ML20206F246
| ML20206F246 | |
| Person / Time | |
|---|---|
| Site: | Maine Yankee |
| Issue date: | 03/31/1987 |
| From: | Whittier G Maine Yankee |
| To: | NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM) |
| References | |
| RTR-NUREG-0737, RTR-NUREG-737, TASK-2.D.1, TASK-TM GDW-87-79, MN-87-41, NUDOCS 8704140192 | |
| Download: ML20206F246 (42) | |
Text
--
MAIRE HARHEE ATOMICPOWERCOMPARUe avavgray,?u"e o",l, (207) 623-3521 e
March 31, 1987 MN-87-41 GDH-87-79 United States Nuclear Regulatory Commission Attention: Document Control Desk Washington, D. C.
20555
References:
(a)
License No. DPR-36 (Docket No. 50-309)
(b) USNRC Letter to MYAPCo. dated December 31, 1986.
Request for Additional Information, Item II.D.1 of NUREG-0737, Performance Testing of Relief and Safety Valves
Subject:
Request for Additional Information, Item II.D.1 of NUREG-0737, Performance Testing of Relief and Safety Valves Gentlemen:
Attachment A contains responses to five of the 12 additional questions, Reference (b), on Maine Yankee's II.D.1 submittal.
During our 8 January 1987 telephone conference with members of the NRC staff, we were informed that questions 1, 3 and 10 required no response.
As discussed with Mr. Pat Sears on 20 January 1987, our response to the remaining questions 2, 4, 6 and 8 is anticipated to be submitted in late June 1987.
Please contact us if you have any questions concerning the enclosed information.
Very truly yours, MAINE YANKEE ATOMIC POWER COMPANY kh OV G. D. Whittier, Manager Nuclear Engineering and Licensing GDW/hbg Enclosure cc: Dr. Thomas E. Murley Mr. Ashok Thadani Mr. Pat Sears Mr. Cornelius F. Holden Q
8704140192 870331
'h l
i PDR ADOCK 05000309
\\'
8571L-GDW P
ATTACHMENT A l
RESPONSE TO ADDITIONAL QUESTIONS ON THE MAINE YANKEE NUREG-0737, ITEM II.D 1 SUBMITTAL OVESTION 5 Dresser, Inc., in March 1976, recommended to Metropolitan Edison Company that the PORV block valve be closed at pressures below 1000 psig to prevent steam wirecutting of the PORV seat and disk.
Testing by Dresser later showed the 1000 psig pressure limit to be overly conservative and that the PORV as designed was qualified to system pressures of 100 psig.
Below 100 psig the deadweight of the lever on the pilot valve was sufficient to keep the pilot valve open. Dresser recommends, if the plant is to operate at pressures below 100 psig, that heavier springs be used under the main and pilot disks to ensure closure. Hithout the heavier springs recommended by Dresser, the PORV should not be used at system pressures below 100 psig. What is the minimum operating pressure at Maine Yankee? If the minimum pressure is below 100 psig, has MYAPCo installed the heavier springs in its PORVs so operating at this pressure is consistent with Dresser recommendations?
RESPONSE
The PORV block valves are not closed to isolate the PORVs from the reactor coolant system at system pressures below 100 psig. Based on previous discussions with Dresser, the heavier replacement springs are only required if leakage and wirecutting of the valve parts has been experienced at the lower system pressures.
Therefore, unless leakage is a problem at low operating pressure, replacing the springs is not warranted at Maine Yankee.
OUESTION 7 Provide more information on the verification of STEHAM, HATAIR, and HATSLUG by Stone & Hebster.
Provide comparisons of the results for STEHAM, HATAIR, and HATSLUG calculations and EPRI/CE data to verify these codes are appropriate tools to evaluate piping discharge transients.
RESPONSE (Stone and Hebster Engineering Corp.)
Attachments 7A, 78, and 7C contain general descriptions and program verifications of STEHAM, HATSLUG, and HATAIR respectively.
RELAP 5/M001 and EPRI test results are used to assist in the verification of both the STEHAM and NATSLUG computer programs.
Hand calculations are used to verify the HATAIR computer program.
8571L-GDH
ATTACPMENT 7A STEHAM 1.
GENERAI. DESCRIPTION The purpose of STEHAM (Ref.1) is to determine forcing functions on piping systems during reeamhammer transients for subsequent input to piping dynamic analysis.
The analysis is based on the method of characteristics with finite difference approximations both in space and in time (Raf. 2).
It calculates the one-dimensional transient flow responses and the flow-induced forcing functions in a piping system caused by rapid operational changes of piping components, such as the stop valve and the safety / relief valve. Flow characteristics l
of piping components are mathematically formulated as boundary conditions in the program. These components include the flow control valve, the stop valve, the safety / relief valve, the steam manifold and the steam reservoir.
Frictional effects are taken into consideration in this program.
i 1
i STEHAM accepts the following as input: (1) the flow network representation of the piping system, (2) the initial flow conditions along the piping system, and (3) time-dependent flow characteristics of piping components. Output consists of time-histories of flos pressures, 21ov densities, flow velocities, inertia, and acaentum functions. Torces are written on tape for direct input to NUPIPE-SW (ME-110).
2.
PROGRAM VERIFICATION The STEHAM model and a piping schematic of the sample probles are diagrassed in Figure 1.
STEHAM is verified by comparing the solution of this sample problem to the results for the same proeles obesined by an independent analytical approach (REI.AP5/ MOD 1, Ref. 3) as shown in Figures 2 6 and by comparison of i
predicted (per NUPIPE-SV, Ref. 6) versus measured support reactions (from EPRI
~
2.
test 1411, Ref. 3) as shown in Figures 7-10.
The input for this problem is indicated in Table 1.
A second problem is defined in Figures 11 and 12 (Ref. 4 and Ref. 5).
STEHAM pressure time-history results are compared to analytical results (Figures 13 and 14) and experimental asasurements (Figure 15). Nedal force results are compared to hand calculations in Table 2.
t The STEHAM generated forcing functions and nodal pressures compare favorably with the RELAP5/ MOD 1 predicted forcing functions and nodal pressures.
3.
REFERENCES 1.
"Steauhanumer Analysis for Piping Systems", ME-167, (STEHAM)
Version 02, Level 01, created 83.047 by SWEC.
2.
" Verification of ME-167 STEHAM Computer Code", SWEC Cal. No. 574.554.1-NP(3)-
090-FB, by A. Esi and L. Loveridge, dated 3/7/83.
3.
" Application of REI.AP5/ MOD 1 for calculation of Saf sty and Relief Valve Discharge Piping Hydrodynamic Loads", Interia Report, March 1932, by Intermountain Technologies, Inc., Idaho Falls, Idaho, Project Laager R. K. House.
I 4
Progelhof, R. C. and Owczarek, J.
A., "The Rapid Discharge of a gas from a Cylindrical Vessel through a Nozzle". A1AA Journal, Vol. 1, No 9 Sept. 1963, PP. 2182-2184
3.
\\
5.
Progelhof, R. C. and Owezarek, J.
A., "The Rapid Discharge of a Gas from a Cylindrical Vessel through an Orifice", ASME Paper No. 63-WA-10.
6.
NUPIPE-5W, ME-110, V03, L14 (created 82.095), " Computer Code for Stress Analysis of Nuclear Piping".
t i
t i
i i
- - --,,--- ----..-.. -_-, -~ -
TABLE 1 INPUT DATA FOR STEHAM
- t TOTAL INSIDE FRICTION PIPE NO.
LENGTH (Fe)
DIAMETER (Ft)
FACTOR 1
16.125 0.408 0.0087 2
12.563 0.5054 0.0059 3
63.562 0.948 0.0077 VALVE CHARACTERISTICS I
0FFICE OPENING DISCEARGE FLOW 2
i ARIA (Fe TIME (Sec)
COEFFICIENT RATE (lbs/hr) 0.0253 0.011 0.94 440,000 i
i UPSTREAM STEAM CONDITIONS PRESSURE PRESSURE (PSIA)
DDSITY (
)
RISE RATE (
)
2410.
7.156 262.5 DOWNSTREAM GAS CONDITIONS i
PRESSURE (PSIA)
TEMPERATURE DESITY (
14.7 70 F (530 R) 0.075 j
i
- l l
NOTES:
- SEE FIGURE 1 FOR SEETCE OF SIEEAM MODEL i
i i
1 i
TABLE 2 FODAL_ZQFCE COMPARIsotr Diameter D=0.25 ft Area A = rDa/4 = 0.0490374.ita Nodal Force = (p +eV afg ) A - p,,,A I
p=
pressure 1b/fta i
i a=
dansity lb/ft8 V=
velocity ft/see g=
gravitational constant 32.2 p,,, = 14.7 x 144 lb /fta At time t = 0.00650 sec.
i Force (lb)
Node Pressure Velocity Density Force (lb)
(Hand fosial (fos)
(1h/ftal LT! TRAM)
Cale.:lation) j 1
a2.523 0.0 0.23954 186.57 186.67 i
5 a2.785 5.7843 0.24076 198.43 198.53 l
10 a4.231 31.219 0.24647 209.00 209.11 15 47.003 78.172 0.25737 230.62 230.7)
I 20 50.214 129.89 0.26979 257.84 257.97 25 52.095 159.43 0.27697 274.93 275.06 30 52.209 161.97 0.27742 276.09 276.23 35 52.168 162.21 0.27731 275.83 275.97 I
i 1
l l'
l l
l J
1 i
)
i STEHAM MODEL OF EPRI SAMPLE PROBLEM i
i s~. ~
SAFETv/atteEF EMPAm8ER OsHFaCE PLATE v&Lvf e
reassumina le wasto j
j eva vgi ic]
svuw is safsvei l
LEGEDID tourestar arsva CEL PIPIleG SCHEMATIC OF EPRI SAMPLE PROSLEM N00E soumers
(,.8I enAncu n
i mussaan W
i (s.si oenecisons i
i,.,i
- 1. i t 1
.mc...
II*II SEGENT 80s.
I II*II gg gy g 2
I f
,, i IB M
is.si (3.,i e-i FGME g 4
l
O
-U S,
m U
8 8w
.m mW W
- 63 W3W wum g
_ 6 Na ee 2I O-f Y O I
I l
9 t
O 4
I
)
Ek w a EE 4
0 I
/
I E
I s'
W I
a I
I 8
8 8
g e
a a
(egl) 33gg lyg3gggg
z9-U33 S
E i
8 8 m
=
5~
,,, 9 >
IE E w
W E
3 e
w sum g
~
o I
U i
st 1
0 l
i O
/k) n2 14 A
-s E
EI
)
,/
I
./
m i
w. - - = - S. - - - -%
g 8a i.
1 1
i.
1 O-(ett) 33uod 21N3MS3S
u a
m.
.a O
5-S w
I 8$
m M
o a s TE2III a
O 1
Ea
. og g
u
.I.
I ey l
I l
I I
l I
c o
I e
i i
i 9
i o
I i
'l
.I:s.
i a=
g s
3 l
N 5
I 8
2 I
I I
r
('90 33W4iAN3NO38 di 4
1
O W
5 t
1 l
- i I
lii i
.,. W I
W I
n 3
5 i
m I
I l
l 1
a I
o i
I I
l ji i
u i
E I
21 0%
I I
I I
S t
I I
I I
I e
i I
4 4
\\
ga 5$
s a
s 1
s, ** % g
- l 9
1 g
I i
i i
-E 8
l 8
8 8
8 4
8 I
8 8
x Osd)3unss3vd f
--7
w I
g I
l 4Iu lO g
35 i
l ErE i
g I
a I
l0 I
.5 t
l*
k 2
l gwv I
i t
21 I
1 i
s I
I 2W
\\
ae
\\\\
\\
j
\\
\\
\\
\\
\\
g I
\\
3 i
1%
1 Ii I
8 g
I g
l W
l w
g i
i i
a 8
8 3
3 g
g o
I I
E 3
g use)3mnssaud
3,000 2,000 -
Soootts I
soo Les usasunto
[i
,'l t.oco -
l l
l l
f', k Ah i
nx o.o f
h5? Y!#
U l N
'd S -i.oco,
i I
I I
-a,ooo
-a,ssoLes usasunto
-3,000 -
j
-4,000 -
0.00 e.to o,30 o,so n,,,
TIME (SEC)
Lastmo murist -se
EP*i it sf as sutts l
Ft0UME 7 COMPARISON OF SEGMENT I SUPPORT REACTION
s,oco
-esoo s. MC ASU,to i
a,oco --
/
7 "'i an'. "O ^:..
.13 in'r'io,", l",,1.
i 2
E l
i i
5 lI ioco i
ll ikf h M
i A
a i
i oo V
'l i l]',h'fi[/Y l
l ll l
lI g i
.ooo llI I
I I
s l
I
-...t...u.u...
,,o,o _
=== =4roo LBS
.oo o.to 0.to
- o. 3o
...o 0..o TIME (stc)
LittNo NuptPt * $ w
_ _ _ o., r e,..uu.
COMPAR SON OF SEGMENT 2 SUPPORT REACTION
l
- THESE FORCE PtANS AND THE ll\\
TlWING DELAY OF THESE PEANS IS Out TO PR00ASLE f
WATER REPORTIO IN l{
SESMENT S l1 s,00o -
lI
. 00 Les
]
l l
I l
q\\
i I
I lg i
5 A000 -
h 0'I l
it 4
d
!. ' k nl\\
in' s
A /\\ /\\
!0 A
i l [ \\l MY \\WV V '
a
\\1 V
b I
II
\\/
3 f
N s I Il y
- 4,000 -
f I
i il 0
l I
\\l S
I il 11
-a,000
- 000LeS y
b-9000L88 MEASURt0
-iS,000 0.00 0.10 0 20 0.30 0.40 0.50 TIME (stc)
LastNO NUPIPC-SW
--- EPRI f t ST RtSULTS FIGURE 9 COMPARISON OF SEGMENT 3 SUPPORT RE ACTION
15,000 14,000 l
I
)
I I
l jl$
l TME FORCE PE ANS AND TME TIMING DELAY OF TMESE PEAMS l
12,000 IS DUE TO PROSASLE WATER lI REPORTED IN SEGMENT 3 1I II ll PREDICTED
- 10,000 OL0w00wN PORCE ll
=
lI k
i S
i l
s
,ix
.000 v
i I
I l
l MEA $unto I
I RCE j
g g
l lil 6,000
=
1 i
. IV T
l l
g I
gi
- eLowo0wN FORCE,Pg l
lll 4,000
=
p p
34, a W
wMERE f
I P ePRt99uRE ATExif
/
- g. AmeiENT PRassuRt
]
g A e AREA 0F Exl?
e e OEN$ TTY AT EXIT 1,000 g
g{
=
l l
v e vg60 CITY AT EXIT l
h fl 1
Li \\
i J t!
l t
0.0 y
,I I
l i
l
$.00 l
~
0,10 0.20 0.30 0.40 0.50 ListND TIME (SEC)
NuPIPt-sw FIGURE 10
--- cPRi Tas? ResuLTs COMPARISON OF SEGMENT 4 5UPPORT RE ACTION l
l
~...~-.
Ag A, GAS I
Po p
L CLOSED OPEN ENO END CASE (A) FOR COIEPARISON WITH ANALYTICAL RESULTS.
WITIAL CONDITIONS:
P /P = 4.72, s / o = 0.80 g
o g
DeteENS40NS:
A,/Ag = 0 6. L = PIPE LENGTH SPECIFIC HEAT RATIO:
Y = C,/C, = 1.4 p = PRESSURE, e SOUND VELOCITY, A = FLOW AREA FIGURE 11 SUDDEN DISCHARGE OF A GAS FROM A PSPELINE THROUGH A NOZZLE
_ ~. _ _ __.-
A A, = O 25A; GAS
+
/
3 28 p
T :79*F
[
p p :0729p; g
f l
To:79*F
- - 5FT
--+
CLOSED OPEN END ENO CASE (B) FOR COtrARISON WITH EXPEResENTAL DATA AfO HAND CALCULAT8001 PRESSURE = p : 14.7 psio, p, = I4 7/O 229 : 64.2 psio AREA =Ag:
10 25)2: O 0491 f s, A, = 0 25 A; =
2 0 0123i82 GAS COttSTANT : R = 5 3 35 It - lb g /lb'R TEaSPERATURE : T : 79'F = 5 39* R, T, = To o
p, = [' = 0.321516 /f t 3
3 DENSITY 2 7 =
= 0 OT36 th/ft FIGURE 12 SUDOEN DISCHARGE OF A GAS FR000 A PtPELINE THROUGH A NOZZLE I
9 W
8 w
$aW w
mW-O 8
'W ge w
5=N L4 8
i5 h
$S*
e80 cv=
h 8
3 8.-
. 'w
/
8 e
m 8
~
J e-E
!i 8
a 8
i e
i i
4;
_ 3 C" 8
8, 8
8 8
8 "s<
u a
e e
u 5-*5
('d/d) ON3 03S010 AV 3WASS3Wd 9
11 a
b l
l
8 l
~
W 9
8 l
' ~
\\
m k@
d 8 8Wx g
- I mE l <{
w
[
8 w
d{
I 3
81
- 2 W
l k-8
)
/
e 8
a u
b
~
5 7
l 8
h l
a 1
a l
5
- I$
a-w i
t 8
3:*
- 8 8
2 9
3 Sa 9
a
~
o o
I
( d/d) ON3 N3dO IV 3WASS2Wd U
l
9 I
8
!o E
I l
h I
a w
/
a 0 3,
/
CD
/
8 $e E
=
I f
w u a:
a 8
I 1
d y
8-r, -
i 2
8 I
~
e5 8
b e
5 I
E 8
i f1 h
Q 2
g' 5
c a
O
=K W
g 3 03S01317 3WrtSS3Wd he 9
I I,
8 s
ATTACIMENT 7B WATSLUG i
1.
General Description The purpose of WATSLUG (Ref. 1) is to determine forcing functions on piping systems during water slug discharge events for subsequent input to piping dynamic analysis.
1 The analysis is based upon rigid body motion of the generally subcooled i
water slug and ideal gas representations of the steam or air using rigid column theory to facilitate tracking the several water-steam or water-air interfaces. The driving force is the steam pressure between the valve and the slug, less friction and other losses, and back pressure. Density changes due to possible local flashing of the water slug are considered.
Having recourse to the control volume theory, the subsequent segment force calculation is carried out.
The input consists of complete piping system geometry, pipe dimensions, valve flow characteristics, valve opening time, detail upstream steam conditions, and initial downstream steam or air conditions, while the output contains forcing functions for each piping seguent based upon flow velocities, pressures, and densities during the water slug discharge event. Torces are written on tape for direct input to NUPIPE-SW (ME-110).
(Ref. 2).
2.
Program Verification WATSLUG is verified by compering its solutions of a test problem to the results for the same problem obtained by an independent analytical approach (RELAPS/ MOD 1, Ref. 3) shown in Figures 3A.3.A-1 and 3A.3.A-2.
NUPIPE-SW (ME-110) generated responses due to the WATSLUG forcing functions on the supports were compared with experimental data from a test run of this probles (EPRI Test 908, Ref. 3) shown in Figures 3A.3. A-3 and 3A.3. A-4 The WATSLUG aodel of the test problem is diagrammed in Figure 3A.3. A-5.
The NUPIPE-SW model is diagrammed in Tigure 3A.3.A-6.
The WATSLUG results and the NUPIPE-SW results generated from the WATSLUG forcing functions compare f avorably with the RELAP5/ MOD 1 results and the EPRI test results.
3.
References 1.
"WATSLUG" (MI-212) computer code by J. S. Hsieh and D. A. Van Duyne, Ver. O Rev. 3 - 12/82 and the related documentation calculation
$76.470.1-NP(B)-038-FD, Rev. 2, " Water Slug Discharge in Piping System (WATSLUG) - Preproduction Version 3", - 3/3/82.
2.
NUPIPE-SW, ME-110. V03.L14, created 82.095.
- 3. " Application of RELAPS/ MOD 1 for calculation of Safety and Ralief Valve Discharge Piping Hydrodynamic Loads" Interia Report, March 1982,'by Intermountain Technologies, Inc., Idaho Tails, Idaho, Project Manager R. K. House.
i j
H
TABLE 3A.3.A-1 INPUT DATA FOR WATSLUG TOTAL INSIDE FRICTION NO. OF PIPE NO.
LENGTM (Fe)
DLutETER (Ft)
FACTOR SEGMENTS 1
16.125 0.408 0.015 6
2 12.563 0.5054 0.315 2
3 63.562 0.948 0.013 3
I VALVE CHARACTERISTICS ORIFICE OPENING DISCHIJtGE FLOW l
AREA (Ft )
TIME (See)
COE7FICIElfr RATE (lba/Te )
0.0253 0.015 0.805 120.83 UPSTREAM STEAM CONDITIONS i
PRESSURE PRISSURE (PSIA) TEMPERATURE DENSITY (
) RIU P. ATE (
2690.
1139 R 8.862
-40.
I DOWNSTREAM GAS CONDITIONS i
DENSITY (g PRESSURI (PSIA)
TEMPERATURE 1
15.
540 R 0.09975 WATERSLUG WEIGHT = 69.768 Lbs.
i l
TABLE 3A.3.A-2 INPUT DATA FOR NUPIPE-S'4 CUTOFF CUTOTF INTEGRATION MODE FREOUENCY TIME STEP TIME DAMPING RATIO 53 433 H 0.0009 Sec.
0.5 Sec.
10%
g PIPE TOTAL OUTSIDE SECTION LENGTH (Ft)
DIAMETER (IN)
THICENESS (IN)
WEIGHT (1b /Ft) 1 4.73 8.625 0.906 74.71 2
12.31 6.625 0.864 53.16 3
12.43 6.625 0.28 18.97 4
69.0 12.75 0.688 88.60 5
1.1 12.75 1.5 6
1.0 8.625 0.322 28.55 7
0.83 6.625 0.432 28.57 0
= YOUNG's MODULUS OF PIPE = 28.3 x 10 PSI EHOT " ECO2 1
STEAM O
Le P
IXI a
GAS a
1 0
if O
WATER
, Le PRESSURIZER 2
VESSEL db
-LF 4
IF N
]
j 3
l i
DISTANCES PROM PRES $URIZER VESSEL:
Lo a SAFsTY/ RELIEF VALVE LG : DOWNSTREAM AREA CHANGE LF s PIPE EXIT IXt TRAILING EDGE OF SLUS 1 4-= FORCING FUNCTION DIRECTION, +
EPRI TEST 908
(, etaW. "& )
SEGMENT NO.
WATSLUG MODEL PIPE NUMBER I PRESSURIZER TO Lo i
2-Le TO LG 3 * 'O IO
FIGURE 3A.3. A l WATSLUG MODEL OF EPRI i
SAMPLE PROBLEM I
4 T
o R
N O
I R
P 5
N 5
4 5
P O
2 5
3 P
0 2
U
/
E I
4 t S
\\ 4 T
0" W3 F
C 2
g'2 E
5~
M O
S 5
VE E
I i 2 2
W L M P
E P
E D
L 0
L 2
2O B i
E R
D 0
2 M O 2 a A
@5 E
O T
R e
M s
- 3. W a%2m@
M M
P w
l a
A S U o S
N P 5
0 3
E 0
9 E
E L 3
E S PI E P P o
D S O A P
o RI
_02 UPM U
S 0
D N M N
N 9i GUA 2
8 IFNS a
s E
1 G oA O
E 5
L 9i l
'4 1
l 3
5 0
[4 91 1
5 3
7 e1 2
s T
2 R
l o
O 2
s1 P
i P
3 l
U 5
3 0
S u/
5 5
5 7
4 o
5 08 a2 0
2 5
e 2
s 6
O yL wE 7
3 1
I 1
i 1
1 W
g 9
- 5 2
/
5 8
l 2
E 2
1 I
W T
l T
R gk R O
O P
P P
d
[k U S
P U
S 1
3 o
/
s 5
o 0
6 s
d 3
t 4
5 E
l 7
0 W
5 5
o 5
4 t
0 3
/w A
'5 3
o, e 2
8 2
i l1
- I l
,l j1i
- 'ii!.
il
4 0
s 2
TN E
M GEN SO I
FT 3 OC 3
N 0 A N U
- 3. S O F A
I 3 R G EAIN RPC UMR G OO I
FCF
)
C E
S 2
(
O E W
Ill i
l l
lllItI l
iiiiIg l! I l
l I
I O
1 D
i i sI O
M iI
/
G 5
UL P S A i
T L
J A E W R D
N m
O E
O G
0 0
0 0
0 0
E 0
0 0
0 L
0 O
p 0
0, 0,
0, 0
0 O
0 0
5 0
5 5
m 1
2 7* wee * $w38"
4 0
3 T
N I
E M
G EN SO I
FT 4OC N
3 N N A. O I
0 U
- 3. S F AI 3R G N
I I OO FCF f7
)
/ /
C 0
ES 2
(
l
\\
I C E M
T I
l
/ *
\\I
\\
g iIill1IllIl I
l tll II81111IiIllll 1
I 0
1 OOM
/
G S
U L
P S
A
^
T L
I A
E W R D
N 0
E O
G 0
0 0
0 0
0 0
E 0
0 0
0 0
0 0
L 0,
0, 0,
0, 0,
0, 0
0 O
0 0
0 8
3 e
3 2
7 1
1 2= wo@L m $w2Om l
4 i'
l 8
.l 11 1
1
180,0 %
}
EPRI PEAK so,ooo A
\\
J ALeh_A c_
^-
o r
N r--
g C
2 9
U
-80,000 m
2 n.
S
-isopoo g
l-2 O
m
-240400 EPRI PEAK
-320,000 I
'U o1 0.2 0.3
- 0. 4 TIME (SEC)
LEGEND NUPIPE-SW FIGURE 3A.3.A-5
EPRI TEST RESULTS COMPARISON OF SEGMENT 2 SUPPORT REACTION
i
.oo.
l f
f 40,000 f,4 2
l O
o
\\
I \\
G I
I\\
g A
f l
, nan-vi j
i j-vv 1
I i
l i
=o
\\
l I
\\ I I
I P/
lI o
I i
I
- 40,000 i
's U
\\
2 i
Oe
-80000 l
l 01 0.2 03 0.4 TIME (SEC)
LEGEND NUPIPE SW
==== EPRI TEST RESULTS FIGURE 3A.3 A 6 COMPARISON OF SEGMENT 3 SUPPORT REACTION
ATTACHMENT 7C l
WATAIR 1.
General Description The purpose of WATAIR is to determine forcing functions induced on piping systems by a hydraulic transient with trapped air (such as water discharging into empty piping) for subsequent input into piping dynamic analysis.
The analysis used rigid column theory to calculate fluid accelerations and velocities, the first law of thermodynamics and the ideal gas law to calculate air properties, including pressure, and the control volume theory to calculate the unbalanced fluid forces on the pipe, segments.
j The input to the program consists of dimensions of the piping system, frictional coefficients of the pipes, valves and fittings, valve characteristics, and the operating conditions of the flow networks.
The ouputs consists of inertial segment forces, positions, and pressure of the air pocket, velocities, and accelerations of the water slugs at indicated time increments of subsequent input to NUPIPE-SW for piping dynamic analysis.
2.
PROGRAM VERIFICATION The WATAIR Program is a computer code developed by SWEC. This program determines the dynamic forcing function induced by the hydraulic transient with trapped air.
Reference 1 presents the hand calculations of the parameters in the program to verify the adequacy of the program.
The comparison of the hand calculated values and the WATAIR Program values showed excellent agreement and thus, the program was verified to be adequate for this analysis.
The verification results are documented in the Reference 1 calculations which is available for inspection at the offices of Stone & Webster in Boston.
3.
REFERENCE 1 SWEC Calculation 13411.16-NP(B)-002-FD, Rev. 1, Pages 126 - 150
f ATTACHMENT A (Continued)
MN-87-41 OUESTION 8 Insufficient detail was received on the key parameters used in the STEHAM, HATAIR, AND HATSLUG thermal-hydraulic analyses.
Provide node diagrams of the thermal hydraulic models.
Provide information on the node spacing, time step size, and choked flow locations used in the analyses.
Discuss the rationale for their selection. Compare the values of these input parameters for the verification calculations requested in Question 7 to the plant specification analyses. Justify any differences. Since the Dresser 31739A valve passed in excess of 118% of rated flow, justify use of 111% of the ASME rated flow for the safety valves in the thermal-hydraulic analyses or provide the results of thermal-hydraulic and structural analyses which account for the larger flows seen in the tests. Also, the flow rates used for the 31709KA valve in the thermal-hydraulic analysis, 228,000 lbs/hr, is less than the rated flow listed for the 31709KA valve in the EPRI " Guide for Application of Valve Test Program results to Plan-Specific Evaluations," Rev. 2, Appendix B.
The flow rate listed was 233,000 lbs/hr.
Resolve this difference in light of the fact that MYAPCo stated steam flows were adjusted to represent 111% of rated flow.
Compare the rated and test measured flow rates for the Maine Yankee PORVs.
If the test flow exceed the rated flow, provide the same information for the PORVs that was requested above for the safety valve.
RESPONSE (Partial)
The node diagrams for STEHAM, HATAIR, and NATSLUG thermal-hydraulic analyses are shown in Figures 1-3.
Time steps used in the STEHAM program are determined by the node spacing and sound speed to satisfy the Courant stability criteria.
Since the minimum node spacing used in STEHAM was about 2.1 ft, the time steps were in the order of one millisecond. Choked flow locations usually occurred at area changes such as a reducer, valve or common header.
These are detected by the STEHAM program as appropriate.
HATAIR and HATSLUG use rigid column theory to calculate water accelerations, velocities, and water / gas interface displacement. Air or steam is treated as an ideal gas. The approach is a lump-parameter method rather than a nodalization scheme. Therefore, node spacing is not required.
The time step used is about one millisecond. Choked flow locations are detected at the relief valve or the exit.
Since the minimum valve opening time is 0.015 seconds, it is believed that one millisecond time steps used for all computer codes is adequqate for the development of forcing functions. This is the same order of magnitude of time steps that were used on the verification problems.
Table 1 shows the fluid conditions and critical parameters assumed for the analysis.
8571L-GDH
STONE & CEBSTER ENGINEEQlNG CORPORATION CALCULATION SHEET
& !C10 68 C ALCUL ATION IDENTIFIC ATION NUMBER J.O. O R W.O. NO.
OlvlSION & GROUP CALCUL ATiON NO.
OPTIONAL TASK CODE PAGE v c 0 9--
F t6 se 1
Stem ^A g(-
5l
@@f"
'~
g i3 J,@ ;
.@,; G m 2
3 2
e, o
o o
o e!k
@~-
@ f,
~
@r e
i j
s OJ 4
4 g
N 21 O
g 22 m
\\P O
n 4
- m W
}
o 27 y
2.
V y
y 2,
19 2
4"
,C m
8 I-O e @><
~
d e
8' 8
5 ao s
s h
0 e
- 2 o
s O
,4 5
DOS 53 YbSID YSIM DAS3D
STONE & CE8 STER ENGINEECING COilPORATION CALCULATION SHEET C ALCUL ATION IDENTIFICATION NUMBER J.O. O R W.O. NO.
DIVISION 0 GROUP CALCUL ATION NO.
OPTIONAL TASK CODE PAGE 3
-,n
- --- n-S (LToP) 7 10
'I kNMd hELSEf VbbNk 13
'H////li
( 3 // m //DX
/
17 8
m> m 20 21 1
I%t 23 24 25 L
j 2.
27 2.
29 30 36 22 f%s s
NmAb Pos m ora on
- WATER, 33 Po smo ra es:.
R.s u EF VAuvE 5*
Lt s
=
79 mat Po ss Tiera CF 2.so Warm 6 ag 35 IM.
37 L
Tomt.
L.tra c,Ts C F-P se trag 3,
40 4
42 43 EssoRE 1 4s 4
STONE & CESSTER ENGINEEQlNG CORPOR ATION CALCULATION SHEET
& 5010 SS CALCULATION IDENTIFICATION NUMBER J. O. O R W.O. N O.
DIVISION 0 GROUP CALCUL ATION NO.
OPTIONAL TASK CODE PAGE i
2 3
WAT st 04 MODEL a
5 6
7 e
9
)
ao si PRSS6&ttiit REwSF vas.vE I
/
'a l
13 h///////
it
)
l l
l
/
is it is i
'S O
i m
m to to 22 N4 23 l '
w j
RS Lp a6 at to 29 30 3'
lo Loc,Amord t"
VAbN'E s
33 v.
t,o cats oeJ Of8-Cov4r4 STR,EAM PsPS AREA c%i 33 34 55 LF TD rN Pges LEq TH 36 37 3e 39 40 ai 42 F\\ c.iaRE S
+3 45 46
t i
l TABLE I (question 8) i i
j Fl.UID CONDITIONS AND CRITICAL PARAMETERS ASST 4ED tolt TIIE ANALYSIS y
Case 4 Case 5 Case I Case 2 Case 3 Transient case 2 relief salves 3 safety valves 3 safety valves LTOP NpSI open descharge open in sequence open in sequence 2 relsef valves 2 relief valves steam discharge stema discharge steam open descharge open discharge with 2 relief Walet, water water, liashing I
valves already f ront s cashine occurs at valves at common peping two phase flow open downstreme Computer Code STENAM STEllAM STENAM WATAIR WATSLUG Used STENAM Valve Opening 0.1 sec 0.015 sec 0.015 sec 0.07 0.06 Tsee (t)
(I)
(t)
(6)
(6)
Flow Rate / Valve 156,420 lb/hr 228,000 lb/hr 228,000 lb/hr 388,000 lb/hr 545,000 lb/hr steam at 2350 pssa steam at 2535 psis steam at 2535 psig water water (2)
(4)
(4)
(6)
(7)
Valve Set 2345 psig ist valve - 2485 psis Ist valve - 2485 pass 485psig 2400 psig Pressure 2nd valve - 2510 psig 2nd valve - 2510 psig 3rd valve - 2535 psig 3rd valve - 2535 psas (2)
(8, 41 (1,4)
(2)
(7)
Peak Pressure at 2464.3 psig 2574.3 psig 2574.3 psis 549.3 psig 2400 psig Valve anlet (3)
(5)
(5)
(5)
(7)
Pressurization 25 psi /sec 63.1 ps:/sec 63.1 psi /sec 12.5 psi /sec 0.0 psi /sec j
Rate (3)
(5)
(5)
(5)
(7)
Back Pressure 2 psag 2 pssa 56.3 psig 2 psig 2 psig Temperature 670*F 670'F 670*F 470*F 568*F (7)
Nel sef Valve Bore De eert er is 1 5/16 in.
j Safety Valve Area is 1.84 an?
All steam flow rates were desated by deve.lang by the derate f at tor el 0.9 4
WOTE:
Nu.ebers en () is reference provided to SWtC by YAEC.
BI-15059-70 2A
ATTACHMENT A (Continued)
't MN-87-41 00ESTION 9 More information is needed on the structural analysis.
Provide information on the lumped mass spacing, integration time step, cutoff frequencies, and damping.
If the cutoff frequency used was less than 100HZ it appears to be too low for piping analysis based on EG&G Idaho, Inc. experience. Most piping analyses use a cutoff frequency of 100 HZ.
Provide assurance the use of the smaller cutoff frequency does not invalidate the analysis performed.
RESPONSE
The structural analysis of the upstream and downstream piping due to safety and relief valve discharge was performed using SHEC's NUPIPE-SH computer program which performs an elastic evaluation of three-dimensional piping systems.
In accordance with this method the inertial characteristics of the piping system are simulated by discrete masses of piping components (including all concentrated and eccentric masses such as valves and valve operators) lumped at selected mode points.
The distributed mass of a piping run is lumped at these nodal points.
Typically, the model will contain at least three mass points between restraints active in the same direction.
The parameters for the time history analysis are selected to ensure sufficient accuracy and dynamic stability of the solution to the dynamic analysis of piping for fluid transient loadings.
The cut-off frequency and mode are selected by a review of the piping geometry and system response characteristics recognizing the fact the typical modes of excitation in this analysis are the higher frequency axial modes.
The total analysis time and integration time steps for the analysis are selected based on a review of the input forcing function and to ensure a stable solution.
The total analysis time is selected to allow sufficient time for the piping system to respond to the input forces and for the response to begin to damp out.
The following analysis parameters were used for the bounding case of steam discharge through the relief valves.
Integration Time Step 0.001 sec Cut-Off frequency 400 Hz Damping 1%
1 8571L-GDH
ATTACHMENT A (Continued)
MN-87-41 OUESTION 11 Pressurizer nozzle loads during safety valve and PORV discharge were not discussed. Compare the calculated and allowable loads for the pressurizer nozzles.
RESPONSE
The following stresses were calculated in MYC-462, Revision 1, for the pressurizer nozzles to the PORVs and the safety valves.
The analysis was performed to the 1982 Summer Addenda of the 1980 ASME code,Section III, Subsection NB-3227.5. HRC Bulletin 107 revised to March 1979 was utilized to determine the three stress categories which were added, in accordance with Table NB-3217-1, to the existing stresses in the pressurizer.
The resulting stress combinations are compared with the stress allowables described in Subsection NB-3222-1.
Nozzle Function:
SV-3001 Line Designation:
3"-RC-22-1504RI Stress Stress Stress Category (psi)
Allowable (osi)
General Membrane (P )
17,500 1.0Sm - 26700 M
Local Membrane (P )
38,820 1.5Sm - 40050 L
Local + Secondary (PL + Q) 43,530 3.0Sm - 80100 (Membrane plus Bending)
Nozzle Function:
SV-3003 Line Designation:
3"-RC-20-1504RI Stress Stress Stress Category Insil Allowable (osi)
General Membrane (P )
17,500 1.0Sm - 26700 M
Local Membrane (P )
38,887 1.5Sm - 40050 L
Local + Secondary (PL + 0) 43,521 3.0Sm - 80100 (Membrane plus Bending) 8571L-GDH
ATTACHMENT A (Continued)
MN-87-41 Nozzle Function:
SV-3002 Line Designation:
3"-RC-21-1505RI Stress Stress Sirgss Category 19111 Allowable (osi)
General Membrane (P )
17,500 1.0Sm - 267C0 M
Local Membrane (P )
38,900 1.5Em - 40050 L
Local + Secondary (PL + Q) 43,363 3.0Sm - 80100 (Membrane plus Bending)
Nozzle Function:
PORV Line Designation:
3"-RC-26-1504RI Stress Stress Stress Cateaorv (psi)
Allowable (osil General Membrane (P )
17,500 1.0Sm - 26700 M
Local Membrane (P )
38,684 1.5Sm - 40050 L
Local + Secondary (PL + Q) 42,151 3.0Sm - 80100 (Membrane plus Bending)
GENERAL NOTES The stresses shown are at the nozzle / pressurizer junction.
The pressurizer pressure stress account for 75% to 90% of the reported stresses.
The remaining 10% to 25% of the stress is a result of loads from the pipes attached to the nozzles, hence a variation in load from pipes produces only a marginal change in the overall stress levels in the nozzles.
l 8571L-GDH j
ATTACHMENT A (Continued)
MN-87-41 l
OUESTION 12 l
1 The Combustion Engineering (CE) Inlet Conditions Report listed the FSAR transients and accidents for each plant which result in a peak pressure greater than the safety valve setpoint. For some plants this list included the l
Feedwater Line Break (FHLB), but for other plants the FHLB was not included.
Maine Yankee was a plant that did not include the FHLB in its list of i
transients and accidents that challenge the safety valves.
From the CE report, it was not clear whether the FHLB was missing because the accident did not challenge the safety valves or because Maine Yankee was licensed prior to the issuance of Regulatory Guide 1.70, Revision 2, and therefore, the FHLB was not analyzed as part of Maine Yankee's design basis. Discuss why the FHLB was not listed for Maine Yankee.
If the FHLB was not listed for the second reason discussed above, it is the staff position that the Maine Yankee submittal is incomplete.
Item II.D.1 in NUREG-0737 specifically requires that PORV and safety valves be qualified for fluid conditions resulting from transients and accidents referenced in Regulatory Guide 1.70, Revision 2.
The FHLB is specifically defined in Regulatory Guide 1.70, Revision 2.
Additionally, from i
the staff review of other plant-specific responses to Item II.D.1, it is clear that for many plants the FHLB accident is the limiting case for providing high 4
pressure liquid to the safety valves, a fluid for which they were not specifically designed originally.
This is exactly the type of concern that 1
NUREG-0737, II.D.1, was established to address.
In accordance with the i
requirements of the NUREG, we require that information be provided to demonstrate that the PORVs and safety valves will function as required to assist in safe shutdown of the plant and will not experience any degradation that would inhibit safe plant shutdown if exposed to the FHLB.
~
RESPONSE
Maine Yankee was licensed prior to the issuance of Regulatory Guide 1.70, Revision 2, and therefore, the FHLB was not explicitly analyzed as part of the design basis.
For this reason, the FHLB was not included in the list of
)
transients and accidents that challenge the safety valves. However, Maine Yankee has evaluated the FHLB accident and determined that it is bounded by l
the Loss-of-Electrical-Load (L0EL) event with respect to system overpressure and challenge to the safety and relief valve.
The Maine Yankee design basis analyses, as reported in the Final Safety Analysis Report (FSAR), assumed that the system response to a FHLB would be j
bounded by either a Loss-Of-Feedwater (LOFH) event or a Main Steam Line Break i
(MSLB), depending on the break location. Containment isolation check valves, located in each main feedwater and emergency feedwater line, prevent any loss i
j of secondary inventory unless the break occurs between the steam generator and i
the check valve. A rupture in the feedwater line upstream of the check valve results in feedwater spilling out the break, following the path of least resistance.
The resulting transient is a complete loss of main feedwater to all steam generators.
The complete loss-of-feedwater transient was specifically analyzed in the FSAR.
For this transient a single emergency l
8571L-GDH a
ATTACHMENT A (Continued)
MN-87-41 feedwater pump is capable of providing sufficient feedwater flow to maintain a heat sink and prevent Reactor Coolant System (RCS) overpressure and challenge to the safety valves.
A FHLB near the steam generator nozzle, downstream of the check valve, will result in a loss of both main and emergency feedwater through the break.
The inventory of the steam generator with the faulted line will be released to containment.
The two intact steam generators will be isolated from the break by the feedline check valves and the Non-return Valves (NRVs) in the main steam lines.
The intact steam generators remain available to remove system heat until the turbine trips. A break of this type is similar to a MSLB, where steam is extracted from one steam generator, resulting in overcooling in the RCS. Under the proper set of conditions, a FHLB downstream of the check valves could result in an RCS heatup and potential challenge to the safety valves. This could occur if the faulted steam generator boils dry or otherwise loses heat transfer capability.
The abrupt loss of heat transfer in one steam generator increases the primary to secondary temperature difference required to remove reactor heat, resulting in a rapid heatup of the RCS and possible reactor trip on high pressurizer pressure.
The reactor trip would result in a turbine trip, isolating the intact steam generators and exacerbating the RCS pressure transient.
This transient is similar to the loss-of-load event analyzed in the FSAR.
The analysis of the L0EL assumes instantaneous closure of the turbine throttle valves without reactor trip resulting in a complete loss of heat removal capability with the reactor at full power.
The pressure spike resulting from a L0EL event bounds a FHLB with steam generator dryout, since two steam generators remain available in a FHLB to remove heat until reactor trip.
Long-term removal of decay heat following reactor trip requires restoration of feedwater to the intact steam generators before the inventory boils dry.
The steam generator low pressure isolation trip automatically performs this function. When the faulted steam generator pressure reaches 400 psig, the Excess Flow Check Valves (EFCVs) in all steam lines, and the isolation valves in the main and emergency feedwater lines to the faulted steam generator close. Once the faulted feedwater lines are isolated, feedwater can be restored to the intact steam generators.
The Low Pressure Isolation System is redundant, therefore, a single failure cannot prevent the restoration of feedwater.
In conclusion, the FHLB event has a potential to challenge the safety and relief valves.
However, because of the design of mitigating systems at Maine Yankee the complete loss-of-load event provides a more serious challenge to these valves. Maine Yankee has specifically analyzed the complete loss-of-load event, and the resulting conditions at the safety and relief valves, demonstrating acceptable results.
8571L-GDH
- --