they will perform their intended function without placing unacceptable loads on the piping system er other components.
NF-4223 Tolerances for component standard supports should be as recommended in Nonmandatory Appendix K, unless otherwise i
specified in the Design Specifications.
I K-1310(c)
Recommended maximum tolerances on component standard sup-ports are 1/16" for linear dimensions up through 12"; and 1/8" for linear dimensions above 12".
Page 4 of 19 I
r As seen, the first two paragraphs quoted are more in the category of good design practices than design criteria. Although they are "Code require-ments," they do not provide a strong basis for acceptance of the areas reviewed herein. Although the Appendix K tolerances are specific, they are so large relative to snubber application, that they also do not provide a strong basis for acceptance. This is especially true for dual snubber installations where it is generally attempted (in the industry) to minimize tolerance differences by procuring such snubbers in matched sets (this was discussed on pages 113-115 of external source document NRCT-17 -- Reference 3.5).
Further guidance in this area is provided in Standard Review Plan ("SRp")
3.9.3 (Reference 3.7).
The following is from page 3.9.3-6 of the SRP:
"The snubber end fitting clearance and lost motions must be minimized and should be considered when calculating snubber reaction loads and stress which are based on a linear analysis of the system or component. This is especially important in multiple snubber applications where mismatch of end fitting clearance has a greater effect on the load sharing of these snubbers than does the mismatch of activation level or release rate.
Equal load sharing of multiple snubber supports should not be assumed if mismatch in end fitting clearance exists."
Although this appears to be somewhat of a reflection of industry practice, the statement apparently was based in large measure on the results of the SSS (see comments by Terao at the top of page 28 of Reference 3.1) which is the subject (in part) of this Engineering Evaluation. As a consequence, it is not considered appropriate to use this as acceptance criteria herein (even though the SRP represents a particular aspect of NRC licensing requirements).
It is also noted that the SRP criteria is not much more specific than the Code in this area, that is NRC interpretation of the recommendations of the SSS is fairly general (see 5.1 and 5.2 below).
Page 5 of 19
O Based on the preceding, the acceptance criteria for this evaluation will be to insure that the general requirements of the Code are satisfied and to insure that there is reasonable consideration of the practicalities of dual snubber installation. More objective criteria are not considered possible with the limited amount of data available in this area (some of which is being reviewed herein).
5.0 EVAWATION The first three sections here (5.1, 5.2, and 5.3) present discussions of those portions of the documents applicable to this evaluation. Specifically, Section 5.1 presents background information within external source documents NRCT-19 (Reference 3.1) and CASE 15 (Reference 3.9) in which reference is made to the SSS (Reference 3.2).
Section 5.2 presents a summary of those portions of the SSS (including a supporting NUREG) which pertain to the area under discussion within NRCT-19 and CASE 15. Section 5.3 summarizes the applicable portions of SWEC procedure CPPP-7 (Reference 3.3).
Section 5.4 presents first an evaluation of the SSS and then an evaluation of how the SWEC procedures address the areas of concern raised within NRCT-19 and CASE-15.
5.1 External Source Documents (Note that the area of unaccounted for rotational restraint of dual struts / snubbers was the primary concern of discussion in the identified portion of NRCT-19.
It is also the primary subject of Reference 3.8 which this Engineering Evaluation is intended to supplement.)
Within the more general discussion of axial / rotational restraints in NRCT-19 (Reference 3.1), specific reference is made on page 27 to the SSS (Reference 3.2),
it is indicated in NRCT-19 that a snubber mismatch Page 6 of 19
(of end clearances) of approximately 0.02", results in a load distri-bution of 40%/60%; and an approximate mismatch of 0.05" results in a load distribution of 30%/70%. All these figures are based on the results of testing associated with the SSS.
As a consequence of these results, it is concluded within NRCT-19 that a 50%/50% load split for dual snubbers is probably not a reasonable assumption to make. The more general conclusion therein, is that dual snubber installations can give rise to load differences caused not only by the rotational restraint capability of dual snubbers, but also caused by snubber end tolerance mismatch (such load differences being with the design loads which would result from modeling such an installation as a single snubber within a piping stress analysis and then assuming a 50%/50% load split between the two snubbers).
Additional discussion regarding the SSS and NRCT 19, is presented in CASE-15 (Reference 3.9).
Reference is made to the portions of NRCT-19 discussed a'oove and to recommendations of the SSS' summarized in 5.2.2 below (see pages 3 to 5 of CASE 15).
These External Source Document summarizations of the SSS and the results of the SSS itself, have potentially significant effect on the current CPSES piping and supports requalification efforts. This is because some of the recommendations within the SSS would add potential conservatisms well beyond what is currently being done in the industry relative to assumed load distribution in dual strut / snubber installations.
This evaluation was undertaken because of this significance to CPSES and the industry, and also because of the limited discussion in external source documents about the basis or validity of the SSS.
5.2 Snubber Sensitivity Study The intent of the SSS (Reference 3.2) is "to develop simplified design and analysis rules for snubber support ^d systems which will bound snub-ber response within acceptable limits." The SSS investigates several l
l l
Page 7 of 19 l
parameters which potentially affect the performance of hydraulic and mechanical snubbers. Of particular interest here is how the SSS addresses the effects of unubber clearance / mismatch for dual snubber installations ("snubber clearance" is the total free pin to-pin movement when a snubber is locked; "snubber mismatch" is the difference between the snubber clearances of dual snubbers).
5.2.1 Snubber Mismatch As a result of the SSS, the following specific recommendations are made regarding snubber mismatch (paragraph B.2.3.1 of Appendix B to Reference 3.2):
(a) Equal load sharing of multiple snubber supports should not be expected if the end fitting clearance exceeds 0.01".
(b)
If the mismatch clearance differential exceeds 0.01" but is less than 0.04", peak loads shall be assumed twice the uniform load sharing value.
(c) Mismatch of end fitting clearance shall not exceed 0.04".
5.2.2 Snubber Clearance The effects of snubber clearances are also investigated within the SSS. Separate but related recommendations (to those in 5.2.1) are made regarding snubber clearances when linear analysis is used to perform seismic analysis of a piping system (paragraph B.2.2.2 of Appendix B to Reference 3.2):
(a) Snubber reaction loads and stresses shall be increased by 100% for clearances greater than 0.00" but less than 0.02".
f Page 8 of 19
(b) Snubber reaction loads and stresses shall be increased by a factor of 4 for clearances greater or equal to 0.02" but 1ess than 0.05".
(c) Detailed nonlinear analysis is required for systems with 0.05" or greater clearance.
5.2.3 Of primary interest here is the basis for the recommendations (in 5.2.1) regarding snubber mismatch.
Paragraph 3.6 of Reference 3.2 indicates that justification is based on the results of a single test program (which is presented in Appendix D to Reference 3.2).
These results are summarized in paragraph 3.6 as follows:
(a) End fitting clearance has a greater effect on load sharing of dual snubber supports than mismatch of activation level or release rate.
(b) Equal load sharing was observed for zero end fitting clearance.
(c) For end fitting clearance differentials of 0.05", 307,/70Y.
load sharing was observed.
(d) Effects of end fitting clearance on support reactions were extremely variable.
(e)
Different trends of support reactions vs. end fitting clearance differential were observed for various support types, configurations, and input. Support types tested were hydraulic and mechanical snubbers, and a rigid strut; support configurations tested were either single or dual configurations; and input was either sinusoidal or seismic-l based.
l l
\\
Page 9 of 19 i
(f)
In all cases for zero clearances, the total reaction for a pair of supports was less than that for a single support of the same type and input loading.
(g) For nonzero clearances, the total reaction for a pair of supports may be greater or less than that for a single sup-port of the same type and input loading.
5.2.4 Appendix D to Reference 3.2, is a summary of the testing which was done.
The full report from which Appendix D was extracted is NUREG/CR 2032 (Reference 3.11). The following paragraphs are based on review of NUREG/CR 2032 since it providere,a more com-plete description of the testing which supports the summary statements and recommendations in Reference 3.2.
Several tests were performed on single and dual strut / snubber installations attached to a short cantilever pipe mounted on a shaker table (although not a standard pipe size, the size was approximately the same as a 6" sch 80 pipe). Two basic input loadings were applied on a displacement basis via the table:
a sine sweep loading with frequencies varying between 1 Hz and 33 Hz, and a seismic loading based on the 1940 El Centro earth-quake. The tested hardware included rigid struts, hydraulic stubbers, and mechanical snubbers. The primary test variables were the end fitting clearances although activation levels and release rates were also tested for the hydraulic snubbers.
Table 5.2.1 (of this evaluation) summarizes the load results from the various tests. For purposes of this evaluation, the results for the sine sweep tests are presented first, followed by the results for the seismic input tests. This should provide a good basis for comparison since all sine sweep tests had nearly I
identical input and all seismic tests had nearly identical input (refer to Section 7.1.5 of Reference 3.11).
It is not, however, reasonable to compare load magnitude results from sine sweep and l
seismic tests.
Page 10 of 19
- ~
- - - - - - - - = - - - -
--- -- --- ~ --- - - - ~~ ~ ~ ~
Certain problem areas which were identified during the testing sheuld be noted here (section numbers are from Reference 3.11):
(a) One of the so called matched pair of hydraulic snubbers was damaged following sine sweep tests.
This snubber was replaced by a reconditioned snubber with'similar charac-teristics for the seismic tests.
It was not indicated what the damage was but it was identified as having occurred after the sine sweep tests but prior to the seismic tests (Section 7.1.7, paragraph 1).
(b) Ovalization of one of the hules in the load path was noted during the sine sweep tests involving dual hydraulic snub-l bers. This was considered to have affected the results for the matched end fitting clearance tests (which occurred following the tests for mismatched end fitting clearances).
It was not indicated as to how much ovalization might have occurred nor if it might have also affected the mismatched end fitting clearance results.
The affected bracket was replaced for later tests (refer to Section 7.3.5).
(c) Three of the five dual rigid strut seismic tests have questionable results. That is, one strut in each of the three tests carried over 90% of the total load and the total load is much less than the total load for the other two tests.
The report suggests this might have been caused by recording instrumentation malfunction (refer to Section 7.4.1).
(d) Ovalization of one of the bracket holes in the load path was noted after completion of the seismic tests involving dual mechanical snubbers, it was not identified as to when this might have occurred nor how much ovalization was present (refer to Section 7.4.4).
Page 11 of 19 1
. -.. -.., - _ ~ -, _ - - _, - -, _ -
(e) One area of possible discrepancy noted during review for this evaluation involves test 17 for hydraulic snubbers.
Section 7.3.5 refers to Figure 7.3.9 for loads from dual snubber sine sweep tests.
From Figure 6.1, it is seen that test 17 involves dual snubbers each with 0.00" "pin clearance." Figure 7.3.9 clearly indicates that the two snubber loads associated with 0.00" and fitting clearance are 2000 lbs and about 780 lbs. However, Table 7.5.2, which also includes loads from test 17, identifies the loads as 1200 lbs and 800 lbs. Similarly, for seismic tests, Figure 7.4.6 indicates that the snubber loads are 1150 lbs and 380 lbs, while Table 7.5.2 list thase loads as 720 lbs and l
480 lbs. The discussion 1, bout Table 7.5.2 is given in Section 7.5 and reference is made there to Section 7.3 regarding the results of test 17.
There is no mention of these discrepancies in the report nor are the associated "raw" test data included.
5.3 Applicable SWEC Methods The following portions of CPPP 7 (Reference 3.3) apply to this evalua.
tien.
Each is taken from Attachment 4 8 to CPPP-7 (the parenthetical i
section/ paragraph numbers are from Attachment 4 8).
(a)
Where riser clamps / lugs are the load transfer mechanism (between pipe and support), and the dual struts / snubbers are modeled.as a single restraint, each strut / snubber is designed for 75% of the total load. There is no discussion in CPPP-7 regarding the intended purpose of this increase although it is addressed in GENX-042 (Reference 3.10) which provides justification for some of the loadings of Attachment 4 8.
The implication within GENX-042 is that the 75% is based on geometric considerations assuming only two of four lugs are acting.
(Section 2.1.2.1, paragraph 3) i Page 12 of 19 i
I
s (b)
When one of two dual struts / snubbers is removed for stiffness considerations on riser clamp / lug installations and the remaining strut / snubber is modeled as an eccentric restraint (thus pro-viding potential rotational restraint), the strut / snubber is designed for the corresponding load from the computer analysis (Section 2.1.2 and Section 2.1.2.2, paragraph 2)
(c)
Where trunnions are the load transfer mechanism (between pipe and support), and the dual struts / snubbers are modeled as dual restraints (per CpPP-7, always done when trunnions are present),
each strut is designed for the corresponding load from the com-puter analysis; if the supports are dual snubbers, each snubber is designed for the corresponding load from the computer analysis multiplied by 1.2.
The reason given for the increased snubber load is to account for possible differential lockup.
(Section 2.2.1 paragraphs 3 and 4) 5.4 Review / Evaluation
(
5.4.1 Of primary interest here is Reference 3.11 (which is the basis for Appendix D to the SSS) since all the recomendations related to dual strut / snubber mismatch are based on this NUREG.
For purposes of review, the data from Table 5.2-1 (of this evalua-tion) has been plotted in Figures 5.4-1 and 5.4-2 for the sine sweep and seismic tests respectively.
Each of the plot points associated with dual strut / snubber tests shows an adjacent two-digit fractional number. This number represents the fraction of the total load carried by one of the two supports represented by the point.
For example, ".73* means that one of the supports carried 73% of the total load represented by the plot point and the other support carried 27% (100% minus 73%) of the total load.
From review of Figures 5.4-1 and 5.4-2, it appears that the reference within NRCT-19 (Reference 3.1) was pertaining to the l
Page 13 of 19 l
l
i sine sweep tests for dual mechanical snubbers and (possibly) dual struts. However, it is potentially misleading to view these results as representative of all the tests. From an applications standpoint, the seismic tests introduced much more of the expected realistic variability than the sine sweep tests. Review of Figure 5.4-2 shows the following:
(a) Dual hydraulic snubbers show an approximate 60%/40% load split when both gaps are 0.00", 0.02", and 0.05".
They also show a 60%-40% split when one gap is 0.00" and the other is 0.05".
They show a 70%-30% split when one gap is 0.00" and the other is 0.02".
(b) Dual struts show a 51%/49% load split when both gaps are 0.02" and when one gap is 0.00" and the other is 0.02" (other plot points for dual struts apparently represent bad data).
[
(c) Dual mechanical snubbers show load splits beginning with l
74%-25% for equal 0.00" gaps and then decreasing to 66% 34%
1 for equal 0.05" gaps, or to 59%-41% for unequal gaps of 0.00" and 0.05".
However, these results may be questionable because of ovalization of a hole in one of the brackets (the load path associated with the ovalized hole was not iden-tified relative to which snubber had its gap varied).
i' l
(d) There are significant variations between comparable single and dual strut / snubber test results.
For instance, the t
single strut load with a 0.02" gap had a load of 2000 lbs while the dual strut cases had total loads of 2750 lbs and 3060 lbs. Of interest is that for the sine sweep test, opposite results occurred. That is, the single strut load with a 0.02" gap was 4000 lbs while the dual strut total loads were 1160 lbs and 1190 lbs.
In paragraph B.2 of l
Page 14 of 19
\\
- - - - -. - - - - - - - - - - - - ~ -
- - - - - - ~ ~ ~ -~ ~
~ ~ ~ ~
t 4 Section 7.3.2, it is suggested that the sine sweep load differences may be the result of stiffness differences, but the seismic test results don't seem to support this.
In addition to the struts, it's observed that the total load for dual mechanical snubbers are generally higher than for a single mechanical snubber while the total load for dual hydraulic snubbers is less than for a single hydraulic snubber. There is limited discussion of these load dif-ferences even though they would appear to be much more significant relative to piping analysis than end clearance variations.
(e) There are significant variations in total loads as end clearances are increased.
In some cases the total load goes up and in others' it goes down.
(f)
In several cases, the load variation is greater for enti clearance changes than for end clearance difference changes.
Because of some of the data problems which occurred with the seismic tests, this is best seen in the sine sweep test
~
results of Figure 5.4 1.
For example, the load for a single mechanical snubber increases from 1000 lbs to 1300 lbs (30%
increase) when the gap is increased from 0.00" to 0.02".
However, the load for the most heavily loaded of dual mechanical snubbers increases from 400 lbs to 440 lbs (10%
increase) when the gap of.one of the snubbers is increased from 0.00" to 0.02" (the other snubber's gap remains at 0.00").
For the case of both snubbers increasing from 0.00" gaps to 0.02" gaps, the most heavily loaded support has 400 lbs (no change). One implication is that end fitting clearance by itself has a much greater effect on support load variations then do the differences between the end fit-ting clearances of dual struts / snubbers (however, quanti-fication based on these results is extremely difficult).
4 Page 15 of 19
This also has implications for all supports with clearances since most such clearances are much greater than those used in,these tests (refer to the DAP Evaluation on the Gaps Issue -- Reference 3.12).
On the bases of the foregoing observations, it is not at all clear how the recommendations of the SSS were arrived at (the recommendations are summarized in Sections 5.2.1 and 5.2.2 of thisevaluation). The type of quantification attempted in the'se recommendations requires far more consistent and reliable data than is presented within NUREG/CR-2032 (noting that the results of some of the tests most pertinent to practical situations -
e.g., dual mechanical snubbers subjected to seismic input -- were questionable because of hole ovalization). There was also very little attempt within NUREG/CR-2032 to identify / quantify the various parameters which appear to have contributed to the sig-nificant load variations resulting from the tests (that is, the effects due to end clearances and end clearance differences could not be easily or conclusively isolated). These parameters include the relative stiffnesses of the supports, the rela-tionship between the frequency of the support and the input loadings, and the ovalization of bracket holes during testing.
There is also a very real possibility for ending up with snubbers much larger than required if the recommendations of NUREG/CR 2175 were implemented.
(One specific potential problem with this can be found within other recommendations of NUREG/CR-2175.
Para-graph B.2.1.3.4 therein states:
"Since the snubber drag loads I
are expressed as a percentage of the snubber rated load capacity, snubbers having a rated capacity much lar.ger than required should not be used.")
Increased snubber sizes could also have related detrimental effects such as adverse dynamic response resulting from the increased system mass, and increased installation dif.
ficulties due to reduced clearancas around snubber hardware.
\\
Page 16 of 19 1
--,v.
n--,
r,_.
m_,
Based on the preceding, it's obvious that end fitting clearance variations and end fitting clearances can have effects on strut /
snubber loads. What's not obvious from the results of the test-ing discussed above, is what these effects might be.
It is reasontble to conclude that dual strut / snubber installations do need to account for potential load variations due to end clearance differences. However, the manner in which this is done cannot be reasonably based on NUREG/CR-2032. This conclusion seems to be consistent with what is presented in SRP 3.9.3 (Reference 3.7) where only general caution is advised without reference to the specific recomendations of NUREG/CR-2175 (see discussion in Section 4.0 of this evaluation). As indicated in NRCT-19 (Reference 3.1), NUREG/CR-2175 provided the bases for the SRP 3.9.3 statement regarding snubber end fitting clearance.
5.4.2 There is no direct acknowledgement within CPPP-7 (Reference 3.3) that strut / snubber end clearances should be accounted for in the design of dual strut / snubber installations. However, for dual r,
strut / snubber installations using riser clamps, the single reaction load from the stress analysis is split 75% 25%. As indicated in paragraph (a) of Section 5.3 of this evaluation, this was chosen in part to address lug load distribution. How-ever, from review of the associated SWEC generic calculation (Reference 3.10), it appears that the 75% value is very con-servative relative to what is required to account for the load distribution. Additionally, this split is consistent with cur-rent industry practice for such installations (for example, refer to the discussion on pages 112 114 of NRCT-17 (Reference 3.5) in which Mr. Doyle of CASE, describes the load splits he has typi-cally applied in dual strut / snubber installations on other 1
projects).
For single strut / snubber or dual strut supports where the associated eccentricity is included in the stress analysis model I
Page 17 of 19
I o
(refer to paragraphs (b) and (c) of Section 5.3 of this eval-untion), the loads from the stress analysis are used directly in the support design process. There is no load increase to account
~
for end clearance variations. However, this is consistent with current industry practice and is considered acceptable.
For dual snubber supports having the eccentricity modeled, the design load for each snubber is the load from the stress analysis multiplied by 1.2 to account for differential snubber lock up. This would be equivalent to a 60% 40% split if the loads were equal from the stress analysis. While less than the 75% 25% split from above, the lower value is considered reasonable and acceptable on the basis of the more accurate representation of loads which result from the eccentric modeling applied for such supports.
6.0 CONCLUSION
6.1 The data and recommendations of NUREG/CR-2175 cannot be applied specifically to dual strut / snubber support installations because there is inadequate data upon which to base the rer.ommendations. However, there is sufficient data to establish that strut / snubber end clearances can have an effect on predicted loads. Until such time as more defini-tive testing is performed, it is concluded that current industry practice as specifically evidenced by the sections of CPPP 7 discussed herein, provides the most reasonable basis for addressing strut / snubber end clearances and satisfying the Code-based acceptance. criteria of Section 4.0.
9 7.0 ATTACHMENTS
'l A.
Table 5.2-1 (4 pages)
B.
Figure 5.4-1 (1 page)
C.
Figure 5.4-2 (1 page)
Page 18 of 18
--. 6--w
7.0 ATTACHMENTS A.
Table 5.2-1 (4 pages) 8.
Figure 5.4 1 (1 page)
C.
Figure 5.4 2 (1 page)
(
f Page 19 of 19 g
-r
b o
Table 5.2-1 t r.c. t e.t o e >
.ert Lo. d i k..
Gt.;;
Enct Lead t.o.
- 1. p ce La ti p t.
i..p-C]roce iibi
- f. t 4 irom NURFG'Cf-1 2
eine i
I..
.i. 0. "
0.,
Fto ~.
.1 j
i
~
cine 2
F-
.. 02 a009 Flo 7.
4 s
- nc i
~
0.uS 40?O rio
~.T..
S canc
~.
P O. V..
5l. 0 F2p
"..4 P
0.00 570
'.Ese Note 1i e
sinc 2
E 0.'i0 490 rag 7.:.4 F.
0.02 700 (sec. Note 1) s)ne
<..> v
_70 c
sc e....
F
.05 7c0 (see Ncts 1.
F U. v..-
,1 v Fig.,.
c ine a
P U.02 550 (reo flote 1) c sine 2
P v.v5 510 Fig 7.~
4 e
f.
0.65 550 s see Note: 14 4
f
- 2 cine J
O.00 4100 f i g 7. ~. 1 12 E:ne 1
H 0.02
!? Xi Fig
~7. :. 1 J.
c3ne
..i v. i.,s r
_.. G f i c.
<.-.1
. s.i Ii nine N
v.no 1000 F1g
. :. 1 J-c 3 n c-3 H
0. '. 2 12.'c 0 F1 ; ".. 1 l
1
- 2 tino N
0.r5 1400 F2 g ; '.'
1,.
3 r:c.
r
... on i
.cv
.u t o.
N u.ou
. :" 5 s ete rat t..
1
- r. :
-Su ila r
0, w-M:.
s.sc t.c t e; 1 2.,c n
iaa
- .. 7 (l
I.
s C.
te sie
- s. e c e " ', t g 1) 1
- i
-2
'i
's.
~ '
+ia
~
~
it
.0.
- 0.*
' i c+ U t..
+ ; n t.
I-0.'".
I
[-
[..
(2cs f.f..t o ;
c.>
Table 5.2-1 (continvoo) a: s e c-Nc.ts.- 6)
Tec.
L C o ci No.
EUDP Enti Lo:id f ro.
Tcrc Evot T.pe C 1 r ric e
( I t.. >
f<c t f roni NUPEG/CF c ris H
0.00" I'GO F:g 7.0.9 H
.'.00 2000 (see hote li
- 3 r:nc 1.
v.Ou SCO Fi g 7. 3. ';
H O.02
- Bi.O csee I;ote 1i 39 c2nd 2
H 8..00 S2O F i o i'.. 9 H
0.05 1620 (see Note 1)
..O nine 2
h 0.92 700 Fig '/ :.9 H
0.v:
2000 (see Note 1) t:ne 2
H O.05
' I '.i F-i o 7. 0. o H
0.US 2430 (tee Notre 1) 17 s:nc 2
H ii. 's :
800 Table 7.5.2 i '
il O.00 1200 (see Notes I cno 2) 1 22 5:ne 2
H 0.00 500 Table 7.5.2 H
- v. 00 480 (see Notes 1 and 5) 25 Eine 2
H 0.00 4 5'.'
Tatsl e 7. 5. 2 H
0.UO 540 (see Note 1 and ': )
- . 7 s i r.c 2
h v.00 480 Table 7.5.2 H
O. t.".*
4 e.0 (see Notc-1 and 29 cine 2
H 0.00 700 l a b l e 7. '*..'
H O.v0 520 tece Note 1 ond 1) t c-: c 1
s e. O..
2250 Fia /.4.1 l
c c-2 :
1 n
- 0. :.:
,;c v,
p ;.a
,,.;, 3 l
4 t '.- :
J R
v.v5 1 2 J.".'
FAO 7.4.1 n.j i -
r v.<..
- a I.
[1, Qi '
l e H_la,.
I p gg ( 'g ( p g j gggg,*
t.:.1 -
v..
n t'*'.
- Fic,
'.4.2
<> 92 1 '! 5a. -
.seo Note 1) 2
- n..>
'*Q V1a
~.4.2 8
O. O'..
J -:' y a c (.e Nr.' r3 1 ie m.
'. o Table 5.2-1 (continued)
(see Note b')
Test Loao No.
Sucp End Load No.
Type Supt Type C1rnce (1b)
Ref from NUREG/CR-2.C2 8
ses s 2
R-0.02" 1500 Fig 7.4.2 R
O.02 1570 -(see Note 1) 9 seiss 2
R O.05 90 Fio 7.4.2 R
O.05 1450 (see Notot 1 ond 5)
't:
11 sets 1
H 0.00 1920 Fig 7.4.1 12 seis 1
H 0.02 1800 Fig 7.4.1 13 seis 1
H 0.05 1750 Fig 7.4.1 4
11 seis 1
M O.00 1570 Fig 7.4.1 12 se)s 1
M O.02 1220 Fig 7.4.1 13 ceis 1
M O.05 1000 Fig 7.4.1 17 seis 2
M 0.00 380 Fig 7.4.6
~
e M
O.00 1150 (see Note 1) 18 seis 2
M O.00 700 Fig 7.4.6 M
O.02 1910 (see Note 1}
19 seis 2
M O.00 1220 Fig 7.4.6 M
O.05 1720 (see Note 1) 20 seis 2
M O.02 800 Fio 7.4.6 M
O.02 1760 (see Note 1) 21 seis 2
M O.05 990 Fig 7.4.6 M
O.05 1900 (see Note 1) 17 seis 2
H U.00 320 Fig 7.4.0 H
0.00 480 tsee Note 13 18 sets 2
H 0.00 250 Flo 7.4.0 H
0.02 600 (sce Note 1) 1 (/
c ol s 2
H 0.09 380 Fig 7.4.3 H
U.05 2l. U (cee Note 1)
M.
501:
2 H
0.O.
230 Fig 7.4.3 H
O.02
.:.50
\\sec Note 11 i
22 sets 2
H O.05 130 Flo 7.4.
H 0.05 200 (cee flot e 1.)
y
- gm F9-v-
'w-pwa g y
,qT
- m r+3 ew-+-
-&--q?-+v T--+Ww---mae-
-F'--
9
-yw'4 wi
-P-w-
g
- O
's
(
Table 5.2-1 (continued)
(see Note 6)
Test Load No.
Supp End Load No.
Type Suot Type.
Clrnce (lb)
Ref from NUREG/CR-200:
17 seis 2
H C.00" 480 Table 7.5.2 H
0.00 720 (see Notes 1 and 2) 20 seis 2
H' O.OO 700 Table 7.5.2 H
O.00 680 (see Notes 1 and
~.1 25 seit 2
H 0.00 760 Table 7.5.2 H
0.00 760 (see Notes 1 and 21 27 sets 2
H 0.00 680 Table 7.5.2 H
0.00 720 (see Notes 1 and 4) 29 seis 2
H 0.00 240 Table 7.5.2 H
0.00 720 (see Notes 1 and 4)
Notes:
1.
In all cases where two struts / snubbers are indiceter.
it could not be determined wnich load corresponded with which support.
2.
Foi sine wave and seismic tests results were presented twice for test 17 using hydraulac snutbers.:
the results did not match and there was no di scuss:cn iri the report about the differences.
0.
For Tests 22 and 25 (sine and seismic). the two hydraulic snubbers had different activation leiels.
4 For Tests 27 and 29 (tine and seismic). the two hydraulic snubbers had different release ratec.
5.
The dets for some of.the dual stiut seismic tes r were apparently bad (ref er to par agraph 7.4.1).
6.
Tests 1.
10, 15. 16. 22, 24, 26, and OS were char act et-i n ct i on t ests on1 v.
No r &f c.ul t 5. are reported.
Test 14 was e series of "resarsec s..uptcr tett4 for which retults werc apparentIv as e ;0cc t ud.
Ne renults aie reported.
Fefer to Figure s.1 cc NUFEG/CP-2002.
e.o' Y
,7y-
.-_m y
.. ~.
e s
')
..)
J w.
5000 _
I SilNI 4000 _
B
^
m
.C v
O
.n M
p ',_
g_-
~ __
-( 2 XfD,10lAl LOAD,SArl CAPS g 3000_
a
.n
~
i % _ _ _' %
y
_ l WTD j
T
.02 2 HYD.IOilt LOAD. VAR. CAPS C1.c.
t a 2000 _
cn a
<c f-A O
'51 d 1 EH F-.
__._._.4-_
5
__ n 2 s1 W is.101n toad.V 2. CAPS g f _ _ _ _ _ _ _ ___.~_
__ U_____J,,,,,,,,,,yt,,,
1000 _
~
y ----- -_-gc - _ _ _ _ _ _ _._._
_._.69 2 E CH.icin t0AD.Va. CAPS
~
.n
.H
,,,2 m.10iu t010.:n CiP:
g l
l l
l l
1 e se s.el e.or e.e3 s.e4 e.e5
-....gs s
n-
,~
j
,\\)
l w..
J.
l l
e
- ::C 5.4-2 S.
s a.
.ll 3000 k
.s
.p,fOttCN.181&LLOAD.SA
/ N t s cu.tolat Loao.vaa. Gaps j
s%
,,s e CAPS
.ll
\\
j
,p
- A'7g'p
/
s
_ 2596.
/
N
/
\\
y,
/
f
't g.
.a
/ //
\\
=
/
\\\\
< e...-
// [-
w\\'N
/
o
\\
J
,/
\\
{
j/
\\
l NTO r,
f g
2 81RWit ieiAL LOAD. VAR. GAPS A
/
A 1590_
//
\\
.,Dr s,Ruis.ietat tono. sam care D
75 in
//
d g
i SiRUT
- lese.
,, c,
.1l
-(
~,
~ ~. _
see.
% S'
' b ! MTD,10lAL LOAD. VAR. CAPS
- m~ e.....< <.....A.
p u.
..i
...a
.. 2
..4
...s END FITTING CLEARANCE (in) l 4
,,.,.a,4r
--w.-,--
-w--,,
,,,-,,.,,e,-,--.-
,,,-~,n-w,,
-. -