ML20138A773
| ML20138A773 | |
| Person / Time | |
|---|---|
| Site: | Davis Besse  |
| Issue date: | 12/06/1985 |
| From: | Williams J TOLEDO EDISON CO. |
| To: | Stolz J Office of Nuclear Reactor Regulation |
| References | |
| RTR-NUREG-0737, RTR-NUREG-737, TASK-2.D.1, TASK-TM 1221, 44572, TAC-44572, NUDOCS 8512120106 | |
| Download: ML20138A773 (10) | |
Text
.
0
/
TOLEDO
%ss EDISON Docket No. 50-346 JOE WiLUAMS. JR.
s va n
-eu..r License No. NPF-3
['$ElEs Serial No. 1221 December 6, 1985 Director of Nuclear Reactor Regulation Attention:
Mr. John F. Stolz Operating Reactor Branch No. 4 Division of Licensing United States Nuclear Regulatory Commission Washington, D.C.
20555
Dear Mr. Stolz:
This is in response to your letter of June 7, 1985 (Log No. 1764) concerning NUREG-0737 II.D.1 Request for Additional Information. Toledo Edison has provided a partial response to the 13 questions in two previous
'submittals dated July 19, 1985 (Serial No. 1171) and October 1, 1985 (Serial No. 1191). With this submittal all questions of your June 7,1985 letter will be answered for Davis-Besse Nuclear Power Station Unit I.
Very truly yours, J elu k J~J n JW: GAB:MGF Attachment cc: DB-1 Resident Inspector J
8512120106 851206 f
PDR ADOCK 05000346 P
PDR THE TOLEDO EDISON COMPANY EDISON PLAZA 300 MADISON AVENUE TOLEDO, OHIO 43652
Docket No. 50-346 License No. NPF-3 Serial No. 1221 December 6, 1985 Attachment Question 3.
The B&W Inlet Fluid Conditions Report analyzes the generic 177-FA plant for a feedline break accident.
In their analysis the PORV was assumed not to operate. The transient data tables and plots were presented in the report for times of less than 40 sec. Liquid discharge was not predicted during these time spans.
The Davis-Besse submittal states that the safety valves will pass only steam with the exception of the feedwater line break event where transition to liquid could occur.
Provide additional information discussing the potential case of safety valve liquid discharge, and discuss the effect on safety valve operability.
Response
The safety valves have been moved from their original location in the valve room to a direct mount on the pressurizer nozzles. Therefore, there is currently no inlet piping to the safety valves. The safety valve discharge piping has also been removed and replaced with a tee.
Because the safety valve is mounted directly on the pressurizer nozzle, any water hammer which may occur on the inlet side will not cause a bending moment en the safety valve and, therefore, will not affect valve operability.
U
- outlet piping amounts to little more than the distance f rva the valve throat to the discharge tee which is 21.75 inch. The transient force which is developed in this short length of pipe can be estimated from the wave force equation F = L Am At where L = pipe segment = 1.8125 ft.
Am = 245,000 lbm safety valve flow rate hr.
At = valve opening time 10 msee This yields a force of 387 lb which causes a bending moment about the valve flange of 4,644 in -lbf (the distance from the valve center line to the valve flange is approximately 12 inches.
While the 4M 6 valve was not tested by EPRI, the 3K6 was and 3
it was found to operate successfully with bending moments of up to 161,500 in-lbf on opening and 152,000 in-lbf on -
closing. Therefore, it is assumed that the safety valve will operate successfully at this much lower bending moment.
Question 7.
The B&W Valve Inlet Fluid Conditions Report stated that liquid flows could exist through the PORV for the FSAR feedline break and the extended high pressure injection events. These same flow conditions will also exist for the block valve. The EPRI. Marshall block valve test program only tested the block valves with steam flow. Since it is conceivable that the PORV block valve could be expected to operate with liquid flow, provide a justification as to how the results of the Marshall tests or other tests can be used to demonstrate operability of the block valves for liquid conditions. Also, evaluate applicability of the test results to the Davis-Besse block valve since the plant valve has a Limitorque SMB-00-10 actuator while a SMB-000-10 and SB-00-15 were tested.
Response
The motor operation of all Limitorque valve controls is the same, basically; however, there is some variation in the manual operation.
The SMB-000 and SMB-00 vary in that with the SMB-000 the handwheel is mounted directly on the drive sleeve, whereas the SMB-00 could have the handwheel directly on the drive sleeve or mcunted on the side using a set of bevel gears to improve the mechanical advantage.
In addition to the Marshall block valve testing, Westing-house tested block valves under liquid, steam and no flow conditions. This report was titled EPRI Summary Report:
Westinghouse Gate Valve Closure Testing Program, WEMD-EM5683 Rev.1, March 31, 1982. The results of this testing revealed that there were no significant differences in valve perfor-mance between liquid flow and steam flow.
Question 9.
Bending moments are induced on the relief valves during the time they are required to operate because of discharge loads and thermal expansion of the pressurizer vessel and inlet piping. The TES report TR-5639-2 did not provide bending moments applied to the PORV. Make a comparison of the predicted Davis-Besse PORV valve bending moments to the
~
tested valve bending moments to demonstrate the operability of the valve is not impaired. _
Response
The bending moments developed at the ends of the PORV, data point (nodes) 240 and 255 have been extracted from TES Project 5639 files and the requested comparison made, as follows:
Using the files on TES Project 5639 & report TR-5639-2 we get:
1.
The THRSAP Node Numbers for the PORV valve ends are 240 & 255 i
2.
The piping flexibility computer results are:
t -
AT N0DE 240 RUN SEQUENCE M
H H
x y
z DESCRIPTION No.
(IN-LBS)
(IN-LBS)
(IN-LBS)
DEAD HX3V4HV
-237
-326
-1,611 WGT.
THERMAL HX3VIDJ 5,334 3,547
-2,425 1
THERMAL HX3VI0B 5,123
-1,223 3,673 2
OBE-X HX30CIJ 1,438 61 150 SEISMIC OBE-Y HX30CIJ 1,456 126 27 SEISMIC OBE-Z HX30CIJ 6,159 697 223 SEISMIC 500 F RESULTANT BLOWDOWN HX32F5F 5,555 3,047 2,930 HR (IN-LBS)
TH1rBLDWN 10,889 6,594
-5,355 13,810 TH2+BLDWN 10,678
-4,270 6,603 13,261 AT N0DE 255 RUN SEQUENCE M
M Mg DESCRIPTION No.
(IN-LBS)
(IN-LBS)
(IN-LBS)
DEAD HX3V4HV
-196
-326
-628 WGT.
THERMAL HX3VIDJ 3,231 3,547 622 1
THERMAL HX3VI0B 2,371
-1,223
-3,587 2
OBE-X HX30CIJ 8
62 10 SEISMIC OBE-Y HX30CIJ 32 125 74 SEISMIC OBE-Z HX30CIJ 338 696 257 SEISMIC 500*F RESULTANT BLOWDOWN HX32F5F 7,485 2,954 2,146 MR(
~L
)
TH1+BLDWN-10,716 6,501 2,768 12,836 TH2+BLDWN 9,856
-4,177
-5,733 12,143
s.
c>
x[
n I
From Table 4.2.1-3a of EPRI PWR Safety and Relief Valve Test Program 4
" Safety >and Relief Test Report" Res. Proj. V102 Int. Rept. April 1982, we observe that the Crosby valve test will continue to operate when subjected 9
to a: moment of 31,600 IN-LBS.
M ) MAX = 12,836 IN-LBS < 31,600 IN-LBS R
Therefore, the valve will be operable with the above bending moment for Davis-Besse PORV.
Question 10.
The submittal provides a list of ' loads that were considered i
'x in the structural analysis and states that the analysis was performed to criteria of ASME Code Section III, Subsections NB, ND, and NF.
It does not, however, identify the load
~ combinations considered in the analysis or the stress limits used for each combination. A list of recommended e
load combinations and stress limits is contained in the jl EPRI PWR Safety and Relief Valve Test Program Guide (by MPR Associates,Inc.).
Provide a list of load combinations and respective stress limits used in the analysis of the inlet piping, discharge piping, and suppotts so as to show how these compare with the recommended combinations and limits of the EPRI Guide.
Response
(a) TiG design of hanger or re'straints (hanger / restraint design ~1oads) are based on the " worst" (maximum)
J combina. tion of thermal (thermal load cases 1 or 2) loads with the deadweight plus seismic (DBE) plus blowdown load epses. These calculations and Ioads are f shown on pages/126,' 146 and 147 of the'TES report i
TR-5639-2.
?
(b) Report TR-5369-2 does not specifically evaluste the Class 1, Upstream' Portion ci the PORV piping nor the SRV connection atop the pres:urizer. As discussed in the report on page 109, previous conclusions'concerning f
"~
adequacy were assessed and judged to be valid. The s
previous adequate conclusions were made in TES Technical Report E-1495-17, Revision A which certified the 2
design to be in accordance witn the 1971 Edition,of Section III of the ASME Code. Generally speaking, the j
blovdown event has been considered as >a normal or
. upset condition and has been considerei in the ASME
]
Code fatigue analysis. The specific list of' normal, 6psbt, emergency and faulted conditionr considered are as follows':
. r
.0
?
q
- D.M 7
.p 2
1 o
s x.
.._ r
\\
\\
..I
~
+
l DESIGN ~ CONDITIONS.
3 REACTOR COOLANT SYSTEM i -
Design pressure, psig 2500 j
Design temperature, F
- For design conditions, the Hot leg 650*
temperature of Pressurizer Relief Cold leg 650 Safety System is assumed 670 F.
TRANSIENT DESCRIPTION DESIGN CYCLES
-NORMAL CONDITIONS Heatup and cooldown at 100 F/hr.
120 Heatup at 35 F/hr and cooldown at 100F/hr 120 Power change 0 to 15% and 15% to 0 1,440 1
Power loading 8% to 100% power 48,000 Power unloading 100% to 8% power.
48,000 10% step load increase 8,000 10% step load decrease 8,000 UPSET CONDITIONS i
Step load reduction (100% to 8% Power) 310 Reactor trip (types A-40, B-160, and C-88) 288 Rapid depressurization 80 Loss of. station power and reactor trip 40 Rod withdrawal accident and reactor trip 40 Control drop 40 EMERGENCY CONDITIONS Loss of feedwater to one steam generator and reactor trip 20 1
Stuck-open turbine bypass valve and-i reactor trip 10 Change of flow 20 FAULTED CONDITIONS Steam line failure and reactor trip 1
Loss of coolant and reactor trip-(LOCA)**
1 TEST CONDITIONS ~
Hydrostate Test Reactor Coolant System 20 (3765 psig, 135 F)
Pressurizer Relief Valve Piping (CCA-8)
[
(3810 psig, 135 F) 20
- LOCA is considered.an Upset Condition, or equivalent, but this. system is not affected by LOCA.
i..
I t
s,
' y,.
-4 4 -
c).
Report TR-5639-2 contains an ASME Class 3 analysis of downstream portion of PORV line which includes piping flexibility analysis for five (5) basic loading conditions or cases:
1.
Deadweight l'
2.
Thermal 1 (a 650*F/450 F case with thE piping downstream of the PORV at 450 F),
y 3.
Thermal 2 (a 650*/120*F case with the piping A
downstream of the PORV at 120 F),
(
4.
Seismic OBE, and 5.
Blowdown (at 500*F).
This analysis includes the following ASME Code stress analysis checks:
NC-3652.1 Sustained Loads, which considers the effect T
l of pressure,. weight and other sustained mechanical i
loads.
~
Eqn (8) = PDo + 751 Ma $ 1.0Sh l
4tn z
1 Ma due to weight, load case 1, above considered.
i i
NC-3652.2 Occasional Loads, which considers the effects of pressure, weight, oth'er susthined loads and occasional loads including earthquake.
1 Eqn (9) = PmaxDo + 0.75i (MA+MB) $ 1.2 Sh 1
i 4tn z
M due to weight,' los,d case 1, above, considered.
A M due to occasional: loads, such as earthquake, load B
i:
case 4, above considered.
NC-3652.3 Thermal ~ Expansion, which considers.the
'/
y s.
i - '
effects of the range of piping thera l expansion
~-
s
,j,3 J
s Eqn (10) iMc $ Sa
)
i z,1 i
Mc is the range of thermal-expansion moments, load y
cases 2 and 3,.atnv,
9 4,
T s.
y s
'}
y.
b:
8.{ */
.g
,y 1' %
O
~
x
)-
~,
-~
,,,t.-,.
...--,.....,,..-..m.
.....--,-,,-,,,~+..-_,.-#,.~.,...
__...m.
m In addition to these Class 2/3 piping analysis checks, two Class 1, Subsection NB checks were made, i.e.,
NB-3652 Primary Stress, which in effect duplicates the Class 2/3 equation (8) check.
Eqn (9) = B1PDo + B DoMi 5 1.5 Sm 2
2t 21 and an NB-3652, Emergency Conditions, primary stress check Eqa (9) = B1PDo + B DoMi 5 2.25 Sm 2
2t 2I d)
Comparing the load combinations and stress checks made in TES report TR-5639-2 with EPRI PWR Safety and Relief Valve Test Program Guide recommendations, it is l
our-observation that TMR (TES) report E-1495-17, Revision 1:
1.
Considered for Class 1 piping, all specified normal, upset, emergency and faulted conditions and compared calculated stresses with the appro-priate service limits, i.e.,
the 1971 Edition of the ASME Code; 2.
blowdown events in the SRV and PORV lines were considered in the Class 1 fatigue analysis and were conservatively called a normal or upset condition; and 3.
that the intent of the requirements of the EPRI Guide, Table I have been met with the performance of the ASME Class 1 piping analysis.
It should be noted that TES report TR-5639-2 does not perform a complete Class 1 piping analysis. Also, the Class 2/3 analysis were performed in accordance with Code rules which predated the definitions (of normal, upset, emergency and faulted conditions for this piping class) used in Table 2A of the EPRI Guide.
4 J
1 1
a Question 11.
The submittal on structural analysis states that the nozzle-to-flange weld of the safety valve and the flange below the safety valve have acceptable stress values.
Provide results from the analysis that support this statement.
Response
Pages 144 and 145 of TES report TR-5639-2 calculates the Eq. 9 stress at the pressurizer nozzle safe end to safety valve flange (structural node 510) due to weight, seismic and safety valve blowdown. The stress was found to be 21,350 psi and the allowable 25,030 psi.
The safety valve nozzle stresses in the pressurizer shell are calculated on page 148 of the report.
While these stresses are not compared to an allowable, the largest stress is 1,556 psi which is small enough to be considered inconsequential.
Question 13 The submittal stated that the safety valves are directly mounted to the Pressurizer nozzles. No piping supports are used on the safety valves or inlet piping. The safety valves discharge directly into tees with their ends closed by rupture discs. The blowdown loads due to a safety valve lift do not appear to be included with the other considered piping loads. Provide additional information addressing the dynamic effects of the blowdown discharge (torsional moment on nozzle if rupture disc burst pressures not equal) and the effects of valve discharge impingement on the tee wall (including of a bending moment on the nozzle) to verify nozzle structural integrity.
Response
Page 145 of TES report TR-5639-2 calculates the possible torsional moment if the rupture disc pressures are not equal. This torsional (My) moment was determined to be 46,648 in-lbf. This blowdown moment was combined with the deadweight and seismic moments on that same page and compared to the eq. 9 allowable of 1.5 Sm, the calculated eq. 9 stress was found to be 21,390 psi and the allowable is 25,050.
The safety valve nozzle bending moments are tabulated on page 151 of the report. These moments are compared to those calculated in the 1973 report TR-1495-10a. The 1973 moments are used as the design moments.
The My bending moment from page 145 was inadvertently neglected from this table. When the My moment from blowdown is added to the My seismic moment of 29,540 in-lbf, the total is 76,185 in-lbf. This resultant My moment is much less than the 120,638 in-lbf moment which is the My moment calculated in the 1973 report.
Templ/E02.