ML20209C147
Text
l d b (k
- 'o g UNITED STATES 8 o NUCLEAR REGULATORY COMMISSION
$ $ WASHINGTON, D. C. 20555
~s*****/
OCT 2 81980 Task Action Plan A-8 Docket Nos.: 50-358,50-352/353,50-367,50-373/374,50-387/388, 50-410, 50-322, 50-397 MEMORANDUM FOR: Karl Kniel, Chief Generic Issues Branch Division of Safety Technology FROM: C. J. Anderson, A-8 Task Manager Generic Issues Branch, DST APPLICANT: llembers of the Mark II Owners Group
SUBJECT:
MEETING WITH !! ARK II OWNERS TO DISCUSS THE LONG TERii PROGRAM CHUGGING LOAD SPECIFICATION (October 8,1980)
Background
In recent months the Mark II owners have made substantial progress in the development of LTP Chugging load specification based on conservative interpretation of Mark Il related tests conducted during 1979, and early 1980 at the modified 4T facility. A report containing the worst chugs
' observed was submitted to the NRC during that meeting.
An attendance list and a copy of the meeting handouts are attached.
Sumary A sumary of the discussion is provided below:
- 1. Introduction Mr. M. Davis of General Electric provided an update of the generic Mark II chugging load specification scheduled for completion by December 1980.
B7020402o9 870120 hbRROO 782 PDR $
209
f i
Karl Kniel 0CT 2 81980
- 2. Chugging Load H. Townsend of GE, described the proposed long term chugging loads. As in the interim chugging load development, the LT chugging load is constructed from the 4T C0 data collected during the 1979 and early 1980 tests.
Seven chug pressure histories were selected as conservative representatives of the boundary chugs observed during the tests. An averaging factor (peak over pressure / average of the peak over pressure and its neighbors) is obtained for each of the seven pressure histories. This factor is applied to account for the magnitude variation in chug strength in a multivent geometry. A bounding Power Spectral Density envelop based on this averaging method is then utilized to determine an average source which will be applied in phase at all vents in a typical Mark II configuration.
This averaging approach is justified based on preliminary observation of recent Mark II full scale multivent test data from JAERI.
Another approach which is under development by the Mark II owners, is to establish a desynchronization window from the multivent test data from JAERI. Monte Carlo method is then utilized to determine the degree of desynchronization between vent pipes in a typical Mark II configuration.
Sources generated from the key seven chugs are then applied out of phase at the vent exits based on the desynchronization windows determined
(-
L above. To confirm the conservatism of the LTP chugging load specification, prediction of available JAERI bounding chugs will be performed utilizing these methods.
NRC Staff Coments Cliff Anderson of the NRC staff summarized the staff's comments of the proposed LT chugging loads.
The staff agrees with the Mark II owners that relevent full scale multivent data supports significant reduction in the highest observed 4T C0 chugging loads for plant application. The load reduction due to either spatial averaging or desynchronization is justified with adequate
- documentation of the methodology and with the prediction of the bounding l chugs in the JAERI facility. ,
f
/ / %.,
Clifford J. Anderson A-8 Task Manager Generic Issues Branch Division of Safety Technology Attachments: As stated cc: A-8 Internal Distribution List A-8 External Distribution List l
LAWRENCE LIVERMORE LABORATORY 1
]A]3' u
-(H NUCLEAR SYSTEMS SAFETY PROGRAM TF80-272 October 28, 1980 Dr. M. Arinobu Toshiba Corporation 4-1, Ukishima-Cho Kawasaki-Ku, Kawasaki 210 JAPAN
Reference:
Visit to LLNL, October 16, 1980
Dear Dr. Arinobu:
This letter is to thank you for your most interesting visit to our Laboratory and for sharing with us your recent research results regarding the MKII containment loading conditions.
It is with gratitude that I tell you that Dr. Kukita has delivered the complete CRT evaluation data set to us for the U.S. NRC (RSR). I thank you for your efforts in making this possible.
As we discussed, enclosed are some reports describing our work in determination of peak forces during the air clearing transient in the MKI suppression system design. I hope they will be of interest to you.
Thank you again for your assistance to our Liaison efforts. I hope that we may meet again on my next visit to Japan.
Yours truly,
.loN Ecsard W. McCauley, P.E., Ph.D.
Thermo Fluid Mechanics Group EWM:km/1439u Enclosures cc:
T.* Lee (NRC-RSR),
M. Shiba (JAERI)
G.Cummings(LLNL)
An Ecy3!Ctqxyttnty Enyk,o s Lnwr:.tyotCaYorru PODon 808Lmxmcvc Ca!,torru91550 TchstnintlI5)J22 It00 Twx910-38 -8339 UCLLL LVMR
k pzu n.fm4- l AN5/A5yE/A!2C k terna11oireil Tsin w k s q _
YfaI $1CmM(-/- tN22a/t ($ NsN40 Sprinf$, Af f 1- '
.hc f. f -g , ,4g STUDIES ON THE DYNAMIC PHENOMENA CAUSED gg gf g/g g BY STEAM CONDENSATION IN WATER ,
M. Arinobu ABSTRACT Dynamic phenomena caused by steam injection into water have been investigated experimentally. These phenomena are supposed to occur in light water reactor pressure suppression pools.
The condensation phenomena are classified mainly by steam velocity and water temperature. In the case of higher steam velocity, periodic pressure oscillation is produced. Its dominant frequency is expressed by an empiri-cal equation. In the case of lower steam velocity, chugging phenomena may occur. The influence of the parameters on the wall pressure amplitude have been investigated statistically. Vent pipe vibrations and methods for reducing the pressure amplitudes have also been investigated.
NOMENCLATURE A1, A2 = Accelerometers . ,
Cp = Specific heat of water at constant pressure [J/kg*K]
D = Vent pipe inner diameter [m]
f = Dominant frequency of condensation oscillation [Hz] i Ja = Non-dimensionalized pool subcooling (= Cp6T/L) !
L = Latent heat of water [J/kg]
s = Steam mass flow density [kg/m 2s]
Pba = Pressure at the pool bottom in the case of (1) in figure 8.
[Pa]
Pm = Reference pressure (= 100 kPa)
Pwl, Pw2, ..., Pal, P1, P2, ... = Pressure transducers Po = Neutral pressure of a oscillating bubble [Pa]
Ro
= Neutral radius of a oscillating bubble [m]
S1, 52, ... = Strain gauges Subm. = Vent pipe submergence depth [m]
St = Non-dimensionalized dominant frequency of condensation oscillation (= fD/V)
Ts = Saturation temperature of steam [*C]
W = Water temperature ['C]
V = Steam velocity at exit of a vent pipe [m/s]
AT = Pool subcooling (= Ts - W) ['C]
K = Ratio of specific heats
= Water density [kg/m3]
pw INTRODUCTION It has been pointed out that steam injection into subcooled water pro-duces unsteady phenomena called chugging Il3 or c~ondensation oscillation. [2]
Research and Development Center, Toshiba Corporation 4-1 t;kishima-cho, Kawasaki-ku, Kawasaki-city, 210, Japan _
l l
nese phenomena have been studied especially concerning[ transient phenomena in the light water reactor pressure suppression s stem. 3}-[6]
It is known about downward venting systems [3 that in the case of low steam velocity, steam bubbles are built up at the vent pipe exit and sub- l sequently collapse rapidly following water rushing into the vent pipe, nese phenomena are called chugging.Ill If the steam velocity is rather high, steam always condenses outside the vent pipe, and periodic pressure oscilla-tion is observed in the pool and vent pipe. The steam-water interface also oscillates. These phenomena are called condensation oscillation.
The present aim is to investigate the characteristics of above mentioned phenomena more precisely. Steam is injected into water from the downward venting system. Test parameters examined here are steam flow rate, water temperature, vent pipe diameter and vent pipe submergence. Pressures and Temperatures are measured at the pool walls and in the vent pipe. Two kinds of test apparatus have been used in order to investigate both condensation oscillation and chugging phenomena.
EXPERIMENT Test facility Figure 1 shows a concep- .
tual drawing of the experimen- A -Aamlerometer tal apparatus for the purpose P-PressureTron=trar of getting general view of the S _.g g phenomena while investigating p condensation oscillation p characteristics. It consists 4d of a cylindrical pool and a * '~
vent pipe. Pool dimensions are 1.0 m inner diameter and g j g 0.7 m height. Two vent pipes, g ys 0.0161 m and 0.0276 m inner -QRT j I
diameter were examined. Con-densation phenomena can be seen through the glass windows @@l in both sides of the pool.
Figure 2 shows the test lS @
facility arrangement. Satu-N l @#
rated steam is supplied from @ "
4 the house boiler in the l@ @
laboratory and steam flow E @ im @ @
rate is controlled by the ci @
! valves of the flow meter. t@
Figure 3 shows the test l j l apparatus used for chugging gg g investigation. Pool dimen-sions are 0.305 m inner diam-eter and about 2.0 m height.
Steam is injected through the header (0.305 m inner diameter, FIG.1. TEST APFARATUS-I
O
)
.r.
- A Stearn Supply -p l
[ : !
7' Flow Meter FIOw COntr01 V0lves ._
xsm
- S FIG. 2. TEST FACILITY ARRANGEMENT h
0.372 m length) and the vent pipe b h 'k (0.053 m inner diameter,1.65.m n U b length). Steam condensation at l the vent pipe exit can be seen - - -
through the glass windows in both @ -
sides of the pool. The steam {_ l supply system is the same as that shown in figure 2.
y[gh --l l
- * ^ ^^
E x riment was conducted with various steam flow rates.
At the beginning of each experi-ment, the air contained in the pipe line was purged by opening the flow con-trol valves. Confirming that the air was purged completely, steam flow rate was set up and steam injection was continued maintaining a constant steam flow rate until the pool water temperature became about 100*C.
Pressures were measured at the pool walls and in the vent pipe by strain gauge type pressure transducers. Water and steam temperatures were also measured at several points in the pool and vent pipe by thermo-couples. Vent pipe vibrations were measured using strain gauges and accelerometers only in test apparatus-I shown in figure 1. These data were recorded by analog data recorder continuously during the experiments. Typical phenomena were also photographed by a high speed camera. .
Examined test parameters are as follows: a Test apparatus-I Steam mass flow density 5.0 % 100 kg/m 2s Vent pipe submergence 0.2 m, 0.4 m Vent pipe inner diameter 0.0161 m, 0.0276 m Water temperature 20.0 % 92.0*C
.y.
.v
Test apparatus-II Steam mass flow density 10.0 kg/m2 s, 15.0 kg/m2s, 20.0 kg/m 2s Vent pipe submergence 0.3 m, 0.5 m, 0.7 m Water temperature 20.0 N 90.0*C EXPERI} ENTAL RESUL'IS AND DISCUSSION Classification of the phenomena Dynamic phenomena were classified as shown in figure 4 by steam mass flow density (steam velocity) and water temperature (Pool subcooling) after the experiments performed with test apparatus-I.
Region I - Steam condenses mainly in the vent pipe. If steam condenses mainly on the inner wall of the vent pipe, self excited oscillation may be induced by the change in condensation surface coupled with water leg oscil-lation in the vent pipe and pressure oscillation caused by condensation. I33 This phenomenon was not observed here.
Region II - Steam flow rate exceeds the condensation rate in the vent pipe. Steam bubbles are built up and collapse. After bubble collapsing, water rushes into the vent pipe and is slowly pushed out by steam. These phenomena are shown in figure 5-A. In this process steam bubbles are some-times left in water or at the outside wall of the pipe and collapse as shown in figure 5-B.
Region III - Because of rather high mass flow rate, water dose not flow into the vent pipe and steam condenses outside the pipe. Periodic pressure oscillation is not observed .in this region.
f A i oc b. . : . .~,.~~. ~.~.5.
.: .. .?
.._ (.: '.:
. M. w % Q , ?V I
a e0 4
, . .....:.. :.:::. .:.{..'.:
- .'. - :. . :. : ..=..::.:.:.
R.. e 4,f ,
W
?
4 , j ,
80
- .:.:'. . :?. V l ?.'+ l
. . . /, i.: ,., :. ::l;... -
\ t v a, y y y /pi W
. i h 70 : , .'.*, . ., i -} ,.: . ,.,'
1 -
- g. p - 1 to : .. . . . . ...-
h,60U ' 9. , .h
~ ~ -
a- .,:.:./, ,
]* ,
./
?/ IV B 50 40 l g '
ei 5 F P.
i gf 30 k .
20 t 40
, k 60 80 h' % ~
0 20 mass f!Ow density m /m's 100 Kg i
! ~~
h U' FIG. 4. CLASSIFICATION OF FIG. 5. CilUGGING PitEt;OMENA Tite FilEMOMEMA ,
Region IV - Periodic pressure oscillation is observed in water and in the vent pipe. Steam condenses outside the vent pipe and steam-water inter-face oscillates violently.
Region V - In this region, vent pipe vibration becomes apparent caused by steam condensation at the outside wall of the vent pipe.
Region VI - Because of high water temperature, steam-water interface becomes very large and a portion of steam escapes up to the open surface without condensing.
- Condensation oscillation (region IV) and chugging (region II) have been mainly investigated, since periodic pressure oscillation may resonate with structures in the pool and high amplitude pressure spikes may be produced by bubble collapse.
Condens'ation oscillation (region IV)
As mentioned above, periodic pressure oscillation is observed in the pool and in the vent pipe.
As shown in figure 6 the dominant frequency of pressure oscillation g, becomes lower with increase in \ \ ,Suh 200 400mn water temperature and decrease 120- \
(
in steam velo' city. The dominant
\
- \.,\ ,
e
, 47 a 60 i frequency is also influenced by . e a 74 the vent pipe inner diameter. rh l 80 - N Steam injection from a larger ok\ % l inner diameter pipe produces c 60-lower frequency pressure oscil-lation. Considering these ex-g 2 40-
- y perimental results, the domi-
- nant frequency is non-dimension- 20-
- }St:0&Jd'
- - ~
g alized by the steam velocity V and the vent pipe inner diameter jo %
D. Figure 7 shows the relation- 20 N 40 b 60 [0 80 90 *C w T ship between the non-dimension-alized frequency and pool sub- FIG. 6. DOMINANT FREQUENCY OF cooling. The pool subcooling PRESSURE OSCILLATION IN A is also non-dimensionalized by VENT PIPE ,
latent heat L and specific heat Cp of water. The straight line drawn in figure 7 indicates fD/V
= (Cp6T/L)1
- 4 Obtained data in this experiment show that the dominant frequency of condensation oscillation may be expressed as f=0. ( ) . (1) .
Considering small vibration of a spherical bubble in water, the vibration frequency is given by 1 1 3kPo f = 2n R- (2)
O Aw b
where Ro, K, PO , and p, mean neutral radius of a bubble, ratio of specific heats, neutral pressure in a vibrat-x104 3 la16.1mm i ing bubble, and water density respec- _
tively. Comparing eq. (1) with eq. 5 - rh (2), the pipe inner diameter and _
, gg l pool subcooling effect's on the'domi- #, ;
nant frequency may be explained as > ~
^ 88 C*a*
follows. The larger inner diameter 3
_ =100 {
of the pipe and the smaller pool subcooling produce the larger K9/m2S steam bubble, and the larger steam ^
bubble vibrates with the lower frequ- I _
ency. The reason why the dominant
~
ID=27.6mm frequency increases in proportion -
o th to the increase in steam velocity 0.5 -
o47 .
is not clear yet. The mechanism .
regarding steam bubble oscillation . .= o60 in water has to be studied more , o a 74 precisely, in order to clarify the -
steam velocity effects on the domi-gkm2 S
. nant frequency.
The vent pipe submergence depth 0.1 seems not to influence the dominant 1 5' 10 20x10i frequency of condensation oscilla-tion as shown' in figure 5'. CPN / h ,
Chugging (region II) FIG. 7. NON-DIMENSIONALIZED Chugging may occur in the lower DOMINANT FREQUENCY OF CONDEN-steam velocity region. The experi- SATION OSCIILATION
~
ments were performed using a narrower pool compared with the vent pipe sectional area, in order to investigate pres-sure spikes. induced by steam bubble collapse in water.
Figure 8 shows the calculated pressure distribution induced by spherical bubble oscillation or collapsing in water, assuming potential flow in the pool.
From the calculation result, if the pool is the narrower and the deeper,' the t
pool walls experience the larger pressure amplitudes. So, the narrower pool l
was chosen to investigate dynamic pressure in water precisely.
i In chugging, the wall pressures generally show a large negative amplitude i
followed by a large positive amplitude. Ill The negative amplitude is caused by rapid condensation of steam and the large positive amplitude may be caused by bubble collapse in water. Therefore the amplitudes are influenced by local water temperature, bubble volume and shape when it begins to collapse, as well as the bubble formation process. Considering these influence factors, the amplitudes are thought to be stochastic.
Figure 9 shows the influence of average water temperature on the positive pressure amplitudes. 'Ihe vertical axis means probability of pressure greater than a given value, indicated in horizontal axis. The larger positive pres-sure amplitudes are produced in the lower temperature water. The water tem-perature also influence the negative pressure amplitudes, with a similar tendency.
}-
1.0 , , . ,,,,, , ,
1.2 g's's a5 0.8 a Tw c o 30 rhsiOHgim2,-
P4 s's N 3 e 50-70 Subm=Q3m
\ s 50.6 -
e 70-90 ~
1.0 - ss 6 ,
s ,
N y 0.4 2
\ Y :
08 s y 8,.nimm_ _ m ,_ _
c
. ?m,
\o x'se s O
' E 3 PIRn
{
0.4 I FIG. 9. INFLUENCE OF WATER W .A i i TEMPERA'IURE ON WALL PRESSURE AMPLITUDE (CHUGGING) 0.2 f I .
l 0 0.2 0.4 0.6 0.8 1.0 1.2 P/Pta ' u Y T o 306PC e 50 70' f'
Ko.2 e 70-90 20 y
T y & a 30 90 --
g
+A -B -C D~ z
/ ,- s Nl m _
gai -
g- -
o
@ @ @9?
~
5 10 , . is ,h 20fHgtm r a 25 f
l (1) (2) (3) FIo. 10. INFLUENCE OF STEAM l
MASS FICW DENSITY ON WALL FIG. 8. CALCULATED UALL PRESSURE O.5 m)
DISTRIBUTION The steam velocity and vent pipe submergence effects on the wall pressures were investigated T w
30-50T
{
o2 20 l
by determining root mean square values of the pressures, as shown p -
- yfj ,
- a 33.go ,,
l in figures 10 and 11. These g r.m.s. values were obtained in t oi - .
~
to the range of water temperature y shown in the figures 10 and 11. "
- j ,
j The r.m.s. value of the wall l
pressure seems to be maximum in o ,
j the region where steam mass flow o 02 or. 08 m g5 i density is about 15 kg/m 2s. The r.m.s. value becomes large as FIG. 11. INFLUENCE OF VENT -
vent pipe submergence increases.
PIPE SUDMERGENCE DEPTH ON l
WALL PPESSURE (CHUGGING ^ -l L.
' Ibis tendency agrees qualitatively with the calculation results shown in figure 8. The reason why the r.m.s. value dose not seem to become zero when the submergence becomes zero, is thought that the r.m.s. value contains the negative pressure amplitudes caused by water rushing into the vent pipe.
Since steam condensation is accompanied with water leg movement in the vent pipe in chugging, as mentioned before, the time intervals between chug-ging events' may be controlled by the frequency of oscillating system composed of the water leg in the vent pipe and steam pressure, as mass and spring, 1 respectively. The intervals between the positive pressure amplitudes greater than 50 kPa were measured. The vent pipe submergence and steam velocity ef-fects on the intervals were investigated, using the measured results as shown in figures 12 and 13, respectively. It can be understood from figures 12 and 13 that the time intervals of. large positive pressure amplitude events is about -
2 Hz . The steam velocity and vent pipe submergence seem not to influence the dominant frequency in the range of the parameters examined here. From these results, pressure change in steam seems to play a dominant part in the supposed oscillating system.
8.
l I l
8 1 rtwl0K9/m23 y -
16 18 c Tw=30-90 C !
Tw=30-90 C f!' I-
' Subm.= 0.5 m 14 16 j p
\ \
'12 f, .
7 14 g
~
Subm' .4 2 10 i y12 -
--o- 10 Kg/m?s -
-o-0.3m c ,
-+- 15
>. * {'
g ! -+- 0.5 a.10 7 --.- 20 g.* D' -+- o 7 d- 6 I e
j 8 !@$ ; .
p
$6 #
e 0 2 4 , ; a @- *
- e 4
" 2 i;
L..___-y,.o , 4T i
0 1 2 3 4 5 1 / e'eA ,,
Interval between Events (sec) 0 1 2 ~~~~ 3
' Interval between Events (sec)
FIG. 12. INFLUENG OF VENT PIPE FIG. 13. INFLUENG OF STEAM NASS SUBMERGENG DEPTH ON CHUGGING m W GUmG EVEWS EVENTS
s
- Vent pipe vibration Vent pipe vibration is induced by lateral loads which are produced ps ION mm by steam bubble collapsing at the y _
m outside walls of the vent pipe. o stS %mr s Figure 14 shows the r.m.s. value 8 -
e 36 j- .
e 66 obtained from strain gauge S3 /
36 -
a100 f shown in figure 1. In the case of higher steam velocity large j4 ,
/)* {
strain and acceleration amplitudes. [ ,
- '-+
have been observed under the con- 2 -
L MW' -
dition of high vater temperature of about 83*C. The signals from 0 10 20 30 [0 50 60 70 80 90*C h strain gauges and accelerometers show that the vent pipe system FIG. 14. VENT PIPE VIBRATION vibrates at an approximately 40 Hz. Pressure signals also show that the dominant frequency is about 40 Hz under the same conditions. The natural frequency was calculated by the dynamic analysis code. The obtained value was about 42 Hz. Therefore, the large amplitude shown in figure 14 results from the resonance .of condensation oscillation with the vent pipe vibration.
Consnents on the methods for reducing the pressure amplitudes '
If steam is injected from many small pipes or many small holes with dif- i ferent diameters, the'1atter method was already 'used in the condencers,' the pool walls will experience much smaller pressure amplitudes. The smaller diameters of the pipes or holes may produce the smaller bubbles. As shown in
- figure 8, the smaller bubbles produce the smaller pressure amplitudes at the wall, if they oscillate out of phase. As the dominant frequency of condensa-
, tion oscillation is in proportion to steam velocity and in inverse proportion to the inner diameter of the pipe through which steam is injected, j each bubble formed at the pipe exits or holes with different diameters may oscillate with different fre-quencies, and out of phase. Al-though these effects are still now
[{w -
o, 200 under investigation experimentally, w one of obtained results is shown 7, '
g in figure 15. When the natural frequencies of the structures are l
a g y ggg
,, i./N,h.-a g.
lower than condensation oscilla- 5 E
tion frequencies the resonance can be avoided by adopting these steam 3 ~~7
/
~
NN,%
- C injection methods. b ._....=J o ,
Steam injection from many 0 20 4.0 60 80 3 100(*C]
small different diameter pipes will also be effective to reduce the wall pressure amplitudes in FIG. 15. DOMINANT FREQUENCY AND chugging, for the same reason an WALL PRESSURE (COMPARISON BE-for condensation oscillation. 'IWEEN ItUECTION DEVICES)
f CONCLUSION Dynamic phenomena caused by steam condensation in water have been investi-gated experimentally.
The phenomena can be classified mainly by steam velocity (mass flow density) and water temperature (pool,subcooling) .
The dominant frequency of condensation oscillation is influenced by steam velocity, pool subcooling, vent pipe inner diameter, and can be expressed by an empirical- equation (1) in the range of test parameters examined here.
Chugging has apparent periodicity and the dominant frequency seems to be influenced by neither steam velocity nor vent pipe submergence depth.
The wall pressure amplitudes experienced in chugging increase in propor-tion to vent pipe submergence and seem to take a maximum value where steam mass flow density is about 15 kg/m 2s.
The large vent pipe vibration amplitudes may be produced if condensation oscillation resonates with vent pipe vibration.
' ~
Steam injection 'from many small different diameter pipes or 'from many small different diameter holes may be effective to reduce the influence of pressure oscillation on,the pool walls and structures in the pool.
ACKNOWLEDGEMENT The author wish to thank to Messrs. Wada dnd Shimamoto for their assistance during the experiments. ,
RERERENGS
- 1. Becker, M. " Dynamic Phenomena during Condensation of Steam in the Pool of the Pressure Suppression System," paper presented at IAEA Technical Commit-tee Meeting, Cologne, Federal Republic of Germany,1976. -
- 2. Okazaki, M., Kukita, Y., Shiba, M. " Effects of Geometry Changes in Vapor Suppression System on Pressure Oscillation Phenomena," paper presented at IAEA Technical Committee Meeting, Cologne, Federal Republic of Germany,1976.
- 3. Saito, T. et al., "On the Unst.cady Phenomena Relating to Vapor Suppression,"
i ASME paper 74-WA/HT-47.
- 4. Narial, H., Aya, I., Kobayashi, M. "Thermo-Hydraulic Behavior in a Model Pressure Suppression Containment during Blowdown," paper presented at i ASME Winter Annual Meeting, December 1978.
'$ . Sargis , D. A. , S tuhmiller, J . H . , Wang. S . S . "A Fundamental Thermohydrau- 1 lic Model to Predict Steam Chugging Phenomena," paper presented at ASME l Winter Annual Meeting, December 1978. f
- 6. Slauterbeck, D. C., Ericson, L. " Nuclear Safety Experiments in the Marviken Power Station," NUCLEAR SAFETY, Vol .18, No. 4, July-August 1977, pp.481-491.*
r' LAWRENCE LIVERMORE LABORATORY NUCLEAR SYSTEMS SAFETY PROGRAM TF80-273 October 28, 1980 Mr. K. Kotani Hitachi Coporation 1168 Morlyama-Cho Hitachi-Shi, Ibaraki-Ken 316 JAPAN
Reference:
Visit to LLNL, October 14, 1980
Dear Mr. Kotani:
This letter is to thank you for taking the time to visit our Laboratory and to share the results of your research efforts in the MKI pressure suppression system modeling with us.
We are enclosing some of our reports which may be of interest to you since they also deal with determination of peak forces during the air clearing transient in the MKI suppression system design.
Thank you again for your most interesting discussions. I hope that we may meet again on my next visit to Japan.
Yours truly,
.10 r Edward W. McCauley, P7E., Ph.D.
Thermo Fluid Mechanics Group Nuclear Test Engineering Division EWM:km/1443u Enclosure s cc:
.T." Lee (NRC-RSR)-
C. I. Grimes (NRC-DST)
M. Shiba (JAERI)
G.Cummings(LLNL)
I f0 An EqualCtystuntyEnplay. lherwtyr'Cavkma POBox 80SLu+Euc.Cablarna94550 Telephone (4tS)J22 IIDO Twx910-336 8339 UCLLL L%fR
_y