ML20147B087

Minutes of the Fluid/Hydraulic Dynamic Effects Subcomm Meeting 780523 at Des Plaines,Ii.Purpose of Meeting Was to Discuss NRC Staff Task Action Plans Re Mark II Containment Pool Dynamic Loads Incl Leak Testing & BWR Temps

ML20147B087
Person / Time
Issue date:	07/01/1978
From:	Advisory Committee on Reactor Safeguards
To:	NRC OFFICE OF THE SECRETARY (SECY)
References
ACRS-1539, NUDOCS 7810030200
Download: ML20147B087 (104)

Text

ex.

g, QMt le Date:

7/6/78 thy.f,. g L(

0v 9 gi m s v.a u

%, f i b.y dQ.:

tbcits- /c39 d

h,l PDR 9 /ac/78

~

M MINUTES OF THE FLUID / HYDRAULIC DYNAMIC EFFECTS SUBCOMMITTEE MEETING MAY 23,1978 DES PLAINES, IL The Fluid / Hydraulic Dynamic Effects Subcommittee of the ACRS held a meeting on May 23,1978 at the Royal Court Inn, Des Plaines, IL. The main purpose of this meeting was to discuss the NRC Staff Task Action Plans related to Mark II Con-tainment pool dynamic loads, determination of relief valve pool dynamic loads and temperatures for BWR containments, main steam line break inside of con-tainment, and containment leak testing. Notice of this meeting was published in the Federal Register on Monday, May 8,1978. Copies of the notice, meeting attendees, and the schedule are included as Attachments A, B, and C, respectively.

No requests for time to make oral statements were received from members of the public and no written statements were received.

One short closed proprietary session was held to discuss material related to the design of the German KWU SRV quencher.

The Designated Federal Employee for the meeting was Dr. Andrew Bates.

INTRODUCTORY STATEMENT BY SUBCOMMITTEE CHAIRMAN Dr. Plesset, Subcommittee Chairman, convened the meeting at 8:30 a.m.,

introduced the ACRS members and consultants who were present and indicated that the day's discussion would be with the Containment Systems Branch of HRR on, topics related to the pool dynamic loads in Mark II's, SRV dynamic loads and temperatures in BWR containments, and main steam line breaks in containments.

If there was sufficient time, containment leak testing would also be reviewed.

Mr. R. Tedesco and G. Lainas indicated that they would review with the Sub-committcc the present :tatus of the generic items listed, and that they had present their consultants from MIT, BNL, and Princeton.

y -

mu-

-p W

T-e F/HDE 5/23/78

.Mr. Cliff Anderson reviewed the program associated with Mark II Containment pool dynamic loads, TAP-A8 (Attachment D). The dynamic loads are associated with LOCA -- blowdown forces -transmitted to the BWR wet well area. The loads encompass pool swell loads produc'ed by the initial air forced into wet well, drag and jet loads from the movement of fluid within the wet well, chugging it. ads from the growth and collapse of steam bubbles at the vents, vent lateral loads from the steam bubble collapse, and impact loads from the pool swell phenomena. The majority of these loads occur over a short time period and are over within about 200 seconds of the start of the LOCA (Attachment D-4A).

The loais affect the design and construction" of the 11 domestic March II faciliti es. The lead plants are Zimmer, LaSalle, and Shoreham. Zimmer's SER should be issued this Summer and the fuel load date is July 1979. Later plants will be licensed through 1983. The Mark II plant, Caorso (Italian),

is now in the startup phase. Also Tokai 2 in Japan is in the startup phase.

The significant loads to the containment pool are related to the air carryover phase where one encounters pool swell, drag and jet loads, and impact loads; and the steam blowdown phase where pool boundary loads and vent lateral loads are encountered. The Mark II Owner's put together a report (The Dynamic Forcing Function Report) to identify all the various loads and submitted it some years ago to the NRC. Since then, several new contributions to the drag and jet loads have been identified. These included an acceleration drag coefficient and a transient jet load. The NRC Staff is continuing its review in this area.

/

In the area of steam blowdown the available data are from the 4T facility with single vent tests. The significant Staff review effort has been to assure that the single vent data was bounding or conservative for the multi-vent situation.

The Mark II Owner's Group program for the resolution of the loading problems has been divided into two parts, a short-term program (STP) and a long-term pro-gram (LTP). The STP is intended to derive and establish conservative loads

.g 1

a F/HDEL

' n/23/78 which are appropriate for the -life of each facility.. Emphasis has been placed on obtaininithese. loads on a schedule commensurate with the licensing schedule. The LTP;ob'jectives are intended to provide data which could be used to justify potentia 12 reduction in"certain of the STP loads. The LTP

could allow some of the later plants to relax their design criteria and loads somewhat.

In response to.a quest % from' Dr. Siess, Mr. Anderson indicated that the advantages of the long. term program would come about as reduced requirementsi for snubbers, and pipe hangers due to reduced shaking of the rest of the plant: structures.

Dr. Constaitine Economus NRC Staff consultant from BNL, reviewed the pool

. swell loads that have been identified.(Attachment E). The phenomena under consideration include vent clearing, air bubble formation, pool swell, pool fa11back', and steam blowdown and condensation loads. The vent clearing

~

phase includes the downcomer lateral loads, jet inads on the base mat, and pressure, loads on submerged boundaries. The initial air from the drywell forms a bubble at each individual downcomer which eventually coalesce and form a blanket of air.under the water. The intial air bubbles produce drag loads on submerged components that are somewhat different from the

~

drag loads that are produced on subn.erged. components during the pool swell phase. During the pool swell one has air bubble pressure loads on sub-merged boundaries, drag loads on the submerged components, impact loads

~

-on wetwell' components, wetwell air ~ compression loads on the boundary, and upward diaphragm loads. Following the pool swell there's a pool fallback where drag loads,on submerged components also occur. Later, steam blowdown and condensation loads associated with the downcomer lateral loads, pressure loads'on~ submerged boundaries, and drag loads on submerged components occur

- as the steam continues-to blow down into.the containment.

Tests conducted by EPRI'on a 1/13. scale show that as the air is, injected the bubbles coalesce.into a blanket and unifoniialy lifts a water slug above the air. The tests were carefully scaled and were done in a 90 l

7 -...

7. -.,

e a

F/HDE 5/23/78 sector simulating the Susquehana plant. About 25-30 downcomers were simulated.

In response to questioning from Dr. Siess, Mr. Kudrick indicated that there are not enought structures within the drywell to significantly affect the pressure field around the annular region, so that blowdown into the wetwell should be fairly uniform. Under these conditions there is no reason to expect significant pressure variations that would lead to asymmetric blowdown into the wetwell.

In response to questions from Dr. Yao, Dr. Sonin indicated that the head losses for each of the downcomers was very high in comparison to the pressure gradients in the drywell area. The high APs cross the downcomers in comparison to the small values of AP around the annular region in the drywell leads one to believe the one will have a uniform blovdown.

In response to additional questions, Dr. Economus indicated that downcomers are typically two feet in diameter and that the downcomer gaps are typically eight feet in diameter leading to approximately a 4 to i ratio.

During scaling of the tests, geometric ratios were preserved., The downcomer area to the pool area was preserved.

Orifices were installed in the downcomers to satisfy scali.ng requirements on the enthalpy flux.

Dr. Economus indicated that the downcomers are arranged symmetrically around the circumference of the wetwell so that one would not expect any asymmetry in the pressuri-zation due to non-uniform distribution of the downcomers.

During the pool swell phase various loads have been identified. These include the drag on structures, such as the bracing system for the vents and the vents themselves. There is a wetwell compression which produces a AP in the upper region, and the impact load on any structure that might be within the containment area. The Mark II Owner's Group has identified y," ~,,

.F/HDEL 5/23/78

a. basis for. specification of the pool swell loads. The basis consists

' of. five elements. The first is an~ analytical model which allows the user to get information regarding the pool velocity as a function of elevation, the air bubble pressure, and the wetwell air compression.

The' basis the GE analytical model is a 4T full-scale, single cell test, the EPRI 1/13th scale multi-vent tests, and the PSTF impact test. Finally, there.was~ an analytical model to allow one to go from the t'ransient effects and calculate the pool drag loads on submerged structures. General features of the pool swell analytical model. (PSAM) includes: one-dimensionality, constant thickness of water slug, an all air blowdown, a bubble temperature that equals the drywell temperature, adiabatic compression in wetwell air space, adiabatic vent flow with standard loss coefficients, and a drywell

. pressure history from the FSAR-temperature history using adiabatic compression.

Study has shown that the one-dimensionality,.the constant thickness of the water slug, and the assumption of an all air blowdown are all conservatwns in the niodel.

Inresponsetoaquestion,-itwasindi-cateu that tre entire phenomena of 'the pool swell takes place in approxi-mately 1-1 1/2 seconds of time at the beginning of the LOCA.

In response to a question, it was indicated that one needs a large LOCA in order to get the pool swell phenomena, smaller breaks provide for slower pressurization 1

of the drywell, better mixing of the air and steam, and smaller loads as the air-steam mixture is forced into the well.

In response to a question from Dr. Plesset, Mr. Kudrick indicated that the 1/13th scale test. conducted by EPRI did not show any oscillations in the pressure.and that they were done with air rather than with an air-steam mixture; however..in the 4T test there was a steam blowdown and they performed I

those tests throughout the transient so that pool swell oscillations were observed at later times.

In. response to a question'from Dr. Zudans, Dr. Economusiindicated that during the 4T tests, they measured the gas temperatures and they found that a polytropic pressure / volume rela-tionship was followed rather than an adiabatic-relationship.. The exponent was approximately 1.2.

Dr._Economus indicated the Staff position would V

(

VD

=

=

5 F/RDE 5/23/78 be that the model must use 1.2 rather than 1.4 as the coefficient.

In response to a question from Dr. Zudans, Mr. Kudrick indicated that there are vacuum breakers on the downcomers which allow air to flow back into the drywell following the reactor vessel blowdown and the steam condensation, i

There have been no observed actuations of these vacuum breakers due to low steam flows and large condensation near the end of the blowdown period.

Qualification review of.the pool swell analytical model is based upon work done by Bechtel and BNL. BNL has found the air bubble pressure acceptable, the wetwell air. compression acceptable, and.the pool acceleration acceptable; however, the pool velocity is not considered to be uniformaly conservative.

It is being required that GE and the Owner's Group add a 10% margin to the calculation of the pool velocity. The Staff is requiring that the standard drag coefficients used in the analysis be corrected for blockage and accelera-tion effects. The Staff is continuing its review of the impact loads. The calculation of the maximum pool swell should be based on the minimum of the pool swell analytical model using a gamma of 1.2 or 1.5 time the submergence depth of the downcomers. The upward disphragm aP should be based on the BNL correlation of the 4T tests and the EPRI tests rather than the maximum observed in the 4T test. Generally speaking, EPRI tests show a esme,fhat higher upward diaphragm AP than do the 4T tests. This is probably due to the effect of pure air that is used in the EPRI tests rather than the steam in the 4T tests.

Dr. Economus then reviewed the LOCA and SRV submerged structure loads (AttachmentF). The phenomena which generate the submerged structure loads include: vent clearing with the water jet which induces a velocity and acceleration transient in the surrounding pool,and drag and impingement loads on-the structure. Fallback of the water slug produces velocity and acceleration transients, and drag loads on the structures also.

Formation of an air bubble produces loads related to the charging of the bubble itself, rising of the bubble through the pool, oscillations of the bubble, bubble coalescence into one large bubble, an induced velocity and acceleration transient on' the structures within the pool, and drag loads. Steam bubble gw-,v,

8 F/HD'E 5/23/78 oscillation and collapse also induces velocity and acceleration transients and drag loads within submerged structures in the pool. The Mark II Owner's Group bases to this submerged structure' loads rest on first principal analytical modeling with some simp 1tfying assumptions.

In the long term a series of scaled' test programs will c6nfinn tne adequacy of the analytical models.

Included are the PSTF 1/3 scale test and simple geo-metry 1/4 scale tests. At the present time the Mark II Owner's Group methodology for the water jet loads includes a 1-dimensional model and no induced flow transients in the surrounding pool. The drag computation includes only structures partly or totally i.ntercepted by the jet and also has the standard drag form wh'ere the drag is proportional to the velocity squared. The jet impingement computation is based upon the momentum transfer.

At the present time the evaluation of the methodology for the water jet loads is continuing, however, the Staff feels that the model is not conser-vative near the jet front and they recommend the use of a non-deforming water j

slug as a conservative upper bound. The Staff also feels that the contribu-tion of induced transients is not necessarily negligible and they suggest the use of a potential flow for a moving source to estimate its importance.

1 The Mark II Owner's Group methodology for the air bubble loads includes a model with a spherical source, using the method of images, and the coales-cense-switch to a 1-dimensional model. The drag computation is based on 2

the equivalent uniform flow at the geometric center. Standard (V ) and acceleration (t) drag coefficients are used.

The model also assumes no interference among structures. The Staff evaluation of the methodology for the air bubble loads is also continuing. The use of the spherical source is probably not adequate. Asymmetry and bubble rise effects can increase the velocity by apprximately 10% near the bubble. The use of the equivalent uniform flow of the geometric center is fine if the structure is small compared to the bubble, but when the structure is extended with respect to the bubbic size, they need to use a value that gives them the maximu'n velocity.

It is also being suggested that the standard drag coefficient be corrected for accelerations and for blockage.

Interference p '.-

F ""

=

. 5/23/78 F/HDE' effects between the various bubbles may also come into play, and for structures closer than 4-bubble diameters the methodology is probably not valid.

The Mark II Owner's Group methodology for the fallback loads includes a freefall of the water from the maximum pool swell height and the drag 2

computation based upon the standard V term plus an acceleration drag.

The Staff is cont 4uing their review in this area and finds that the model is generally acceptable except that the drag coefficients should be corrected for blockage.

Dr. Economus then reviewed the steam condensation and chugging loads (Attach-mentG). The origin of the loads is the motion of the steam-water interface at the vent exit. Displacement effects of the interface motion creates pres-sure transients in the suppression-pool which are transmitted to submerged boundaries. The behavior of the source of the loads depends primarily upon the vent steam flux rate; at high steam fluxes the motion of the bubbles is

~

essentially sinusoidal and the amplitude is relatively constant; at low steam fluxes the motion is unstable and the bubble collapses and grows with a frequency of occurrence and amplitude that is random but can be bounded. The charater of the source is also sensitive to the air content of the steam and-the global pre'ssure level witHn the containment as well as some dependence on the suppression pool temperature. The Mark II Owner's Group basis for the chugging load specification on the submerged boundary is based upon the' direct application of the maximum loads observed in r

a bounding full-scale,. single cell test (4T). The application of the loads provide additional conservatism in that an exact vent synchronization was assumed as well as a selection of frequency content that maximizes structural Staff review of the evaluation of the steam condensation and chugging response.

load specification is continuing. The Staff still has some questions on an issue that has to do wi,th the fact that the load specification has been evaluated by measuring pressures on the surface of the 4T facility which has its own structural characteristics and thus the fluid / structure inter-action'must be considered. The other question revolves around whether or not single vent data is bounding with regard to multivent situations.

p

(,

a 4

1 F/HDE 5/23/78 Foreign licensee multivent data indicates that the current specification is probably bounding. i. e long-term program will look at additional multi-vent tests and provide confirmation that the multivent effects on chugging loads are bounded by the current specifications.

Dr. Scanlan, NRC Staff consultant from Princeton, reviewed the 4T tests that determined the effects of bubble collapse on the Mark II sub:nerged wetwell boundary. The origin of the phenomenon is the unstable and rapid collapse of the steam bubble at the vent exit. The collapse produces a sharp underpressure and then an overpressure transient at the vent which is transmitted throughout the pool to the boundary.

In order to properly design the containment, one needs to know the pressure magnitude at both the vent and the wall on the Mark II Containment. GE has based the load specification on measured wall pressures in the GE 4T test facility. Some

-l questions have arisen as to whether or not this is entirely appropriate i

due to the geometry of the 4T facility.

Following the bub'le collapse b

there is a sharp overpressure created which propagates outward toward the Mark II Containment wall. This rebound pressure can be affected by the geometry of the vessel itself,and in fact in the 4T facility the ringing out effect occurs with a frequency that is characteristic of the geometry. The 4T facility modifies the ring-out by lowering the frequency. The Staff has suggested that GE go back and look at the integratinn of the pressure pulse over time in the containment in the 4T test vessel and consider that to representative of what happens at the vent.

In this way it is hoped that interference from the vessel walls will be reduced or eliminated.

Dr. Scanlan indicated that there was a chance that GE would find that they have a lesser load to be used on Mark II Containments than the one that they are presently using if they go back and reanalyze the pressure pulse using the method of an integration over time for the impact.

In response 1

F/HDE 423/78 to aLquestion from Dr. Zudans, Dr. Scanlan indicated that his idea involved taking a conceptual sphere of a particular diameter around the vent and then the pressures at a point would be integrated out over time during a particular chug which would be chosen to represent the worst case.

In response to another question, Dr. Economus indicated that the Mark II Owner's Group have a multivent Mark II model which uses a Monte Carlo technique and assigns different strength signals to different downcomers in order to superimpose the wall loads-.from the various downcomers in a multivent facility. Mr. Anderson indicated that they are using a Monte Carlo method because they eventually want to take. credit for the distribution of chugs when calculating the wall loadings.

Dr. Sonin, NRC consultant from MIT, reviewed the lateral loads of the downcomers that are produced by the collapsing bubbles during the chugging phenomena (Attachment I). He indicated that during the later stages of the blowdown at low steam mass fluxes steam bubble growth and ' collapse occurs in a generally asymmetric manner around the downcomer vents. The asymmetric collapse of the bubbles causes the water slug to impact on the end of the vent producing lateral loads in the downcomers. He explained that the lateral loads are impulsive with loading that ranged from 1-10 milliseconds at an interval of approximately 2 seconds. The downcomer lateral frequency. is generally on the order of 100 militseconds. The direction and magnitude are generally random. Adjacent downcomers appeared to be unaffected by operation of adjacent downcomers that are at least 3 diameters away. Test results are available from three different facilities, the KWU facility in Germany (GKM) used.a 24 inch downcomer, the GE test facility at San Jose, and the 4T test facility with a 20 inch downcomer, as well as an additional KWU-Karlsteir facility with a 24 inch downcomer..The maximum static equivalent loads measured for these three facilities were 8,800 lbs., 2,400 lbs., and 31,500 lbs.,

respectively. The original specifications by GE proposed the use of the higher of the first two values, 8,800 lbs. The Staff has some concern with this because of the large value produced in the Karlstein tests. There were differences in the geometry of the test facilities relating to both the V

P m,p.

F/HDE 5/23/78 sr5 mergence and pressurization of the downcomers and the wetwell area. The major difference in the tests is the structural natural frequency of the downcomer. Dr. Sonin indicated that the static equivalent load should be proportional to the frequency of the structure and the impulse load from the collapsing bubble. When the highest static equivalent load is plotted against a natural frequency, the two German tests lie on a line which indi-cates that they have been associated with approximately the same hydrodynamic pulse and a simple model was derived which agrees reasonably well. Tne GE data point lies considerably lower than the German data. However, there are some reasons why GE's data point might be expected to lie louer. First, they did not report the absolute magnitude of the load but rather the components of the load which would increase the load by a factor of, perhaps, 50%.

Secondly, GE's test did not go up into the high temperature region that the KWU tests were run at and where the Germans found higher loads.

Finally, GE's data was with a somewhat smaller downcomer and when the down-comer size increases one would tend to get somewhat higher loads once again.

At the present time, the recommendation is, that for downcomers with a natural frequency below 7 hertz, the lateral load of 8,800 lbs. be used. For natural frequencies above 7 hertz,it is recommended that a line based on.

a load frequency plot that bounds the available data be used.

In response to a question, the Staff indicated that the Mark II frequencies for the downcomers generally lie somewhere in the range of 2-10 hertz.

Dr. Maise, NRC Staff consultant from Brookhaven, reviewed the impact loads that are due to the pool swell impacting upon structures located above the pool and within the wetwell portion of the containment (Attachment J).

Dr. Maise reviewed the dynamic forcing function report ' impact methodology.

In order to do a structural, dynamic analysis one needs to have a force history for the load of the object. One needs to select the pulse shape, the maximum pressure of the pulse, and the duration of the pulse.

In the DFFR the maximum pressure is correlated as the function of the velocity both for pipes between 0 and 20 inches diameter and for "I" beams between 0 and 5 inches and 5 and 20 inches across the flats. The impact pulse duration has been taken to be 7 milliseconds and is considered to be constant 4

4 p

t F/ HOE 5/23/78 for each of the different structures impacted. Tests were conducted in the GE pressure suppression test facility by holding a target above the water pool and injecting air into the pool. The force applied to the target by the impacting water was then meas'ured. The force was measured in two different ways:

there were pressure transducers on the bottom of the target which were integrated to determine the force, and forces at the support were measured. There were also accelerometers on this target.

The major area of concern that the Staff has involves & slight curvature of the pool at the time of the impact. A flatter pool produces higher peak pressures and a shorter impulse time. There is some feeling that perhaps the maximum pressures from PSTF may not in fact be representative of the bounding pressures one could get in a Mark II pool which might be flatter.

Dr. Maise has looked at some tests where impacts have be done on flat pools (Naval Impact Data) and has found that the flat pool impacts generally produce higher pressures and shorter impulse times. When the ratio of the stresses of impacts by flat pools to impacts in the DFFR were calculated, it was found that at natural frequencies above approximately 140 hertz, the stresses from the impact of a flat pool were significantly larger than thost encountered in the DFFR'. A plot of target size vs. natural frequency for both flat targets and pipes indicating conservative and non-conservative loadings has been produced (Slide J-7).

In summary, Dr. Maise indicated that the Staff believes that the current DFFR methodology for impact may not be conservative for all structures, that the Mark II Owner's Group has been made aware of these concerns,and the NRC is awaiting the Owner's response which should take the form of a modified methodology.

Dr. Zudans questioned whether or not the flat pool testing method took into account the relationship between the duration of the impulse load, the frequency of the structure, and the transferred momentum to the structure.

He. indicated that for situations where the natural frequency of the structure was similar to the frequency of the impulse load there would not be nearly as much energy transferred to the structure and the stresses would not be as high. Dr. Maise indicated that the curves that appear on Slide J-6 tend to peak near the frequency where the pulse duration and r

o a

h/HDE 5/23/78 the frequency of the target are comparable. Dr. Maise indicated that the method that they use may not be entirely correct but they feel that it was definitely conservative. Mr. Kudrick indicated that the approach taken was an initial approach and the Staff has still not firmed up its definite opinions on this point. They have not reached a final resolution and they are still discussing this with the Owner's Group.

Dr. Nelson Su NRC Staff, began the discussion of Task Action Plan A-39 which involves the dynamic loads 'and temperatures in BWR containments pro-duced by safety relief valve operation (Attachment K). The objectives of this Task Action Plan are to establish the $RV load criteria and pool tempera-ture limits for the Mark I, II, and III Containment systems.

In general, this a generic' problem for all three.BWR containment types. The phenomena are generic. the methodology is generic, and the pool temperature limit is related to the type of discharge device rather than the type of con-tainment. The initial safety relief valve discharge pipes were straight pipes that ended in the wetwell containment fluid. Subsequently, the Mark I Containments have been changed to a ramshead device which is essentially two elbows connected together. More recently, GE has developed a quencher which has four arms extending radially outward from the central downcomer.

Each of the four radial arms has small holes in it. The intent of the small holes is to break up the big air bubbles into a lot of small bubbles or a cloud of bubbles during the SRV operation. The effects of the small bubbles tends to reduce the load substantially and it improves the pdrformance of the steam condensation process. Two types of T-quencher have also been developed. One, a KWU design; the second, for Mark I Containments.

In response to questions from Dr. Zudans and Dr. Siess, Dr. Su, Mr. Huber, and B. Sobon (GE) indicated that the quencher arms were designed to accept a torsional load resulting from the arbitrary blocking of all the holes on one side of the arm and the discharge of steam from the holes on the other side of the arm.

In reality.-there is a uniform spacing and number of holes on both sides of the quencher arm and there should be no net torsional effect or rotational effect en the downcomer discharge device, Dr. Su indicated that

7. -

v--

o-FIHDE 5/23/78 there has been several operational experiences with stuck opened safety relief valves. The Huergassen Event in Germany in 1972 occurred when one safety relief valve remained open for 30 minutes. - The discharge was through a straight pipe and vibrations started when the pool temperature reached 160 F.

The pool. metal liner separated from the reinforcing beam in this event.

In the KrJ4 Event in Switzerland in 1972, discharge from a straight pipe with a 2-valve test indicated that vibrations started when pool tempera-ture reached approximately 170 F.

Damage included displacement of catwalk sections and the failure of an instrument line within the wetwell area.

Various stuck-open valves have also occurred in U.S. plants. These include:

Peach Bottom, Browns Ferry, and Hatch 1.

pool temperature reached 122 to 165 F.

There was damage on safety relief valve discharge supports but there was no vibration observed in the wetwells.

In these instances, the safety relief valve downcomers had ramshead type devices on them. Dr. Su indicated that the mechanism for producing loads during SRV discharge loads occurred

~

when the steam bubble at the discharge point becomes larger and larger due to the reduced heat transfer as the pool water temperature increases.

The large bubble eventually breaks away from the discharge point and collapses within the wetwell region. The collapse of the bubble produces the loads on the structure.

In order for this to happen, two criteria must be met:

(1) one needs a large steam mass velocity (on the order of at least 40 lbs/ft/sec),

and (2) one needs a high pool temperature.

Dr. Huber, NRC Staff consultant from MIT, reviewed the SRV hydrodynamic loads, the air bubble loads, and two of the mitigation devices that Dr. Su had mentioned, the ramshead and the GE quencher (Attachment L).

Dr. Huber indicated that the safety relief valve is connected to the main steamline. Typically there are a dozen or fifteen of these within a Mark.II reactor system and the safety relief valve line runs into the containment pool. At the end of the pipe running to the pool, there is one type or another of mitigation device mounted. Generally, the pipe runs are on the order of 100-200.ft. in length 3

and there are about 100 ft of air in the line initially between the safety relier valve and the water and then another 20 ft. of water at the bottom of the pipe. The chronology of the event begins with the actuation of the w

e F/HDE 5/23/78 SRV. The frequency of actuation is approximately six times per year. Steam is discharged through the relief valve into the line compressing the air that's initially in the line ahead of the steam. The water begins to clear from the line and forces its way into the pool. After the water clears (on the order of a second or two after the valve opens), the air follows, and finally after the air, the steam flows into the pool. Dr. Huber indicated that he would be discussing the loads associated with the air clearing and the air bubbles that are formed from the discharge device.

Depending upon the type of discharge device used, one has two or more air bubbles formed in the pool.

The air bubble, pressures are higher than th,e ambient pressure of the pool.

The bubbles rise to the pool and expand as they rise. There is a tendency for them to over expand and oscillate as they rise. Of interest in determining the effects of the bubbles on the pool walls is the knowledge of the peak positive and negative pressure of bubble and the frequency of oscillation of the bubble.

Its rise time and its initial position when it is formed are also of interest. Several different events must be considered when analyzing safety relief valve actuation. Single valve actuation tends to pressurize one of the safety relief valve lines to produce loads associated from the actuation of that one valve. One can also have consecutive valve actuations where the single valve opens a number of times causing loads each time steam is blown down into tne pool. With a consecutive valve actuation, one may have different initial. conditions in that the line may be full of steam rather than air or that the steam-air mix may be in different proportions. One may also have a different water leg length contained within the discharge line and thus one may get higher initial bubble pressures into the pool. One can also have the actuation of a leaking safety relief valve which also tends to give you different initial conditions within the safety relief valve discharge line.

Finally, one can have multiple valve actuation where adjacent safety relief valve discharge lines are discharging into the pool at the same time, and one has the effects of multiple loads upon the wetwell system. The approach used to model a phenomenon for the ramshead was an analytical model with a number of submodels within it. The model includes a line transient y.,

r --,

F/HDE 5/23/78 calculation, a one-dimensional model for the steam discharge, a bubble formation model, and a bubble dynamics model. Once ons has the bubble dynamics, a classical method of images is used to determine the loads on the pool boundary. The model was developed and presented in the DFFR and was compared within plant tests at Quad Cities and Monticello.

Empirical inputs were determined which allowed the model to fit the observed in-plant results.

In particular, a bubble formation efficiency which in a sense measures how well the ramshead is doing its job is one of the inputs.

The initial conditions were adjusted via. the bubble formation efficiency to give bounding model predictions for both the positive and negative amplitude of the loads. At the present time, the acceptability of the ramshead model is limited to the in-plant conditions actually tested.

These conditions involve the Mark I plants and tests at Quad Cities and Monticello where more '

50 tests have been run. Line lengths and air volumes of in-plant tem, are not typical of the Mark II plants.

Dr. Huber then reviewed the GE quencher program. The model is generally applicable to all of the quencher designs. The model for the quenchers tends to be a purely empirical one. Governing parameters include:

the line volume / quencher area ratio, the water leg length within safety relief valve discharge line, the pool area / quencher area, and the pool temperature. Seventy-three tests have been done in-plant in Germany.

There have been extensive German reduced scale tests, and further in-plant tests are underway at Caorso. Results indicated single valve actuation peak pressures range between.6 BAR positive pressure and. 4 BAR negative pressure, consecutive valve actuation peak pressures are on the order of 1.7 times the single valve actuations, the frequency range of the bubbles range between 5 and 11 hertz, the babble rise time is on the order of 3/4 of a second, and the pressure attenuation and multiple valve actuation load super-position has been determined.

It was found that the quencher loads are substantially lower than the ramshead loads.

It is intended that the-data base will be extended and the GE quencher methodology 4

4

~

? Y rr - -

F/UDE 5/23/78 17 Will be applied for Mark III applications. The Mark III application will be based on a statistical model utilizing full scale and reduced scale boundary load data for single valve actuations, consecutive valve actuations, leaking valve actuations, and multiple valve _ actuations.

The superposition of multi-ple valve actuation loads is based on a square root of the sum of the squares principal.

P0OL TEMPERATURE LIMITS FOR SRV QUENCHER DESIGNS (CLOSED PROPRIETARY SESSION _)

Dr. Sonin reviewed data that had been submitted to the NRC by GE in support of operating temperatures and mass fluxos for various designs of quencher device. Proprietary d6ta was obtained by GE'from a foreign source and is being held in confidence by General Electric and the NRC. The iata includes a graph (Attachment M-3) that indicates the local pool temperature and the steam mass fluxes at the time various tests have become unstable and have produced large pressure oscillations in the wetwells. A bounding line is indicated on the graph for which stable operation of safety relief valves have been demonstrated. A quencher performance map (Slide M-4) that indi-cates steam mass flows at the nozzle outlet vs. water temperature for dif-ferent regions of stability for the quencher design. The data on Slide M-5 indicate the negative and positive pressure amplitudes vs. water temperature for four different quencher designs.

It shown that with the proper design 0

of the quencher one can reach almost 100 C water temperature in the pool while maintaining pressure oscillations below plus or minus 3 psi.

In their conclusions, the NRC Staff indicated that ramsheads have not been proven for operation at pool temperatures above 100 F for full reactor pressures. The KWU type quencher has been approved for use up to pool temperatures of 195 F and it has been shown that the pressure pulses are less than plus or minus 3 psi for high reactor pressures and discharge rates and that the values of AP are less than plus or minus 8 psi over the entire range of reactor discharge pressures. The NRC expects that other variations which follow the KWU quencher design guidelines can be shown to have similar condensation performance.

t 4

I M

• WW g

wn

. 1 F/HDE. 5/23/78

- The meeting was returned to open session at this point, i

In response to a question fYom Mr. Etherington,.Mr. Tedesco and Mr. Lainas

. indicated that the Staff was feeling very confident about the adequacy of the technology in the area.of safety relief valves and quenchers and the loads that'have been derived from the testing of these devices. The Staff will be looking at additional in-plant tests and a good deal of operating experience'willj;be.recordedwiththesedevices. Based upon the present

' understanding of'the phenomena involved, the Staff does not expect that any new unknown phenomena will appear. There are a few areas where the Staff has not yet' reached complete agreement with the Owner's Group and Geaeral Electric. They do note that the Owner's Group and that the General Electric Company programs are built around analyses and tests, including in-plant testing, and that these programs are well-defined and that when they are conducted and completed they should pro-vide the NRC with an adequate assurance regarding the overall design mar-gins available for General Electric Boiling Water Containments to accom-modate loads adequately'.

MAIN STEAMLINE BREAK' INSIDE OF CONTAINMENT EVALUATION FOR ENVIR0UMENTAL 1UALIFICATION OF INSTRUMENTS Mr. Baranowsky, NRC Staff, reviewed the present status of the main steamline breakinsideof. containment (AttachmentN). The concern for the main steam-line break inside of. containment ' arose when it was realized that the tempera-tures within the-containment can reach 400 F for short periods of time.

Currently.Jenv.ironmental' test programs for containment equipment are LOCA-envelope oriented.

In general, LOCA temperatures are on the order of 300-34'0 F. 'This program will cover calcuations of the containment response, component thermal response.and' testing requirements for simulation of reasonably conservative containment environmental conditions for postulated main,steamlinebreaks-(MSLB).

The program plan. includes an interim nosition which was completed by the Staff;in February 1978, the evaluation of the

' technical ~ bases for _a" Staff position which is targeted to be completed in' September of 1978,'the establishment of analytical critoria which is t

0' p:r-TN n

-.m1,-w y,~

44,-

s

F/HDE-5/23/78 targed for completion in:0ctober of 1978 and the establishment of require-ments for test simulation whieft is also targeted for completion in October of 1978. The high temperatures within the containment for a MSLB accident are higher than those for a LOCA because of the high enthalpy of the steam released during this accident. During a LOCA a large amount of internal energy of the primary fluid is used up in transferring energy from the phase to the gaseous phase. When all of the energy sources and heat 4

sinks are considered, one may typically end up with 60 psi pressures and 700 BTUS per pound mass energy content in containments for a LOCA which yields a temperature of approximately 290 F.-

For a main steamline break one may end up with 60 psia pressures and an energy content of 1200 BTUs/

lb-mass yielding approximately 565 F temperatures, tir. Baranowsky reviewed the best estimate evaluation model that the Staff had developed to help in their review of the steam generator blowdown and containment temperature analysis. They considered steam generator blowdown models with different break sizes and different itquid entrainment fractions in the effluent. They looked at containment heat sink condensing heat trans-for coefficients. They luked at heat sink condensate revaporization.

The condensate revaporization term is a factor which takes into account inefficiencies at the boundary layer where the steam and the air mixture is. undergoing the condensation process.

After a containment atmospheric temperature profile was obtained, they evaluated the heat transfer to various components within the containment considering voth condensing and forced convection to get a component temperature response. They look, in a quali-tative manner, at the spacial temperature distribution and heat transfer variations within the containment that might affect the outcome of the calculation. They took the analysis of various typical components and compared them with the qualification test ranges. The Staff's intent is to be able _to come up with a better specification for qualification testing of equipment that needs to be furctional within a containment.

6 p %.

5S

~

F/HDE

- 20L-5/23/78' In response to al question from Dr. Zudans, Mr. Baranowsky indicated that the Staff does not anticipate substantial changes in their qualification requirements.of equipment within the containment. He indicated that in a couple of instances they have found marginal qualification of equip-ment and then in those instances people have requalified the equipment.

In response to a question from Dr. Lawroski', the Staff indicated that they would make available to the Committee, a Brookhaven report, that discusses containment leak testing. This topic will be discussed at a future meet-ing of the Fluid / Hydraulic Dynamic' Effects Subcommittee.

Dr. Plesset thanked the participants and adjourned _ the meeting at 3:32 p.m.

NOTE:

For additional detailed information concerning this meeting, a complete transcript may be found in the NRC Public Document Room at 1717 H St., NW, Washington, DC or may be obtained from Ace-Federal Reporters, Inc.,

444 North Capitol St., NU, Washington, DC 20001.

1 i

7.-,

i w, w.

1

n NOTICES 19789 Dated at Wrethington. D.C. this 28th and Indemnity Group, Offlee of Nu.!

written statementa may be prc:entrol day of Aptt! 1378.

riest Renctor llenulntion, on or beforc l by meinet rs of, tite pishlic recorrtincs :

Uret rto SrAt M Nuct.ran July E IM The rectunt. should bc will be permittert otily csurtne tho.;c ;

portions of the meeting when a tran.

filed its ennne ction witt Dorket Nos. l heript is beurt kept. nmi que.: mm mav j RKcutatony Cor.tr.stutorv*

SI'N U.500.A and tit'N 50 59.l. A.

I SAMUts.J. Citt!.x*

The Environmental if.coort we ten.;

be asked only by mcrnbert; of the Sno. I dcred but Imtlally rejecteti and is ex.1 corumittec, its consultants, and Sintf. l pected to be recubmitted on or before j Per::ons de:.arinc to make oral state. i (TR Doc. 78-12177 Filed 3-5-78; 8:45 am!

September 1. 19711. A urpo ste nottec g ments should noufy the Desit:nnted *

• of receipt and nvallability for thls rc.j Federal Employec as far in advance as '

anaining portion will be puultshed at practicabic so that appropriato ar.

N590-01.]

that time. A desditnc for filing of rangements can be matic to allow the (Docket Nos. STN 50-532 and STN 50-5033 other content!or.3 relatin:t to matters neccatry time during the meetmg for.

covered in the omittcd material will be il such statements.

i AA!20NA PuttlC $i2VICZ Co.,IT A1.

cstaclished by the floard subsequenti The agenda for subject meeting j (PALO Yttog HUCtf AR CEHERATING to acceptance of the Environmentalj Shall be as follows!

g STATION,113413 4 Ano 3)

Report for a detailed review.

g rocsoAy, naAv :s. wie t

Atter the Environmental Report has f 4:Ja a.rn. Unitt the conchutors e/ Business i

ge<.ipt of Antitrust inform : en end Appue.

been received and analyzed by the i

.De Subcommittee mar meet in caceutive Nea fee Censerv< tion reemas end onhaniait Commission's Director of 14w: lear Rc (

t!<enses: Time fee hbminaien of Views sa actor Regulation or his designee, af bI pr cNat.nN en to u ore i Aa'it'ud M*H*'e draf t envircnmental statement will bc:

their preliminnry opintons terratuin.i c:at.

Arizona Public Sc vice Co. on behalf preparsd by the Commission's staff.:

ters whien snould be eenndered durce tr.e of itscif and ten joit app!! cants--

Upon preparation of the draf t enyt.

meettnc anri to formuMte a report'nnd ree.

Southem California Ed!:en Co.. El ronmental st.atement. the Commission 1 ommenondons to the full committee.

Paso Electric Co. San Dicco Gas and will cause to be published in the Pro.q At the conclusion of the 'r:xceuuve Ses.

a t

o avai bill 4 g t,he Sube i i

e Ih o s 7 Rcets Elcetric Co Nevnds Power Co De, y,

I.utes of the ;rnC Staff, the c.mcret Elec.

partment of Water and Power of the t

cornments frorn interested persons on 4 city of Los Angc!cs, c:ty of Anaheim.

the drnit statement. Upon considera*l.tric Co., and their consultants, pertment to a city of Durbank. city of Glendale, city t

t tion of comments submitted with rc*l' his review.The succommittee rney thers enueus eo de !

of Passdena, and city of Riverside, Caltf.-(the applicants), pursuant to spect to the craf t emironmentni statc9 termine whether the mnttr:rs idenurled tr.

section 103 of the Atomic Enert y Act rnent. the staff will prepare a finsi en.j the truttAl ScSMon hnYe b*en Wiccuntcly eCV. *

• of 1954. as amended. filed partions of vironmental statement, the availabd., cred and whether the istolect is rency for j

their application. There pt:rts which ity of which will be noticed in the Fro-j review by the fuit couumstee.

consist of the Safety Analysts Report. Ea AI. RIcts tra.

In addition. It may be necessary for general and financial informntion Copics of the Individual portions of h the Subcommittee to hold one or more.

were accepted for docketine on March the appilcation, u noted above are j closed sessions for the purpose of ex.

31.1978 nr.d are asstimed Dochet Nos. availabic for public examination and4 ploring matters involving proprietar7.

STN 50-502 and STN 00-503.

copying for a fee at the Commission's a information. I have determined. !n ac.

In addition a portion of the applien. public document room.1717 H Street d cordance with Subsection 10(d) of i llon illed contains the information rc, NW., Washington D.C. 20553 and at J Pub. L. 33-4C3 that, should such ses- :

t! c Phocn!x Public Library. Scienec t hIcDowell Rond. Phoenix. Artz. 85004.] sluns be required. It is quested by the Attorney Ocncrnt for and Industry Section. 12 East closc these set:stons to protect proprt.

the purpose of an antitrust review of l

ctory information (S

U.S.C..

the app!! cation a set forth in 10 CISt ter ard!ng li Part 50. Appendix L. and was also ac.

Dated at Bethesda, Md., this 20th 'i 552b(c)(4)).

repted for docketime and is asstrued day of April 1073.

Further Infortnation Docket Nos. STN 50-502 A and STN toples to. be discum:cd whether the -

50-503-A.

Fon Trre Nect.zAn RtctetATonY meeting has been cancclied or resche-I The ripplication is for nuthort:ntion C9%ttsstoN.

duled, the Chairman s ruline on re -

quests for'the opportunity to preant :

to construct and operate two pressur' STot.s JottN F'Rcacto'rs oral statements nnd the titac idlotted hed water renetors designated as the i

i C7 tic / I,fgh t Water

&r can M otunuM by n pNu Lc!cphone call to the 13e::tenated I,M Palo Verde Nuclear Ocucrating Eta.

Branch No.1. Diptsfon of Prof. t tlon. Units 4 nnd S on the apphennts.

e d.,,

8CI "8785#"O crnt Employec for this mcctinn. Dr. $

site in Marleopn County Aris. The rc.

actor 13 desilmed for operation at a IMI Doc. 78-12170 Flhd 5 5 70: 8:45 am!

Andrew L.13ntes telephonc 202 634-core power Icvc! of 3000 menn wattsI' [7590-01]

1010 between !!:15 a.m. and 5 p.m.,

thermal, with nn cruuvalcut net elec.

c.d.t.

ic ott mL of approximately 1307 Aovisory Comnitt ON Rf ACTOR 5AFE.

DalCd: MBy 3.1078.

A Notice of I!carinc setting forth.

WAtos snComum oN nmy.

JonN C. Hom the radiologient loues to be considered:

ORAUUC DYNAMIC LULCl$

A4f pfsoty Comm t flec mnagentent O//tcen durtnti the resiew ts betu:: put:liched '

g,,g,8

. separately. A dnte for submitting p*Lt.,

Int Doe. 781:547 tWd 5 5-in; n% nm)

I tions for Iceve to intervene on Indiolo.(

The ACitS Subcommittee on Fluid /

ttcal imuca is set forth in the Notlee of' Ilycrnulle Dynamic Effects will hold n

[3110-0ll

IIcaring, mccitun on May 23,1tritt nt the Itoynt OFFICE Or MAtlAGEMENT Af1D Any perr.on who wir.het to have his Court Inn,1750 S. Elmhtnst Road.

BUDGET

?!cwn en the antitrust. ntatters of the!

1)es P!n!nes, llL. to dia.cuu the sta'ts topilention presented to 6he Attorney 1 items sclated to the Mnrk I.11 and 1H CLEARAlict of Rii* ORTS UCneral for conahleration t.tunt!d Containment Systems, ilsts of R gvests submit such viewn to the U.S. Nuclear In accortt.mec with the procedures The following is n list of renuests for Itci:ulatory Conunt.wlun. Wehinr. ten, out!!nmi in the la nruAt.1&mun on elenrnnce of reports intended for use D.C. 20M5. Attenhom Chief. Antitrust October 31, lin*l. pace 14971 or.tl or ! !n collecting information from the s

(,

ffDfRAt REGISTtt. '"L 43. NO. 89-MONoAY. MhY 8.1978

FLUID /HYLMULIC DYNAMIC EFFECTS SUBCOMMITTEE MEETING DES PIAINES, II.

MAY 23, 1978 ATTENDANCE LIST ACRS NRC STAFF M. Plesset, Chairman C. Anderson C. Siess H. Ornstein H. Otherington' J. Fudrick S. Lawroski J. St.apaker H. Sullivan, ACRS Consultant R. Tedesco L. S. Yao, ACRS Consultant G. Lainas Z. Zudans, ACRS Consultant T. Su A. Bates, Designated Federal Employee W. Butler P. Baranowsky G. Minor GENERAL ELECTRIC COMPANY l

I L. Sobon-ENERGY, INC.

G. Niederauer BROOKHAVEN NATIONAL LABORA'IORIES C. Tung ILLINOIS ATIORNEY GENERAL C. Economos R. Scanlan S. Sekuler P. Huber G. Maise E. Dowell S'IONE & WEBSTER A. Sonin J. Metcalf SARGE!ff & LUNDY J. Dukelow O

w v

w.

=

-2up, r

PRESENTATION SGEDULE

FLUID / HYDRAULIC DYtWIIC EFFECTS SUBCOMMITTEE MEEfING DES PLAINES, IL TUESDAY, MAY 23, 1978 I.

CONTAINMENT LEAK TESTING 8:30 a.m. - 10:00 a.m.

Task Action Plan A-23 II. DEfERMINATION OF RELIEF VALVE POOL 10:00 a.m. - 11:30 a.m.

DYNAMIC,f. DADS AND TEMPERATURE FOR BWR CONTAIIMENTS Task Action Plan A-39 LUNG 11:30 a.m. - 12:30 p.m.

III. MARK II CONTAINMENT POOL DYNAMIC 12:30 p.m. - 2:00 p.m. ~

LOADS Task Action Plan A-8 IV. MAIN STEAM LINE BREAK INSIDE OF 2:00 p.m. - 3:30 p.m.

CONTAINMENT Task Action Plan A-21 1

I I

h s

I r

\\

l

y/t&w III. MARK II CONTAlfENT POOL DYtWIIC LOAD, TAP A-8 A.

INBODUCTION (C. AtOERSON) 15 MIN.

B.

LOCA Pcot DYNAMIC 10 ADS 1.

POOL SWELL LOADS EC0f0MuS) 10 MIN.

2. DRAG AND JET LOADS (ECONOMUS) 15 MIN.

3. CHUGGING LOADS 6CONOMUS, SCANLAN) 25 MIN.

4. VENT LATERAL LOADS (SONIN) 10 MIN.

5.

fbOL SWELL IMPACT LOADS.(MAISE) 15 MIN.

4 4

D~I

1 k

e LOCALOADSOVERVIEW SEQUENCE OF EVENTS SIGNIFICANT REVIEW AREAS MARKIIOWNER'SGROUPPROGRAM STP,LTPANDDAR SUPPORTING PR0eRAM NRCA-8REVIEWPROGRAM

.- REVIEWPROGRAM REVIEWSCHEDULE MARKIIPl.ANTSCHEDULES LEAD PuwTS LICENSINGSTATUS FUELLmDDATES O.

4 6

I

s.

e% '

r;;- -

u i i

i

s. '

1

~

)*.

C'1

/

\\

)

L*.'d

4. ;

e dy

.h l

- -l j

6' p-6

,q.

.f

[

Y

..,(

f>

r; t,

0 *:

REACTOR

.*.o.

'1 p

Q f

e c.

MAIN STEAMLINE

'.g.'.'

[

's i N A W T_c 6 G4 N.' m a

d.

.f d

q o-0.

5AFETY RELIEF VALVE 6.f j

6-

=

i

'9

a. -

'.**.c c*,'

r

,;/

h

=

h

. JET S AMMG i ' f,..'

DEFLECTOR $

{.

I/\\\\

-+s-g J,2 b

}%-

?:.,

L c

r

..h [,b,U,*7MIM 'ISD*h Z

f a

pp

- DOWNCOMER s.c a.

?g h

?$('

&n.,a.

s

+

~

,s,

+

• a,,,,s

,s,.

w s

+

'e e

.i

'Q1 4

g=

ho').

D

'74 6

p

^

^

".l Froth Impingement on Structures l if l

Fallback l---4>{

Fallback Loads l

V Air / Steam Flow Continues Wetwell Pressurized Post-Swell Wave Loads V

i steam condensation H

Pressure oscillations l

1r Blowdown Over Lateral Loads on Downcomers Due to Chugging if l

ECC5 Reflood M

Negative Pressure l

I II Long Term.ieatup Thermal Loads Second Pressure Peak D-Y a

..__._._m

A LOCA Sequence of Events Time Phenomena Potential Dynamic Leading Conditic.-

Loca Occurs

- Sonic Wave 0

- Compressive Wave lI Uowncomers cleared of Water

- Water det Loads and Air Flow Starts

- -p.

- Reaction Loads on Downcemers

- Bubble Load 0.8S

- Lateral Loads on Downcomers y

Fool dwell in a bulk Mode

- Impact Loacs

- Wetwell Compression Dra; Loads on Submerged

0. 8+1. 55 Structures 1f l

ureaKthrougn l

1.55 V

l Fool Swells in Frotn Moce b--- pf - - Frotn impingement on Struc. i l V

1. S+5S I

ra i lback M

- Falloack Loacs v

Air / Steam Flow Continues

- Wetwell Pressurized 1+205

- Post-Swell Wave Loads V

I Steam condensation H

- Pressure oscillations 4+2005 (cond. oscillations) 3r w wn ver 20+ 500S

- Lateral hoads on Downcomers ;_

(chugging) to Chugging 605 (end blowdown) jf 110S

)

ECCS Reflood F--- ->j

- Negative Pressure II Long Term Heatup

- ThermaI Loaos 4

_(1-4)x10 S

- Second Pres,ure Peak Peak drywell and wetwell pressure SOS O-k Maximum diaphranm A P down 0.7S Maximum diaphraama P up 2.0S

- p 5

l

m a

m e

- 4 4

9 e

' b 9

0 SIGNIFICANT REVIEW AREAS

/

AIRCARRYoVERPHASE

- Poot SWEu. LaADS

- DRAG AND JET LOADS

- IMPACT l o" DS A

1 STEM BLOWDOWN PHASE

- Poot BOUNDARY LOADS

- VENT LATERAL LOADS e

e

\\

1 l

1 emme 4eo seeeeesepee se epus m. p en se d as

. eusses suggenome som M een esema*

= e...

a****

• gre 4 p-

.. -... ~ -

1 l

l.

sat TERM PROGRAM (GENERIC)

- ESTABLISH CONSERVATIVE LOADS APPROPRIATE FOR THE LIFE OF EACH l

M4aKIIFACILITY

- EMPHASIS ON DEVELOPING LOADS C04ENSURATE WITH LEAD PLANTS

. LICENSING SCHEDULES LONG TERM PROGRAM (GENERIC)

- PROVIDE ADDITIONAL INFORf% TION TO JUSTIFY A' REDUCTION IN SPECIFICSIPLOADS

-CONFIRMATIONOFSPECIFICSTPLOADS O -L

_. - p. -.

o

\\

s DESIGN ASSESSMENT REPORT (RM UNIQUE)

- PLANT UNIQUE APPLICATION OF GENERIC POOL DYNN4!C LOADS AND METHODS i

- ESTABLISH POOL, DYN#41C LOADS EXCLUDEP FRCN THE GENERIC PROGPMi

- EVALUATION OF EACH MARK II PLANT (STRUCTURES PIPING AND EQUIPMENT)

TO POOL DYNAMIC LOADS 9

9 4

e

' - ~ -

e-6 e

KEY MARK II SUPPORTING PROGRM IASKS AIRCARRYOVERPHASE

-4TTESTS

- Poot: SWEU. MODEL

- EPRI 3D TESTS

-DRAGANDJETMODELS

- DRAG LOAD IESTS (LTP)

- PSF TESTS

- EPRI UNIT CA f IESTS (LTP)

STEM BLOWDOWN PHASE

-4TTESTS

-LICENSEEIESTS

-FSISTuDIES

- SINGLE CELL MODEL (LIP)

- MULTIVENT % DEL (LTP)

-MULTIVENTTESTS(LIP)

-DYNAMICLATERALLOADS(LTP)

- MONITOR WORLD IESTS (LTP)

/D - P

..--.3..

{4

. ~..

MARK II PEVIEW MILEST0fES APRIL,1975 NRCMARxIILETERS

• NOVEMBER,1975 GENERIC LOAD REPORT APRIL,1976.-

Pa>N:: 1CUESTIONS OCTOBER,1976 FCUND20UESTIONS

%Y,1977 STP/LTPPROGRAM SUTEMBER,1977 TAPA-8

%RCH,1978 REVISED SUPPORTING PROGRAM REPORT

• APRIL,1978 STP DOCUMENTATION CmPLETE

~

JUNE,1978 STPACCEPTANCECRITERIA AucuST,1978

. LTP REQUIREMENT SEPTEMBER,1978 STPSER

• APRIL,1980 LlP DOCUMENTATION COMPLETE OCTOBER,1980 LTPSER

• MK II 0.G. MILESTONES O-9 4

e

b 1

.I p

g-l i

~

~

li

~

NRK II FACIUTY SCHEDULES 1976

-1977

, 1978 1979 E80

, 1981 L 1982

. 1983 L

g g

R

• ZimER 1 O

b b

s

• LaSAU.E 1 & 2 0

O O

D i

• SmRBW1 b

b l

HANFORD 2-C.

/1 O

. Susoua nta 1 & 2 O

t.'h

.O O

is ru r

l LIMERICK 1 & 2 G

o o

Cy NIN=gigE

.}

Q e

a aiu.y i

O FSAR b

O SER 1

O FUEL LDAD

• LEAD PuwTs n

'.9 d

9

^

.i l

MARK II CONTAI M ENT - SUPPORTING PROGRAM 3

LOCA - RELATED TASKS l

TASK TARGET BATE

' NUMBER ACTIVITY ACTIVITY TYPE COMPLETION DOCUMENTATION DOC /5UBN A.1 "4T* TEST PROGRAM Phase I Test Report Completed NED0/NEDE 13442P-01 5/76 - 5/76 LP SER l

Phase I Appi Memo Completed Application Meno 6n6 - 6/76 LP SER Phase II & III Test Rpt Completed NED0/NEDE 13468P 12R 6 - 1/77 LP SElt t

Application Memorandue Completed NEDE 23678P In7 - 2n7 LP Sell f

A.2 POOL SWELL MDDEL REPORT Model Report Completed NED0/NEDE 21544P 12n 6 - 2n 7' LP SER I

l A.3 IMPACT TESTS PSTF 1/3 Scale Tests Completed NEDE 13426P 8/75 - 9B S LP SER I

Mark I 1/12 Scale Tests Completed NEDC 20989-2P 9n5 - 11n5 LP SER A.4 IMPACT E DEL PSTF 1/3 Scale Tests Completed NEDE 13426P 8# 5 - 9/75 LP SER Mark I 1/12 Scale Tests Completed HEDC 20989-2P S n5 - 11# 5 LP SER, A.5 LOADS OM SUBMERGED LDCA/RH Air Bubble Model Completed NEDO 21471 9/77 - In8 LP SER STRUCTURES LOCA/RH Water Jet Model.

Completed NEDE 21472 9n7-1#8 LP SER Applications Methods Coepleted NEDE 21730 Quenc. Air Bubble Model 2Q 78 NEDE 21570P

'12 # 7 - 1 # 8 LP SER Quenc. Water Jet Hodel 3Q 78 NEDE 23539P Steam Condensation Model 3Q 78 NEDE 23610P Appl. Memo. Supp.

3Q 78 NEDE 21730 Supplements Simple Geometry Tests 3Q 78 Report 1/4 Scaling Tests 3Q 78 Report Model/ Data Eval.

4Q 78 Report A. 6 CHUGGlHG ANALYSIS AND.

Single Cell Report Completed NEDE 23703-P 9/77 - 11 # 7 LP SER TESTING Multivent Model Completed NEDE 21669P 2n8 - 3/78 4T F51 Report Completed HEDE 23710-P 4#8 - 3n8 LP SER l

A.7 CHUGGING SINGLE VENT CREARE Report 2Q 78 NEDE 21747P A.9 ERPI TEST EVALUATION EPRI-4T Comparison Completed NEDO 21667 8n7 - 9/77 LP SER*

EPRI 1/13 SCALE TESTS 3D Tests Completed EPRI HP-441 4/77 - --

LP SER*

EPRI SINGLE CELL TESTS Unit Cell Tests 2Q 78 Report A.11 MULTIVENT SUBSCALE TESTING Preliminary MV. Prog Plan Completed NEDO 23697 12R7 - 1#8 LP SER.

AND ANALYSIS Phase I Scaling Anal.

2Q 78 Report Phase I Scoping Tests 4Q 78 Report Phase II HV Test &

Final Prog. Plan IQ 79 Report Phase II MV Test Rept.

IQ 80 Report NV Tests - Final Rept.

3Q 80 Report I A.13 SINGLE VENT LATERAL LGADS Dynamic Analysis 2Q 78

. Report

' A.16 CHUGGING LOADS IMPROVEMENT Ringout Removal Analysis IQ 79 Report g

"Subeltted in response to MRC questions.

.(.

4

lL i

'i l-c

.r l

MARK II CONTAINHENT - SUPPORTING PROGRAM SRV - RELATED TASKS

-TASK i

(

NWEER ACTIVITY TARGET DATE ACTIVITY TYPE COMPLETION DOCUMENTATION DOC /508M 8.1 QUENCHER EMPIRICAL MODEL DFFR Model Completed ~

~NED0/NEDE 21061-P Sn6 - Sn6 -

Supporting Data:

Completed NED0/hEDE 21078P Sn5 - 7R5 B.2 kAH5 HEAD MODEL DFFR Hodel Completed NED0/NEDE 21061-P 9n6 - 9n6 LP SER Supporting Data Completed NED0/nEDE 21062-P 7n5 - 10n5

. Analysis Completed NED0/NEDE 20942-P Sn5 - 7n5 B.3 -

n)NTICELLO IN-PUUIT Preifainary Test Rpt.

Completed NEDC 21465P 12n6 - In7 LP SER 5/RV TESTS' Hydrodynamic Report Completed

.NEDC 21581P 8n7 - Sn?

B.4 C0!!SECUTIVE ACTUATION Analytical Models 2Q 78 Report TRANSIENT ANALYSIS k.5 5/RV QUENCHEls IN-PLANT Test Plan Completed

'NEDM 20988 'Rev. 2

-CAOR50 TESTS Test Plan Addendum Completed NEON 20988 Rev. 2. Add l' 10n? - 3n8 12R6 - 3n7 Prelle. Test Report 3Q 78 Report Final Report.

IQ 79 Report B.6 THERMAL MIXING MODEL Analytical Hodel Completed NEDC 23689P 3n8 - 3n8 B.7 SRV LINE CLEARING Analytical Model.

3Q 78 Report i

B.10-MONTICELLO FSI Analysis of FSI 2Q 78-Report LP SER B.11 DFFR RAMSHEAD MODEL Data /Hodel Comparison Completed NSC-GEN 0394 9D7.- 10n7 TO MONTICELLO DATA B.12 P.AMSHEAD SRV METHODOLOGY

. Analytical Methods Completed NEDO 24070 SUK'.ARY lon7 - 11n7 LP SER B.14 QUENCHER EMPIRICAL MODEL Model Confirmation 2Q 79 DFfR Rev.

UPDATE B.15 QUENCHER HULTIVALVE MOCEL Statistical Hodel IQ 79 Report 4

B.17

'Q ANALYTICAL HODEL Model Development IQ 79 DffR Rev.

Q B.18 Q FORCING FUNCTION Data Evaluation IQ 79 OffR Rev.

i I

B.21 Q EMPIRICAL MODEL Statistical Evaluation IQ 79 DFFR Rev.

IMPROVEMENT l

8.20 Q ATTENUATION Data Evaluation IQ79 DFfR Rev.

i 1

• s-a LL L L L L

l

=

W W W W W W

W W

2 3

$ $ 2 h

3 R

~~ ~ ~

R. R. R. R.

3 R R. R R.

R. R. O. R. R. R. R.

~~~~~~~

g m

~

W' 2*

R*

R

~~ ~

R R

RRRRRRR 23 33 2

33 3 3 3

h333333

~,

h-d'

.t

.I!.

N N

NN

. ~.,. I 1 1 1.

11 i

~

4=== =.

mm ttttttt 5

a $$s 3

2 0

2 E2

-3 $$

EE *2:2200 SSSSSW

~~~ ~ ~

s~ ~ ~

~~ ~~~~~~~

g 2

20 Was W W SW W a a0 Wa*****

y i

• • ttt t t "t t t 5% 555555%

e 88'888 8 8 88 8 8 88 8888888 a

3 m

m g

8 WW WWW WW 5W W W WW WWWWWWW e

a 83,g 77 77 7

77 7 7 7

am ac.

1.*g #$

II II'e I " II $ I Ie'. IIIIIil m.I T

3 o

m om-om~

m ~ mm o o m~

mummmmm a

We au EW 8

3 j

--}

~~

4""

E O

E g

-~" gig a 5g! g=22!! a.

as es e

=

==

E m

-k ee

-a a

ma

.=

.g si a

.2

.g a

h e

sss 4,

4"e e e es ce 4

I""

m

^^

5

t%

I" 53-3 W,

.$$ TTT 55 34 55 5

?"

• C"E p

~

aae se an Ea e

64 aa4 2 w

ss:

a e

g L

u

?

O E

2 g

B C

y E

C g

5

~

~

.E o

m ss 8

e a

as 8

W a

=

~

3 S

a M

=

m

• 4 8

E s

2" C

g-g es 2

gg g_

g W.

e i

n i

88 3

2

]

-g I

.o O-t3 V

1.

Dr. s -

POOL SWELL LOADS

• REVIEW CHRONOLOGY OF PRIMARY LOCA LOADS

' ORIGIN OF LOADS AND STRUCTURES AFFECTED J

' BASIS FOR SPECIFICATION

• DESCRIPTION OF ANALYTICAL'MODEL

' QUALIFICATION OF ANALYTICAL MODEL

• EVALUATION OF ANALYTICAL MODEL-

• LOAD SPECIFICATION AND EVALUATION WL d E 03allE g _.,

U.050 I

CHRONOLOGY"dF PRIMARY TOCA LdADS

' VENT CLEARING

DOWNCOMER LATERAL LOADS

JET LOADS ON BASE MAT

" PRESSURE LOADS ON SUBMERGED BOUNDARY

• AIR BUBBLE FORMATION 8' DRAG LOADS ON SUBMERGED COMPONENTS

' POOL SWELL

AIR BUBBLE PRESSURE LOAD ON SUBMERGED BOUNDARY

- i

DRAG LOADS ON SUBMERGED COMPONENTS

" IMPACT LOADS ON WETWELL COM'ONENTS P

" WETWELL AIR COMPRESSION LOADS ON BOUNDARY

UPWARD DIAPHRAGM LOADS

• POOL FALLBACK

DRAG LOADS ON SUBMERGED COMPONENTS

' STEAM BLOWDOWN AND CONDENSATION

DOWNCOMER LATERAL LOADS

" PRESS 0RE LOADS ON SUBMERGED BOUNDARY

" DRAG LOADS ON SUBMERGED COMPONENTS liiiin

$0..S.il.!l

~

UM wet L. f.aAD -

p m

--i 4

q.

4

.4, ift

.,D * *f,.7 l

l..,Y..'..fr*~4

>.. v

,e!

~

.9.'

- ORWELL i

REACTOR G..

.g r

v n'

\\,

g.

4

/

-M

,[W/d e.

DIAPHRAGM

@$(

/

UPWARD t

'/

op g

6

.y A % :-

4.,

- Q::4 1

'. - ~

A A

,.k. +

30 5

.?.t; t

i

...r$r j

A 4 J- -

WETWELL AlR IMPACT

?..

i hW C COMPRESS!ON s

/

a, tg,.

}

!. {

I.

t

,,-. DRAG l

d".bche;' O 4 $ 4Tdip; c

Eas lM)D!Q..)

WETWELL AIR-

?F W

?!',

M

dffQqh, M

3.

Ei j.!TJ d 5jj'W@b1

.As',.M:..j Zi i % j

?

]pdlfd. I IR BUBBLE rg @-

5

%y:!

+ Q COMPRE$SION M[

M.S* f,$

RESSURE.'-

2:

' = " --

PLUS HYDROSTA-W, i

i M,m

. A k *d.[p.

A h;c'ov@

sp4g:

AIR BUBBLE M&,q *q +F is A.,.6..i.[. 2

$. $...$[.#@..ldf;;L'45, $,

R.. f.d.d. nz.

M n

G RESSURE PLUS

?,... 4;,,#<

1. ,."...w g,f,yL.:t.

?

e.e., - :.y~' !,E. m.

.a

,.g,4 >..n.g..,q

.S-9 y:.G;e.Jit ia -

HYDROSTATIC 6

d.u M,a.. v,r-, qx.

C cm.rt

,sS.

,,y.< 3 2, gq

.,.g.,,.

.y;

~

. c,,,

s a

• l' I

LOCA POOL S. HELL LOADS 6 -,3 G

.O 8

g I

e c...

MARK II OWNERS GROUP BASIS FOR SPECIFICATION OF POOL SWELL LOADS

' GENERAL ELECTRIC POOL SWELL ANALYTICAL MODEL (PSAM)

" P0OL VELOCITY VS ELEVATION

POOL ACCELERATION VS ELEVATION

" AIR BUBBLE PRESSURE

" WETWELL AIR COMPRESSION

' 4T FULL SCALE SINGLE CELL TESTS "QUAliFYPSAM

" MAXIMUM POOL ELEVATION

" DIAPHRAGM UPWARD AP

• EPRI 1/13 SCALE MULTIVENT' TESTS

QUALIFY PSAM

' PSTF IMPACT TESTS

" PRESSURE-VELOCITY CORRELATION

" PULSE DURATION

• SUBMERGED STRUCTURE DRAG ANALYTICAL MODEL

.O.... M._.. 9

.g_

e

.g' GENERAL FEATURES OF PSAM -

• ONE DIMENSIONAL

• CONSTANT THICKNESS WATER SLUG

~

' ALL AIR. BLOWDOWN

' BUBBLE TEMPERATURE EQUALS DRYWELL TEMPERATURE

' ADIABATIC COMPRESSION IN WETWELL AIR SPACE

~

' ADIABATIC VENT FLOW WITH STANDARD LbSS COEFFICIENTS

• DRYWELL PRESSURE HISTORY FROM FSAR-TEMPERATURE HISTORY FROM ADIABATIC COMPRESSION 03)III)0

_O.SEl g-f y

QUALIFICATION OF PSAM GE 4T

'EPRI sE X

BECHTEL X

X X

BNL X

X' EVALUATION OF PSAM AIR BUBBLE PRESSURE

- ACCEPTABLE WETWELL AIR COMPRESSION - ACCEPTABLE POOL VELOCITY

- NOT UNIFORMLY CONSERVATIVE POOL ACCELERATION

- ACCEPTABLE BASED ON MEASURED DRYWELL PRESSURE HISTORY bDM E~b O.A.']

3 4

EVALUATION OF POOL SWELL LOAD SPECTFICATION LOAD OR-MARK II OG NRC EVALUATION PHENOMENA SPECIFICATION OR SPECIFICATION AIR BUBBLE PSAM ACCEPTABLE PRESSURE POOL VELOCITY PSAM 10% MARGIN REQUIRED POOL ACCELERATION PSAM ACCEPTABLE DRAG ON SUBMERGED PSAM + STANDARD STANDARD CD'S TO STRUCTURES CD + ACCELERATION BE CORRECTED FOR DRAG VOLUME BLOCKAGE AND ACCELERATION IMPACT-PSTF PRESSURE-STAFF REVIEW VELOCITY CORRE-CONTINUING LATION AND FIXED PULSE DURATION WETWELL AIR PSAM ACCEPTABLE COMPRESSION MAXIMUM POOL 1.5 X SUBMERGENCE MINIMUM OF PSAM SWELL WIIH Y = 1.2 OR 1.5 X SUBMERGENCE UPWARD DIAPHRAGM MAXIMUM OBSERVED ENL CORRELATION OF AP IN 4T TESTS 4T AND EPRI RESULTS 4

E-7 H.M.0

Du. G u ~ -- - -

LOCA AND SRV SUBMERGED STRUCTURE LOADS

' ORIGIN OF LOADS

• BASIS FOR METHODOLOGY

• MARK II OWNERS GROUP METHODOLOGY

• EVALUATION'0F METHODOLOGY t

A A

^

I555Il F -I G.MI.1

,-,a

PHENOMENA WHICH GENERATE SUBMERGED STRUCTURE LOADS

~

' VENT CLEARING WATER JET

" JET VELOCITY AND ACCELERATION

" INDUCED TRANSIENT IN SURROUNDING POOL

" DRAG AND IMPINGEMENT LOADS

• WATER SLUG FALLBACK

" VELOCITY AND ACCELERATION TRANSIENTS

" DRAG LOADS

• AIR BUBBLE FORMATION

" CHARGING (ATTACHED) BUBBLE

" RISING (SEPARATED) BUBBLE

" OSCILLATING BUBBLE

" BUBBLE C0ALESCENCE (POOL SWELL)

" INDUCED VELOCITY AND ACCELERATION TRANSIENTS

" DRAG LOADS

• STEAM BUBBLE OSCILLATION AND COLLAPSE

" INDUCED VELOCITY AND ACCELERATION TRANSIENTS

" DRAG LOADS aui

S MARK II OWNERS GROUP BASIS FOR SUBMERGED STRUCTURE LOADS FIRST PRINCIPAL ANALYTICAL MODELING WITH SIMPLIFYING ASSUMPTIONS

' SUBSCALE TEST PROGRAM TO CONFIRM ADEQUACY

-l OF ANALYTIC MODELS

• • PSTF 1/3 SCA!.E TESTS

• • SIMPLE GEOMETRY 1/4 SCALE TESTS 9

H..W

MARK II OWNERS GROUP METHODOLOGY FOR WATER JET LOADS

~

' JET MODELLING

" ONE DIMENSIONAL

" NO INDUCED FLOW TRANSIENTS "i SURROUNDING POOL

• DRAG COMPUTATION

DRAG ONLY ON STRUCTURES PARTLY OR TOTALLY INTERCEPTED BY JET 2

• • STANDARD DRAG ONLY (D ~ V )

' JET IMPINGEMENT COMPUTATION

" M0 MENTUM TRANSFER p

HM0

EVALUATION OF METHODOLOGY FOR WATER JET LOADS

'REVIEWCdNTINUING

• MODEL NOT CONSERVATIVE NEAR JET FRONT -

RECOMMEND USE OF NON-DEFORMING WATER SLUG AS CONSERVATIVE UPPER BOUND

' CONTRIBUTION OF INDUCED TRANSIENTS NOT NECESSARILY NEGLIGIBLE - USE POTENTIAL

- l FLOW FOR MOVING SOURCE TO ESTIMATE IMPORTANCE k

e e

"~'

N 001AJ!!J

MARK II OWNERS GROUP METHODOLOGY FOR WATER JET LOADS

' JET MODELLING

" ONE DIMENSIONAL

" NO INDUCED FLOW TRANSIENTS IN SURROUNDING POOL

• DRAG COMPUTATION

• • DRAG ONLY ON STRUCTURES PARTLY OR TOTALLY INTERCEPTED BY. JET 2

STANDARD DRAG ONLY (D ~ V )

j

' JET IMPINGEMENT COMPUTATION

" MOMENTUM TRANSFER

~

p IDM?

1 EVALUATION OF METHODOLOGY FOR WATER JET LOADS

' REVIEW CONTINUING

• MODEL NOT CONSERVATIVE NEAR JET FRONT -

RECOMMEND USE OF NON-DEFORMING WATER SLUG AS CONSERVATIVE UPPER BOUND

' CONTRIBUTION OF INDUCED TRANSIENTS NOT NECESSARILY NEGLIGIBLE - USE POTENTIAL FLOW FOR MOVING SOURCE TO ESTIMATE IMPORTANCE-l

!3 33I O.S.U.!]

MARK-II OWNERS GROUP METHODOLOGY FOR AIR BUBBLE LOADS

• BUBBLE MODELLING

SPHERICAL SOURCE

METHOD OF IMAGES (BOUNDARIES)

" C0ALESCENCE-SWITCH TO ONE DIMENSIONAL

• DRAG COMPUTATION

" EQUIVALENT UNIFORM FLOW AT GEOMETRIC CENTER 2

STANDARD (V ) + ACCELERATION (V) DRAG

" NO INTERFERENCE AMONG STRUCTURES

)= - (,

DBY

[D.MU

1 EVALUATION OF METHODOLOGY FOR AIR BUBBLE LOADS d

' REVIEW CGNTINUING

• SPHERICAL SOURCE - ASYMMETRY AND BUBBLE RISE EFFECTS CAN INCREASE VELOCITY ~10% NEAR BUBBLE

' EQUIVALENT UNIFORM FLOW - FOR STRUCTURES LARGE COMPARED TO BUBBLE USE. UNIFORM FLOW AT MAXIMUM VELOCITY

' STANDARD DRAG COEFFICIENTS

" CORRECT FOR ACCELERATION

CORRECT FOR BLOCKAGE

' INTERFERENCE - FOR STRUCTURES CLOSER THAN FOUR BUBBLE DIAMETERS. METHODOLOGY NOT. VALID 4

Ml3.$

MARK II OWNERS GROUP METHODOLOGY FOR-FALLBACK LOADS i

• MODELLING

FREEFALL FROM MAXIMUM POOL SWELL HEIGHT

• DRAG COMPUTATION

" STANDARD (V2) + ACCELERATION (h DRAG S

10. Mil

EVALUATION OF METHODOLOGY FOR FALLBACK LOADS

' REVIEW CONTINUING

' STANDARD DRAG COEFFICIENTC

" CORRECT FOR ACCELERATION

" CORRECT FOR BLOCKAGE

!AA.Q.0 l

pu. sw n

\\

i STEAM CONDENSATION AND CHUGGING LOADS

• GENERAL FEATURES

• MARK II.0WNERS GROUP SPECIFICATION

" CONDENSATION LOADS

CHUGGING LOADS

' BASIS FOR SPECIFICATION

• E ALUATION w L-. :/ c.

Did O..Sl0

^

GENERAL FEATURES OF STEAM CONDENSATION AND CHUGGING LOADS

' ORIGIN OF LOADS IS MOTION OF STEAM-WA'TER INTERFACE AT VENT E)(IT.

' DISPLACEMENT EFFECT OF INTERFACE MOTION CREATES PRESSURE TRANSIENTS IN SUPPRESSION POOL WHICH ARE TRANSMITTED TO SUBMERGED BOUNDARIES.

' BEHAVIOR OF " SOURCE" DEPENDS PRIMARILY ON VENT STEAM FLUX RATE.

.l

" HIGH STEAM FLUX - MOTION IS ESSENTIALLY SINUS 0IDAL - AMPLITUDE RELATIVELY CONSTANT

LOW STEAM FLUX - MOTION IS UNSTABLE (STEAM BUBBLE COLLAPSE - CHUGGING)

FREQUENCY OF OCCURENCE AND AMPLITUDE RANDOM BUT BOUNDED.

• CHARACTER OF SOURCE ALSO SENSITIVE TO AIR CONTENT AND GLOBAL PRESSURE LEVEL; SOME DEPENDENCE ON SUPPRESSION POOL-TEMPERATURE.

D 0110

• ~'

(D.1M

MARK II OWNERS GROUP SPECIFICATION FOR STEAM CONDENSATION LOADS

' STEAM FLUX AB0VE CHUGGING THRESHOLD

" SYMMETRIC LOAD ONLY

SINUS 01DAL' PRESSURE FL.UCTUATION APPLIED TO SUBMERGED BOUNDARIES UNIFORMLY BELOW VENT EXIT, LINEAR ATTENUATION OF AMPLI-TUDE TO POOL SURFACE.

TWO VALUES OF PEAK-TO-PEAK AMPLITUDE ARE SPECIFIED DEPENDING ON STEAM FLUX RATE.

RANGE OF FREQUENCIES IS SPECIFIED (8100

MARK 11 OWNERS GROUP SPECIFICATION FOR CHUGGING

~

LOADS ON SUBMERGED BOUNDARY i

• TEMPORAL VARIATION

+

\\

(

6 1)

E~

m.

m m

m..

O*.

l

a. -

r MAXIMUMAMPLITUD$$ARIESWITHLOADAPPLICATION

~

e e

TIME e

6

.e t

e e

e G-4 e

MARK 11OWNERSGROUPSPECIFICATTONFORCHUGGING LOADS ON SUBMERGED BOUNDARY

' SPATIAL VARIATION j

MARK 11 Pt.AN VIEW 2708 J

y 00 1800

/

1 t6 CRCSS SECTION V V 90*

A. ASYMMETRIC LOAOlNG CONDITION 8.UNIPORM LOACING CONDITgN e

g t

a.--- -

FOLLOWS OBSERVED j

o S

MAXIMUM PROBABILITY DISTRfBUTION 0BSERVED

-~

g n-k

-[____

$E

.__/____.____._.___.__ _

g m

MEAN VALUES x

',i MAXIMUMOBqERVED 0

90 M

270 350 AZIMUTHAL ANGLE f -s 6

MARK II OWNERS GROUP BASIS FOR CHUGGING LOAD SPECIFICATION ON SUBMERGED BOUNDARY

' DIRECT APPLICATION OF MAXIMUM LOADS OBSERVED IN A BOUNDING FULL SCALE SINGLE CELL TEST FACILITY (4T)

" CONSERVATIVE DATA SELEC'TED PREFERENTIALLY

• • SMALLPOOLAREAR$LATIVETOPROTOTYPE

" LOW AIR CONTENT

' LOADING APPLICATIONS PROVIDE ADDITIONAL CONSERVATISM

EXACT VENT SYNCHRONIZATION

" FREQUENCY CONTENT OF SELECTED PRESSURE SIGNATURE MAXIMIZES STRUCTURAL RESPONSE

" CONSERVATISM DEMONSTRATED ANALYTICALLY i3 83[

g_ _ c 10..M.3 c

EVALUATION OF STEAM CONDENSATION AND CHUGGING LOAD SPECIFICATION

\\

' STAFF REVIEW CONTINUING

' LEAD PLANT PROGRAM

ACCPETABILITY OF CURRENT SPECIFICATION REQUIRES RESOLUTION OF 4T FLUID /STRUC-TURE INTERACTION CONCERN

FOREIGN LICENSEE MULTIVENT DATA INDICATES CURRENT SPECIFICATION IS BOUNDING

• LONG' TERM PROGRAM

" SUBSCALE MULTIVENT TEST PROGRAM AND ANALYSIS WILL PROVIDE CONFIRMATION THAT MULTIVENT EFFECTS ON CHUGGING LOADS ARE BOUNDED BY CURRENT SPECIFICATION U.SM

D v. S' u J :.

4T FSI EFFECTS ON LOAD SPECIFICATION FOR MARK II SUBMERGED WETWELL BOUNDARY

ORIGIN OF PHENOMENON - UNSTABLE AND RAPID COLLAPSE OF STEAM BUBBLE AT THE VENT EXIT

• COLLAPSE IS MANISFESTED IN SHARP UNDER - THEN -

OVER PRESSURE TRANSIENT AT THE VENT WHICH IS TRANSMITTED THROUGHOUT THE POOL AND.T0 THE BOUNDARY

' DESIGN REQUIREMENTS

" PRESSURE MAGNITUDES AT THE VENT

-i

RESULTING WALL. LOADS ON MARK II CONTAINMENT

• LOAD SPECIFICATION IS BASED ON WALL PRESSURES MEASURED IN GE 4T. FACILITY - HAVE THE MEASURED PRESSURES BEEN AFFECTED BY THE INTRINSIC CHAR-ACTERISTICS OF THE FACILITY?

HOW HAVE THEY BEEN AFFECTED7

• MARK II OWNERS GROUP BASIS FOR ANSWER IS ANAMET PHYSICAL STUDY OF 4T CHARACTERISTICS COMBINED WITH ANALYTICAL STUDIES EIS gy GMQ

., l

MARK II OWNERS GROUP EVALUATION OF 4T FSI EFFECTS

' 4T GE0 METRY AND ELASTIC CHARACTERISTICS AFFECT WALL PRESSURE-TIME HISTORIES BY " RING OUT" AFTERCH0GPULSE

' FREQUENCY SPECTRUM OF WALL PRESSURES BELOW 20 HZ NOT GREATLY AFFECTED BY 4T CHARACTERISTICS

' PRESSURE AMPLITUDES IN 20-50 HZ RANGE ARE AMPLIFIED BY 4T CHARACTERISTICS

' ALL EFFECTS ON PRESSURE AMPLITUDES FOR FRE-QUENCIES ABOVE 50 HZ ARE LESS THAN 10% OF M,AXIMUM

' FOR 10-20 MSEC CHUG PULSE DURATION 4T ACTS LIKE PROTOTYPE WITH ALL VENTS IN EXACT SYNCHRONIZATION

• FOR 1-3 MSEC DURATION 4T DOES NOT REINFORCE PRESSURE AMPLITUDES

' FSI EFFECTS ARE LESS THAN 10% OF TOTAL

' 4T WALL LOADS ARE BOUNDING FOR PROTOTYPE l

[0.M3 l

EVALUATION OF MARK II FSI ACTIVITIES

' SIMILARITY OF 4T FACILITY TO PROTOTYPE SINGLE CELL

VOLUME SIZED CORRECTLY BUT GE0 METRY MISMATCH EXISTS

" HEAVY ELASTIC CYLINDER SURROUNDING DOWNCOMER DOES NOT PRESERVE CORRECT SIMILITUDE

" 4T GE0M$TRY CREATES DISSIMILAR PRESSURE EFFECTS

4T ELASTICITY CREATES. DISSIMILAR PRESSURE EFFECTS

• 4T TEST INSTRUMENTATION

" INADEQUATE TO INFER CHUG SOURCE PRESSURES EXPERIMENTALLY

• ANAMET TESTS AND ANALYSIS

SUCCESSFUL IDENTIFICATION OF 4T SYSTEM CHARACTERISTICS BUT DEMONSTRATE IMPOSSI-BILITY OF IDENTIFYING INTRINSIC CHUG SOURCE CHARACTERISTICS FROM WALL PRESSURES

• 4T FS1 REPORT - DOES NOT PROVIDE CONCLUSIVE DEMONSTRATION THAT 4T FSI EFFECTS ARE CON-SERVATIVELY FACTORED INTO LOAD SPECIFICATION

'-4T DATA CAN BE USED TO DEVELOP IMPROVED CHAR-ACTERIZATION OF CHUG PULSE FOR APPLICATION TOMARK11PROTQTYPEBYEVALUATINGTHENET CHUG IMPULSE Fdt 030M y_3 0.M]

I

STAFF RECOMMENDATIONS FOR RESOLUTION

• DETERMINE NET IMPULSE FOR 4T " WORST" CASE BY INTEGRATING MEASURED WALL PRESSURES VS TIME OVER A SPHERE WITH RADIUS EQUAL TO 4T TANK

' DETERMINE PRESSURE RESPONSE TO THIS IMPULSE ON MARK II GE0 METRY USING REPRESENTATIVE PULSE SHAPE'AND DURATION

' COMPARE MARK II RESPONSE WITH CORRESPOND.7.NG 4T RESPONSE TO DEMONSTRATE CONSERVATISM _0F CURRENT LOAD SPECIFICATION

• MARK II OWNERS GROUP IS CURRENTLY IMPLEMENTING THESE RECOMMENDATIONS 9

i H -4 03UlI[

H10?

F-

+e e-w-r-9 D

/-

s on.

Sw~

LATERAL LOADS ON DOWMCOMERS A.C. R.. S.

CWcage ; 25 May IME I

_^

AA.S.

s-,

ORIGIN OF LATERAL FORCE OuqiNG CHUGGIMG 9'

S 4

m

-w--_

^

m h

smuunnus emuuhumu.

r g

N

~

~

4

^

N%

w

{f t

+

4 1

s NET A

LATERA L Fom.C E 8

b

.+,

+,

+2.

NATURE OF LATER. A L LOAD S IMPULSNE' 1 - 10 rnseC LoAolNG

~

OcWNCO M ER LATERAL FREQUENCY ~ l O O msec.

LOADNG INTEEVAL ~ 2. sec.

OtRECnOM RANDO M MAGNI TU DE.

STOC.1%ST1C.

U M A irl:EC.TED SY OPERATION OF AQlACENT DOWNCOMERS

'S Dt A.' s AW AY, o

y

_3

e.sn n

e a

q m

sn

+

>~s.

+w.

Tl:ST FAC.1 LlTN MAX. STAT 1c_ EQUN ALEMT LO A O MEASURED KWU -

GKM (24")

980o LBF GE (2e) 24 o o ter kWU-KARLSTEIN (24')

31 500 LEF 4

l

- ---.=--.-..._,,_... _ _,_,_, _

e

,m i

i s'.'

4 F(t) -

!. 2%p(+)

ad k[ I = [Fdt-

~

^

DYMA Hi<

o LOAD

~'

u?

OEFLECt104 3 et STRAIN 0 V v sl MN u

.A EQuNALEVP ummumet

muem,

@qEST srATic. EQUIVAl.ENT LOAD MEASURED (LBF)

/

l Js l Kwo-Karlsk 2A" i

e i

'30,000

/

KWU-GKH-E u

r p

Q

\\

/.

20,000 F~i.N'. s,sco _r_

tb;

~

SPEClRCAnqN

( 7gg KWU-GKM to,000 3,300 tbf '

G Eii 0j 1

O 10 20 so Y,

NAT. FREQUEMCy ' (.gz},

SUMMARY

o p:- o474 I

i

i Oc. A L

~

j IMPACT LOADS

• DFFR METHODOLOGY

' BASIS FOR METHODOLOGY NRC CONCERNS WITH EXISTING METHODOLOGY raiir p

M3 G

4 5

i<

w-w

e DFFR IMPACT METHODOLOGY P

L MAX I

' \\

t MAXIMUM PRESSURE i

O MAX o

0 a

• s o

s 4

4 0 - 20" D 0 - 5" 5 - 20" V

t V

V y

Y PIPE I

BEAM PULSE DURATION t

7 MSEC (CONSTANT)

=

7t e,- r

f e

e F

h I

F,(t)

F,( t )

AIR

>](

l y

TARGET y

AV

/

AIR j

y

_. s /

/

/

%s

~-

/

PRESSURE SUPPRESSION TEST y,3 FACILITY (PSTF)

u-m

>va w,

-~~~

9 HYDRODYNAMIC IMPACT 4

e iv

/

Ii

(

H

\\

v

[Fcit MV I

=

=

g OR PER UNIT AREA Fl r CYLINDRICAL D

OR W

J4 e

v

,-,,s.,

a v

FLAT - POOL METHODOLOGY IMPULSE FLAT

)

y

/

I

=

p PIPE D

OR W

-l DURATION 7

FLAT POOL D1

( OCHI AND SCHWARTZ )

D2 V

MAXIMUM PRESSURE b

U max Pmax Y

21 p P

=

MAX p

_y.

7/

~ -

. mtmm-bra- ' ! l. ! '... -

~

1:nt...L ' !.' !........ '. m..

9:; iI:ii:

.r

.. -l t 1 -.-

- s + i 8- ---

r- -

8-s ll

.jlll i 11.~s::

!st gi..

l.

.;. 7.l - 't' e. -; !

gj

• t-i n i jj

-:j i

m

..t S* -y,i i

~

q il!ir 1

t i

vi e;"

q" y::;,

1 i:. "

til:

T..

.. '{ :.-m*rmg q l.. e'p: a.

,i. i.. f.A...:...

.. _~ l -

ji: i.

-r};:

+

si

.

Ratio of stresses by flat Pool method to o

r "l

u

.1.~

t i"

r-i,i:

4

ii.?.

@h.i.n

l.i- ;9 'DFFR method for a 10 inch pipe at various

i. ill

c:C'b

+.e

-..l.i d

2-

~.

~ q, -l m

..i

.. i l ~n l

.2

.I :i

.I..,

4

-f

,*44--l.m.e.

..p impact velocities.

i f.

i 4!..

r7

. i.

- t-1..

4 g

-l-

.i

. (c

.i.!

j.

.t.

.i....:.

.iel p:n ~~~

91 a.

.i

.t..

g..

$ l a---

a-

,e...

l,--

_.4r j 12.

.:l: :.p ;;; ;-.p i :. ; i:,.

.g.

i q..

+

3

. Ll.

}.

ll ;

, l:; j j i;;r. ;4.

4 i;p J..

4 l

i i

,i: q ii.;

.:i.

u

.. + -. e in

.a

.~.,i.

t.t 1

g j,r.y

.,t -

i..

.i. ;

3,..

s.

,, 3..,

,l g,.:.,

,,j ! t,:.

.;....l ! g 1.

.c t

a....n

.1

-. __. 1

_t._

1

. ; g.

1l._}.

l' i g..

i

! g-s.

c.. I..c

_g..

t

,;l)

~. ;il, t

.,..,,, : 1..

,u, ;;;

2:.

2.

.g lp.

2

.t.

hl-t~

-; p

. lc o

l

;i

; ;;. :

i:..!-::(.. i i,i :

1,;.

. 3

i.

a i.

.:.g 3

l ji. Pil h;i ;'. I

- l1;,.?;.'i 8

.. ; l,.,

i

,.p 4,...,

ll!

i:

x f.

. '. '.. i ;

.I; -.

py gd - d.

i g

--t,,

• E I

i.'

.i,;-

1;:

-y.

-d--

!)/

=.

i i

i; i

f.

=. ggi 9 ; *.

.. -1.t-

.8-

-I-1 t

.;..,r

.., -+:. ; g-i.

l Ii b.

+

.i p

l ii

.lI

.t. ; ~h ;

i l : e,- : ::. !.

4i-i:..

. g :.

[

,1l: nn 1

t..

!J L... h. i 3;

,3p si..*L 0

~

,::.L. ir.

. I, 1

i ! c,

. 2.

~

Ii

.i :.2. A

i g

..lI

..n

..a ',;..i.

ni

i-.. i.:

1..;

L;.

s. ' J i a.

. L'. - y

-. I ' I

.N i

.t

l,.:

.._i I.

l p.1 p.;

y'

. k'h-.cQ%

%, eN e"..

i 8

i l

.l:

.. ll

j

-.3.;

4,, ;+m...

ij.g L,

. f.,

g;.

g..]

... l....

g j1,:

gi,-

g i r..

i

..,,,. n.

l.:.

1 J):i j.1

. Li

.i.

.[

i-

- (.

j.;

ia. f M.

,a p;

,;. i.

F ".i /-.:

jl.

-lg

!!ll i

! pt

. p.

.g

.r

. A..-

j-e-

l-l m

tili

/... _.

.__]-_

.2_

.. q la : 9 :, y, ;

i. o.

l ;I

, ll1.;::..:.t.

.,t:,;!.--

1.

I.

j.

.t ti.

e;;

i g,;;

... /;.

l

~.

; :- l.

p i;s i.;. e 3 r.1

..l i.,

. l.. !

-l i I p im

.l,8

,.,/.....

I.

i

..._ ji

a. l

a a

e l....

..y in 1 -

.s

-..,:.;i

.l.

..,,. l i.

p 2

- f.A

_~

., ; i I. p:. '1

.u, c

. i.

5.[N,,

. l t_

- W:.... '

- 2,.

.+.

i7 a.p 9;. 2

.i 0p 3

_'1

.;.

t

.i 4,

+

.. i i;t th.;.;];

. /

-.f f

.g

..p 4;!

pi1[,gf l

g 7[d"'h:.,..

.h

..! q 7,,

... j i;.

t..

i,

,. -- 4

, y l.

I

.iI:

A s.

i,.,g..

.j.

..g

. k. l u s,

a.

4 g.

..g

'*i

..i

._.cr9:

- - --- p i.' l

.1.

',I l 4'

1. 2'k. - ' -.#,

'p.

L.

2

.g_

g..

• .k,i."d,/

... i.O.

F... -'.

_ 3

. c' m :

!.'n. - -

i i..l..!:

l; J

l

___.c I

1;i-y:.i..

i i

. al

._...fl.m,_..

tilt i

-7_-

.lh.

l

- i i

].t.;

i

.l--

.l_

3. :r-g;.[i].j ilp jy y.t.1; dl y).2"

__.*.'.I,. d n:.i.

..g.

r ui 4..

,j.,

i;.

I.-

_...t-.

.i,

i

..y a.g 3 i 2

_ _.1..

a i

'. -i

. 4

!, pl.p, g

.s, 8.

L.

i si,.,..3../s..9, 3 ry

-3

+

..j~

I g.,;g 6,,

I.I

y...

......_..+w, p

p,

..8 1

i..: C, sI.

j;:

l i

.. a,;;.J,...:

gli In

,t

t..g l -

o,

i..

9

.t,.s l

.i.

jl;t n,;,,...

.l

+

:.! l. :

1:

" I~:

i,

.. i-i

_a

... u. l. e, ;;i,,.g..

l.l

.j p..g u ':.

i g

._

._ _.,,_i_

g. _..

9

. i. la:

_.A I:;j.j,;.p p>;i.:

a i..,

i.

g pi.

q...j.

. !.1 ;

.j j,

..fl. l.lf.:.g,.j.

4, -

..{.)

.... j. ;_...'.

g :i i.

. t.,:

7,-

e,

,i4, 3..

J 4..

J...a u [.

-.ll; ;;r

.I 1,.

e

.i. y.

.i.$...

li

p.l 3i,

-. y

. l _. g,

, 1.

.. i,.

1-

.4

.. Lq.;

ii

;r;;.

p!q. j:

. __. i _,..

..c

+l-p I

i

.i.

, s.,

_. j ;

.. t..

..... [ : ;.

9. :;t i

!:l y'

g.:.

.7 3

[.

s

._g y jLq.. gl..._ ;.

[.... -

.o

_p

- J.3 i

s

...l -

.F. isl p-

$,4 l-.

!.. I.

ee

I.-

e,.lr

.8 *.

...i'-l.

ilb.

tr t

r-ll.

sr;;

_L.m..L;

.j,..9

.. t

.g a.

.l s.

ll. j;,.!',.l.g..,

4

.l...

g

. ;.q.. i.. a.

i.i

.:i; 3,i

.f

,g..

e l. i,

.l

.g

.i

.l l. l4l l

i.p,.p,

. l.. f

.2

1 i..

1:

ia r.

.{

• p t.

..,t.

.. I i.

i

~.I.

I

i..

. si

'lj:

.j,,,i g.

[1..;..,.'

h l...

l-

..l -

8 I i.l. M....g i ;..,

. ;.t-

. ';i.

R.

,!.h-.

I;

- '....j.

l.

l.lr

.i

_i

!.i

. i.

3..,.l.i ai g.

,.. m t

-l I.

i jl.

4.y.I,.t M:;r t

gp :...

i

.. j..

. :1 !,

.1 i.

.i.

l. l.

i:t.-

r l.l.

.u._. 4 :d'5

,l

,g, p.

.h.

.cj

.!. 4,:...

..I.

.q.

1

~....

+

...q.

. _14 _

t 2

,. 4 6.

- Ult.i[.i

.....,.l..._

_g G,

.I :

-l :

. t.:

tl c.s,.

a,, :. 4 ig a.

y.,.

g.

i m... :. e

,3..

..y..

g.

r., !.

i ii,

i. :

r e

e

,i'

}.*

!ij [.. h[. $.,o b,

~

e

!Ii M.

u,..f u,m.. A;U ~

h

' a' [;.u._k.m.,M h..![,....o 'I M O/9

....._; _..._ ' M.M. dd, b,.

i

.l.

I [... i.

d F

i m

g

/

/0

/00

/000

/O. 6 m

e 6

I 1000 m

4m.rr.--

..m.: n-e -

y tr -

.m.

r

. s ;.-

_t t

.g.

.s et :

r-r - v s r o ;a sen s-

.2 p".tr C L4.4 a r.'i i

,.t.

! y r..

r}

? '

p(.

4 r...' :- & :. ' - ). i.

4 6

?id.*t. ~= 7,=.:;r&_.--

,=.i.:

.:q i..,3 '- b.*.. ::ra._

~

f. i; be.qTW I

.9 b*

E6E6=1+45241:- c=.n4 4=8 :=wM-!=14W= - i==Ki= :4.

j

_ u.=-

_=_..r.: _:-_ m...r.:,... c.q.. :..:. :. __. u.. _;.

.. m. m y

- =,-4.r c. n w :

p x.c

+, um., 6 0,= ~:

T M

W. *q_+_TDF.:EP *=*=. -:+?:.=i. h _:n:-9 PN'-P

=c Pad

.. w

.=

c H:=t===x-=

-arnofnB[gNAUVEw.-ra#===r

= ;.- a h s.

p

_W_. m=1.sML:2x n#rde_=~g: e=RG.k1 e

.. _.. ; =g p_w...=.

__........= _

..=_..... _. _. _.

=-

.== _= Fnppp:--

~

. t _. - w. :.. _

-._......___.w---.----.-

=-

.t-L!

O

~

i,

......_.t,.._.

.S 100

./

~

.c.

< - - - - + -

m w _,%

.. <.,y

x..,...

., t Ll

[.: ;.. - T *a s

~

' )-.

r-- ad., p..a. 2., -- ; r.T -rs.r. n j w h- /; s j=..ji,s,.

y

=~_-~5 M-1" ;- =L;;

--N =i' ~;" hi tf N ;;-2:~..

r. -

_--; + 3._2.

=

g

.-p..

. ; = :t;ri A ;.zp. : y.=-g m._.N.

- Y

===:.=. :..-t. -.. :.p:.

n. f..-- ; : : - : n. - -

-c.:.. - - - --. :

~:

C

---

.r*-

% W.e.lEF E t.g',! ".' N. 7.u &. :

~

L*-rR& 1 h_i + ' - -

",W

==

r w>

.m -.

m s -t % e:

=ws.:v y..=: u :

%.,w g.: y _. Df:Q r.c.

g 2.- g t.l."; r. _- 5.ir +-,. =

_ =E.i4=-5EtM&dI"_i-FI_i_-5MhM:M M3J 0 K.;.. g.. g.;=,-

i..

kr M

.._... _.._. _,...,...... __ _ _= m__.

a

==_._-._E_....

-__..===._,=:==.i_

.=....:~....= _::,1 4- --- : : :.r = = :7. - :r :.. ~...: :..

t

..:= =-- - ---.~ :._- t - - -.

2

_._ a 10 r-0 5

.10,

15 20 SIZE OF TARGET, INCHES MAP '0F REGIONS WHERE DFFR IMPACT METHOD IS CONSERVATIVE AND WHERE IT IS NONCON-SERVATIVE FOR PIPES AND FLAT TARGETS.

D Dil0 G....._E......l

'hJ===

SUMMARY

1)

CURRENT DFFR METHODOLOGY FOR IMPACT MAY NOT BE CONSERVATIVE FOR ALL STRUCTURES.

2)

MARK II OWNERS GROUP HAS BEEN MADE AWARE OF THIS CONCERN.

3)

NRC IS AWAITING OWNERS' RESPONSE IN THE FORM

_l OF A MODIFIED METHODOLOGY.4 4

4 e

(0

-...1.9 8

'T~- Y

s.

~ - -

E.

TAP *A-39 OlnLINE

-TAPA-39 OVERVIEW (35 MINS.)

I. SU

-DISCUSSIONONHY[R0DYNAMICLOADS(20 MINS.)-P','HuBER

- DISCUSSION ON Po0L IEMPERATURE LIMIT (15 MINS.)

/-l. SONIN

- STARIS/ CONCLUSIONS (20 MINS.)

I.SU l

d 4

e W /(

e

Oa)ECTIVES

1. TO ESTABLISH

-SRVLOADCRITERIAAND

.-POOLTEMPERATURELIMIT FORMARKI,IIAND111.

2.. TO ESTABLISH POOL TEMPERATURE LIMIT FOR ATWS APPLICATION

- GENERIC FOR MARK 1, II AND,III

- PHENO 4ENA ARE GENERIC

-MET 10DOLOGYISGENERIC

-POOLTEMPERATURELIMITISRELATEDTOTHETYPEOFDISCHARGE DEVICE) NOT TO THE TYPE OF CONTAINMENT.

-INTERFACEWITHOTHERTASKACTIONPLANS Po0LTEMPERATURELIMIT POOL TEMPERA'RJRE LIMIT 300LIEMPERATURE SRVLDADLRITERIA SRV LOAD CRITERIA

_IMIT A-7 A-8 MARK I

[hRK ll Xu

rO i

8 I

O l

\\

l

/

\\

/

i s

/

k

%1 /

5 j(

STRAMHT DOWN PIPE PMSHEAD 8

e i

I

+

I I

I W

Else

)< - 3 1

e 300 (TYPl i

1200 (TYPl

\\

/

~

i

./

\\

4

/s-

/

\\

\\

gd i

omVWELL WALL

//

,\\

~

s

~ * '

• h-12" $/80 (TYPl

~~Vl

/

vtNT HOLES (TYP)

L*

\\

33'4 " O.

%.= ~

\\

N

/,

~

/

103.so yorg til Quencher Plan View (Typ.) 238 Plant Arragawnt gown Il H

d y'

N l

\\

i

(

j

(

l KWU T4]ENCHER

.T T

9 T

n F M T b %1b} M k & -erJ- -Y Q _-Z w A h= m ~ & [ S Olh-t%RK I I-0UENCHER

)<- b

.s OPERATIONAL EXPERIENCE

- WUERGASSEN EVENT, GERMANY G972)

- STRAIGHT DOWN PIPE

- @ E SRV REMAINED OPEN 30 MIN.

- VIBRATION STARTED WHEN POOL TEMPERATURE REACHED 160 F

- POOL METAL UNER SEPARATED FROM REINFORCING BEAM

- l@l EVENT, SWITZER UND G972)

- STRAIGHr DOWN PIPE VALVETEST 0

-VIBRATIONSTARTEDWHENPOOLTEMPERATUREREACHED170F

- DISPLACEMENT OF CATWALK SECTION AND FAILURE OF AN INSTRUMENT LINE

- VARIOUS STUCK-OPEN VALVES IN U.S. PLANTS (RAMSHEAD) 1

-PEACHBOTTOM

-BROWNSFERRY

-HATCH 1

-POOL,TEMPERATUREREACHED122TO165F

- DAt%GE ON SRV DISCHARGE SUPPORTS

-NOVIBRATIONOBSERVED Id G

.LJ%

,s SRV HYDRODYNAMIC LOADS -

,' INTRODUCTION

• METHODOLOGY AND TEST' PROGRAMS

" RAM'S HEAD

" GE QUENCHER I

e e

l])))[

L -I Qf)

6

.s a

m p_

a.R. m

\\

w x

C']

/

\\

)**.

a'-

)*,0 cic e*f '

, r.9

• l a

4 i

i 4.0'.

V. *.

.s

- */

M'

_ g. g IQ f

• *g 'n',a.,

~

• • q b.,

REACTOR p

p:

T h

.I O-MAIN STEAMLINE

.,d y

oj 4

f,C

%.c

.. <,~

e q

f.

6 f

$-*.a T. evs.w..sco g > s. vn1 %

SAPETY

• a d

j t.

[

.C RELIEP g

VALVE

,'a-g g,

O' o

q a

o'.

e..

S, f o>

[,:{

(

i..?

D..c

,. q 9

3.< /

( C.

%.'a M 6'2 % 'd'*"*$ g,-

a

%2

{

• [

j JET d~

k

,o g*,

OEFLECTORS

.g Nf

\\

a f&

]i.

hY'2.5 T6'.'.

{

{ Q',-

,e

[.,.,(.

$e..

5.

aD, a

ps4

[

x'Y, -- DOWNCOMER

.(

b; b,

?.'Of h;

?, =

bN e :: _ _ :::AN

??

'.*a s

n

~

r

+

~

~

P2 h

4!

sk*:

5' d'?.'"

SAF ETY RELIEF g;l D

^^'**'

+

4

.,Q VALVE

+E x

~

w W

y.E 75.

DISCHARGE W.

~'

's,i Oi

--- SUPPO RT

'c DEVICE %'**1 U.

7

%.... k

$Y kb

=,,

y g.g h'

h

• b ibk i

).

6 Typical Mark II Pressure Suppression Containment LINE l.ENGTH =l5 OFT hlR Vol UME.

1 70 FT SU6MER(7Phlcg a 20 FT L -t-6

r, METHODOLOGY

• PEAK POSITIVE AND NEGATIVE BUBBLE

' PRESSURE POOL

• OSCILLATION FREQUENCY

> nme> BOUNDARY

' RISE TIME LOAD

• BUBBLE POSITION

• SINGLE VALVE ACTUATION

" SINGLE VALVE SINGLE ACTUATION (SVA)

CONSECUTIVE VALVE ACTUATION (CVA)

" LEAKING VALVE ACTUATION (LVA)

' MULTIPLE VALVE ACTUATION (MVA) 13ill8 La O.S.!l.:)

RAM'S HEAD METHOD 01_0GY ANALYTICAL MODEL LINE TRANSIENT v

BUBBLE FORMATION +L -

." EFFICIENCY" = 0.1 (1

BUBBLE DYNAMICS u

INFLUENCE COEFFICIENTS

.. l INITIAL BUBBLE POSITIC' METHOD OF IMAGES 4== _. ' 4 (FOUR) FEET FROM

/\\

RAM'S HEAD

. 4 ON + LOADS lg nLOAD MULTIPLIERS" SVA, LVA.

. 6 ON - LOADS.

}

CVA y

LOAD SUPERPOSITION!+-

AS 1D v

MVA~

03030 L-N N.50.0

l RAM'S HEAD TEST PROGRAM.

• MARK I IN-PLANT TESTS

" QUAD CITIES 17 VALVE FIRINGS

" MONTICELLO 55 VALVE FIRINGS

' COMPARISON WITH MODEL

" BUBBLE FORMATION EFFICIENCY = 0.1 FAVORABLE

INITIAL BUBBLE POSITION 4 FT COMPARISON FROM RAM's HEAD j

WITH SVA DATA

" BUBBLE FREQUENCIES, RISE TIME IN ERROR

~

~ 100%

" CVA DATA DETERMINE LOAD MULTIPLIERS

LVA DATA MARGINALLY BOUNDED BY SVA PREDICTIONS

" MVA DATA SUGGEST LINEAR, IN PHASE LOAD SUPERPOSITION

• LIMITATIONS - LINE LENGTHS AND AIR VOLUMES OF IN-PLANT TESTS NOT TYPICAL OF MARK II 030M Gua L-3

GE QUENCHER TEST PROGRAM

' GOVERNING PARAMETERS

" LINE VOLUME / QUENCHER AREA

" WATER LEG LENGTH

POOL AREA / QUENCHER AREA

• • ' POOL TEMPERATURE j

' TESTS

" 73 GERMAN IN-PLANT FIRINGS

" EXTENSIVE GERMAN REDUCED SCALE TESTS

FURTHER IN-PLANT TESTS (CA0RSO MARK ID UNDERWAY

' RESULTS

" StVAPEAK PRESSURES ~ 0.6 BAR, -0,4 BAR

CVA PEAK PRESSURES OF THE ORDER OF 1,7 TIMES SVA'S

FREQUENCY RANGE 5 - 11 HZ

" BUBBLE RISE TIME ~ 0,75 SEC

PRESSURE ATTENUATION AND MVA LOAD SUPER-POSITION DETERMINED

' QUENCHER LOADS SUBSTANTIALLY LOWER THAN RAM'S HEAD. LOADS.

03DM O..U.i j

~

OA3

201, I

l

..}

DA7 MEASUREMENT l

stb T10 l1 L_..

4L DA2

[ k*

'~ % --

T13 Tl2 T11 II T2 I

y T3 i

[iK, DA3/13

~

y g

( )

DA6/16/12

\\.

DA4M4M1 T6 T8N e

9 DA 9/10/17/I8 l

~

b PRESSURE MEASUREMEN' T

1. TEMPERATURE MEASUREMENT

~

A5 T1 MERIDIAN 1050 TEMPERATURE ME*!URING POINTS ON 105"

~

VEf1TICAL CENTER SECTION (EXCEPT T13)

DA PRESSURE TRANSDUCER -

T THERMOCOUPLE Large Scale Relief Valve Tests - Pressure and Temperature Measuring Points i

E

e

- OUTER SHELL IB 8

CORE SPRAY SYSTEM CONTAINMENT.

T S

l

' NW200 h

Q FLOW LIMITER p

pf7 NW25 OWSPRAY MAIN-STEM COOLING SYSTEM g

N PIPE NW25 PR ESSURE J

RELIEF i

SYSTEM NW25 HIGH-PR ES$UR E INJECTION SYSTEM

.g x

x S

MANOMETER Pg i

f 8

StMULATED WET WE LL

/

I I

I v

v w

w T@

e 4

I

,...... < >......e

. sg g

, @Gn P

PR G

b t

~

L-Y

GE QUENCHFR METHODOLOGY (FOR MARK III APPLICATI0HS)

' STATISTICAL MODEL BASED ON FULL SCALE AND REDUCED SCALE BOUNDARY LOAD DATA SVA

"' MULTIPLE REGRESSION DATA CORRELATION,95 - 95 CONFIDENCE CRITERION TO PREDICT PEAK POSITIVE BUBBLE PRESSURE

" BUBBLE DYNAMICS MODEL + PEAK POSITIVE BUBBLE PRESSURE + PEAK NEGATIVE PRESSURE

" FREQUENCY 5 ~11 HZ

" BUBBLE RISE TIME ~0.75 SEC

' CVA

" MEAN PEAK POSITIVE PRESSURE-1,714 TIMES SVA

" 95 - 95 CONFIDENCE CRITERION APPLIED

' LVA

" NO DATA AVAILABLE

" WILL BE MONITORED IN CA0RSO TESTS

' PRESSURE ATTENUATION

" NONE IN VICINITY OF QUENCHER (WITHIN 2Ro)

" 2Ro/R DECAY

' MVA SUPERPOSITION

" PEAK LOADS DO NOT EXCEED BUBBLE PRESSURE

" BETWEEN VALVES "SRSS" SUPERPOSITION ACCEPTABLE 030]l

" LINE OF SIGHT CRITERION O.10.0 4_9

c.

G$ QU$NCH$R'ON-GOING TESTS

• CA0RSO (MARK-II IN-PLANT)

" TOTAL NO. OF FIRINGS 102

" SVA NO 0F FIRINGS 34

" CVA NO OF FIRINGS 32

" MVA NO. OF FIRINGS 9

MONITOR FOR LVA 1381'1

$0.S.3.3 L -t b 2.

s Q C JY T!1-l N S f f. o s n i st.-

a n n. r u o n e au sLinas M-3,M-v H

S~> j-t - q ~

POOL t t=MPERATURE UMt i A.C. R. 55 Chica3e,

Mag 25,

1 ) 7 S' 6

e N- /

A. A.s.

1 4

4 A

I-I SRV OISCWARGE NOZZLES

~

I I.

I I

Q j

V y

($ STRAIGHT PtPE (6:>) RAM'S W EA D

?

j PElitFORA"T1C4 5 C1.csec

• 8 LMOs

~. /. :L.;

~G*i!

.. ?.".

l PliiRI: CRATED

'l Y

E.ND S

'ti

2. U l (:': ' s

..*g**

i Me G:) QUENCHER U

.,,c

200 40 LBWFT*-SEC gg's KEC&lMENDATION X-h b___

17a*F.';;Tws ).

g,

'bN 3

tt.

R 160*F O

X

-XX g,.150 --

m g

a

-f u r

'P R-t q

a o

X --

,A u-77- -

s

, 7 tv N

\\,

~ 0' 1

OhlSTRATED

~

~

E~ D(c) o A

N

-1 s s a.

W N

N

.l s.

x

~~

4.

N N

A

a. 100

~

'w N ' 5 --- -

~

~

x 2

~

STARTCc)

~n ENDIA) s STAET (A 8

GERMAN DATA

~

O 2.0 cni8"y staqLE PIPE RAMSME/\\DS A 15 cH I P) SlA141.E PIPE X ISCkt Ib") D006LE PIPE '

~

• M^R' 1 S I

FROM AIEDE -2io78 4

't i

i go i.

i cJ 0

50

/00 150, ~

200 i

g i

i nsrs, t taver2 sac)

Ij,,

i i

.j-1 Figure 2

k

% /.gt*sa.<.

Io 1s so soo 2.00 100 C i 80

/

MtCHPM8QUENCY DiseuAacs wrru susen'meAi.

$ N Rsc7:ucY

,/ TRAN$mCN AEGION PRt35URE RAffC Ots HA ACE wnTH so.4 kWan2 34$$kWem2 SUPRAcMITtCAL.

las.7ss)

(7.5 omi g

/ 4 40 N PRt350M8 A ATfQ

$=200 H1 a2,,tWem2 e42 8-10C concensarson IN P Pt At kWon2

/

? tin 1:9 psi 1 2. 9 pci

,0 t,5 gsi max.

/

mmx mg

/// f// b o

u s

t 0

10 20 30 40 50 40 40 t00 200 300 400 600 400 t000 MASS Pt.CW OEN817Y AT NCud CUTLET (twm2d g

g Q U EMC.R ER. PERFO R M AMc E HAP h

f /t O (' Y( i l=. 'T'M-R Y

~

~

e

~

-~.m v

u, e

1 i

MAX. POS ANO NEG. PRESSURE AMPLITUDES AT FLCCR 21 -

P. I 18 -

i f5- ~

r 12-9 l

j.,

6-l l

3-C

-.% " a

• a t d ' o -

• e c r==., ax a

on 5

op

.a ao g,, o m

,g, og se o*+ a r=

3

-6

\\'

\\

-9 20 30 40 50 60 70 80 9C-C ICO

& WATER TEMPERATURE CONDENSATICN AT HIGH MASS 51 OW CENSITY, MEASLRED.

MAXIMUM. PRESSURE AMPUTUDES (QusMcasa.. i<wu),

9 8

4 w

i' 90-80 1 %

• F-yn M"ESI

(

20 operating canc.e ho

\\

h h

30 80 MAX. PRESSURE.

F004o N i eu s.,

2. 2. PSI-10 '

0 3co 450 600 750 SCO 1050 120 0 o

ISO REACTOR PRFSSURE (es;A)

FU L L-SCALE QOEW C H ER.

1eSTS (14 SI TU),

f f$ l l Yb"S Y fO 4

O

CONCLUSIONS :

i l

1.

RAMSHEADS No r PROVEN F~o R O PE R.ATIO N AT-ABOVE C:a.

IlO F A FU LL R EAC.% R.

FRETT U RES.

2.

KW U - TYPE QUENcuER CK ro Ic)5 F AP<

• 3 psi

'4-h3h reac4er p recsvre s

d. E' ps i cVe_r Whole, ra y1 e 3.

EXPECT 74AT-OTHER VARI ANT 1 W HtCH FOLLOW kWU QUENCHER OEClGN GuloELINES CAed S E' SHowlN To H AV E StutLAR coNoeMS AT10 AJ PEE!, FORM ANC E -

9 9

H-7

)

)

m --

.s MAIN STEAM LINE BREAK INSI:.2 CONTAlfNENT:

EVALUATION FOR ENVIRotNENTAL 9 M IFICATION

- CONCERN - ADEOUACY OF ENVIR0tNENTAL TESTING OF SAFETY RELATED EQUIPMENT FOR POSTULATED MSB ACCIDE TS INSIDE CONTA!?NENT.

• CURRENTLY, EtN!R0tNENTAL TEST PROGRAMS ARE LOCA ENVELOPE ORIENTED.

M33 ACCIDENTS HAVE THE POTENTIAL RESULT OF CONTAltNENT ATtOSPHERE TEMPERATURES ON THE CRDER OF M FOR SHCRT PERICDS.

- THIS IECHNICAL SAFETY ACTIVITY WILL ADDRESS THE CONCERNS REGARDING THE EVALUATION OF APPROPRIATE ENVIRONMENTAL CONDITIONS FOLLOWING A

^

POS11) LATED MSG ACCIDENT INSIDE CONTAltNENT AND THE METiODS BY WHICH THE CONDITIONS ARE SIMULATED DURING COMPONENT QUALIFICATION TESTING.

- THE PROGRAM WILL COVER (1) CONTAlfNENT RESPONSE, (2) COMPONENT THcM RESPONSE, AND G) TESTING REQUIREMENTS FOR SIMULATION OF REASONABLY CONSERVATIVE CONTAlfNENT EtNIR0tNENTAL CONDITIONS FOR A POSTkJLATE MSB. THIS INCLUDES COMBINATIONS OF TESTING AND /NALYSIS.

9 e

,,v.

, _, -. _. - -.