ML20245B908
Text
--
E
- y..
{
i pl,:
ty.
f i
2 Jul. M/F 4
. p.
- F
% q, E l lir. Voss ficore Assistant Director for Environmental projects i
L' (1.5. fluclear Regulatory Connission
- Washingto.n, D.C.
20055
,M l
Dear Mr. 14oore:
=
The fluclear Regulatory Commission (NRC) recently published an environiaental state-ent entitled " Addendum to the Final Environmental l
Statement.for the Operation of the Diablo Canyon Nuclear plants Units 1 and P.."
The Environmental Protection Agency (epa) has reviewed the Addendum and our comments on it follow.
t On page 11, conclusion 4.b.(3) states, "The risk of accidental radiation exposure has been addressed in depth in the Conmission's v
Reactor Safety Study (WASH-1400, NUREG-75/014) and found to be accept-t j ably low." A more appropriate conclusion for the Diablo Canyon Adden-
] l dum could have been "...and found that the risk from reactor accidents 9
will not undulv jeopardize public health and safety or the environment,"
a or as indicated in other environmental statemopts, "The risk associated yith accidental radiation exposures. is very low" (Holf Creck).
We believe that it is premature to make a judgment such as in the
""p r Diablo Canyon Addendum s'ince the Reactor Safety Study is only the first 1!
step in determining reactor risk acceptability.
In order Lv reach a
' y, societal judgment on rist acceptability at least two other factors must 4
he included:
(1) a cost-honefit assessment on the adeauacy of present e
reactor safety as related to rick; and (2) the societal perception of d
risks and meae.ures of acceptability. This also was reflected in Mr. Saul Levine's testimny nn June 11, 1976, to the House Subcommittee on Energy and the Environment.
- l In order to determine hnw society perceives rist
- and its acceptabil-ity, fpA helieves that it is essential to obtain the viewpoints of various j
components of our society's interest.
Further, it appears that the URC conclusion indicates that the Com-mission has embraced the 4eactor Snfety Study as their study. This is g-inconsistent with the recent contention of fir. Saul I.evine who informd g
epa that the Reactor Safety Study was under the sponsorship of the Com-l t:c r.npor g M g fMy_c.roun_hsAdad byJ-rArr.an mie inn tmt 'that 4 t 4e n
}
q
~
4 8707010442 070610 f.
8 if PDR l
I'*#
THOMASS7-40
e, j "...
(
/
t w
If the Cormission has now adopted, Rasmussen, not of the Conmisslott.
rather than.just accepted, the study report, the point is an importantOth l
..one and should be clearly mm e.different officials of the !!P.C are publicl
,,j positions.
He believe that it is important that these points be clarif the record.
me.
Sincercly yours, Oridnnt :liCoed b'/
1 0
.,y, p. H v;:
U. D. P, owe, Ph.D.
Deputy Assistant Administrator for Radiation Programs (AW-458) q
- i vC C. A' IQ 4,. > 7 t o c N 4.v f );*(l, f(.h i. bh u $r kAM r
i a
I h
L' A
J d
V' j
\\
4 '.
I h
s
a
y3 9'-
(
(
UFFICIAL USE ONLY DATE ISSUED:
9 Jul41977 MINUTES OF THE ADVISORV COMMITTEE ON REACTOR SAFEGUARDS SUBC0fWTTEE ON EMERGENCY CORE C00L"4G SYSTEMS WASHINGTON, DC APRIL 22, 1977 On April 22, 1977, the Advisory Committee on Reactor Safeguards, Emergency Core Cooling Systems Subcommittee met to discuss fundamental research as applied to ECCS related phenomena.
The notice of the meeting appeared in the Federal Register, Vol. 42, No. 67 - Thursday, Ap il 7,1977.
There were no requests for oral or written statements by members of the public and none were made at the meeting. Attachment A is the meeting agenda.
The attendees list is Attechment B.
The tentative meeting chedule is Attachment C to the minutes.
Attachment D contains selected slides and handouts used at the meeting. A complete set of slides and handouts is attached to the office copy of these minutes.
OPEN SESSION (8:30 a.m. - 6:37 p.m. ) INTRODUCTION Dr. Isbin, Subcommittee Chairman, called the meeting to order at 6:30 a.m.
The Chairman said the Subcommittee had not attempted to define explicitly what work should be included under the term fundamental research.
Dr. Isbin said the Subcommittee submitted four questions to the NRC Staff:
(1) what is your interpretation of fundamental research; (2) what research is needed; (3) what fundamental research programs are being conducted by the l
NRC Staff; and, (4) how are research results expected to be applied in such areas as NRC licensing activities, reactor design and reactor operation.
$Db5Ydb 0FFICIAL USE ONLY s
7
o r
c l
0FFICIAL USE ONLY ECCS 4/22/77 Dr. Isbin reviewed the NRC agenda (Attachment C), and said the Subcommittee's goal is to establish reliable and effective ECC systems by review of the NRC Reacto
- Safety Research (RSR) programs. The Chaiman also said that the Subcommittee should examine the question of how detailed the code neadels have to be to describe the ECC phenomena under study. He also noted that the reactor fuel research will not be discussed.at this meeting.
Instead, a separate meeting will be devoted to that subject in the future.
Dr. Isbin explained the procedures for conducting the meeting, pointing out that Thomas G. McCrekss was the Designated Federal Employee in attendance. The Chairman introduced Ron Scroggins from ;!RC RSR to begin th.e day's discussion.
NRC FUNDAMENTAL RESEARCH PROGRAM - R. SCROGGINS - RSR Mr. Scroggins began by reviewing the agenda for the day's meeting discussion (Fig.D-1). He said that NRC notea that the phrase fundamental research is subject to many definitions. NRC divided the research procram into three separate areas:(1) model development studies, (2) separate effect tests, and (3) integral tests. NRC feels the model development studies area encompasses fundamental research. Mr. Scroggins rev'ewed the topics for the discussion under the title of "model development studies".
These topics included heat transfer correlations, interfacial mass and momentum exchange, phase distribution, and pump behavior.
PHENOMEN0 LOGICAL NEEDS FOR ANALYSIS (C0DE DEVELOPMENT) - S. FABIC - RSR Dr. Fabic began by stating that his discussion would dwell on the topic of what is needed in the areas of ccde development; he would not discuss code development or the reason for their development.
0FFICIAL USE ONLY
\\'
9
~
~
(
0FFICIAL USE ONLY ECCS 4/22/77 Dr. Fabic reviewed the steps involved in developing the research codes in use today.
He said the first step is the development of mathematical models to describe phenomena of interest. These mathematical models are combined to form modules that describes components of various reactor systems such as the downcomer, steam generator, Upper plenum, etc.
Dr. Fabic said that the systems codes in use today are basically an assembly of modules that simultaneously computes all the component parameters of the reactor system under investigation.
Dr. Fabic said what is lacking today is basic understanding of complex physical phenomena. He said const'itutive relations are needed for such phenomena as transfer of mass momentum and energy phenonena across interfaces.
flow topology, and transfer of energy at the wall on internal structures.
Dr. Fabic reviewed in a general fashion the equations used in the current two fluid (liquid / vapor) code models. A slide of the general equations used in the two fluid models were shown (Fig D-2), and the various constitutive relations needed for the equation were also reviewed (Fig D-3).
Dr. Fabic also said it is imperative to understand the effect of scaling in the development of the basic code models.
Dr. Fabic discussed two areas of the LOCA transient that fall under the heaoing of multi-dimensional fluid formulation needs. These two areas of interest are:
(1) downcomer penetration of ECC water, and (2) core reflood.
Dr. Fabic also said that current codes such as TRAC can repre-sent the reactor as a three-dimensional system, while the associated reactor equipment such as the steam generator accumulator, pumps, etc.
are represented one-dimensionally (Fig D-4).
RSR wilf be focusing their efforts in the area of development of systems equations that adequately describe the physical phenomena taking place.
If this is done, the improved codes could be extrapolated model large scale geometries for which experimental data may be lacking.
0FFICIAL USE ONLY
t
/
(
0FFICIAL USE ONLY ECCS 4/22/77 In reply to a question from Dr. Zaloudek, Dr. Fabic said he believed that multi-dimensional best estimate codes will be used primarily to quantify the degree of conservatism present in the codes used for licensing applications. In reply to another question from Dr. Zaloudek, Dr. Fabic said that RSR hopes to have a version of the multi-dimensional TRAC code available for public use by the end of this calendar year.
Replying to a comment from Dr. Isbin, Dr. Fabic said that there are studies under-way to determine the sensitivity of various assumptions and phenomena described in the code equations to avcid becomir.g bogged down in a detailed description of the physical phenomena irvolved.
In reply to a question f rom Dr. Catton, Dr. Fabic said the TRAC coh will be a best estimate code, and will not be incorporated into the RELAP licensing code.
HEAT TRANSFER (CORRELATION AND EXPERIMENTS) - DR. Y. Y. HSU - NRC-RSR Dr. Hsu began by stating the objectives of the RSR research in heat transfer.
These objectives are:
(1) to provide heat transfer equations for best estimate (BE) codes during blowdown, (2) to supply the thermal hydraulic package for the reflood phase of the heat transier code.
The basic heat transfer parameters under study by RSR include I
blowdown heat transfer (which consist of transient CHF, CHF, and post CHF heat transfer regimes).
RSR is also studying work undemay to develop heat transfer correlations for the reflood phase of the LOCA transient.
Dr. Hsu reviewed the programs underway in the areas of blowdown heat transfer and reflood heat transfer.
The overall program has been divided into two areas:
development of heat transfer correlations, and experi-ments to" verify the correlations (Fig. D-5).
Dr. Hsu said Drs.
Henry, Chen, and Griffith would give presentations at today's meeting 0FFICIAL USE ONLY am
~
e
(
0FFICIAL USE ONLY ECCS 4/22/77 on the work they have underway in the area of heat transfer.
Dr. Catton asked how the heat transfer programs at GE and ORNL ere factored into the NRC RSR heat transfer research program.
Dr. Hsu said that the GE and Oak Ridge work is being used to verify the correlations developed by the RSR group.
Dr. Hsu briefly reviewed some of the work he is doing in the area of transient CHF heat transfer.
Dr. Hsu showed data of predicted versus actual results from semiscale tests conducted at INEL (Fig D-6).
Dr. Hsu noted the extensive data scatter, and said the cause of the scatter is two fold:
(1) the CHF phenomena is stochastic in nature, and (2) local bundle conditions have not been accounted for.
Dr. Isbin noted that the data scatter may indeed be due to the use of average instead of local bundle condi-tions; he felt that other re' searchers at GE,for instance, have developed more effective correlations by including the use of local bundle conditions in their calculations.
The Chainnan asked the NRC Staff if any effort has been made to make some of the proprietary research data available in the open literature to preyent needless duplication of research work. Mr. Ross (NRC) said that in general he believes repetitive work can be benefical in that it serves as a check on the research work of the organization by another. He also said that because of Commission rules on the protection of proprietary information, NRC has no choice but to treat these materials as proprietary.
Dr. Hsu said that RSR is attempting to establish a data bank of all the vendor's test data to assure that the data are based on the same assumptions to allow a uniform comparison of this information.
Dr. Hsu reviewed some of the research work being conducted on the post CHF heat transfer regime.
This low flow rate, low quality regime is of; particular interest at this time because of the paucity of data 0FFICIAL USE ONLY
~
o 0FFICIAL USE ONLY ECCS 4/22/77 available. Work conducted by Groeneveld and Cheng was reviewed (Figs D-7 -8).
Dr. Hsu said i.he dip seen in the quenching curves is now an object of study to determine if any conduction effects are being seen.
BLOWDOWN HEAT TRANSFER PROGRAM - R. HENRY - ANL Dr. Henry began by reviewing the five main areas of the ANL blowdown heat transf er prograni. These areas are:
(1) two-phase critical flow, (2) transient critical heat flux, (3) post CHF heat transfer, (4) rewet criteria, and (5) energy removal during rewet.
Dr. Henry said that for the cold leg break LOCA case, experiments to date have indicated that no CHF is seen during flow reversal, and that if nucleation by surface cavities is precluded, CHF is seen when the fluid passes through the annular flow transition regime.
Reviewilg tests conducted at the semiscale facility, Dr. Henry said he believes his work in the area of homogeneous nucleation has shown that CHF can be delayed by decreasing the system pressure. Showing results from two seriscale tests, i.e., test S-02-7 run at 2260 psia, and test S-29-2 run at 1750 psia. Dr. Henry said the core is cooled for approximately one order of magnitude longer at the decreased pressure.
Reviewing a plot of temperature versus axial length for the two tests (Fig D-9), Dr. Henry said the amount of heat removed by lowering the system pressure to 1750 psia represents one-half full power second for the entire core, or two and one-half full power seconds for the high power axial segment measured in the test. Dr. Henry also said these results are conservative.
0FFICIAL USE ONLY
~
e
~
(
g 0FFICIAL USE ONLY ECCS 4/22/77 In order to understand the mechanisms involved in the above test results, ANL has constructed a high pressure freon blowdown test loop to simulate the above discussed semiscale tests (Fig D-10).
Dr. Henry cited results from two tests conducted in the freon blowdown test loop at simulated water
- ressures of 2250 psia and 1750 psia. He said the results showed that for lower pressures the onset of subcooled boiling occurred at a lower position in the core and as seen in the semiscale tests, CHF was greatly delayed.
l Dr. Henry said ANL is studying the degree of margins to DNB under steady state conditions. Preliminary results indicate that for flow transients at full power it appears that there is more margin to DNB at low pressures.
l 1
Dr. Henry reviewed work involving the interpretation of semiscale data
~
for the post CHF heat transfer regime. Using the data obtained from the core inlet turbine meter during blowdown, Dr. Henry said ANL calculated the
$ eat transfer coefficient of the exiting steam.
It was found that by parametrically varying the steam heat transfer coefficient (h), a very l
good fit could be obtained to the semiscale temperature data if it is assumed that h is equal to 0.5h of steam (Fig D-ll).
Dr. Henry said he believes this indicates that post CHF heat transfer data from semiscale can be interpreted based on the steam flow through the core alone.
j Dr. Henry discussed the work under way in the area of rewet criteria.
Reviewing the interface temperature equation, Dr. Henry i
stated that he believes it is impossible for a liquid to wet a hot i
surface when the interface temperature contact between the coolant and the tot surface exceeds the critical temperature of the coolant.
{
l i
Present work on the rewet phenomena centers on determining the thermal conductivity of the clad surface after
- has become oxidized.
Using 0FFICIAL USE ONLY I
C e.
(
f 1
0FFICIAL USE ONLY l
)
ECCS 4/22/77 l
data from semiscale test 5-0-34 (Fig D-12), Dr. Henry said the results show that a good estimation of clad rewet capability can be obtained by assuming 3 wall temperature between the range of homogeneous nucleation and critical temperatures expected for fluid contacting oxidized cladding.
Following further discussion, Dr. Catton asked if the following statements are a fair assessment of the work conducted by Dr. Henry. Dr. Catton said that if the interface temperature is greater than the homogeneous nucleat'on temperature, then nucleation phenomena controls whether or.not liquid-solid contact would occur; conversely, if the interface temperature is less than the homogeneous nucleation temperature, hydrodynamic phenomena would control liquid-solid cLntact.
Dr. Henry concurred with Dr. Catton's statements.
Summarizing his work in the area of rewet, Dr. Henry said that once the rewet temperature is known a straight forward analytical solution can be obtained as to how fast heat energy is being pulled out of the clad wall. Dr. Henry said the rate of heat energy release is limited by the internal thermal resistance of the fuel in the cladding.
PDST-LHF HEAT TRANSFER - J. CHEN - LEHIGH UNIVERSITY Dr. Chen began by defining the two existing post-CHF flow patterns (Fig. D-13).
These two flow patterns are dependent on the percent quality at time of CHF (high or low quality).
Dr. Chen described his efforts to model the post CHF heat transfer process.
This work involves the determination of the following quantities:
(1 )-
effective heat transfer coefficient, (2) effective heat flow rate, (3) effective area fraction, (4) effective vapor temperature.
Initial work centered on the determination of the vapor heat transfer phenomena.
0FFICIAL USE ONLY
'\\
V
L.
s<
0FFICIAL USE ONLY ECCS
.g-4/22/77 Dr. Chen said he has chosen to describe the vapor convection coefficient by use of a momentum analogy in terms of effective two phase pressure drop.
He said this is a departure from the normal method of heat transfer modeling which involves the use of the Dittus-Boelter relationship.
Dr. Chen said in order to calculate the vapor heat transfer the actual quality rather than the equilibrium quality...ust be determined, i.e.
the vapor temperature rather than saturation temperature must be known.
In order"to determine the actual quality, Dr. Chen examined some of the mechanisms that govern equilibrium quality formulation. He said the balance between heat transfer to the vapor ~and to the heat sink governs the amount of equilibrium quali.ty seen. A rough functional dependence between the actual and equilibrium quality can be determined by studying the mechanisms that govern equilibrium quality.
Dr. Chen shown a plot of actual quality calculated by the correlation described above versus actual quality points derived from the data obtained from various experiments.
A good agreement between calculated and actual quality data is shown (Fig. D-14); the average deviation of data points is approximately 5%.
Dr. Chen described the development of a model to determine the liquid heat transfer mechanisms which is based on work by Dr. Gill at MIT.
Dr. Chen reviewed three mechanisms for liquid heat transfer from a hot wall (Fig. D-15).
These three mechanisms are:
(1) super heating, (2) bubble growth and disruption, and (3) evaporation of residual film.
Dr. Chen reviewed the series of equations that predict the amount OFFICIAL USE ONLY o
+
(
l l
l 0FFICIAL USE ONLY l
ECCS 4/22/77 l
of heat transfer for each of the three mechanisms descr' bed above (Fig D-16). These equations were combined to give an average heat transfer for each liquid-wall contact. The area of liquid contact is assumed to stochastically vary with driving potential for super heat raised to some power.
Dr. Chen said he developed an empirical correlation to determine the l
time average fraction of hot surface that is under liquid contact (H).
Using F, and the correlation to determine the vapor L
L heat transfer contribution described earlier, Dr. Chen said the heat transfer can be determined during post-CHF for a given set of conditions.
Dr. Chen showed a slide comparing measured and predicted heat flux for post-CHF heat transfer. Measured data were obtained from the NRC Data Bank of post CHF heat transfer experiments conducted at INEL. The average deviation shown in the data is 14.2% (Fig 0-17).
l Dr. Chen said at present, he is attempting to obtain informat' ion on vapor super heat and vapor quality by conducting direct measurement experiments of vapor super heat temperature.
Results so far have been encouraging.
REFLOOD HEAT TRANSFER - P. GRIFFITH - MIT Dr. Griffith began by describing the past and present work in the development of the REFLUX code model.
The REFLUX code model is used to determine the heat transfer calculations for reflood.
Dr. Griffith said that the first two objectives in the development of the REFLUX model was:
(1) demonstrate that it is possible to model the entire 1
0FFICIAL USE ONLY
k1 t
~
OFFICIAL USE ONLY l
ECCS 4/22/77 l
1 boiling curve as a function of mass, velocity, quality,and pressure and calculate the time and teraperature for a quench experiment; (2) demonstrate that the phenomenon of axial conduction can be neglected.
Dr. Griffith reviewed a series of slides that compared REFLUX code predictions versus actual data from boiling tests involving the use of liquid nitrogen.
The slides show compt.. isons for prediction of quench, and wall temperature (Fig D-18 and 19). Good agreement is shown. Another slide shown indicates that the effects of axial conduction can be neglected (Fig D-20).
1 Dr. Griffith described the reflood flow regimes seen for low temperature regions (Fig D-21), and high temperature regions (Fig D-22).
He also described the development of the carry-over criterion used in the code.
Dr. Griffith said that future work is centering on the area of downflow heat transfer during upper plenum draining. He said that at present there has not been any criterion developed to determine when downflow actually begins.
The Subcommittee and consultants questioned Dr. Griffith on the deltails of his REFLUX code model.
INTERFACIAL MASS EXCHANGE PROGRAMS - N. ZUBER - NRC-RSR Dr. Zuber briefly reviewed the programs underway to study interfacial mass exchange. The objective of these studies is to determine the rate of change of void in two phase mixtures.
Programs are underway.
in the following areas:
(1) evaporation during flashing, (2) condensa-tion in separated flows, (3) macroscopic and microscopic studies in degassing.
0FFICIAL USE ONLY
-- _ -- -- s J
g t
e
(
UFFICIAL USE ONLY ECCS 4/22/77 Dr. Zuber reviewed the condensation tests underway by Bankoff at Northwestern University.
The objecti ns of the tests are to determine the steam con-densation rate on a water surface, and to develop conductive tracer techniques and other devices to measure local condensation rates.
Dr. Zuber reviewed the degassing programs underway to determine the release rates of nitrogen from water. A macroscopic degassing study is underway at Argonne National Lab. The objective of this study is to determine the effect of gas release on total flow during blowdown. A slide of the test facility for the study was shown (Fig D-23). A microscopic degassinq study is underway at NASA. The objective of tnis study is to determine the rate of generation of gas during transient depressurization. A slide was shown of the facility being used to study the phenomena of bubble growth rate, the effect of partial pressure, common nozzle shape, etc., on the degassing phenomena (Fig D-24).
Preliminary results from the NASA study have shown that the degassing rate dramatically increases when liquid is saturated or super saturated.
It has also been shown that internal bubble surface area greatly aids the c'.:gasting rate, and that it takes a long time to reach true saturation conditions.
Mr. Ebersole asked if the potential.'f gas binding in U tube steam l
generators is being studied.
Dr. Zuber said NRC is studying the problem from a standpoint of noncondensible gases as they affect the accumulator, but he said this infonnation can be applied to other potential problem areas, such as steam generators.
Dr. Isbin asked if the research programs dis-j.
cussed above represent unsolicited proposals, or has NRC explicitly requested this research work be carried out. Mr. Scroggins said that in general most of the work is solicited, but some of the work is in response to areas of general interest to the NRC.
I 1
0FFICIAL USE ONLY
~
r
g 4
0FFICIAL USE ONLY ECCS 4/22/77 VAPOR GENERATION DURING FLASHING - 0. J0NES - BROOKPAVEN NATIONAL. LAB (BNL)
Dr. Jones began by saying the work being conducted at BNL is related to analytical efforts directed towards the development of nonequilibrium vapor source terms for the post dryout and flashing regimes.
In addition, experimental work is underway with the objective of obtaining baseline data on the flashing flow phenomenon.
Dr. Jones reviewed the analytical effort involved in deriving an equilibrium vapor source term (Fig D-25 and 26). He said that at present, BNL has developed the expressions to describe the vapor source term and a check against available data is now in progress to verify the analytical work.
Dr. Jones showed slides comparing results of calculations for predictions-of actual quality versus derived data of actual quality from tests conducted at MIT and Harwell (Figs D-27 -30 ).JBNL said that the present correlation tends to over predict vaporization rates and as a. result under predict actual wall temperatures. BNL also compared the correlation to data obtained by Swenson. Results of the comparison show fairly good agreement with the Swenson data.
Dr. Jones reviewed the experimental effort underway, the objective of which is to veternine the vapor source term as it relates to the phenomenon of flashing.
Vapor source term determination is being attempted by measurement of the axial quality gradient.
Dr. Jones reviewed the measurement concept used to determine the axial quality gradient. Two coincident probes are used to measure the fluid quali ty. The vapor source term is derived from the quality measurement.
Dr. ' Jones described the optical probes used to obtain the above quality measurements (Fig D-31).
The probes used in the experiments had outside diameters ranging from 200 to 400 microns.
Two of the probes used in the I
experiments were shown to the Subcommittee.
Slides of the test rig used, j
0FFICIAL USE ONLY
~ -
s.
g.
1 4
l 0FFICIAL USE ONLY
)
1 I
ECCS 4/22/77 i
and the electronic traces from the experimental probes were shown (Figs 1
D-32 and 33).
BNL has also developed a test rig to provide a global reference for the measure-ment of void fraction.
Results shown (Fig D-34) indicates +. hat the BNL test rig offers a simple accurate way to measure void fraction.
Dr. Jones summarized his discussion by sayirg that the analytical work has progressed to a point where the post dryout analysis is almost complete, and the flashing analysis is well underway.
For the experimental work, BNL believes that the loccl void probe will provide accurate measurements of void fraction.
Construction of test facilities for measurement of void fraction is underway (30-40% complete).
MOMENTUM EXCHANGE - N. Z'JBER - RSR Dr. Zuber began by describing the need for modeling interface momentum exchange, the modeling requi'rements, and the problems encountered in i
modeling interface momentum exchange (Figs D 37).
The modeling program is concerned with the following topics:
(1) flow topology l
(dispersed, separated, or mixed), (E) the forces acting on the fluid flow, and (3) the effect of geometry and boundaries on the momentum exchange model used.
Dr. Zuber reviewed the seven research programs underway in the area of interface momentum exchange (Fig D-38). The work of Dr. Ishii in the area of drift flux modeling, Dr. Houghton on dispersed flow, and Dr. Hemlick on downcomer interactions was reviewed by Dr. Zuber.
The objectives of Dr. Ishii's work in drift flux modeling are two fold:
(1) derive a one dimensional drift velocity for bubbly, droplet, annular, and annular mist flows; (2) derive a two dimensional drift flux veiocity for dispersed flows.
Expressions have been derived to describe one 0FFICIAL USE ONLY e
s I
OFFICIAL USE ONLY l
ECCS 4/22/77 dimensional drift velocity for bubbly, droplet, and annular flows.
Corre-lations has also been dertved to describe the distribution parameter, C.
o Results of comparisons of calculated and experimental drift velocities for annular and droplet flow were shown (Figs D-39 and 40).
The work by Dr. Houghton on dispersed flow was reviewed.
The objective of this work was to mearure and model transient forces arising during motion of a particle in a fluid.
Slides of the experimental apparatus were shown (Figs D-41 and 42).
Tests were run using different size and shape particle geometries.
Two significant results in the experiments are:
(1) the proximity, of a concentric spherical wall around an accelerating sphere increases the coefficient of induced mass; (2) particle shapes have a pronounced effect on the coefficient of induced mass.
Dr. Zuber reviewed Dr. Hemlick's work in the area of do'.:ncomer interactions.
The objectives of this study include providing experimental data for verifi-cation of multi-dimensional codes developed at Los Alamos Scientific Labora-tories (LASL), and to transfer the instrumentation technology developed for other LASL programs to NRC's creas of interest. NRC is also initiating a contract with LASL to obtain data in correlations to predict flooding conditions at the upper core plate, and to detennine the mechanism for entrainment of droplets above the plate.
SEPARATED FLOW STUDIES - A. DUKLER - UNIVERSITY OF HOUSTON Dr. Dukler began by saying that there are two programs underway involved with the study of phase distribution phenomena.
The first program involves the study of interactions in separated flows.
The objective of th.is. program is to develop a mechanistically based understanding of the phenomena involved.
This program is divided into three components:
(1) modeling of the mechanism for flow reversal (flooding); (2) modeling of the interfacial OFFICIAL USE ONLY r
k-e y
i,
0FFICIAL USE ONLY ECCS 4/22/77 momentum tenns involved in up-flow; and (3) modeling of fluid interaction when droplets are present.
The second program under Dr. Dukler's direction is involved with work on flow regimes.
This program has two goals:
(1) modeling of up-flow transitions; and (2) modeling of fluid in an unsteady state for a horizontal configuration.
Dr. Dukler detailed his work on the separated flow studies program. A schematic of the flooding up flow test loop used for the studies was shown and its operation was explained (Fig D-43). Dr. Dukier reviewed experimental results showing the rate of entrainment for various liquid flow rates (Fig D-44).
Dr. Theofanous questioned Dr. Dukler concerning the parameters of the experiment.
After extensive discussion, Dr. Dukler remarked that he has a basic disagreement with Dr. Theofanous regarding the modeling of the entrainment mechanism.
Dr. Dukler discussed the problems associated with the measurement of fluid particle velocity in the test apparatus. The velocity measure-ment involves the use of a hclium neon laser.
Problems were beir.g encountered with the measurement of the velocity of large fluid particles.
l At present, the problem appears to be solved after a modification was made in the measurement system.
Dr. Dukler reviewed the program undenvay to develop a mechanistic basis for the understanding of flow pattern transition.
He detailed the analytical development of correlations to describe annular transition from bubble l
to slug flow, and churn to slug flow, as well as flow transition from slug to churn flow (Figs D 47).
l 0FFICIAL USE ONLY a
m
(
(-
~
0FFICIAL USE ONLY ECCS 4/22/77 FLOODING INVESTIGATIONS - H. RICHTER - DARTMOUTH COLLEGE Dr. Richter said the objective of his rescarch was to study the flooding j
phenomenon to determine which physical mechanism is responsible for flooding under various conditions, and to examine the competing scaling laws in an attempt to determine the correct scaling parameter for use in modeling two phast flow in large scale pipes.
Dr. Richter reviewed his work in the study of flooding for large pipe cases. A slide of the experimental setup was shown (Fig D-48). The preli-minary results of his expe'iments indicate.that the kutateladze parameter might be the proper scaling even for large diameters.
This function is shown on Fig D-49.
VENDOR PRESENTATION ON FUNDAMENTAL RESEARCH - J. LEATHERMAN - GENERAL ELECTRIC Mr. Leatherman began by stating the reasons why GE is involved in ECCS research.
Among the points he cited were:
(1) to increase understanding of BWR behavior; (2) assure safety; (3) to communicate safety to the public; and, (4) increase the availability / reliability of the prciuct line nuclear plants.
Mr. Leatherman emphasized that he believes research must have a payback.
This payback can take the form of reduced conservatism in regulatory requirements, and reduce licensing review time.
Mr. Leatherman noted that research can indicate both good and bad resuits, therefore equitable judgments have to be made with regard to the effects of research infor-mation on licensing activities, providing for credits as well as penalties.
0FFICIAL USE ONLY r
{
(
(
0FFICIAL USE ONLY ECCS 4/22/77 Mr. Leatherman reviewed three areas of the 10 CFR-50 and Appendix K calculations that GE believes contains excessive conservatism that could be reduced. The areas highlighted included:
(1) decay heat I
(ANS plus 20%), (2) Moody and critical break flow model, and (3) ex-cessive metal-water reaction conservatism (full Baker-Just inside and outside the fuel rod),
Drs. Theofanous and Catton questioned Mr.
Leatherman regarding the concept of changing 10 CFR-50 and Appendix K in a piece-meal fashion without evaluating the effects on the overall model conservatism by the reduction of conservatism in one or two specific Mr. Leatherman agreed that would be preferable, but he does not areas.
believe that reduction of conservatism in the ebove mentioned areas would Dr. Plesset remarked that the degrade the rest of the evaluation model.
new ANS standard has not as yet been officially released, and that the new Moody critical flow model is still under evaluation by the NRC Staff.
Mr. Leatherman introduced Mr. Sozzi to make a technical presentation regarding GE fundamental research. Mr. Sozzi said that he would highlight the research areas that GE believes can provide new infor-mation to give a more realistic prediction of plant LOCA response.
The research areas discussed included:
(1) critical two phase flow, (2) low and countercurrent flow phenomena, (3) film boiling / steam cooling heat transfer, (4) rewetting, (5) spraf heat transfer, and (6) subcooled mixing in two phase flow.
Mr. Ebersole asked G E if successful cm cooling for BWR's recuires both bottom flooding and core spray.
Mr. Sozzi said that both
~
bottom flooding and core spray are required for successful core cooling.
Mr. Ebe.rsole also asked General Electric if they are confident that core spray distribution will be satisfactory for a LOCA situation.
Mr.
Leatherman said that a long term core spray test program is in progress j
at the Vallecitos test facility. He also said that successful OFFICIAL USE ONLY i
-. ------- --- a e
r.-
0FFICIAL USE ONLY ECCS 4/22/77 confirmatory tests of core spray distribution in an.ir environment have been corducted.
WESTINGHOUSE ACTIVITY IN FUNDAMENTAL RESEARCH - J. CERMAK Dr. Cennak began by responding to Dr. Isbin's four questions as described in the Chairman's opening statement. He said the }], definition of fundamental research is resaarch dealing wit.h the understanding of physi. cal processes.
Westinghouse feelsthat there is no work that is really needed in the area fundamental research, but Dr. Cennak did cite several studies that would be useful. These areas include work in flooding and entrainment phenomena, and hydraulic mechanical interaction affects.
Responding to the question regarding W's cctivities in the area of fundamental research, Dr. Cermak cited five specific programs areas:
(1) drift flux modeling, (2) LOCA hydraulics modeling using the BART-BASH code, (3) steam-water mixing. phenomena, (4) rewetting phenomena, and (5) and fluid structure interaction. During a discussion of the above items it was learned that Westinghouse and NRC have been holding annual'research and development meetings. Dr. Isbin requested the he be kept cognizant of any such future meetings.
Westinghouse described their efforts to reply the results of fundamental research to licensing activities. Westinghouse submittals of research results in the areas of steam water mixing, drift flux model modifications, and rewetting phenomena have not as yet been accepted by the NRC Staff for use in licensing applications.
Dr. Cermak remarked that a planned ECCS Subcommittee meeting in July of this year will discuss the continuing work on.the drift flux model and review progress regarding NRC acceptance of the model modifications proposed by Westinghouse.
0FFICIAL USE ONLY
0 l
(
q 4
0FFICIAL USE ONLY ECCS 4/22/77 Westinghouse is currently at work on their MULTIFLEX code. This code models pressure vessel forces seen after a pipe rupture in the vessel nozzle region.
Dr. Cermak believes the MULTIFLEX and BART-BASH codes will grectly aid the W licensing effort.
COMBUSTION ENGINEERING ACTIVITIES IN FUNDAMENTAL RESEARCH - B. GOODWIN -
J. HOLDERNESS Mr. Goodwin led off the CE presentation by stating that CE's recent efforts in the area of fundamental research has centered on the development of a reflood heat transfer code model called THERM. The TliERM code has been proposed for use by CE to generate reflood heat transfer coefficients for LOCA licensing analysis.
CE topical report CENPD-228 documents this application.
Mr. Goodwin introduced Dr. Jim Holderness, author of the THERM code to detail some of the fundamental models used in the code.
He said this code is a phenomenological type code in that it combines mathematical models and correlations to describe the fundamental processes of liquid entrainment, thermal nonequilibrium, and two-fluid heat transfer mechanisms.
Dr. Holderness reviewed the basic heat transfer and hydraulic regimes considered by the code (Fig D-50). He said the dispersed flow heat transfer regime is the principle regime of inportance for calculating peak clad temperature.
To model the dispersed flow heat transfer two things must be known:
(1) LOCA flow rates of liquid and vapor, and (2) the heat transfer mechanism between cladding, liquid, and vapor.
Dr. Holderness said that the entrainment phenomena governs the LOCA' flow rates in the dispersed flow region.
0FFICIAL USE ONLY J
v.
f 0FFICIAL USE ONLY ECCS 4/22/77 CE reviewed the THERM liquid entrainment model (Fig D-S1). There was considerable discussion between the consultants and CE regarding the formulation of the liquid entrainment fraction.
Dr. Holderness said the liquid entrainment fraction has been semiempirically derived using FLECHT test data, and that the definition of the liquid droplet spectrum curve is an area of interest for future fundamental research.
CE reviewed the six heat transfer mechanisms considered by the THERM code for modeling in the dispersed flow region (Fig D-52).
Dr. Holderness said that the THERM computer program has been verified against test data from the FLECHT program, semiscale tests, and single tube tests. CE said the mejority of the veri.fication has been for the integral effects of clad temperature, collapsed liquid level, and exit flow rate.
I FUNDAMENTAL RESEARCH CONDUCTED BY BABC0CK AND WILCOX - C. MORGAN Dr. Morgan began by briefly discussing research work conducted by B&W in the area of analytic investigation of loads generated during subcocled blowdown. B&W compared the CRAFT-2 and RELAP-4 computer codes with the method of characteristics solution of a classic water hammer problem, the quick clesing of valve. Dr. Morgan said the results show that CRAFT-2, RELAP-4, and the method of characteristics solutions in one dimension are equivalent.
The i
results of a comparison of CRAFT-2 with method of characteristics solution for the valve pressure was shown (Fig D-53).
Dr. Catton thought the code results were amazingly accurate.
Dr. Fabic (NRC Staff) felt that B&W had made a mistake,and the results shown were a matter of good. luck.
Dr. Catton concurred with Dr. Fabic's observation.
l 1
1 0FFICIAL USE ONLY
h_
..s
(
0FFICIAL USE ONLY ECCS 4/22/77 Dr. Morgan reviewed another research program conducted by B&W-a series of clad swelling and burst experiments. B&W felt that the clad rupture and swell test conducted at Oak Ridge, whi..h were used as a basis for the development of the ECCS flow blockage model were non-prototypic of an actual accident situation.
B&W ran a series of tests, which they considered more prototypic of actual accident conditions (Fig 0-54).
B&W showed test results that they said indicated the clad rupture angle is almost random in nature.
Dr. Morgan said that the Oak Ridge test had shown a preferential rupture angle. Analysis of the data is not as yet complete.
Citiag a study done by Dr. Bingham (B&W) on the frequency of clad rupture taking into account such parameters as clad thickness, variations in pellet heat generation, etc., Dr. Morgan said
~
the B&W test result agree fairly well with the study conducted by Dr. Bingham (Fig. D-55).
B&W feels Dr. Bingham's model conservatively calculate clad rupture for a LOCA situation.
DISPERSED FLOW STUDIES - R. LAHEY - RPI Dr. Lahey began by saying that the objective of his research in the area of phase separation and distribution is to determine the mechanism for onset of annular flow.
Dr. Lahey reviewed the equations used to derive an analytical expression for radial void distribution in a pipe (Figs D 57). He said that he can ar.alytically predict void distribution in a pipe for an air-water flow si.tuation. The voids are driven by turbulence of the liquid phase fluid.
Dr. Lahey reviewed his analytical work directed towards the determination of the interfacial drag term BE (Figs D 59).
Dr. Lahey said the 0FFICIAL USE ONLY r
A-i t
0FFICIAL USE ONLY ECCS 4/22/77 interfacial drag can be independently modeled by the code develoners.
Dr. Lahey went on to describe his program for the study of phase separation analysis. He said the objective of the program is to derive an acceleratica term for a two phase flow situation. The accelerati.on term is. to be objec-tive,that is, it will not vary under the rotation of a coordinate ssystem.
Dr. Lahty said that comparison of the virtual mass acceleration force against the interfacial drag for~ce has shown that the acceleration term is significant and must be included in the phasic momentum equations.
Work is still continuing in this area.
'Dr. Lahey reviewed some of the upcoming RPI experimental work in two phase flow that is assigned to verify the analytical work described above. A slide of one of the test facilities was shown (Fig D-60).
There was extensive Subcommittee discusssion concerning Dr. Lahey 's derivation of the virtual mass tenns.
PHASE DISTRIBUTION AND SEPARATION PdEN0MENA (PDSP) - N. ZUBER Dr. Zuber began by saying that phase distribution separation phenomena must be modeled through realistic predictions of a LOCA. He said that the objective of PDSP research is to be able to predict phase distribution for " Tee" junctions and nozzle configurations (Fig D-61).
Dr. Zuber reviewed the work conducted on droplet flow phenomenon by Drs. Lee and Azbel which were carried out at Stony Brook College.
4 The objectives of the work are:
(1) obtain experimental data on the effect j
of heated surfaces on droplet motion and distribution in mist flow; i
0FFICIAL USE ONLY i
_ __ _ _ _ _C____
b
(
i 0FFICIAL USE ONLY 1
ECCS 4/22/77 l
l l
l (2) analyze experimental data and describe phenomenological models to predict the effects of heated surfaces on mist flow dist'ibution. Slides showing r
l the results of the work to date are attached for your information (Figs l
D 66).
The recearch work conducted at Los Alamos and Harwell on droplet entrain-ment and de-entrainment was reviewed (Figs D 68).
l 1
l Dr. Isbin asked if there will be feedback between the work conducted at Los Alamos and Harwell (Enaland).
Mr. Zuber said that the instrumentation developed at LASL will be made available for use at Harwell.
TWO-PHASE PUMP PERFORMANCE RESEARCH - W. BECKNER - NRC Mr. Beckner said he would review some of the research underway in the area of two-phase pump preformance.
NRC is interested in this work for two reasons:
(1) study the effect of the intact loop on fluid flow daring blowdown; and (2) study problem of pump overspeed.
Reviewing an example of B&W test data to determine the two-phase friction multiplier, Mr.
Beckner said NRC has three criticisms of test results to date:
(!)large scale steam water tests of pumps are lacking, (2) data produced to date have a lot of scatter and/or uncertainty, and (3) NRC does not believe the data obtained to date have been properly correlated.
l l
NRC reviewed the effort of MPR Associates, a consulting firm that aids the NRC in following the progress of two-phase pump programs of ERPI and l
the NSSS vendors. MPR Associates is also developing a pump data bank.
The two-phase pump research program of EPRI was reviewed (Fig D-69).
EPRI has four programs underway in conjunction with Combustion Engineering, 0FFICIAL USE ONLY 8,
. yECURSE DN$
l ECCS 4/22/77 l
1 B&W, CREARE, and MIT. The work of CREARE and MIT was reviewed in detail.
The CREARE/EPRI test objectives were reviewed (Fig D-70). A slide of the CREARE Lest loop was also shown (rig D-71). The loop can model pump; in three separate environments: air-water, freon-water, and high pressure steam-water. The pump used in the test loop is a 1/20 scal + model of a full size reactor pumo. Future plans include the scale modeling of a 1/3 scale B&W type pump.
In reply to a question from Dr. Plesset, Mr. Beckner said tests would be conducted in both steady state and transient modes.
The joint MIT/EPRI analytical study of two-phase pump performance was re '
viewed (FigD-72). This study will attempt to model the cause of pump bead degradation.
INSTRUMENTATION DEVELOPMENT - Y. HSU - NRC Dr. Hsu said the objective of the instrumentation research and development (R&D) program is to develop instrumentation devices and accompanying techniques to enable the measurement of two-phase flow parameters with increased accuracy. The current needs of the R&D program were reviewed, and a test matrix assigned to address these needs was shown (Fig D-73).
Dr. Hsu cited examples of two-phase flow parameters that must be included in two-phase flow models to give accurate results.
He said the inclusion of the slip phenomena and modeling of velocity and void distribution pro-files in the radial direction will greatly reduce the errors seen in two-phase model correlations.
Dr. Hsu reviewed the development of new techniques in the area of instru-mentat, ion measurement (Fig D-74).
NRC future work will focus on non-instrusive measurement techniques and more refined data interpolation techniques,
,,,, r. 6 o
'eme y-
0FFICIAL USE ONLY ECCS 4/22/77 l
FOREIGN RESEARCH EFFORTS - S. FABIC - NRC Dr. Fabic briefly reviewed the basic experiments underway in foreign
{
countries that he said would be useful for model development (Fig 0-75).
]
During discussion of the above programs, Dr. Fabic said the question had been asked whether there would be any tests aimed at improving understanding of hydroelastic phenomena.
He said that the HDR test facility in Germany will conduct blowdown tests utiTizing a flexible core barrel to study vibration and/or deflection phenomena.
In response to Subcommittee ques-tions, Dr. Fabic said the HDR facility is nearly full scale (3 meters in diam'eter - 9 meters in length). He also said that there would be two independent measurements of critical flow in the break nozzle. Mr.
Etherington asked if the foreign resaarch programs are leaning towards large scale phenomenological type test. Dr. Fabic replied that while the foreign facilities lack a LOFT sized experiment, they do have fairly large scale integral test experiments undenvay such as the reflood tests conducted in Gen.:any and Japan.
FUNDAMENTAL RESEARCH PROGRAMS SUMMAfY - R. SCROGGINS - NRC Mr. Scroggins began by saying that NRC believes they have a responsible model development test program in place. NRC reviewed the separate effects tests undenvay in the areas of model correlation development and verification (Fig D-76).
Responding to the question of how the results of the research will be used, Mr. Scroggins said that the main emphasis of the Office of Research is the development of best estimate codes as well as advanced codes such as TRAC.
NRC research also intends to turn over research results to Nuclear Reactor Regulation for consideration of modifications to Appendix K, if modification is warranted.
0FFICIAL USE ONLY s-
l 0FFICIAL USE ONLY l
ECCS 4/22/77 Dr. Fabic reviewed rerearch programs in the areas of heat transfer, fluid dynamics, and instrumentation that are receiving emphasis at this time and most likely will in the future (Fig D-77).
Dr. Isbin noted that Dr. Tong conducted a task force study last year that addressed the necessity for full scale testing, and concluded
)
that it was not necessary.
The Chairman asked Dr. Tong'if he feels experiences of the past year have changed his conclusion regarding full scale testing.
Dr. Tong replied that there has been no change in the attitude toward full scale testing. Reactor Safety Research feels that if system effects and scaling effects can be verified for advanced codes such as TRAC, full scale testing will not be necessary. Dr. Tong said that scaling effects can be verified by j
separate effect tests, and systems effects can be verified by
)
integral tests.
Both types of tests are being conducted either in I
the U.S. or aboard.
EPRI RESEARCH PROGRAMS - T. FERNANDEZ - ELECTRIC POWER RESEARCH INSTITUTE i
I Mr. Fernandez began by noting the areas of fundamental research that should be addressed.
These areas include:
(1) physically based two-pha:,e flow and heat transfer analysis methods, includi.ng supporting development test and qualification experiments; (2) two-phase flow instrumentation; (3) two-phase scaling principles; and (4) fluid / structure interaction.
l EPRI also reviewed a diagram of the LOCA ECCS research subprogram (Fig D-78).
i
\\
An extensive list of EPRI sponsored projects in various areas of basic phenomena research was reviewed.
Projects were highlighted in such areas as two-phase flow and heat transfer, critical flow, decay heat, reflood, component thermal hydraulics, fuel rod transient response, and advanced systems investigations.
0FFICIAL USE ONLY i
s -
k, y,.
c, OFFICIAL USE ONLY ECCS 4/22/77 During the above discussion, a CREARE research program entitled " Water Reactors Safety Research Support: Task B" was highlighted by Mr.
Fernandez.
In response to questions from Dr. Isbin, Mr. Fernandez said this program involves examination of alternate ECC systems.
He also said that a final report on this project is due to be released shortly.
Dr. Isbin evpressed an interest in cbtaining a copy of this report. The Chairman also asked EPRI if they have any opinion on whether or not improved ECC systems are necessary.
Mr. Fernandez replied that EPRI's approach is to gain an understanding of the current a
systems s'o as to be in the position of replying to the question:
"Are alternative ECC systems really in improvement to what we currently have?"
Mr. Ebersole asked EPRI if they have a program (s) addressing the subject of identifying the minimum configuration of dynamic or active systems at a plant which must be sustained in a safe shutdown after a trip. Mr. Fernandez replied that he. had not shown a complete list of all the EPRI nuclear power division research programs, and therefore he did not know if such a project,or projects,was under way. Mr. Fernandez did provide a list of research and Govelopment projects under the control of the Nuclear Power Division of EPRI (Attachment D-79), but he said that it was not an all inclusive list of EPRI R&D Nuclear Projects.
The meeting was adjourned at 6:37 p.m.
NOTE: A complete transcript of this meeting is on file at the NRC Public Document Room,1717 H Street, N.W., Washington, D.C.
Copies are also available from Ace-Federal Reporters, Inc.. 444 North Capital Street, N.W., Washington, D.C. 20001 (202/347-3700).
0 0FFICIAL USE ONLY t
s, ACRS SUBCOMMITTEE ON ECCS Fundamental Research - ECCS Phenomena April 22, 197~
8:30 AM Introduction - R. M. Scroggins, NRC 8:45 AM Fundamental Research Requirerants - S. Fabic, NRC 9:15 AM Heat' Transfer Cer. relations Introduction 2.Y. Y. Hsu, NRC Transient CHF - R. Henry, *Jil Post-CHF Heat Transfer - J. Chen, Lehigh Univ.
Reflood Heat Transfer - P. Griffith, MIT 10:30 AM Break 10:4a AM Interfacial Mass Exchange Introduction - Y. Y. Hsu,'NRC l
Vapor Generation During Flashing - 0. Jones, BNL
~
11:30 AM Interfacial Momentum Exchange Introduction - N. Zuber, NRC Dispersed Flow Studies - R. Lahey, RPI/N. Zuber, NRC Separated Flow Studies - A. Dukler, Univ. of Houston Flooding Investigations - H. Richter, Dartmouth College 1:00 PM Lunch 2:00 PM Phase Distribution Introduction - N. Zuber, NRC Phase Distr. & Separation - R. Lahey, RPI/N. Zuber, NRC
/}TT)?c H m E N T C w
4."
A_
r.
s.
2-2:30 PM Pump Behavior Introduction
- Break 3:45 PM Foreign Research - S. Fabic, NRC 4:10 PM Sumary - R. M. Scroggins, NRC 4:30 PM EPRI & Vendor Presentations 6:00 PM Adjourn Se M*
M O
9 9
e Oe ea e a e e.
e e
4
)
l l
1 M\\ atw&
I