ML17221A607
| ML17221A607 | |
| Person / Time | |
|---|---|
| Site: | Saint Lucie  |
| Issue date: | 02/05/1988 |
| From: | Woody C FLORIDA POWER & LIGHT CO. |
| To: | NRC OFFICE OF ADMINISTRATION & RESOURCES MANAGEMENT (ARM) |
| References | |
| RTR-NUREG-0737, RTR-NUREG-737, TASK-2.D.1, TASK-TM L-88-59, NUDOCS 8802090101 | |
| Download: ML17221A607 (42) | |
Text
V REGULATO NFORMATION DISTRIBUTION TEM
( RIDS )
I ACCESSION NBR: 8802090101 DOC. DATE: 88/02/05 NOTARIZED:
NO FACIL: 50-335 St.j Luc ie Plant>
Unit 1 i Florida Poeer 5 Light Co.
50-389 St.
Lucie Plant> Unit 2> Florida Poeer 5 Light Co.
AUTH. NAME AUTHOR AFFILIATION NOODYi C. O.
Florida Pacer Rc Light Co.
RECIP. NAME RECIPIENT AFFILIATION Document Control Branch (Document Control Desk)
DOCKET 0 05000335 05000389
SUBJECT:
Responds to 870610 request for addi info re NUREG-0737I Item II. D. fi "Performance Testing of. Relief 5 Safety Valves. " Combination of-hydraulic 5 other loads mill result in lower piping stresses 5 support loads.
DISTRIBUTION CODE:
- 046D COPIES RECEIVED: LTR J ENCL J SIZE: 9 @
TITLE:
OR Submittal:
TMI Action Plan Rgmt NUREG-0737 5 NUREQ-0660 NOTES:
REC IP lENT ID CODE/NAME PD2-2 LA TOUR IQNYI E INTERNAL: AEOD/DOA ARM/DAF/LFMB NRR/DEST/ADS NRR/DREP/EPB ILRB REG F 01
/DRPS DIR COPIES LTTR ENCL 1
0 1
1 1
0 0
1 1
f 1
RECIPIENT ID CODE/NAME PD2-2 PD AEOD/DSP/TPAB NRR/DEST/ADE NRR/DEST/MEB NRR/DREP/RPB OGC/HDS2 RES/DE/EIB COPIES LTTR ENCL 5
5 1
1 1
0 1
1 1
f 0
f EXTERNAL:
LPDR NSIC 1
1 1
NRC PDR TOTAL NUMBER OF COPIES REQUIRED:
LTTR 23 ENCL 18
0 I
P. O. BOX 0, JUNO BEACH, FL 33408.0420 FEBRUARY.
5 1988 L-88-59 U. S. Nuclear Regulatory Commission Attn:
Document Control Desk Washington, D.
C.
20555 Gentlemen:
Re:
St. Lucie Units 1 and 2
Docket Nos.
50-335 and 50-389 TMI Action Item II.D.1 Re est for Additional Information By letter dated June 10, 1987 (E.
G.
Tourigny to C.
0..
Woody),
the NRC identified additional information the staff required to continue its review of TMI Item II.D.1 of NUREG
-0737, "Performance Testing of Relief and Safety Valves."
By letter L-87-339 dated August 14, 1987, Florida Power
& Light Company (FPL) provided a completion date of February 5,
- 1988, for response to St.
Lucie Unit 1 questions 2.C.,
2.D., 3.,
4.,B.
and 4.C.,
and St.
Lucie Unit 2
questions 1.C.
and 2.B.3.
The purpose of this letter is to submit the response to these questions.
The remaining St.
Lucie Unit 1
and Unit 2
questions were addressed in FPL letter L-87-448, dated November 6, 1987.
Should there be questions regarding this
- subject, please contact us.
Very truly yours, C
. Woody Executive Vice President COW/MSD/gp Attachment cc:
Dr. J.
Nelson Grace, Regional Administrator, Region II, USNRC Senior Resident Inspector, USNRC, St. Lucie Plant MSD/002.RAI 8802090101 880205 PDR ADOCK 05000335 P
LPDR an FPL Group company
Page 1 of 7 NUREG - 0737 ACTION ITEH II.D.l ADDITIONAL QUESTIONS ON ST.
LUCIE UNIT 1 SUBHITTAL uestion No.
2 Insufficient detail was received on the key parameters used in the RELAP5/HOD1 thermal-hydraulic analyses.
Additional information is needed on the following items; C.
D.
2C Since the PORVs are used for low temperature overpressure protection (LTOP),
they will be required to pass subcooled and saturated water during these transients.
Was the PORV piping analyzed for liquid discharge?
If not, justify not analyzing this transient condition or perform the analysis and provide the results for our review.
If the transient was not analyzed because it was considered to be boundqd by steam discharge, compare the calculated steam discharge loads with estimated water discharge loads to verify the steam load is bounding.
Provide the comparison for our review.
What PORV flow rate was used in the analysis?
~Res ense The PORV piping was not analyzed originally for liquid discharge.
However, for the purpose of comparing the hydraulic forces during the low temperature overpressure protection (LTOP) transients with the design steam discharge
- loads, a simplified RELAP5/HOD1 analysis consisting of the PORV section of piping alone was performed.
The LTOP conditions considered in the study are listed as follows:
LTOP CASE SETPOINT PRESSURE (p>>e) 530 530 350 WATER TEHP 303 270 190 PRESSURE RATE (psi/sec) 55 55 55 HAXIHUH PRESSURIZER PRESSURE (psia) 1040 660 460 These conditions are based on the applicable Technical Specifications, overpressure mitigation system setpoints, operating procedures and the latest LTOP Report prepared by Combustion Engineering.
Since these LTOP conditions were determined for only one PORV in operation and there is no symmetry in the PORV piping geometry, each transient was analyzed twice (once for each PORV).
An isometric drawing showing the geometric configuration of the system is attached. as Figure 1.
The comparison of the resulting peak hydraulic forces against those of the SRV/PORV steam case has been summarized in Table 1.
Although some forces on the legs upstream of the PORV are slightly higher during the LTOP transients, the overall hydraulic forces on the entire PORV piping are less during the LTOP transient than those during the steam transient.
The LTOP transients also occur at lower temperatures and pressures.
Therefore, the resulting piping stresses due to pressure and thermal expansion will be lower.
- Also, the support loads will be lower due to less thermal expansion.
This combination of hydraulic
Page 2 of 7 NUREG - 0737 ACTION ITEM II.D.I ADDITIONALQUESTIONS ON ST.
LUCIE UNIT I SUBMITTAL loads and other loads will therefore result in lower piping stresses and support loads than the steam discharge loads.
The steam discharge load is determined to still be the bounding transient.
2D The PORV initial steam flow rate was 153,000 lbs/hr and calculated by the RELAP5/MODl computer code for the August, 1982 analysis was 190,800 lb/hr at the end of the transient.
NUREG - 0737 ACTION ITEN II.D.l ADDITIONAL QUESTIONS ON ST.
LUCIE UNIT I SUBMITTAL Page 3 of 7 uestion No.
3 The prediction of new loads and comparisons of these loads with the loads used in the design analysis is a valid engineering practice provided that the frequency content of the load sets are basically the same and that the piping and supports are identical in both analyses.
Provide new and design load curves for several highly loaded locations to verify the frequency content of these loads is basically the same.
~Res oese Based on a review of the input data it has been determined that it is not possible to readily compare the frequency content of the forcing function used in the design analysis with the RELAP5/MODI analysis.
The design analysis calculated the loads using a
modal analysis technique.
The RELAP5/M001 analysis calculated the loads at each piping segment.
Since the modal points and segments do not coincide, no direct correlation between the two, at any specific points, is possible.
A condensed report of the analysis performed by Ebasco in August, 1982 is attached as Appendix A to this study.
The report compared the results of the
. RELAP5/HODl analysis'oading with the design analysis loading.
For that comparison, the transient force time histories for the RELAP5/H001 analysis loading was reported at the same location as chosen in the design analysis loading by EDS.
From the curves in the two analyses, it is observed that in the majority of the cases:
i)
The RELAP5/MODl analysis forces typically dissipate faster than the design analysis forces.
ii)
The peak value of the forces calculated in the RELAP5/HODl analysis are lower than those calculated in the design analysis.
It is therefore concluded that the piping system and its supports are adequate for the newly calculated transient forces.
Further comparison would require a
very detailed analysis of each calculation to produce comparable data.
Page 4 of 7 NUREG - 0737 ACTION ITEbl II.D.1 ADDIONAL QUESTIONS ON ST.
LUCIE UNIT I SUBMITTAL uestion No.
4 Provide the following information on the design analyses used to determine pipe stresses and support loads are within allowables:
B.
C.
What dynamic structural piping computer program was used in the design analysis?
Provide verification of this program for use in SV/PORV blowdown piping analyses.
Use of EPRI or similar test data and verification of predicted loads/stresses with measured values is required.
The dynamic piping model used affects the accuracy of the predicted stresses/loads.
Provide the following information on the model used:
1.
Haximum and minimum lumped mass spacing used.
2.
Calculational time step used.
3.
Haximum or cutoff frequency used in the analysis.
If this is less than 100 Hz, verification should be provided to show this is conservative or accurate in predicting peak loads/stresses.
The lumped mass spacing and calculational time step should be consistent with the 100Hz cutoff frequency also or verification provided to show accuracy or conservatism.
4.
What damping factor was used in the analysis?
Typically 1% for upset and 2% for emergency conditions are the maximums allowed; if greater than these values were used, justification is requested.
~Res ense The Impell (EDS Nuclear in 1974) computer program EDSGAP (version 7/20/74) was used in the steam hammer structural analysis portion of the St.
Lucie Unit 1
Pressurizer Safety and Relief System 'evaluation.
EDSGAP has been verified for a comprehensive set of example problems in accordance with Impell Corporation quality assurance procedures.
4C1 4C2 4C3 The EDSGAP structural model used for the blowdown evaluation was compared to the PISOLID (seismic analysis) geometry and found to be essentially identical.
The PISOLID computer results for the pressurizer relief system analysis (revision 4) show flexural maximum mass point spacings of 73.5 inches (for 10 inch Schedule 20 pipe),
72 inches (4 inch Schedule 40),
and 68.25 inches (6 inch Schedule 40).
The results also show a
minimum spacing of 2.25 inches (2
1/2 inch Schedule 160).
In addition to the fundamental structural response modes of the piping, high frequency longitudinal response modes were determined.
A sufficient number of modes were included in the analysis to ensure that both the fundamental structural and the relatively high frequency longitudinal responses of the piping would be accurately represented.
The forcing functions extracted from the hydrodynamic (steam hammer) analysis had a time period of 0.00141 seconds.
EDSGAP Direct Integration Time History Analyses do not have a specified maximum cutoff frequency, but use the integration time
NUREG - 0737 ACTION ITEM II-D.I ADDITIONAL QUESTIONS ON ST.LUCIE UNIT I SUBMITTAL Page 5 of 7 step and the forcing function time step to control the structural response.
Using a forcing function time step of.00141
- seconds, and assuming a minimum of five points to define the wave
- form, the forcing function would transmit a
maximum frequency content of over l00 Hz, to the EDSGAP structural model.
In addition, comparison of the EDSGAP and PISOLID structural models showed that the EDSGAP model was at least as detailed as the PISOLID model.
Results of the PISOLID computer analysis showed that modes up to 100 Hz (mode number
- 75) were available for subsequent dynamic analysis.
Therefore the EDSGAP structural model had the detail required to respond to dynamic loadings up to 100 Hz.
4C4 One-half percent (I/2%) structural damping was used in the safety relief system blowdown structural evaluation for St. Lucie Unit l.
NUREG - 0737 ACTION ITEM II.Del ADDITIONAL QUESTIONS ON ST.
LUCIE UNIT 2 SUBMITTAL Page 6 of 7 uestion No. I Insufficient detail was received on the key parameters used in the RELAP5/NOOl thermal-hydraulic analyses.
Additional information is needed on the following items:
C.
Since the PORVs are used for low temperature overpressure protection (LTOP),
they will be required to pass subcooled and saturated water during these transients.
Was the PORV piping analyzed for liquid discharge?
If not, justify not analyzing this transient condition or perform the analysis and provide the results for our review.
If the transient was not analyzed because it was considered to be bounded by steam discharge, compare the calculated steam discharge loads with estimated water discharge loads to verify the steam load is bounding.
Provide the comparison for our review.
~Res ense The PORV piping was not analyzed originally for liquid discharge.
However, for the comparison of the hydraulic forces during the low temperature overpressure protection (LTOP) transients with the design steam discharge
- loads, a simplified RELAP5/HODl analysis consisting of the PORV piping section alone was performed.
The designation of the piping segments used in the model are shown in figure 2.
The LTOP conditions used in the study are:
PORV Setpoint (PSIA)
Water Temperature (of)
Pressure Ramp Rate (PSI/SEC)
Max Pressurizer Pressure (PSIA) 470.0 192.0 80.0 535.0 These conditions are based on the applicable Technical Specifications, overpressure mitigation system setpoints,.operating procedures and the latest LTOP Report prepared by Combustion Engineering.
Since these LTOP conditions were determined for only one PORV in operation, and there is no symmetry in the PORV piping geometry, the same transient was analyzed twice (once for each PORV).
The attached Table 2 summarizes the peak hydraulic forces during the LTOP transients and the PORV/SRV steam transients.
Although some upstream piping forces are higher during LTOP, the overall hydraulic peak forces are smaller during the LTOP transient than the steam transient.
The LTOP transients also occur at lower temperatures and pressures.
Therefore, the resulting piping
- stresses, due to pressure and thermal expansion will be lower.
Also the support loads will be lower due to less thermal expansion.
This combination of hydraulic forces and other loads will therefore result in lower piping stresses and support loads than the steam discharge loads.
The steam discharge load is determined to still be the bounding transient.
NUREG - 0737 ACTION ITEM II.DeI ADDITIONALQUESTIONS ON ST.
LUCIE UNIT 2 SUBMITTAL Page 7 of 7 uestion No.
2 Provide the following information on the PIPESTRESS 2010 analyses used to determine pipe stresses and support loads are within allowables:
B.
The dynamic piping model used affects the accuracy of the predicted stresses and loads.
Provide the following information on the model used:
(The figures provided were not adequate or legible).
3.
Maximum or cutoff frequency used in the analysis.
If this is less than 100 Hz verification should be provided to show this is conservative or sufficiently accurate in predicting peak stresses and loads.
The lumped mass spacing and calculational time step should be consistent with the 100Hz cutoff frequency also or verification provided to show accuracy or conservatism.
~Res esse 283 The cutoff frequency used in the analysis was 72 Hz.
The contribution of loads from frequencies beyond the cutoff frequency was taken into account by utilizing the left out force option in the computer program.
Therefore, the stresses in the pipe and the support loads calculated, will not be significantly different from those calculated using a cutoff frequency of 100 Hz.
From a practical standpoint the contribution from higher modes cannot be realistically calculated using a linear elastic analysis model because of the following physical characteristics of an actual piping system:
a)
The non linearities in the snubber
- supports, such as dead
- band, friction, different spring constants in compression and tension, etc.
b)
The gaps in certain categories of rigid restraints.
It is recognized that linear elastic analyses are conservative and these non-linearities will tend to r educe the response of the piping system.
We therefore consider that piping analysis performed is adequate.
The mass point spacing used in the analysis for various sizes of pipes is adequate for analysis up to 100Hz.
The calculational time step used is consistent with the cutoff frequency of 72 Hz.
It is also adequate for calculations up to 100Hz.
SP/NRC-0737.NRC.0003
3
.rC 44 I
4'.
oX 4
yl
~
I~4
,~ $
45)
, Ql~
ssC isa sl>>
40" 0
~45.'
~,I JrJ ~
JtJ.
~l
.oS I
I~ 4.+
l..
r
,~.ir J
~
75",
'4'g7:.
gg r 4s.
5 gc rp O~
4r>. ~ i+~>
H b
~
O'
~~.
o+
ge r
f4 44c '. r '+w>
~
ri
~ %
g)
~4 Pr go~+ OQ Jr
'.s 1
~$
ry~
~
~
I
~i.+i,. q J C I Qlo 1
~
s 1
'Op.
ff 74
~
~
CAl.PLoT QH ~. Slum S6saee'C Pie Best Available Copy I
470 10
)J
~
l re.
l,.~,,
qt;..s:sS Ci. i 4 ~
~.
ayl+
rl g1%
jr)a
~
4J 40 '
~Q 04
~cl
~A p<"
O
~ ~
fq$
~.sss-fa
~'r-
~4~
')4 4
~ r.t
+p jCc.
r 24 B
t~
gb lJ e
~7
~
'ac.: j" gV sr IO fl
+
p~
g sF g0 rblrS7aS 2S isa C
71 47 ~ ~ L7
.31$
.Z37 4.425 6')$
')00 65$
675 tg
~
tar
~. Uelm reit I
~~IZED ggl,l~ I)~
PICQRI so)bC-bZZ IO.7jb
.ZSO
Vl0 Q3 Q3 1
V1201 0
03 Q13 014 I
~
~
~
~
4 4
Q1$
CALfLOffIIIOENGMATlONS Of LEO NtNNERS N
V1202 0"
Q~ 0 Q
0 Qlt Q33 V1I7<
0 Q
V1175
, 31 1
3
~
Q7 I
ST. LuCia USiT Z PRESSURIZER RELIEF SYSTEM FIGURE 2
ST.
LUCIE UNIT I NUREG 0737 ACTION ITEM II.D.l ADDITIONALQUESTIONS TABLE 1 PIPING LOADS COMPARISON (lbs)
IEbasco RELAP5/MODI IAnalysis August, 1982 I
I LEG PORV 8RV max min max min I
17 I
194
-75 106
-86 19 I
186
-25 20 I
46
-6 21 49
-4 104
-72 22 115 "4
268 "383 I
23, 117
-3 224 -388 I
24 267 W
448 "791 25 211
-3 388 -553 26 115
-1 259 -307 27 68 28 i 44 "8
29 I
88 "13 30 I
231
-22 31 I
37
-3 32 I 79 "4
192 -226 33 112
-3 216 -348 34
~
128
-3 242 "409 35 i 250
-5 417 -779 36 I 242
-3 384 -684 37 i
105
-1 192 "496 I
I Ebasco LTOP Analysis December, 1987 I
I I
V 1402 V 1404 I
CASE Cl CASE 2 I
I max min max min I
I 296 "8
I I
I I
I I
I I
76
-2 116
-1 219
-1 I 528 0
i 108 0
I 307 0
202
-15
, 176
<<44 I 212 0
)136 0
39 0
I 302 535 147 236 270 188 235.
123 46 141
-8 304 0
0 0
-2
-47 0
0 0
-2 78 119 226 545 111 228 142 128 117 67 25
-8 309 552.
151 185 185 120 147 61 73
-2 145
<<]
-1 0
0 0
-19
-21 0
0
-2
-9 200
-7 0
0 0
3 29 0
0
-2
-2 54
-2 84
-1 164 0
389 0
8O O
ll3 14
-1 8
-2 10
-3 9
-3 5
-1 I
I 204
-7 I
406 0 I 112 0 f ll5
-7 173 0
8
-2'l 8
-2I 5
1O3
~
-2
~
I I
- l I
l I
I I
I I
I I
I I
V 1402 V 1404 V 1402 V 1404 I
CASE Al CASE A2 CASE Bl CASE B2 max min max min max min max min I
NOTES:
l.
2 ~
3.
4.
For Leg Identification See Figure l.
Leg 827 is a combinationof Leg 18 and 27 in the Report Piping upstream of PORV: 17, 19, 20, 21 (Partial),
27, 28, 29, 30, 31, 21 V 1402 and V1404 represent PORV valve numbers.
ST.
LUCIE UNIT 2 NUREG 0737 ACTION ITEM II.D.1 ADDITIONALQUESTIONS TABLE 2 PIPING LOADS COMPARISON (Lbs)
LEG I
Ebasco " LTOP Analysis
- December, 1987 v1475 I
via 74 max min I max min lEbasco RELAP5/MOD1 Analysis K)ctober, 1982 gn..v'spv n.ax min max min I
I I
I I
I 17 19 20 22 23 24 25 27>
28 Q 0 32 35 36 37 NOTES:
486 262 161 252 674 1134 826 77 242 389
'172 354 795 1300 965 396 I
~
I II 7
I II 9
I I
I I
I
-14 I
0 I
9 I
II
-11 II
-28 II
-1 I
I I
-6 I
II
~ 0 I
II
- 0 II
,,-8 II
-19 II
-13 II
-24 II
-.'14 II 15 I
II I
1 I
I
-. 90 1277 1257 1817 1626 1507 1449 1150 1713 1701 2073 18az 1802 1498 980 II I
377 II I
II
-1825,'
I
-1926,'
I z377 I
I
-2016,'
I
-za&2',
II
-16o8,'
I
-1092 I I
I 133 II 146 I
386 I
I 519 I
-2401 I
71 II
-2715'7 I
-2845) 36 II 2836 I
19 I
-3180 I
37 I
-2930I 25 II
-173al 13 I3 10
-4 "13 I
I 327 I
I 538 II 71 I
I 26 II I
II 22 II I
I I
I II 12 II 282 I
I II I
I
~ '
II II II II II I
II II II II I
II II I
I I
I II 7
II p
I II p
I II I
I 9
I II I
II
-11 I
I 8
I I
3 I
II 0
I I
II II I
II I
I II I
I II II II II II II II I
w I
II II II I
~
I 1.
2 ~
3 ~
4 ~
For leg identification See Figure 2
Leg 427 is a combination of Leg 18 and 27 in the Report Piping upstream of PORV:
17, 19, 27, 28,29, 3'0 (Partial)
V1474 and V1475 represent PORV valve numbers
ST LUCIE NUCLEAR POWER PLANT UNIT NO 1 APPENDIX A Pressurizer Power Operated Relief Valve and Safety Valve Discharge Piping
TABLE CONTENTS Page l.0 Xntroduction
2.0 System Description
3.0 Description of Analysis 4.0 Conclusions from Analyses
APP ENDIX A
1.0 INTRODUCTION
This attachment summarizes the study performed in August, 1982 to evaluate the adequacy of the pressurizer relief piping and its supports at the St Lucie Unit No 1 nuclear power plant, as required by NUREG 0737 Item II.D.1.
The analysis was performed utilizing the RELAP5/MOD1 computer program.
The valve discharge loads on the pressurizer relief piping system can be induced by the opening of either or both of two power operated relief valves (PORV), the opening of any or all of the three spring loaded, self activated safety valves (SRV) or a sequenced combination of PORV and SRV.
Actuation of these valves allows discharge of high pressure steam from the top of the pressurizer into the discharge piping causing pressure and momentum transients throughout the piping system.
These transients create significant time varying unbalanced fo'rces in each straight segment of the piping until steady flow is achieved.
The time histories of the discharge loading are determined throughout the system by employing a hydraulic model.
This hydraulic model is suitable for execution with the RELAP5/MODl computer program.
Forces in each segment are computed by a post processor
- code, CALPLOTF III.
This postprocessor has been written by Ebasco to develop forces from the output of REALP5/MODl for each of the piping segments.
The necessity for writing the postprocessing code arose from the unavailability of an EPRI developed postprocessing code for interfacing with RELAP5/MOD1.
APPENDIX A
- 2. 0 SYSTEM DESCRIPTION The pressurizer relief piping system for the St Lucie Nuclear Plant Unit No 1, consists of two power operated relief valves, three spring loaded self actuated
- valves, the interconnecting discharge piping, and the quench tank.
An isomertic drawing showing the geometric configuration of the system is attached as Figure 2.1.
Also shown on this drawing are the location and types of the supports and restraints.
Superimposed on the isometric drawings are designations of the piping segments employed in the CALPLOTF IIIhydraulic force model.
The component numbers of the REIAP5/MOD1 model used to calculate the thermodynamic properties and flow conditions are shown in Figure 2.2 The same drawing also provides the design pressure and temperature of the various piping segments.
Information on the capacity of the individual supports shown in the isometric drawings is provided in Table 2.1.
In the event of an abnormal transient causing a sustained increase in pressurizer pressure at a rate exceeding the control capacity of the pressurizer
- spray, a high pressure trip level is reached which trips the reactor and opens the two power operated relief valves.
If the relief provided by the opening of the two PORVs is insufficent to limit the pressure rise from reaching the setpoint of the SRVs, the safety valves will open as, needed to relieve the overpressure.
Table 2.2 provides a list of the transients which can result in opening of either the
Xa FSAR analyses, no credit is taken for PORV operation for pressure relief.
Consequently the results of the analysis are conservative from the standpoint of peak pressure and pressure ramp since in reality the PORV will actuate, and the SRV setpoint may not be reached.
As a consequence two cases considered to be bounding plus a third realistic case have been investigated to determine the hydraulic loads on.the relief piping and supports.
The first two cases assume either actuation of both PORVs simultaneously, or of all three SRVs simultaneously; the third case considers actuation of the two PORVs followed by a simultaneous actuation of the three SRVs.
The time sequence adopted is given below:
Opening of PORV Pressure Rise Ramp "Opening of SRV Time 0.0 64.4 psi/sec Time at which pressure reaches 2485 psig + 3X The two PORVs are located on two separated lines and are connected to the pressurizer nozzle through a common 4 inch line.
Two power operated block valves are provided in series with the PORVs to achieve isolation if necessary.
Ihe three safety valves are each independently connected to the pressurizer safety valve nozzles.
The discharge piping from both the PORVs and SRVs are connected to a 10 inch common header which leads to the quench tank.
The power operated valves are set to open automatically at 2385 psig + 3X.
Each of these valves develops a flow rate (steam) of 153,000 lb/hr.
The safety valves are set to open at 2485 psig + 3X.
Each valve has a capacity of 213,000 lb/hr (steam) flow rate at the set pressure.
n322~jnngqM
~ 4'
APPENDIX A
- 3. 0 DESCRIPTION OF ANALYSIS The pressurizer relief piping system is modelled as a network of fluid control volumes connected by Junctions.
Ihe designations of the component volumes are given in Figure 2.2.
The thermodynamic properties in each control volume and the flow conditions at each Junction are computed as a function of time following valve actuation utilizing the RELAP5/MODl computer program.
These are in turn inputted in the postprocessor program CALPLOTF III which computes the forces at each segment of the pi ping.
Three analyses have been performed corresponding to the following scenarios of valve actuation.
a.
Two PORV open simultaneously.
SRVs closed (Case A) b.
PORVs do not open.
All three SRVs open simultaneously (Case 8) c.
PORVs open.
Pressurizer pressure continues to increase and SRVs open simultaneously (Case C).
Opening times for the PORVs and SRV were chosen to'be 110 msec for the representative Dresser 31533VX-30 PORV and 6 msec for the Crosby 3K6 SRV.
APPENDIX A
4.0 CONCLUSION
S PROM ANALYSES Results of the analyses are shown as typical transient force time histories in Figures 4.1 through 4.6.
The three cases which have beea examined are labelled A through C as previously stated.
Consistent with the approach taken which is to compare results of the present aaalyses with prior analyses which had resulted in the present
- design, the transient force time histories are reported at the same location chosen ia the KDS analysis.
Thus the transient hydraulic forces given Ia the figures represent:
Horizontal force acting oa the first pipe segment after a safety relief valve - Segment A in Fig 2.1.
Horizoatal force acting on the next straight segment past the first elbow in the same line segment 3 in the same line.
I Horizontal force in segment upstream of PORV designated as segment D in Figure 2.1.
Horizontal force oa segment downstream of di.fferent PORV, designated as segment F in Figure 2.1.
Vertical force in straight segment connecting SRV discharge to the header - designated as segment C in Figure 2.1.
Vertical force acting on the segment of piping which runs from the pressurizer to the T connection to the two PORV piping segments "
designated as segment E in Figure 2.1.
Strictly speaking, direct comparison of the hydraulic forces developed in the prior analyses and the present anaLyses is aot meaningful per se, since prior
,analyses had assumed that the highest hydraulic loads would occur under the assumptioa that all valves, PORVs and SRVs, would open simultaneously.
That condition has aot been analyzed in this report.
The hydraulic loads resulting from the simultaneous actuation of all of the valves had been employed in the design of the pressurizer relief piping and its supporting system.
Comparing them with the hydraulic forcing functions from the present analyses to those Mich resulted in the system desiga can yield information on whether the design is adequate for the presently estimated forces.
Under the assumption that the system would behave in linear fashion, comparison of the hydraulic loading functions yield direct information on whether the system is still adequately designed'
APPENDIX A The assumption of linear behavior is considered appropriate since the prior analysis was an elastic analysis wherein geometric nonlinearities introuduced by the presence of gaps in the snubber type restraints had been linearized, provided that:
a)
The magnitude of the hydraulic forcing functions in the piping segments are not so significantly different than the previous magnitudes so that material nonlinearities will not occur, i.e.,
portions of the piping system will not be stressed past yield; and b)
'The frequency content of the forcing function is also sufficiently similar to that of the forcing functions derived in the previous analysis so that unexpected different resonant modes of the system are not excited.
For case A,
PORV actuation only, the comparison between results of the present analysis and EDSs prior analysis indicates that the hydraulic forces in the vicinity of the PORVs are generally lower in the new analysis (see Table 4.2).
To a great extent this is due to the fact that the prior analysis assumed an opening time of 20 milliseconds instead of the present 110 msec.
To a much lesser extent the differences may be due to the differences in the methods followed by EDS and Ebasco in extracting forces from the thermo-hydraulic computer program output.
Ebasco's CALPLOTF III program computes the time varying force for a complete segment of piping.
Hence when viewing the force in segment F of Figure 2.1 as shown in Figure 4.7, one can visualize it as the force which as a function of time acts on that entire segment (at its center of mass).
EDSs forces are not given for complete
- segments, and only on rare occasions do their plots of forcing function time histories approximately coincide with the force on an entire segment.
One such instance occurs for segment F (Figure 4.7). In most instances, however, their plotted forces are not the force on the entire segment.
0322M/0039M
APPENDIX A For instance, the EDS force shown in Figure 4.8 does not represent the time history of the force acting on the segment of piping from the SRV valve outlet to the first elbow.
The forcing function number 5 acting on node 55 of EDSs structural model, is only one of the forcing functions in the segment of piping between the SRV valve and the first downstream elbow.
Since recasting the entire time history of EDS forcing functions at individual points within segments into an overall segment forcing function would be extremely time consuming, Ebasco has chosen to compare the present analysis to the approximate peak in the segment forcing function, as deduced from the HS's computer output, whenever direct comparison is not possible.
This comparison is shown in Table 4.1.
wherever direct approximate comparison between the ED's and Ebasco's computed forces is possible, the time histories of both calculations are shown; as for instance in segments F of Figure 2.1.
Unfortunately this direct comparison basically means that only the PORV reactions can be directly compared.
The magnitude of the newly computed forces are generally lower.
We frequency content cannot be directly compared to the EDS analyses.
However, the newly calculated forces typically dissipate in a shorter time (lower energy content).
It is therefore concluded that an accurate stress analysis of the St Lucie 1 pressurizer relief piping system would indicate that the piping system and its supports are adequate to cope with the PORV actuation transient.
For Case B,
SRV actuation only forces computed by the present analysis are equal in magnitude to those computed by EDS.
The results are shown in Table 4.1 for those segments which are essentially affected by the SRVs.
Case C has n'ot been analyzed specifically since the sequential opening of the PORVs followed by the SRVs would not cause the most severe loading on the discharge piping system.
So stress analysis has been done for the St Lucie Unit 1 pressurizer relief piping system under the influence of these forces.
Table 2.1 however, lists the expected values of the reaction forces upon the systems supports and restraints, and these forces are generally well within the capacity of the restraining system.
For those instances in which the expected forces exceed the listed capacity, we rely on our experience with mechanical snubbers.
Testing has shown these types of snubbers to be capable of accepting at least twice the load for which they are rated (tests showed in excess of 4 times the capacity).
The other restraints and clamps would show greater capacity than listed in Table 2.1 if faulted limits were used as allowed by EPRIs recommendation.
0322M/0039M
~'
APPENDIX A In estimating the reaction loads on the restraints we have made use of the peak hydraulic forces which have been calculated, and the fact that the characteristic time duration of each oscillation in the load is shorter than the fundamental period with which the pressurizer relief piping system would respond, i.e., since duration of each oscillation varies between 10 msec and 30 msec, and the fundamental period is about-200 msec (from EDS analysis).
Consequently, the dynamic amplification expected in the response of the piping system should be close to unity.
Review of the EDS reported stresses reveals that there is generally in excess of 30 percent margin to the stresses developed for combinations of sustained loads, hydraulic and seismic loads.
This margin should accommodate any of the increased loads due to the revised hydraulics, since except in a few cases the new hydraulic loads are nearly the same or lower in magnitude than the old hydraulic loads.
Based on the above, it can be concluded that the St Lucie 1 pressurizer relief piping system and its restraints are adequate for the presently calculated hydraulic loads from discharge of the SRVs and PORVs.
.v h.p~W~
10 P~j 5
oX
~~
4 j'
p~
50 40 ass'>
S
~ 4 5
~
40 5
~
C q5
)s'.
S'gt:.
jl'-
r Cl leS g)%
<<C 11 ~
Ilh
~ 11 '>Jpo l
yWai.
rk
/rg op
+ W+~
O'S
$1
~r~47
~O
$7&i
'4 CI I'
Cj C4 (pht
+'
+~+
L 1i II 5004 0jj Il 0 Ch+4p, 001r 00 Ch I y$0 I
Qo rV~)
.Il o I
~~i.1I'jjr.yet 4 Il rjo yQ
<
'll'-
J r
l..V' J
ST. %CIA mrs? t SSNSURItSI agl.ggt gZSTtSS SICtNR g,L ZS
~ tC C
11 01C Cli
<05tC btt h.$00 4.055 10.1 00
.237 280
.2$ 0 700 675
o C
oP+g tp d
~ e~
fO
~..
C Itg.
~
~
~
N 4'
I or
~
Pm x~
r l1l oil f1
'.l0 <<P elk
~0 ~
BEST OBTAIIVhBLEspy
~+po4 + 4 oewl~
Q
~~LPgr eoeeL
..I'C, r oo
~
~tR tt%
'lr~
fN Q ~
J+
ttet
~ I%
Il 4r A
~r j
Ptp
~
V~
~+
o fo (i
~p 11~
11
~
ip yO r
~
C
.sl I
of?
to,'L~:
L3~.f"
~ I
?
ot) i.
I gg%
i fl ~
~Ia
~
r
~
k( t5Cfr ts Bf~
~
C I? ~ 5 0 ~ o?
2.075
.375 4.spo 237 2Bo S5 Cys yoo 655 675
~ etf oit ~
go ~
fort rt~
~ ae Ipf<<C BZZ t0.7$ 0 ZSO tt
~Jg Cf O~
I Jp N
f.O5R R ~
or~
t+
~...i t MRS O~>
t+ '4p.
~ g)~ pp p
~
eO gV "50
~
'oo l
- Ego,
?S CQ
~ P g?I tr l
~
~ 11 AS"
'+to o
/
t.z.
SB Py d rt iver r(
g p
p+t.
~
~
1 QuattctlaR, pf
>?
i1%
tt gp o
~ 0 of
- .14'-
I
.~
'-P
~
o tt
~0 r.
.r F'Agf 'l ST. URER IltITI SRESSOllIXCR ItELICt SIST%I FIClNK 2.2 Pt ~ C llPfe
- ftttoaeo, ottt ~Ill, it
TABLE 2,1 ST LUCIE 1 SUPPORTS/RESTRANTS IN PRESSURIZER RELIEF PIPING Support/Restr.
Designation (Refer to Fig 2.1)
RC-005-Sise Ca pacity Expected Design (kips) hpprox. Max**
Loads Normal 6 Upset Hydraulic Load (lbs)
Snubber Clam (lbs) 12h Y-Snubber PShlo 12B (2)YZ Snubber (2)PSA3 12C S.H.
34B Y Snubber PSh3 34h Z Snubber PSh3'4C S.H.
36 X-Snubber PShlo 40 Y"Stop RSSA3 113 XY"Stop Frame 105 S.H.
103 X"Stop Frame 101 Z-Stop RSSA3 100 S.H.
98 Y Snubber PSh3 97 S.H.
96 Z Stop Special 89 Y Snubber PSA3 90 X Snubber PSA3 88h Z-Stop RSSA3 88B S.H.
86 S.H.
70 S.H.
66 Z-Stop RSSA3 65 S.H; 63 X-Stop L-Frame 62 S.H.
62h Y Snubber PSh3 61 Z Stop L-Frame 55C X"Snubber PSA10 55A S.H.
55B Y Snubber PSA3 53A Z-Stop RSSA3 5lh S.H.
+3677
+3051
-1682
+685
+2126
-1432
+5702 1126 592 769(-1267) 1290(-922)
-199
+835 741 978(-1584)
+828
+2402 1195(-1254) 381 223 588 1901(-1288) 221 695(-1418) 765
+831
+2354
+3846 F78
+1686 2192(-1581) 312 15 6 Ea.
6 6
9 2.5 About hbout 3
6 hbout 6
6 3
3 hbout 6
6 15 6
3 10 3 Ea.
3 3
10 3
3 2
3 6
1.7 3
3 3
3 2.5 3
3 10 3
3 10,000 10,000 7,000
.7,000 2,000 Small 1,000 600 600 1,000 500 700 600 500 500 1,000 500 500 S.H. - Spring Hanger
- X Stop 1919(-1512)
Y Stop 1329(-2179) e*
Based on a Dynamic Load Factor of approximately F 0 10
TABLE 2.2 Calculated Pressurizer Safety and Power Operated Relief Valves Inlet Fluid Conditions During Pressurization Transients(1)
Pressuration Event Loss of Load Loss of Feedwater Flow Loss of hC CEh Ejection Valve e
PORV Safety PORV Safety PORV2 Safety PORV Peak Pressurizer Pressure (PSIA) 2562 2506 2534 2477 Pressure Ramp Rate
( PSI /SEC) 60.0 64.0 12.0 27.0 52.4 64.4 8.0 Fluid Conditions Steam Steam Steam Steam Steam Steam Sequence of Events for Pressurization Transients Which Actuate Safety and/or Power Operated Relief Valves(1)
- TQK, SECONDS Pres suriza t'ion Transien t Events During Transient Loss of Load Loss of Feedwater Flow Loss of hc CEA E ection Event Initiation Reactor Trip:
0.0 0.0 0.0 0.0 1.
High Pressurizer Pressure 7.75
- 28. 8 2 ~
Low Flow Trip Opening of PORV Opening of Safety Valve Peak Pressure 7.35
- 8. 95 11.0 26.9 32.4 32.8
.86 (2)
- 5. 55 7.4 1.53 4.0 Safety Valve Closing Closing of PORV
- 13. 4 15.8 33.6 36.9 12.0 (2) 5.87 (1) EPRI Safety Relief Valve Test Program
. (2) PORVs do not open under Loss of hC conditions 11
TABLE 4.1 SRV
& PORV ACTUATION HYDRAULIC LOADS (lbf)
Segment St Lucie Unit 1 Identification RELAP5/MOD I St Lucie Unit 1
- EDS, RELAP3 SRV (1)
Actuation A
max 36 min
-1687 B
max 31 min
-1214 C
max 80 min
-3145 D
max 85 min
-83 E
max 86 min
-106 F
max 383 min
-268 PORV (2)
Actuation 134
-116 80
-745 176
-216 231
-22 75
-194 SRV-PORV (3)
Simultaneous Actuation 531
-897 3192
-3143 1816
-1015 597
-123 423
-1786 120
-723 NOTES:
(1)
SRV Actuation Only - Opening Time 6 msec (2)
PORV Actuation Only - Opening Time 110 msec (3)
& PORV Simultaneous Actuation - Opening Time 20 msec h9')'illIhh') OM
TABLE 4.2 CALPLOTFIII RESULTS FOR PORV ACTUATION LEG NO.
1 2
3 4
5 6
7 8
9 10 11 12 13 14 15'6 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 MAX. FORCE
- 49. 2 100.9 92.4 186.8 1060.8 597.6 178. 6 505.Q.
49.2 68.0 186.4 49.2 27.6 115.8 744.8 216.0 193. 7 45.4 185.5 46.3
- 48. 98 114. 7 116.7 267.4 211.2 115.1 22.4 44.1 88.0 230.8 36.6 79.0 112.4 127.8 249.7 242. 4 104. 9 298. 2 370.8 MIN FORCE 50.4 150.1 113.0
- 224.3
-1072. 6
- 899.2
- 307.2
- 447.2 50.4
- 84. 7
- 155.1 50.4
- 39. 2
- 134.0 80.4
- 176.1 74.6 9.7
- 24. 5 6.0 3.8 4.3 3.1 5.8 2.8 0.99 5.8 8.1 13.4 21.5 3 ~ 2 3.9 2.9 2.9 5.1 3 '
0.9 2.5
- 236.7 13
TABLE 4.3 GQZLOTFIII RESULTS FOR SRV ACTUATION LEG NO.
1 2
3 4
5 6
7 8
9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 MAX. FORCE 862. 1 1413.8 1157.0 2818.1 6495.4 6599.0 4954.4 5172.4 862.1.
1446.6
~
2784.7.
862.1, 543.2 1686.7 1214.1 3144.6 106.4 16.98 29.6 13.9 103. 7 267. 7 224. 4 447.6 388.0 258. 8 14.95 24.0 51.1 84.6 10.8 191.6 215.9 242.2 417.2 384.8 192. 2 759.3 889.7 MIN. FORCE
- 55.5 25 ~ 2
- 21.1
- 61.7
-477.7
-494.8
-394.8
-623.5
- 55.5
- 19.2 63.8
- 55.5
- 69.0 35 ~ 7
- 30.6
- 80.3
- 86.4
>> 15.1
- 43.7
- 16.1
- 71.8
-382.9
-388.3
-790.7
-552.7
-307. 4
- 12.5 25 '
- 52.1
- 83.1
- 10.1
-225.9
-347.9
-409,3
-778.8
-684.0
-495.6
-1343. 0
-1542. 3.
14
<~LSTL ] SRV3 SRV Pi/25/82RELSSEGNENT3SV1703-1707 CASE B FIGURE E4.1 LEG14 CD CD CD CV CD CD CD CD CV BEST OBTAINABLECOPY CD CD CD CD C)
CDO No CQ CV QJ Q
O
~ O Lc o CD CD CD CD CD C.
1C 20 p
QC TINE (SEC) 1%
C. 40 C. 50 n
FPLSTL15RV3 SRV 07/26/82RELSSEGNEhT35%1708-1710 O
CASE B O
FIGURE 4. 2 LEG15 OO O
fV OO OO BEST OBTAINABLECOPY OO O
OCD (nO Qj O
~ CD Q
OO IC O OO O
CV OO O
O P4I
- e. eo
- o. 1e O. 20
- 0. 3e T ISE ~SEC)
- e. 40
- e. so
- e. ce
~PLSTL 1 SRV3 SRV 07/26/82RELSSEGHCNT3SV1711-1901 o
CASE B FIGURE 4.3 LEG16 oo oo oo oIll (V
BEST OBTAINABLECOPY oo oo
~ CV oo Q3 v)
DJUo Q
CD
~o
<o o
CD o
Lll o
C)
CD oo
- e. eo
- 0. 10
- 0. 20 C. 3C T IVE (SEC)
- 0. ae
- c. Se
- c. ce
~ORV o
C) o OGt CGi B~PFL55KC'ICK I PP" ce i eeeC CASE A FIGURE 4.4 CI C) oo CV oo o
(0 8EST OBTAINABLECOPY o
o P4 I~Co C
~ CCI (Q
CC Q.o C) o oo o
oo o
I oM I
e
..e e
T)HE (~E 18
- e. ~e
- e. De
FPLSL1PQRV2 PQRV 06/08/82REL5SECHENT2PV35020000 CASE A FIGURE 4. 5 LEC22 oo CD Al oo oo oo o
CO (A~o
~ o
~ CO C)
Q o CD CD EST OBTAIINABIE pgF,>
CD CD oAl CDo o
oo o
(V
'e. 00
- 0. 10
- 0. 20
- 0. 30 TINE (SEC) 19
1 c'v.f
~ >
.y
/
. FPLSL1PORV2 PORV 06/08/82RELSSEGNENT1PV3006-3007
- WO'o CASE A FIGURE 4.6 LEC30 oo o
CV oo oo CV oo o
CO Ko
~ o CV CC E3 oo CD CO BEST OBTAIIVABLECOPY oo o
CDo CD oo o
C. CC C. 10 0
20 0 30 T INE (SEC) 20
- 0. 4C
- 0. SC C rC
Y I
) W g
KC
Q. I ~
BEST OBTAINABLECOPY EDS RV 6 S 9 0.020 sec Openings 4(
I 6
IL) <a ~
L I
St Lucie 2 - PORV 6 200 msec SRV 8 40 msec ie 1<<POKV 6 110 msec opening on1y Segment P
Pigure 4.7 C.2 0.3 a..~
T f ME I Sf-:f.')
0-7 8'~" 0-."~T -
l UCH:
EDD 3'l00001-00SR FORC>NG FUNCTION NO SS FIT NODE: 176
BEST OBTAINABI.ECpF y tO 0
'.C 0.0 tu C3 6 0.:0 0
~ 40
.o en~I EDS' PORV 6 SRV 8 0.020 sec openings, Segment A
Figure 4eg
>>0.80
~ St.
-1.0 0 f 0-9 rncfe 1 - SRV Q 0.006 sec openin6 anil(
+.
0."
0.3 os T lt"eE.
f 5f:C) 0-5 0.6 0-8 t
1) t I
tBr.!SrD->T.
~ UCrF' tr r q.'I pir'Qsi --po~q F'QRC
> H~ FUNCTION N,R g,-,T NODL bg
-s.8
(I