ML20236F107
| ML20236F107 | |
| Person / Time | |
|---|---|
| Site: | Trojan File:Portland General Electric icon.png |
| Issue date: | 07/27/1987 |
| From: | Cockfield D PORTLAND GENERAL ELECTRIC CO. |
| To: | NRC OFFICE OF ADMINISTRATION & RESOURCES MANAGEMENT (ARM) |
| References | |
| TAC-65726, NUDOCS 8708030215 | |
| Download: ML20236F107 (18) | |
Text
- _ _ _ _ _ _
m ammum 8N WWY M
David W. Cockfield Vice President, Nuclear July 27, 1987 Trojan Nucicar Plant Docket 50-344 License NPF-1 U.S. Nuclear Regulatory Commission ATTN: Document Control Desk l
Washington DC 20555
Dear Sir:
Support Design Verification The Nuclear Regulatory Commission (NRC) letter of July 9, 1987 requested information pursuant to Part 50 of Title 10 of the Code of Federal Regu-lations, Section 50.54(f) (10 CFR 50.54(f)].
Portland General Electric Company (PGE) was requested to address three concerns regarding supports in safety-related systems and describe actions taken or to be taken to resolve deficiencies, including appropriat.e justification for operation if longer-term action is to be taken. PGE responded'to the July 9, 1987 NRC letter on July 10, 1987 and clarified that responso by letter, dated July 15, 1987. On July 16, 1987 PCE met with the NRC staff in Bethesda, Maryland, to discuss this subject. During the period July 21-23, the NRC staff I
reviewed the subject program in PGE's Architect-Engineer's (A-E) offices.
j This letter further clarifies the PGE response to the NRC lotter and provides additional information requestod during the July 21-23 NRC Staff review.
1 The throo NRC concerns identified in their July 9 letter and the PGE ros-l 2
ponses are as follows:
(1) Inadequate Supports. The NRC expressed a concern that supports were inadequate to meet design loads.
As described in the PGE letters of July 10 and July 15 and discussed in the mootings with the NRC, a total of 453 supports l
have been reviewod and evaluated, and only ten were found to be inadequato. These ten were the responsibility of the A-E Civil Engineering Group and are located in the Main. Steam System.
Furthermore, these ten wera part of the fourteen supports which had been designed late in the construction of the Trojan Nuclear Plant to account for the turbine trip dynamic loads.
8708030215 870727 DR ADOCK 05000344 0
g l
PDR I
121 S W Satrnor. Street Portland. Oregon 97204
Portland General ElechicCcarvaly
)
j U. S. Nuclear Regulatory Commission July 27, 1987 Page 2 j
(2) Original Design Loads. The NRC was concerned that the original design loads were derived inappropriately.
As a result of the Support Verification Program, it has been determined that the original design loads for the fourteen anfety-related main steam line supports were found to be based upon an j
assumed pressure-time relationship Which was based upon data from
)
another nuclear plant with a similar turbine.
Since plant
)
conditions were not identical, conservatism were applied to the
]
Trojan design values. Tests on the Trojan system in 1976 showed the design values and assumed pressure-time relationship to be quite conservative.
It was not unusual in 1975 to obtain design information in this manner. The design loads have now been reevaluated using the appropriate data for the Trojan Nuclear Plant.
(3) Documentation. The NRC was concerned that there were no original design calculations.
It is not entirely accurate to state that there were no original design calculations. Complete calculations did not exist for all of the fourteen safety-related main steam line supports designed late in 1975 by the Civil Group but partial calculations did exist.
It is important to recognize that during the time-frame when Trojan was being designed and constructed, the requirements for preparing detailed calculatiori were not as specific as at the present time. As a result, the number and detail of calculations performed for present-day plants is far more extensive than in the early 1970s.
It is for these reasons that PGE initiated the Support Design Verification Program on a short-term basis and has committ'ed to verify designs for other safety-related supports on a I
long-term basis. The long-term program will be initially scoped
)
and developed by September 1, 1987, and will be similar to the other programs PGE has underway to document the design basis of Trojan such as the Design Basis Document Program.
In regard to the additional information requested by the NRC identified as concerns, Attachment I discusses how PGE is assured that all Civil-designed supports will be included in the Support Verification Program. addresses why the rock anchors installed at Trojan are acceptable for continued operation. describes how the loads were reverified and determined to be satisfactory. Attachment 4 provides the basis for the adequacy of the seismic loads. Attachment 6 describes the manner in Which the welds were determined acceptable.
The verification of the final 15 anchors identified as pending in the July 10 letter has been completed. All anchor designs are acceptable. A revised Summary of Pipe Support Verification Action Plan is provided as.
i i
k i
l Portland General BectricCompany U. S. Nuclear Regulatory Commission July 27, 1987 Page 3 In conclusion, PGE would like to reemphasize that the Support Design Verification Program has identified the problem as unique to A-M Civil Group designed safety-related supports on the Main Steam System. The Support Design Verification Program has developed documentation for supports where none could be located and has provided assurcnce that Trojan is safe to operate. We trust this information will assist the NRC in arriving at the same conclusion.
Sincerely, l
Attachments c:
John B. Martin Regional Administrator, Region V i
U.S. Nucicar Regulatory Commission Mr. David Kish, Director State of Oregon Department of Energy Mr. R. C. Barr NRC Resident Inspector Trojan Nuclear plant Subscribed and sworn to before me this 27th day of July 1987.
OMyCommission w(,,
/,
G
-c Notary Public of Orekon CM A+[
Trojan Nuclear Plant Document Control Desk Docket 50-344
. July 27, 1987 License NPF-1 Page 1 of 2 l
ARCHITECT-ENGINEER CIVIL DISCIPLINE RESPONSIBILITY FOR PIPING SUPPORT DESICNS I
In the original design of the Trojan Nuclear Plant, pipe supports were generally designed by the pipe support vendor. Bechtel's Civil discipline became' involved on an exception basis, participating in the design of less than 10 percent of the pipe supports. The principal areas in which Bechtel designed pipe supports were as follows:
1.
For seismic supports inside containment added late in the design (the 265 snubbers and restraints).
2.
For supports which carried loads of relatively high magnitude or were characterized by relatively complex structural design Which.
were beyond the vendor's capability (main steam and non-safety-related structures in the Turbine Building).
3.
For pipe anchors inside containment Which, due to material lead times, required early design (the 33 pipe anchors).
In order to identify the Bechtel Civil-designed pipe supports, a search was performed of the following documentation:
1.
Civil Discipline calculations 2.
Civil Design Drawings 3.
Civil Field Change Request Drawings This search encompassed the original civil design activity on pipe supports at Trojan.
4 The following table gives a breakdown of tne number of pipe supports in each system which have been verified -with totals for each type of support.
As a result of a subsequent review of the civil design drawinFs and field change requests, 39 additional pipe supports were identified on. July 27, 1987 which may contain civil discipline design elements.
These designs
- were included on building structural drawings as notes and details. We are in the process of establishing design responsibility for these supports.
If these supports are found to be civil-designed, their design will be promptly verified. Based on the'results of the support
~
verifications completed thus far, there is reasonable confidence that these supports will be d~monstrated to be adequate.
s
Trojan Nuclear Plant Document Control Desk j
Docket 50-344 July 27, 1987 License NPF-1 Page 2 of 2 ARCHITECT-ENGINEER DESIGNED
- PIPE SUPPORTS FROM THE ORIGINAL DESIGN Number of:
(Verified / Total)
System Snubbers Restraints Hangers Anchors Total A.
Safety-Related Chemical and Volume 33/33 1/1 0
8/8 42/42 Control Safety Injection 27/27 6/6 0
4/4 37/37 Pressurizer Spray 20/20 0
0 3/3 23/23 Pressurizer Relief 26/26 0
0 0
26/26 Component Cooling 6/6 12/12 0
0 18/18 Water Reactor Coolant Loop 8/8 0
0 4/4 12/12 Drain Reactor Coolant 36/36 13/13 0
0 49/49 Instrumentation Residual Heat Removal 27/27 1/1 0
7/7 35/35 Reactor Coolant Pump 27/27 10/10 0
7/7 44/44 Seal Injection Steam Generator 12/12 0
0 0
12/12 Blowdown Main Steam 8/8 22/22 0
0 30/30 Subtotal 230/230 65/65 0
33/33 328/328 B.
Non-Safety-Related Moisture Separator 6/6 18/18 12/12 0
36/36 Reheater Relief Valve Discharge Turbine Stop Valves 3/3 0
0 0
3/3
)
l Main Steam Bypass 9/9 2/2 0
0 41/41 1
Subtotal 9/9 20/20 12/12 0
41/41 GRAND TOTAL (A+B) 239/239 85/85 12/12 33/33 369/360
- Known designs by the Civil Discipline.
1550P I
i i
Trojan Nucicar Plant Document Control Desk Docket 50-344 July 27, 1987 License NPF-1 page 1 of 7 ROCK BOLT CONCERNS During the NRC review of the Support Verification Program, several con-cerns were raised regarding the adequacy of rock bolts in support anchorages.
NRC CONCERN 7:
i In calculation J-307, sheet 15, a calculation is provided to check concrete for rock bolt pull out.
In area calculation credit seems to be taken for over lapping areas. Is this interpretation correct?
How does this calculation correspond to that presented on sheet 126 in calculation H-19?
Response
i The approximation used to evaluate the concrete area defined by the intersection of projected shear cones and the surface of the concrete does recognize the overlapping areas and does not take credit for
_l them. The concrete area from this approximation (see NRC Concern 15) is 282 in.2 The exact concrete area defined by.the intersecting circles is 337 in.2 This shows that the approximation.of the concrete area in Calculation J-307, Sheet 15 did not include the overlap and is conservative. The method used on Sheet 126 of Calculation H-19 is a simplified approach used for an initial j
evaluation to see if additional refinement is warranted.
In this particular case, the calculated embedmont requirement was 1.68 inches and B inches were provided. Therefore, the designer concluded no 4
additional refinement was warranted.
I 1
NRC CONCERN 15:
In calculation J-307 the calculation of concrete area shown on sheet 15 appears to be in error. The parameter D in the equation is
'3 treated as a diameter when in fact it is a radius equal to the' embedmont depth.
The calculation should be revised.
1 1
Response
tion,theshearconeareanowbecomes282in.gthesameapproxima-The Calculation J-307 has been revised. Usin (See response to
'NRC Concern 7.)
NRC CONCERN 10:
1.
Why was an 8-inch embed selected? Were embeds less than 8 inches ever used?
f
__..__._._._.m__m__.__.
__.-_____..___.m__.___
k l
Trojan Nuclear Plant Document Control Desk Docket 50-344 July 27, 1987 License NPF-1 Page 2 of 7 2.
In MS calculation J-301, pages 68 thru 70, we have the following comments:
a) Why was the l value added back into the calculation for
)
es embedment length. Since the 1978 Williams manual and the 1
1984 Bechtel design guide calculates shear cono subtracting the full "Y" dimension.
NOTE: By not adding the l value back into the es length, a reduction in pullout capacity would
]
result approximately equal to 17 kips.
l 3
b) Justify using 4 =.85 rather than.65 c) It appears that no bond calculations were performed between I
the bolt and the grout. This calculation would result in an j
embedment greater than 8 inches. Also the 78 Catalog and the j
Bechtel design standard have embedment length of 18 inches l
and 14 inches. Please justify this difference, j
3.
The original design strength calculations for the type IV generic support (part of 265 Bechtel designs) used a value of 22,200 i
I pounds for an allowable rather than the 21 kips value. Please determine what was the design allowable used for the original design.
4.
Design calculation J-301 for SS-80 and SS-83 Rockbolt edge i
distance is only 6 inches.
Is this adequate to enable full rated load on the rockbolt?
I S.
Pipe Anchor 16 used an allowable of.67 x 37 = 24.79 kips rather l
than 11 kips.
j i
Resronse to 10.1 In the early 1970's, embedment depth recommendations for rock anchors, such as are currently found in the manufacturer's catalogs (viz. Williams Cats-log No. 448) were not available.
The 8-inch embedment for the 1-inch diameter rock anchors was l
originally selected based on a rational interpretation of the applicable code stress criteria in the early 1970's.
As mentioned in FSAR Section 3.8.4.2.1, the Trojan Plant design criteria included American Concrete Institute (ACI), Standard No. 318-63.
Section 1707 of this ACI Code addresses shear as a measure of diagonal tension which represents the stress regime in an embedded anchor concrete i
cone type tensile failure mechanism. The expression for ultimate I
shear stress, v, to be applied to the projected concrete area e
l l
l l
Teojan Nuc1 car Plant Document Control Desk Docket 50-344 July 27, 1987 License NPF-1 Fage 3 of 7 (planar), representing the cone, for concrete pullout capacity determination is given by the expression:
vc=44 ff '
e in which 4 is the capacity reduction factor, which for diagonal tension, bond, and anchorage is specified in ACI Section 1504 to have a value of 0.85.
f ' is the specified concrete compressive strength.
e The effective projected concrete area, A gg, was. determined based on e
the pipe support base plate anchor bolt configuration with consideration of shear cone projected area overlap, as ' applicable.
During the 1987 pipe support reevaluation effort,.Bechtel's Civil Calculation J-301 was. performed to demonstrate that thelembedment depth of 8 inches for a 1-inch diameter rock bolt, as was used in the original design, will develop.the tensile stress capacity (33 kips) of the rock bolt (calculated as T = 4 Tbolt yield, with 4 = 0.90).
The metho6elogy to calculate pullout capacity of concrete is consistent with the embedment criteria specified in Section B.7 of ACI 349-85 which specifically addresses concrete expansion anchors.
These methods are considered to be applicabic to rock bolts, although.it should be noted that in addition to anchorage development through the expansion shell mechanism, the annular space between the rock bolt shaft and concrete drill hole is completely grouted.
This grouting provides-added assurance of rock bolt development.
The individual rock bolt tensile capacity calculated as described above, i.e.
T=Agg'44ff',
e
\\
e was further reduced to arrive'at a value for working stress design for normal plant operating conditions, thus allowab10
- f " f = 21 kips T
In the absence of tensile test data specific to a 1-inch diameter rock bolt embedded 8 inches, an indication of the lower bound rock bolt tensile capacity through concrete cone development is available by review of test data representing wedge anchor (maxi-bolt) capacities.
A wedge anchor engages the concrete thrcagh an expanded shell near the base of the anchor in a manner essentially the same as a rock bolt because of the presence of the grout.
Trojan Nucicar Plant Document Control Desk Docket 50-344 July 27, 1987 License EPF-1 f
Page 4 of 7 The wedge anchor test data considered representative as a lower bound j
are for 5/8-inch diameter wedge anchorc embedded 7-1/2 inches. The effective depth of engagement is the depth into the drill hole in the concrete to the base of the rock bolt expanded shell.
The effective engagement depth of the 1-inch diameter rock bolt averages about 7-3/8 inch, based on measurements. The expanded shell diameter of the 1-inch rock bolt is approximately twice that of the 5/8-inch wedge anchor Which should engage more effectivo concrete area.
Tensile test results for 5/8-inch diameter wedge anchors embedded 7-1/2 inches are as follows:
Test Series 1 Tension Tests: Pittsburgh Testing Laboratory, June 11, 1980.
Test Sampics: Drillco maxi-bolt wedge anchor, S/8-inch diameter with 7-1/2-inch embedment.
Concrete:
Design 28-day compressive strength of fe' = 3500 psi.
Test results:
Maximum Load sample (kips)
Failure 1
30.05 Anchor 2
29.86 Anchor 3
28.88 Anchor 4
29.39 Concrete 5
30.88 Concrete Test Series 2 Tension tests: Tennessee Valley Authority, CEB Report No. 80-64, October 24, 1980.
Test samples: Drillco maxi-bolt wedge anchor, S/8-inch diameter with 7-1/2-inch embedmont.
Concrete: Design 28-day compressive strength f ' = 4000 psi e
(interpreted from test report).
4-Trojan Nuclear Plant Document Control Desk Docket 50-344 July 27, 1987
. License NPF-1 Page 5 of'?
1 Test results*
Maximum Load
.j Sample
~(kips)
Failure j
9 1
30.45 Anchor 2
29.33-Anchor 3
29.83 Anchor j
4 30.45 Anchor-
]
Based on the above data, the average value of maximum tensile loads sustained by the anchors without concr9te failure (anchor failed) is:
Tb
= 29.84 kips (Average of 7 samples);
- ave, max.
The average value of maximum tensile loads for which the concrete.
+
-i cone failure mechanism was exhibited is:
T
= 30.14 kips (Average of 2 samples) 1 eave.
l l
max.
If this value is extrapolated in terms of concrete compressive design strength:
T'
= 30.14 5000 = 36.02 kips eave.
3500 max.
Comparing these to the basic allowable tension value'for rock anchors Which was used in the re-analysis,~Tallowable = 21.0 kips (see
.Bechtcl's Civil Calculation J-301), results in the-following ratios:
Tb
= 29.84 = 1.42 (lower bound)
Tallowable 21.00 T'
= 36.02 = 1.72 (interpreted to be representative e
Tallowable 21.00 low value.for concrete shear cone development).
The actual concrete strengths are historically _ higher than the specified design strengths and this will increase the margin. The.
above ratios.are considered valid indicators of concrete pullout strength.
l l
Trojan Nucicar Plant Document Control Desk Docket 50-344 July 27, 1987 License NPF-1 Page 6 of 7 j
l i
Thus, we have concluded that the 8-inch embedmont for 3-inch diameter rock bolt provides adequate safety margin. A rock bolt embedment depth of 8 inches was the minimum used in the anchorage design for pipe supports.
The basis of our conclusion is summarized below:
1.
The original criteria for rock anchor design allowable loads was not unreasonable and was based on a rational interpretation of applicable code guidance at the time.
I 2.
The criteria used for the reanalysis of existing rock bolt anchorages of pipe supports ensure that the design loads are adequately transferred into the concreto.
3.
Test data for anchors which have embedment Jc.pths essentially l
the same as the minimum for the rock bolts, and which develop
{
l anchorage capacities in a very similar manner as the rock bolts, provides reasonable assurance that the rock bolts will develop the relied upon ressistance with adequate capacity margins.
l Response to 10.2 1.
Response to Item 10.1 provides the rationale for determining the embedmont length which is concistent with the embedmont criteria of Section B.7 of ACI 349-85.
2.
Response to Item 10.1 provides the basis for using & = 0.85, rather than 0.65 in the original design.
3.
Bond calculations are not performed for the interface between the bolt and the grout since bond is not relied upon to transfer the l
tension load to the concrete.
The load is transferred to the l
concrete by engagement of the rock bolt expanded shell with the l
concrete and grout through bearing.
Concrete shear cone resistance governs the rock bolt capacity development, not bond between the grout and bolt.
4.
The justification for 8-inch embedmont is given in response to Item 10.1.
The 8-inch embedmont provides an adequate safety margin and is consistent with available design methodology during the early 1970's.
William's Rock Anchor '78 Catalog and current version of Bechtel design guide C2.34 did not exist in the early 1970's when the Trojan Plant was originally designed.
-)
Trojan Nucicar Plaat Document Control Desk Docket 50-344 July 27, 1987 Licence NPF-1 l
pago 7 of 7 l
I Response to 10.3 The original design allowable rock bolt tensile capacity for normal operating loads was based on 0.6 (yield strength) of the bolt.
This is q
consistent with the AISC Specification where the allowabic tension I
stress is 0.6 Fy.
Therefore, for the 1 inch diameter rock bolts allowable tension load was taken as 0.6 (37,000 pounds) = 22,200 pounds.
Response to 10.4, Edge dircances of 6 inches for 1 inch rock bolts for supports SS-80 and SS-83 '.tave been evaluated for the applied. loads, and have been dete:. mined to be adequate. Calculation J-301, Sheet 7 addresses the ef'.ect of this reduced edge distance.
i Response to 10.5 As stated in Calculation J-301, Sheet 68, the allowabic shear load for rock bolts is 2/3 (yield strength of bolt). This criteria was used in the original designs. Furthermore, the loads that were considered in the evaluation of pipe anchor No. 16 were " failure loads" of the pipe.
The allowable shear load of 11 kips for normal plant operating conditions is used for pipe supports which are not required to resist pipe failure loads.
SAB/djh 1550P.0787 i
l l
1 i
Trojan Nuclear plant Document Control Deck Docket 50-344 July 27, 1987 l
License NpF-1 i
page 1 of 2 l
LOAD DEFINITION CONCERNS During the NRC review of the Support Verification program, the following concerns were identified.
pGE responses follow each of the concerns.
1 NRC CONCERN 6:
In calculation J-307, sheet 84, the loads used to verify support SS-1259 and SS-1263 do not agree with the loads presented in Impell support load summary (RDC 82-055) for these supports. What is the reason for the discrepancy?
Response
The reason for the discrepancy is under investigation and will be determined as a pset of the post-startup program.
Calculation J-307 has been redone using the loads presented in the Impe11 load summary for RDC 82-055 with satisfactory results. The support was acceptable as-is, no modifications were necessary.
NRC CONCERN 12:
In the civil qualification calculations for the 265 supports, some supports were qualified to only the OBE seismic load icvel.
For these i
supports, the calculations must be revised to qualify the supports to the SSE seismic load level.
Response
There were 57 out of 265 supports that had been verified using the OBE i
seismic load instead of the SSE seismie load. All of these supports have been reanalyzed using the SSE seismic load, and all were found to be satisfactory.
{
1 NRC CONCERN 13:
I In the qualification by Civil of supports for the pressurizer Blowdown System, loads would be taken from the support drawings.
The loads on the support drawings correspond to the Rev. O pipe stress calculation not the Rev. 1 pipe stress calculations. Justification for the use of Rev. 0 loads or revised evaluations for Rev. 1 loads for these supports should be provided.
i H
J
1 Trojan Nuclear plant Document Control Deck Docket 50-344 July 27, 1987 License NPF-1 page 2 of 2
Response
For the pressurizer Blowdown System, analyses were re-performed using the Revision 1 loads in place of the Revision 0 loads, and all supports were found to be satisfactory.
NRC CONCERN 16:
In the civil evaluation of 265 supports, no verification of the correctness of support loads was made. Without this verification there can be no confidence that the supports have adequate safety j
margin. A verification of all support loads should be made.
Response
The Support Verification program was expanded to include a verifi-cation of the support loads for the 265 safety-related piping supports j
j designed by the A-E Civil Engineering Group. Because many of the support loads were originally provided by the Nuclear Steam Supply Systcm (NSSS) ven6or, this information was requested. The NSSS vendor has verified und provided loads for 42 of the supports to date. The A-E already had load information for 85 supports. Reviews have been perforned for these 127 supports for which verified loads have been provided with all supports being satisfactory.
Based upon re-verification of the 127 out of 265 supports and support loads (approximately 48 percent) with all supports being adequate as designed, pCE has a high degree of confidence that the supports are conservatively designed for the loads, and that the loads are accurate.
For that reason, and the fact that verification of the remainder of the loads is not likely to be cumpleted by the NS3S vendor for several days, it is proposed t' hat this remaining load verification action item be transferred to the long-range program to be completed after startup.
1 l
GAZ/pr 1555P t
J
Trojan Nuclear Plant Document Control Desk l
Docket 50-344 July 27, 1987
{
License NPF-1 Page 1 of 2 ADEOUACY OF SEISMIC LOADS The NRC reviewed the PCE response to IE Bulletin 79-14 and its application to the Support Verification Program.
Several of the IE Bulletin 79-14 load analyses were reviewed.
Two concerns were identified by the NRC.
The two concerns are as follows:
1 NRC CONCERN 17.b.1:
The methodology for combining two-dimensional loads used was the Square-Root of the Sum of the Squares (SRSS); this may not be adequate.
j Responso Section 3.7.3.6, Three Components of Earthquake Motion, and Subsection 3.7.3.6.3, Piping Systems, of the Trojan Final Safety Analysis Report (FSAR) specifies that for flexible piping systems the results of the horizontal analysis were combined with the results of the vertical analysis by means of the SRSS.
3 NRC CONCERN 17.b.2:
l Significant reductions were noted in the seismic loads for several supports without an accompanying explanation or assessment.
f
Response
1 Seventy-three IE Bulletin 79-14 calculations have been reviewed.
Only l
five have been found to have significant (>10 percent) load reductions on numerous restraints. The seventy-three calculations include piping in the following systems:
Chemical Volume and Control System (CVCS)
Component Cooling Water System (CCWS)
Containment Vent Monitoring
- Diesel Generator Fuel Oil System
- Spent Fuel Pool Cooling System (SFPCS)
Main Steam System Miscellaneous Small Pipe (<2 inches in diameter)*
Reactor Coolant Pump Seal Injection System Residual Heat Removal System (RHRS)*
Safety Injection System (SIS)*
Service Water System (SWS)*
i
^
3 Trojan Nuclear Plaat Document Control Desk i
Docket 50-344 July 27, 1987 l
License NPF-1 Page 2 of 2
- l The systems identified with an asterisk (*) are those in which the five calculations were found'to have significant load reductions. The l
load reductions for each calculation were evaluated and determined to be justifiabic and reasonable.
Reasons for the load reductions are as
-l follows.
j l
1.
Grouted penetrations have been included in the piping analysis which has the not effect of adding more restraints to the original analysis.
2.
Actual As-Built geometry was included in the analysis instead of the proposed design geometry.
l 3.
The reanalysis used a response spectra that more closely represented the spectra for the center of mass of the piping system than the original conservative response. spectra.
1 a
l
)
GAZ/djh 1555P.0787 J
l l
l t
i L______________._____________________________._.___.
j
l eor c
a a
n rco a
f s
f o
c p
f s
e s
i i oop s
i r
i t
mt t u i
t u
t u
a s
t a
t a
l nsd a
s c
s o
yd n3 s
u s
d aa s
ry s
e o
f s
t t r t
R dl so r
r so r
e t
o o
t o
zwry h
p ec p
yeoc c
p va p
l npa n
u if u
a pu a
s f s s
nd uq i
aese l
l lt l
es d
l l
l a l
R u7 a A
A A s A
y l
se t td N
rae A
roun L
epqg 0
0 d
P b pei 10 0
0 0
0 1
e mud s t
N USA e a
O N
D l
I f t e
T oo r
C N
A y
t N
r e
O yo
)
)
f I
bt l
l a
T c
a) a)
s A
rd a iy iy C
eer tl tl n
I b nt rl rl o
F mgn au au n
I uio pf pf R
N sc
( (
( (
e E
eb 0
r V
D u 04 60 4
a S
33 0
0 0
25 1
5 T
s R
t d O
l r n P
i o a P
y - vg
)
)
p U
bti n l
l p3 S
cCi a
a u
rd e r
i i
s s E
eet re t
t e
P b niee r
r gi I
mgh en a
a nr P
uicni p
p i o N srig
(
(
ng F
eA gn 5
9 i e O
D nE 0
3 6
1 6
6 at E
3 3
2 4
2 3
ma Y
eC R
R A
l h
M a
t M
td
.o U
f oe db S
osT rn 5
9 6
e t / ea 6
2 0
i n rrddl 4
3 2
1 2
6 f i eoeiP 6
3
/
4
/
/
i b pws
/
/
5
/
3 re mpenn 4
3 6
1 6
5 er uuioi 6
3 2
4 7
4 va NS vCe ss R
t t d s r r ee o o t
t r c
pp n
au i
pp n
e ei l t m
uu n
i p
pa s
ec a
s s o
ae i
ir r
R u n
i M p P
Pt e
- r y
dd t
i s
b yt D
ee p
dP d
d ed m t S t t i
e e
eR ne e
h aa r
t m t
t aM f t t
l l c
ae a
ad ar i
ee s
l t l
l nsl S o W
rr e
es e
eaea
- p D
R ys R
R gr np s
yy St
- s
- rau ou t
t t t
y r
yr yert NS r
L ee r
t mo t o t b oc os A
f f o
eap eh ebh u ee pd T
aa p
f ep f c f ucr vp pa O
s s l7 p
at u an annt ii uo T
a u
7 2
/3 B7 A4 1
2 3
4 5
S1
4 Trojan Nucicar Plant Document Control Desk Docket 50-344 July 27, 1987 License NPF-1 Page 1 of 1 l
WELDS NRC CONCERN 19:
The support calculation for SR-305 did not contain any calculations relating to welding.
Also the civil drawing did not contain any welding symbols.
Are the welds adequate and what size were used in j
fabrication?
l
Response
The important welds on SR-305 are full penetration welds (butt weld with backing bars or backup fillets) and therefore, have the strength of the structural member. The structural member was evaluated in Calculation J-75 and found to be adequate, indicating the weld is adequate. The full penetration welds for this particular support are not called out explicitly on the support drawing. The welds are covered by notes and reference to other drawings. The welds for SR-305 have been field verified to be full penetration welds.
The same method of identifying full penetration welds is used for other similar supports. Welds that are not full penetration welds are generally shown on the support drawing.
l l
1555P.0787