ML20127L611
| ML20127L611 | |
| Person / Time | |
|---|---|
| Site: | Vogtle  |
| Issue date: | 09/30/1984 |
| From: | WESTERN CANADA HYDRAULIC LABORATORIES, LTD. |
| To: | |
| Shared Package | |
| ML20127L594 | List: |
| References | |
| NUDOCS 8506280070 | |
| Download: ML20127L611 (37) | |
Text
....
.-.-.-.._.-......-a--
t WESTERN CANADA HYDRAULIC LABORATORIES LTD. g i
s t
.I ORGIA POWER COMPANY H
FINAL REPORT HYDRAULIC MODEL STUDIES q
J OF ALVIN W. VOGTLE NUCLEAR PLANT FLOW CONDITIONS TO CONTAINMENT EMERGENCY SUMPS FOR BECHTEL POWER CORPORATION LOS ANGELES, CALIFORNIA 4
BY WESTERN CANADA HYDRAULIC LABORATORIES LTD.
PORT COQUITLAM, B.C.
73010 SEPTEMBER, 1984 8506280070 850326 PDR ADOCK 05000424 E
- - - ~ -
..--.-._~..u WESTERN CANADA HYDRAULIC LABORATORIES LTD.
5 TABLE OF CONTENTS I
Pane No.
b I
1.
PURPOSE I
2.
SUMMARY
2 3.
CONCLUSIONS 3
4.
INTRODUCTION 5
5.
TM MODEL 7
5.1 Model Construction 7
5.2 Model Scoles 8
5.2.1 Consideration of Froude Number 8-5.2.2 Scale Relationships -
9 5.3 Instrumentation 9
6.
TEST PROGRAM 10 7.
TEST PROCEDURE 13 7.1 Model Operation 13 i
7.2 Determination Of Augmented Discharge 14 7.3 Flow Potterns And Velocities 14 7.4 Velocity Reducing Modifications 15 7.5 Gilsonite Bottom Movement Technique 15 8.
FLOW COtOITIONS WITH ORIGINAL LAYOUT 17 9.
DESIGN MODIFICATIONS 19 9.1 Refuelling Canal Droins And Woll Openings 19 9.2 Relocotion Of Intakes 20 l
l 9.3 Spray Flow Distribution 22 l
l
-.e.e
._m ww 2-
'"7' 7
.gmwf >
=.e.Jh.gesdmWE3Iern em1_ - - - "
LIST OF FIGURES i
1.
'MODEL LAYOUT 2.
MODEL AS CONSTRUCTED 3.
DETAILED MODEL DIMENSIONS 4.
SELECTION OF AUGMENTED DISCHARGE 5.
FLOW PATTERN AND VELOCITIES - ORIGINAL LAYOUT 6.
GILSONITE MOVEMENT - ORIGINAL LAYOUT 7.
ADOPTED MODIFICATIONS 8.
FLOW PATTERN AND VELOCITIES - ADOPTED MODIFICATIONS 9.
SURFACE AND BOTTOM FLOW - ADOPTED MODIFICATIONS 10.
FLOW PATTERN AND VELOCITIES - ADOPTED MODIFICATIONS - TRAIN A ONLY 11.
FLOW PATTERN AND VELOCITIES - ADOPTED MODIFICATIONS - TRAIN BONLY 12.
SINGLE TRAIN OPERATION - ADOPTED MODIFICATIONS 4
0
. ~. ~. _ _.
- _- ~,
WESTERN CANADA HYDRAULIC LABORATORIES LTD.
1.
PURPOSE The hydraulle model studies described in this report were proposed by Western Conodo Hydraulic Laboratories Ltd.'s quotation of 10 March 1977 and were commissioned by Georgio Power Company's Purchase Order Number PAV-695 of 15 April 1977.
The purposes of the studies were:
1.
to examine opproach flow conditions to the RHR and spray pump Inlets at the Alvin W. Vogtle Nuclear Plant; ii.
to test modifications to the original design concept with the aim of developing approach flow conditions to the sumps which were os uniform os practicably possible.
The opplicable NRC Regulatory Guide 1.82 (June 1974) offers guidance on the selection of an acceptable design. However, compliance with the regulatory guide position that the design coolant velocity at the inner screen should be approximately 0.2 fps (calculated with the assumption that one-half the vertical screen area is blocked) does not insure that the sump will perform os required after a LOCA as flows approaching the sump may not be symmetrical and high velocity streams could carry excessive debris to the sump structure.
I t
i
.a- -.-.
/ -
w 2 :_.-
~=
i l
WESTERN CANADA HYDRAULIC LABORATORIES LTD.
i 2
i!
2.
SUMMARY
A hydraulic model study covering the entire containment floor areas between el 171.75 ft and el 178.00 ft has been corried out at a 1:7 scale to assess and improve i
flow conditions to the emergency core cooling system intakes at the Alvin W. Vogtle Nuclear Plant, being constructed near Augusto, Go. Flows from the RHR sumps were l
recirculated to the model through a well chamber set below the moc'el floor at the opproximate reactor location. Spray flows were divided and introduced around the l
model in accordonce with calculated prototype distributions. No direct flow occurred I
between the reactor or spray flow inlets and the sumps. Tests were corried out at on ougmented discharge of 1.5 times the Froude scaled value. Tests examined flow conditions at low and high water levels for both single and two train operation of the RHR and containment spray systems.
4 It was found that flow conditions around the intake coges with the original design were dominated by three main flow streams which entered the intakes ar~ a e
with velocities between 0.6 fps and 1.9 fps. The strength and directions of these streams were governed by the structural configuration of the containment walls, the location of the refuelling canal drains, and the postulated water depths and discharges.
Due to these main flow streams, opprooch velocities to the screen coges were not a function of screen crea.
Maximum opproach velocities to the intokes of the original design ranged from 0.4 to 0.9 fps, Figure 5.
Relocation of the refuelling canal drains so os to discharge vertically l
downwards into the possage west of the west secondary shield wall reduced the strength of the streorn created by the drains and relocation of the intoke sumps to l,
positions out of the main flow streams following the containment walls, reduced velocities directly approaching all intakes to 0.5 fps or less, Figure 8.
The intent of Regulatory Position 7 of NRC Regulatory Guide No. l.' 2 is to minimize the possibility of screen blockage. This,'ys been ochieved for the Alvin W.
I Vogtle Nuclear Plant containment through utilizing a hydraulic model for optimum layout of the ECCS.
,------w,
~, -.,. ~ _ -,,, -,,. - _., - - - - -, - - - - - ~ - - _.
n-, - - - ---w s.
= - -
-a..=
WESTERN CANADA HYDRAULIC LABORATORIES LTD-3 3.
CONCLUSIONS The conclusions which were drawn from this study are:
l 1.
The flow conditions in the intakes area as originally designed were dominated by three main flow streams, with velocities of 0.6 fps or greater.
The streams, shown on Figure 5, were located as follows:
1.
along the east containment vessel wall with a velocity of 0.90 fps on entering the Spray A intake area, ii.
along the west containrnent vessel wall with a velocity of 0.57 fps on entering the RHR B intake area and iii.
from the refuelling canal drain ducts with velocities up to 1.90 fps on entering the Train B intakes area.
2.
The strength and direction of the streams resulted from the structural configuration of the containment, the location of the refuelling canal drains, and the postulated water depths and discharges.
3.
Flow velocities approaching the intakes in their initial design positions were sufficiently high to transport debris which could block the screen coges and reduce the effectiveness of the intakes. The approach velocities to the intake screens were entirely dependent on the characteristics of the main structure and not on the screen area of the intakes.
4.
Regulatory Position 7 of NRC Regulatory Guide No. l.82 could have been satisfied by construction of screen cages which would have provided suffi-I cient screen surface area to pass the required flow with a computed average velocity of less than 0.2 fps. However the intent of NRC Regulatory Guide No. l.82 would obviously not have been met, i.e. debris would not have a low velocity area in which to settle out.
i t
c
~ _ ~ -
.... --~ - -
WESTERN CANADA HYDRAULIC LABORATORIES LTD.
4 l
5.
To achieve the intent of NRC Regulatory Guide No. l.82, the strength of the approoch streams had to be reduced and the intakes had to be relocated away from the main streams into creas where debris would be dropped out of the flow before reaching the intakes. This was ochieved through the model study, in which relocation of the intakes and refuelling conel drains to the positions I
shown in Figure 7 produced the most suitable intake approach conditions consistent with structural and access limitations.
6.
With the intakes installed at their odopted positions, flow velocities to the int &es were generally much lower when the system was operated at high water level or with single train operation than when it was operated in the low water - two train mode.
7.
The findings of the I:7 scale model study would be applicable for prototype flow conditions with water temperatures up to 254 F and water levels up to el 181.25 ft.
8.
The majority of tests were run on the assumption that the spray return flow vrould be evenly distributed. Introduction of spray flow around the contain-ment vessel walls with a distribution different than 50 percent on each side did not affect the model study results.
r
j WESTERN CANADA HYDRAULIC LABORATORIES LTD.
5 4.
NTRODUCTION t
The Georgio Power Company is constructing Units I and 2 of the Alvin W.
Vogtle Nuclear Generating Station near Augusta, Ga. Bechtel Pcwer Corporation is the prime consultant for design of the plant.
The Nuclear Regulatory Commission (NRC) requires that attention be addressed in the plant design to the satisfactory operation of both the emergency core l
cooling sumps (ECCS) and the containment spray sumps (CSS) following a loss-of-coolant accident.
The Residual Heat Removal (RHR) pumps in the Alvin W. Vogtle Nuclear Plant will withdraw 4500 USgpm during single train operation and 3550 gpm per pump during two train operation. The CSS pumps will each withdraw 2600 U5gpm during either single or two train operation. The minimum postulated water depth above the containment floor during operation would be 2.67 ft.
The NRC Regulatory Guide No. l.82 recommends that design flow velocities at the ECCS and CSS intakes should be maintained below 0.2 fps to allow debris with a specific gravity of 1.0S or greater to settle before reoching the screen surface. The design flow velocity is to be calculated on the assumption of 50 percent blockage of the fine inner screen. The basic assumption behind this regulation guide is that the approach velocities will be a function of screen area. However,in many containrnents this is not the case, particularly where the approach flow is regimented into streams by the structural arrangement within the containrnent walls. An intake could be located within on approach stream velocity of greater than 0.2 fps and while the letter of the Regulation Guide would be met by having a large screen area, the intent of the Guide would not. Debris would be carried to the intakes by the approach streams.
The intent of the NRC Regulation Guide No.1.82 would be best met by locating intakes away from strong approach streams and, if possible, reducing the strengths of the approach streams where their velocities are independent of screen area.
,-e-,'
r.
---..----,-,,----e---
w
,-e
-r-,
t l
WESTERN CANADA HYDRAULIC LABORATORIES LTD.
e Since the flow pottern and associated velocities within the containment are
(
not amenoble to numerical prediction, a hydraulic model was used to determine the best locations for the intokes and minimize the potential for blockage by debris.
b t
9
..=_a
.. w -
.1
- =.
WESTERN CANADA HYDRAULIC LABORATORIES LTD.
7 i
5.
TPE MODEl.
l 5.1 Model Construction The model was constructed to on undistorted scale of 1:7 on on octogonal plywood base raised 2.5 ft above the Laboratory floor, Figure 1.
A 12 in. high f
waterproofed plywood outer wall was constructed around the edge of the base platform. The base platform was constructed in four separable sections for conven-ience in storage offer completion of the project.
The circular outer wall of the containment vessel was represented by 16 gauge galvanized sheet steel erected inside the outer plywood model wall.
The secondary shield wall and major structures located between the shield wall and the containment wall were represented in 3/4 in. plywood or sheet metal, Figures 2A and 2B so that flows from the reactor or spray discharge areas would pass through scaled flow passageways to the intakes. The model was constructed from floor el 171.75 ft to el 178.00 ft in accordance with dimensions given on Bechtel Drawings Nos.
IX20488004 Rev. B to IX2D488007 Rev. 8, i139E31 and IX2D4F002 Rev. A.
l Screen cages of 3-l/2 in. square grating and 5 ft square plan area, prototype dimensions, constructed of plastic grid and capped by a solid metal top at el 174.08 ft, l
were placed over each of the four proposed sump locations, Figure 2C. Sump flows were drawn off vertically downwards through 14 in. prototype diameter openings located in the model floor under the screen coges. Flows from the RHR sumps were returned to o baffled RHR return well located below the model floor in the reactor vessel arco near the centre of the model. Flows from the CSS sumps were pumped to a flow-dividing weir which returned 10 percent of spray flow to the reactor area, 20 percent of spray flow to discharge into the refuelling canal and 70 percent of the flow to be distributed for flow around the sides of the containment.
Water was drawn from the intakes by two 2 HP and two 1/2 HP pumps through flexible 14 in. prototype diameter hosing. The flexible hosing permitted ready interchange of intake locations.
- - u.=
.~
T WESTERN CANADA HYDRAULIC LABORATORIES LTD.
g j
Filling or draining of the model was accomplished by 2 in. diameter lines connecting to the central model area inside the secondary shield wall.
Model dimensions of containment vessel structures and initial intake positions as approved by Bechtel Power Corporation engineers prior to testing are shown on Figure 3.
Important dimensions were checked by the Project Engineer after i
l construction and were within 3 1 prototype in.
5.2 Model Scales t
5.2.1 Consideration Of Froude Number The flow patterns in a free surface body of water such as the flooded containment vessel depend upon both flow inertia and the effects of gravity octing on the water surface gradient.
This results in the free surface hydraulic modelling l*
requirement that the Froude number, relating inertial to gravitational effects, be the same in the model and prototype. However if flow velocities, and therefore surface gradients, are small, momentum effects dominate gravitational effects on the flow streams and the flow pattern is dependent upon only the relative velocity between 4
points. in this situation, fulfilling Froude number similitude is not necessary to replicate flow conditions and the primary modelling consideration is that the model i
Reynolds number must be large enough to ensure that the flow regime is turbulent. It was not necessary to reproduce heated water conditions of up to the postulated post 4
j LOCA temperature of 254 F in the model as model Reynolds numbers at room temperature were sufficiently high to ensure that flow was in a turbulent regime.
Froude numbers characterizing flow conditions in the plant following a LOCA will be generally in the range of 0.01 to 0.07.
In the cose where momentum effects dominate gravitational effects, the model dischorge may be increased, within limits, above that required for Froude similitude without changing either the flow pattern or the relationship between i
l I
velocities at different locations.
This technique facilitates photography of flow patterns in a model and enables more occurate,-sadings to be mode of the flow
'~
I J velocities. The technique of augmenting model discharges above Froude scaling was adopted in this study.
l
WESTERN CANADA HYDRAULIC LABORATORIES LTD.
y 5.2.2 Scale Relationships I
An undistorted scale of 1:7 was odopted for the study. This scale gave the following model to prototype relationships on the basis of maintaining Froude similitude in the model and of using a discharge augmented to a factor of 'A' times the Froude scaled discharge.
i Model Scale Description Scale Relationship Froude Similitude Auomented Dischorae Length L
1:7 1:7 2
Area L
g,g9 g,49 5/2 Discharge L
1:129.6 A:129.6 I/2 Time L
1:2.65 A:2.65 i/2 Velocity L
1:2.65 A:2.65 5.3 Instrumentation The water depth in the model was measured by a Lory Type-A point guage occurate to 3 0035 prototype ft.
0 Discharges from individual spray and RHR intakes were measured by pressure drop across machined brass orifice plates. Pipe discharges were calculated using formulae and tables given in the American Society for Testing Materials Manual No.
25 and were considered accurate to 1 2 percent. Calculations for and plotting of orifice discharge curves were independently checked.
Flow velocities in the model were measured using a Nyeric Low Speed velocity meter and were checked using a Gloster mini-propeller meter. Each meter was calibrated before and offer the documentation stage of the study. The occuracy of the Nyerpic velocity meter reading was 2 9 percent in the range between prototype velocities of 0.3 fps to 0.9 fps.
t
. -.... ~ - -. -
l WESTERN CANADA HYDRAULIC LABORATORIES LTD.
10 i
i 6.
TEST PROGRAM The test program was divided into three phases:
Phase 1.
The purposes of Phase I were:
1.
to establish a flow scale for the model which would produce higher model flow velocities and would therefore permit more occurate velocity measurements than could be achieved at Froude similitude but which would not result in any change to flow patterns near the intol<es.
ii.
to examine flow conditions in the model and identify any poten::al problem creos and 111.
to assess whether these conditions worronted design modifications.
The test sequence followed in Phase I was:
Test No.
Operating Trains Discharge Scale Water Depth, Ft.
I A&B Froude' 2.67 2
A&B 2.65 x Froude 2.67 3
A&B 1.5 x Froude 2.67 4
A l.5 x Froude 2.67 5
8 1.5 x Froude 2.67 6
A&B l.5 x Froude 5.33 in which Train A was the combination of Spray A and RHR A sumps operating together and Train B was the cornbination of Spray B and RHR B sumps reperating together.
..,. _ _ _.. ~.. _ _ _ _ _ _... __
WESTERN CANADA HYDRAULIC LA! ORATORIES LTD.
gg Phase 2.
The purposes in Phase 2 of the study were to determine the optimum locottons for the spray and RHR pump intakes and to examine other possible structural modifications.
The test sequence followed was determined on an iterative basis. Numerous intake positions were examined in varying detail. Proposed intake locations and other modifications were discussed with Bechtel engineers as testing progressed to ensure I
that the optimum arrangement was found with regard to approach flow conditions and to structural considerations. Test conditions in Phase 2 were:
1.
low water level, el 174.42 ft., with 2.67 ft flow depth to produce the highest velocity condition, 11.
model discharge of 1.5 times the Froude scaled value, lit.
two train operation, and iv.
70 percent of spray discharge split 50 - 50 percent between the east and west sides of the containment.
Following testing under these conditions, the adopted intake locations were examined to ensure that no undesirable effects would occur in the prototype during the following flow conditions:
I i.
water level, el 177.08 ft, with 5.33 ft flow depth t
l
[
11.
single train operation of either Train A or B, lii.
the 70 percent portion of the spray discharge restricted to flowing entirely around either the east or west sides of the containment alternatively, i
iv.
model discharge of 1.5 times the Froude scaled value.
1 J
,..--,-.,_.n.
,__,n__,---
n
m _
.. =_ _
~
WESTERN CANADA HYDRAULIC LABORATORIES LTD.
IL Phose 3.
I The purpose in Phase 3 was to develop details of, and to document, the adopted design alternatives.
t l
Docurnentation was corried out for the following conditions:
Flow intoke Water Augmentation Operating Loyout Depth,ft.
Factor Trains Original 2.67 1..
A&B j
Original 2.67 1.5 A&B 4
Original 2.67 1.5 A
Original 2.67 1.5 B
Adopted 2.67 1.5 A&B Adopted 2.67 1.0 A&B Adopted 2.67 1.5 A
Adopted 2.67 1.5 B
Adopted 5.33 1.5 A&B i
l l
f i
1 i
.a
.. r,:
.a
-m-f WESTERN CANADA HYDRAULIC LABORATORIES LTD.
i l
7.
TEST PROCEDURE t
7.1 Model Operation Water was recirculated through the model by four pumps, one per intake.
Flow was controlled by manually operated gate valves located downstream of orifice meters in the pump discharge lines. Flow through the orifice meters was determined i
from visual readings of Meriam gauge fluid (SG of 2.95) monometers. Water level was controlled in the model to 1 0005 f t by manual operation of filling or draining valves.
0 Independent readings were made by two observers of all manometer or point gauge readings during the Phase 3 documentation stage of the study.
Most testing was carried out at a low water level of el 174.42 ft to obtain the maximum velocities that would occur in prototype. Check tests were run at a water level of el 177.08 ft to verify that there were no adverse effects generated at deeper water.
All water drawn from the two RHR pumps was recirculated to a well set below tl.e model floor near the centre of the model, at the opproximate reactor location. Spray flows were divided up by a fixed elevated weir box which discharged to the model by gravity flow with 10 percent going to the centrol well, 70 percent being directed to the north end of the model, and the remaining 20 percent being I
returned through an elevated 2 in, diameter line which simulated spray flow caught by the refuelling canal drains, Figure 1. The 70 percent spray flow directed to the north end of the model was further split by weirs into two co>ol parts flowing around the east and west sides of the containment vessel walls. Plasticized fib ~re matting was used downstream of the weirs to reduce turbulence.
Discharge through each spray intake was at all times equivalent to 2600 USgpm. Discharges through the RHR intakes were 4500 USgpm during single train operation and 3550 USgpm during two train operation.
i l
Tests were run during Phase 2 of the test program to determine whether there was any effect produced by distribution of the 70 percent spray flow other than 50 percent to each side of the containment. The entire 70 percent spray flow was
j WESTERN CANADA HYDRAULIC LABORATORIES LTD.
14 directed alternately down both the east side and the west side of the containment j
vessel walls by blocking off appropriate weirs.
7.2 Determination Of Augmented Dischsge Tests to determine on appropriate augmented discharge level were run at low water level in the model.
The purpose of flow ougmentation was to achieve flow velocities which could be more occurately measured without distorting the basic flow pattern.
f Flow patterns produced throughout the intakes area during two train oper-otion with Froude-scaled discharges were photographed and documented. The' "intokes areo" was considered to be the southeast to west sector of the containment vessel i
between the elongated chamber to the east of the easternmost occumulator tank and i
the refuelling cavity structure, Figure 3. The intoke flows were then increased to o I
level of 2.65 times Froude scaled discharge, producing prototype velocities in the model. Overhead photographs and visual observations showed that the surface flow I
patterns, especially around the pressurizer relief tank and in the areas of train B intakes vorled from that recorded with Froude-scaled discharges.
5 The discharge augmentation factor was reduced to 1.5 and photographs again taken of flow patterns. Comparison of overhead photographs, Figure 4, and visual observations showed that flow conditions were similar for discharges scaled to both 1.0 and 1.5 times the prototype Froude number.
l An augmented discharge of 1.5 times Froude scaled discharge was used for Phase 2 testing and for documentation of flow velocities in Phase 3. Tests were also run at Froude scaled discharge in Phase 3 to co.. firm that similar flow potterns j
developed at the two discharge levels with the original and odopted intake layouts.
i q
7.3 Flow Patterns And Velocities Surface flow patterns os indicated by,t;onfetti traces were photographed throughout the intakes crea, using two 4 in. x 5 in. press cameros with overlapping i
fields of coverage. Exposure time was generally 2 seconds.
t t
.,__-...._.___,,...,,,__.,__.._,,-__,,_.,_._.,,,_.,,--_,-.,,,_,-..-...,__._-,m, m
.-u...
w
-... =.
}
WESTERN CANADANYD2AULIC LABORATOXIES LTD.
g3 Subsurface flow patterns were observed using permanganate and titanium I
dyes. Bottom flows were studied using gilsonite tracer material.
Flow velocities were normally measured at mid-water-depths in regions of I
maximum flow. Fast flows along the surface or bottom were also recorded if noted to be difforent from those at mid-depth.
Velocities approaching the intakes were measured at a distance of 5 prototype ft from the intake centerline. Velocities were not generally measured at fixed positions in the model as the regions of maximum flow velocities varied in position with the intake or structural configuration being studied.
Minor localized velocity increases which were due to structural constrictions, obstruc-
- l tions or to intake proximity were noted but were not shown in the overall flow pattern as they did not represent the debris-carrying capacity of the system.
7.4 Velocity Reducing Modificatians I
i Following investigation of flow conditions with the original design layo0t, j
Figures 48, 5 and 6, it was concluded that modificotton of the initial layout was j
required to place the intakes in low approoch velocity areas. These modifications took the form of structural alterations, relocation of the refuelling canal drain outlet and q
relocation of the intakes.
Structural modifications were carried out using plywood and sheet metal.
Relocation of intakes was offected by drilling through the model floor and mounting appropriate pipe fittings. Alternate positions for any intake were connected to a single flexible hose on the pump suction line. During testing, unused openings were sealed off with thin plastic cover plates. The intakes in use were covered with centered 5 ft square coges of plastic grating. The intake cages provided visual scale for flow observations and eliminated air core vortices from the model.
7.5 Gilsonite Bottom Movement Technique Gilsonite, a hydrocarbon with a specific gravity of 1.04, was used in the model as a tracer material to indicate areas of flow concentration near the model I
floor. The material had a mean size diameter of 1.5 mm and a uniformity coefficient,
.m..-.
WESTERN CANADA HYDRAULIC LABORATORIES LTD.
l6
- being the ratio of diameters for the 60 percent finer size to that of the 10 percent
}
finer size, of 1.9.
j f
Tests were. prepared by filling the model to the correct water depth without i
circulating pump 9:: tion and spreading a 1/8 in. thick layer of gilsonite uniformly over the model floor in the "intokes area", Figure 6A. Overhead photographs were taken of I
the initial gilsonite layer. The appropriate pump trains were started and further i
photographs taken at real time intervals of 10,30,60,120 and 180 minutes from the stort of the test. These photographs were compared following the test to determine f
areas with bottom velocities higher than that required to induce motion to the gilsonite.
k A movement calibration test was mode with measured current velocities over gilsonite bonds ocross straight channels to determine the minimum flow velocity required to move the gilsonite across the model floor. It was found that the gilsonite wos sensitive to bottom turbulence but that with a flow ougmentation factor of 1.5, velocities equivalent to between 0.35 and 0.48 prototype fps were needed to clear the gilsonite from the floor.
During testing it was noted through diffusion of permangonote dye, that turbulence near the modelled columns and structures was greater than had been experienced in the calibration facility. It is estimated that the creas not cleared of gilsonite in the photographs of the bottom movement tests were subjected to velocities of less than 0.35 fps, in the neighbourhood of columns, localized eddying or upwelling produced clearing of the tracer material though measured overage velocities were os low as 0.30 fps.
No apprecioble gilsonite movement was noted with tests at 1.0 times Froude discharge scaling as flow velocities were too low.
=
a _. =.
WESTERN CANADA HYDRAULIC LABORATORIES LTD.
ii
~
8.
FLOW COPO(TIONS WITH ORIGINAL LAYOUT Two train operation of the model at low water levels with the model as initially designed showed that flow patterns in the intakes area were dominated by three fast flowing streams with velocities of 0.6 fps or greater, which maintained their identity to the intakes, Figure 5.
These streams were developed by:
1.
the flow around the periphery of the containment vessel wall on the east side being constricted as it possed the chamber entrance at the southeast sector of the containment vessel, Feature H on Figure 3, and entering the intokes cree with a prototype velocity of approxi-mately 0.9 fps near the outside wall, I
II.
the flow coming around the periphery of the containment vessel wall on the west side being constricted as it passed the refuelling cavity structure, feature A on Figure 3, and entering the intokes crea with a velocity of about 0.6 fps, l
iii.
the flow from the refuelling canal drain ducts which discharged in a southerly direction from el 181.25 ft directly above the fon duct outlet producing southerly flows away from the fan duct exit of approximately 0.9 fps. Local bottom velocities measured near the foot of the falling stream were up to 1.9 fps.
The peripheral stream along the east wall produced flow velocities of l
opproximately 0.65 fps approaching directly onto the Spray A intoke cage. Flow l
velocities along the containment vessel wall a few feet away were up to 0.9 fps.
Flows approaching Spray A from the north, around the east occumulator tank were slow, with velocities of opproximately 0.2 fps.
l Both the confetti traces, Figure 48, and the gilsonite migration photos, Figure 6, showed that the east peripheral stream retained its strength as it flowed along the wall until it was largely drown off by RHR A intake. Velocities impinging j
l I
~. - -
WESTERN CANADA HYDRAULIC LABORATORIES LTD.
lg directly onto the RHR A screen coge from the east were about 0.5 fps with velocities of up to 0.4 fps being measured approoching the intake from the north.
The discharge from the overhead refuelling cavity drain ducts produced a flow from the fan duct towards the Spray B intake. Velocities approoching the Spray B Intake were about 0.7 fps, Figure 5. The direct stream path of the overhead discharge towards Spray B intake is shown clearly by both the confetti trace and gilsonite photos, Figures 4 and 6.
The residual of the main stream opprooching Spray B intoke from the northeast was turned to the west around the food bearing column north of the intake and approached RHR B intoke with velocities up to 0.4 fps. The peripheral flow around the west wall approoched RHR B intoke with velocities of between 0.4 to 0.5 fps. These two currents set up a general clockwise circulation around the two intakes with slower eddies being generated away from the main circulation.
The flow patterns in the intake area were basically the same at the high water level as they were during the low water tests. Flow velocities were lower during the high water test than during the low water tests.
A review of the containment structural geometry indicated that no adverse effects to flow patterns, nor increases in flow velocities would be experienced with surface water levels up to el 181.25 ft.
Single train operation was investigated at low water levels.
Discharge l
through the RHR intakes was increased to 4500 USgpm for single train operation as opposed to the 3550 USgps drawn through each RHR intake during two train operation.
The maximum velocities approaching the RHR A intake were nbout 0.4 fps. Flow velocities radially approaching RHR B intake were between 0.3 and 0.4 fps with higher velocities of up to 0.5 fps flowing circumferentially around the intake at a distance of opproximately 5 ft, reflecting a locally induced circulation. Flow velocities approach-l ing the spray intokes were lower at all times during single train operation than during two train operation.
l I
u
~. -
= __._
..2..._.
WESTERN CANADA HYDRAULIC LABORATORIES LTD.
- 9 G
9.
DESIGN MODFICATIONS I
9.1 Refuelling Coal Drains And Wall Openings it was noted during the Phase I tests that the three main flow streams entering the intakes oreo maintained their identity until they reached the intakes, resulting in direct Impingement on the intake screens of flow velocities exceeding 0.5 fps. Two methods were investigated to improve this flow conditiom 1.
to reduce the strength of the flow streams through structural changes and j
I ii.
to relocate the intckes away from the main flow streams.
Numerous small openings 2 ft 6 in, high by 6 in. wide were constructed in the west, southwest, south and southeast secondary shield walls to permit some flow fro'm the central reoctor oreo to reach the intakes without having to flow around the containment vessel wall. Flow through these openings had little effect on either the flow potterns or on velocities near the intakes. The limited effect of the openings was not considered sufficient to justify their odded structural costs and they were closed off.
The refuelling canal drain ducts were relocated to eliminate the strong southerly stream towards Spray B intoke. The ducts were initially moved to discharge northward from the north refuelling cavity wall, post the fan intake. This arrange-ment resulted in a strong circulation north through the fan duct, south along the west containment vessel wall and easterly post the stairs near RHR 8 intake.
The momentum of the westerly peripheral stream was slightly increased. Flow velocities j
near Spray B intoke were reduced but those near RHR B intake were increased to 0.6 fps during two train operation at low water level.
l l
-~
~ - -
WESTERN CANADA HYDRAULIC LABORATORIES LTD.
20 i
The refuelling canal drain ducts were then relocated to discharge vertically downwards into the possage west of the west secondry shield wall near the opening to the centrol reactor area, Figure 7.
By discharging the ducts downwards, the momentum of the discharge was dispersed and there was no longer on identifiable single stream from these ducts to the intakes.
In particular, the approach flow velocities to the Spray B intoke were greatly reduced. The L shaped constriction at j
the south side of the south opening in the west secondary shield wall was removed to permit easier possage of flow from the central area to the intokes. The opening was moved to o distance of 6.5 ft south of the building centerline, Figure 7.
9.2 Relocation Of Intakes o
The original Spray A intoke position, Figure 5, was near the region of strong flow possing along the peripheral wall from the east. This intoke was moved to a location between the two accumulator tanks, Figure 7. The position was remote from the main flow stream and was not exposed to strong currents.
Maximum flow velocities approaching the intake during two train operation at low water level, Figure l
8, were reduced to about 0.4 fps.
Flow approoched the revised location more uniformly from all sides than had occurred at the initial Spray A intake position, Figures 9,4B and 68.
Due to structural requirements, it was not practicable to relocate the RHR A intake very for from its original position, Figure 7. The intoke was shifted northwords os for os possible away from the main strength of the peripheral current. Flow velocities approaching the intake in its revised location were between 0.4 and 0.5 fps i
during two train operation at low water level, Figure 8.
l Relocation of the refuelling canal drain ducts eliminated the strong southerly I
stream which had initially impinged onto the Spray B intake with velocities of about 0.7 fps. The Spray B intake was moved eastwards and opproximately 3.5 ft further away from the containment vessel wall to permit more room for development of l
uniform approach conditions to the intake. A third Spray B intoke position tested closer to the pressurizer relief tank, Figure 7, indv7 strong flow through the passage d
I south of the intake between the tank and a nearby column.
Flow velocities l
opprooching the Spray B intoke in its adopted position were 0.5 fps or less during two train operation at low water level, Figures 7 and 8.
l
-+- -.
a..
WESTERN CANADA HYDRAULIC LABORATORIES LTD.
2I The RHR B intake was initially located in the path of the peripheral stream flowing along the west wall and in the circulation pottern set up by dischorge from the overhead refuelling canal drain ducts. The intake was relocated to o position 4.00 ft south of the refuelling cavity structure and to the east of the stairs, Figure 7.
Subsurface flow velocities approoching the intake, measured at a distance of 5 ft from the intake, were between 0.3 to 0.5 fps, Figure 8.
i A test with two train operation at higher water level confirmed that flow velocities in the intakes area were approximately 20% to 45% lower at all points of observation with the deeper flow.
Tests at low water level, with either trains A or B operating individually, Figures 10, II and 12, showed that the flow velocities approaching the intokes during single-train operation were not increased over those measured during two train operation. The higher velocities anticipated near the RHR intakes os a result of increased discharge during single train operation were compensated for by the reduc'ed strength of the main flow streams and the more uniform opproach flow conditions to the intckes.
It was apparent that the relocation of the intokes had resulted in a significant reduction in opproach velocities to Spray A and RHR B intakes, Figures 5 and 8, which l
together with the reduction in velocities at Spray B intake produced by relocation of the refuelling canal drain ducts, had produced on overall improvement in opproach flow conditions and a reduction in the potential for blockage, Figures AB and 9A, compared to conditions with the original design.
The relocation of RHR A, Figure 7, hd been restricted by structural and equipment constraints. A reduction in opprooch velocities to this intake through its relocation had not been achieved to the sorne extent as at the other intakes. It should be noted, however, that original approach velocities were generally lowest at the RHR A intoke, Figure 5.
Furthermore, the relocated intakes produced opprooch velocities which were generally the some for all intakes. That is, conditions were no better, or worse, at the RHR A intake than at the other intakes.
l l
l l
l l
- = -. _.
f WESTERN CANADA HYDRAULIC LABORATORIES LTD.
22 1j Within the constraints imposed by structural considerations and the locottons i
j of large pieces of equipment, it was concluded that no further reductions in approach flow velocities could be ochieved through relocotton of the intakes.
The flow
)
conditions were largely established by the momentum of the strooms entering the intakes area. The intake locations developed through the model studies are considered optimum under these conditions, with the intakes being located away from the high l
velocity zones of the approoch flow streams.
increasing the screen areas around the relocated intakes would not have further reduced approach flow velocities.
l 9.3 Spray Flow Distribution The 70 percent of the spray flow introduced to the north end of the model had been evenly divided to flow around both the east and west side walls to the intakes. Tests were run to assess the effect, if any, of introducing all of the,70 percent spray flow down either the east or west side of the containment vessel.
When the 70 percent spray flow was discharged in its entirety around one side or the other, there was a compensating effect from the RHR flows delivered to the central reactor area. The majority of the RHR flow from the central area exited to the outer portion of the containment vessel through the possoges in the secondary shield wall on the side of the structure furthest removed from the spray flow. No appreciable increase was noted in the flow velocities, compared to those which had been measured during normal testing, when the 70 percent discharge was introduced to only one side or the other, indicating that the model test results are not sensitive to j
the distribution of the spray flow.
t 1
h W.A. McLaren, P. Eng.
Monoger
... - - -. ~ _ - -..:.-
.n.
FIGURE I
91tEli LEVEL GAUSE 9
Q ggMA4 WhL M ERP400F M NTAug CASE OUTEft anLL
-r I
I 'Pf" e-fIWAsg I
l es
-4 1::WR T.
".2-
_. R - w --
Sm
=
SCHEMATIZED CROSS-SECTION 20'-0" S
WALL OF l
N00EL di INTARE s,
IMTAKE
'/
+
l/
la 5
l Flow Distateutine wEin
/f
//
/
RNA A
//
iMTant a
.L.GAUSE i ',
/ /
~*4 g
RETistas h
wEu.
s%,
,i
$g ii ji g Q
- SE008tOARY SHIELD wALU l s%%
FL0w Il (shown schemeticelly ) ll g gigygggggggg it o wEin
\\g O
1l 4,
\\
A 1
%Q "TAKE
,8 N%
l 3
i Y
g 1
- 2* d FLEXIskE i
6 g.WETAL u
Pumum EastNLL I I $.I PUMPS 8I l y
wCR s0X %
ggig I
C 2* d OleCNARGE LINES I
m 6'-0" MIN.
f-O"__
Q'I asETERIMS CAiFICES PLAN NOTE e F0ft WEL Amo STRUCTURAL oimEnsions usto insiot Tut ConTAimestui vEsse..sEE rie. 3 SCALE: I "a 4' ALVIN W. VOGTLE NUCLEAR PLANT I:7 HYORAULIC MODEL STUDY MODEL LAYOUT WESTERN CANADA HYDRAULIC LABORATORIES LTD.
scil 3687A-WCH
= _
l 1
Figure 2
,7 S
g. __
](^N i
..{[. 5 3F '~
A
~
j l
14t s
~. '
l w
v A
>{
>. ;[ ',,.
Y.
.. r, -
as
,C A.
View of model structures without water I
I I
I 1
!l
,I a.,-
' s -t I
m.
t
-.c..
I B.
Refuelling canal drains located above fan duct opening I
p 74 l
l C.
5 ft. square screen cages with alternate i
intake positions sealed off MODEL AS CONSTRUCTED I
SN i
T Y
O 3
r
- e N
e D
I
. m.
.i A U S
L T N
E s
s.
P S a
t E
R n
a
'M u
c R L I
U t
. n.
A a
G w.
E a.
E r
D D
FI F
s.
L e.
O
.u C
. ~..
..i U M r
s.
N L
o e
=.
.c I.
O R c
w.
C E
e e.
- e. e 6 E iL D
so e
L
.m i.
ti U
O T
G A a
M e
co
. r o i.n V D
- e.. f
. Y I
L n
e@e@eeee ae WH D
s.
E N
V :
IL I
7 L
3 A
A T
E D
~
).
~ i:
y -
3-11 I
1=
(
i
' I I
m I
94 I
q
_N t
i f./
.f., '
g
+
T
=
r o
m.
w
~
4 Ij
.y3 f'
y.
Nl.i*
7 3
1'
.. g c.
N7
=
e lq
,.L m
=
s J.9 ~
. h i'
{
g g
e t
l y'
- p~
i h
,t -
t'
~
a g
OII pg.,g-
+
^ -
i_'
u T
.u e
h t
=
e c
J u
N I
m sa t
i
- =
s D
3 oc TL S
E IRO TA R
O B
\\
t g-y g-A L
C IL U
A R
D Y
H AD A
N A
C N
R E
TS E
W I
i'
Figure 4 g
MI 3,
.a
~
' -)
,3 4:
/j
's.
f I
?h \\
fM
,s
... x hs-3 l
~,
her::=zenc, 4 A.
Surface flow patterns produced during two train operation with original layout at Froude scaled discharge, low water j
level.
i
.. -g
. ~.
?.
~
\\
U
';
- i i '.
v t-I a
1?
'a l.
B.
Surface flow patterns produced by two train operation at a t
discharge scale of 1.5 times Froude, low water level.
1 i
SELECTION OF Al'G:!ENTED DISCH ARGE
li e
N S
O e
I E
I TA T
T 5
N Y I
RE A D C
P L U O
O E
P TS L
T IN R
R E U A
U A L Vd R
G E E T
I L D Y
F C O D A O U M N L W N
A T
I IL NA E C L -
LT U R N L
G A
O EI E
R V. D T G V
I E
Y T R L
W H
A O R N
P g
I 7
E V
T L
A 1
W AW O
/
L W
F O
L
/ud l
I V
i
. I' p
e n
E 3
G E
L N
ll v
A.
s o
x h
/D l
f
)
%A
~
4 HR I
/%
'1, h
'/
\\
/
s m
Y
@U D
L
,,'N T
\\
L f
S E
e J
I R
O
)Iy s:!l
.):J T
A i
~/
R
.t !
6 E
O L
B n
A C
A L
a.
S i
l
//-
IL C
e, l
,e AU f
,e R
1 D
1 9
1 l
Y l
H 1
l A
1
\\
D
\\\\
?
A N
e A
e C
c s
N v
R E
T S
E T
E o
W so p
Figure 6 W-
"3.
-4 c,
't, =..*,.e.
' ' ", ~,.g us,g Ah., f 'h', y ;
b
..s.4.
-~
r
. %.;.y;;
4.
.t
.y. -
.,t 2:
..{.
=
9,
.t. p,.
.s 3
. ; rt:c$5
?hh'?
5
)
-)k,c.ik 1.y h;[!' & %'
./
~
'g
's
\\
s
'K.._
N_
km Nj 4
s
. - r.r-
.t=_
A.
Model with original design layout seeded with gilsonite prior to test.
s..
f*,
{
{p' <j 5
g i
f
\\g
,j t
N s
'A I
F.
's L.
%; w
^
%s
- r. %
_w J
t w
~,,
- ,.., mm B.
Gilsonite swept from floor areas af ter 30 mitiutes of two train operation at low water level.
i GILSONITE MOVEMENT - ORIGINAL LAYOUT
1 1
5a I
s' E O
w-a 4
D
< a 9
I Uo N
t -.
oO u
w o3 3a g
y
=l ga y E O
. o w
N$
W Q.
O i
.s g
O irie-7 y5 w
3 l
4, B
=
=
.i. u n
3mnionuis mn 3
[
-=
x
=
l g
. /
5; e r 1-!
e3U NEg a
ga g-
,' s s
r li a
V
,N,s',
E t
w
{
if Ig
~
o
-a i
gg
=
i
$l A J' k-w 9
I O!\\
o
%$.d D
l
-(
2-I O
E "I
3
/
i twr o
l JIB ir w
O I
k k~
g g. !.
f E
l
{-
},
V u
62 g
e 5 5; 2
2s o
O y
a g
a 5
sT oql
==
o s
i
=
5 ss
)
o
'5 3 E ej'l j
oEN5 O
9}j I
t 8~
=
N i5 2
g m
0OO y
=
g.,
i s
i 4
+~
O L
l l
l N
S O
I E
T I
C A
T 8
TI N P
N Y S
A D C
L U OO O
E P T R
S L
I N
R ET I
U A L V A A
G FI E E C
R L D I
T C O D F U M N
I o
N A D w e
O T
E C L L N M I
T it G
R O A L
V R E D E
W. D T E V
Y T T LE H
A P
N P
I 7
O R
V E
L 8
D T
A 1
W A A
O W
L F
W O
L b
i.
1 r
J)
.i.
,s
. i..,
.e i
..i.
i D
.e N
E w
.m.
G E
r.
e.
/
g L
..,s e
.i e
N1 e..
1 e
.,.t, A
e A
/
VARP S
l i
f 4_
I A.
Q R
MR 6
i ce
/
rr
/ '- f
/
o ta Y
n e
AE x
R G eR t
S A t
%C N
H er 5
0, o
2D sJ
\\
h il sN
/ /
D p
T
\\
/
L gt
/
f o^
S k
t~
IE R
f
\\
j/..^h Iu$
O TA R
O E
O L
B A
o, ~
A
,'vE R
IL C
L S
C
% V m
/"-
A f
U R
r D
a ll d a
H l
Y l
o wC l
l1 N A
1 D
A s
NA N
o R
E TS
,E E
o W
6 I\\1 l\\lll L
l Figure 9 1
=
4,4,mmens f
\\
4 4
l w
l wg-> W
~-
~
e g
A.
Surface flow patterns produced at two train, low water level operation with the adopted modifications.
orvz l
}
I N'f 4
f,
.! fi.
g l
.)
s I
r sa y
- ~,
,EE' B.
Gilsonite swept from floor areas after 30 mins. of two train operation at low water level.
SURFACE AND BOTTOM FLOW - ADOPTED MODIFICATIONS
S e
E 0
T I
Y T
1 N Y I
S L
A D C N o N
L J E
P T'
O O R
S L
A I
U R
E T G
A L V A N
FI E E C
AI L D I
R C O D F T
U M N I
N A D e
C O
L E IL N M E
LT U
V G
R E
O A L
V R E D
. O T E R WY T T E
H A
T P
A N
P I
7 O W V
D L
A 1
WA w O
o L
LF I
l e
i.
i
\\
m.,, ii..
. e.
i o
s c
8 E
.e i.
G i.
E
. i.
,s L
.i.
. i..
N t,u,.
m.
s v
d g
u o
g C,
j s
I
)
3 i
l i
i!
Ah g
g
..R M
s#
R n
/,,
T TO
/
T Y
O A E R
R P
G P
S.R T
H E
- k C E
oS F
I tD 0
\\
3; Q
,/
D
/ /'
T L
1 e
o-S t
E RI R
O s
TA n
R O
E Il a
L B
A 0-A 3
iI c
C 1
L l
S m
c e
iL e
l_l U
s A
e R
l D
l i
Y l
Oi l
l H
l l
A l
D n.
S*
A N
A s.
C i.
se N
v
- 0. t R
E T
S E
T E
O W
N i
m a
e S
E e
I Y
1 T
TI S L
N Y 1
A D C N O N
L U O O E
P TS L
I B
R R
E T U
A L V A IN G
E E C
A L D FI C O D I
R F
T U M N ID N
A O
l e
E CI L
L L N M E
T V
G U R
E O A L
V R E D T E R W. O Y
T T E
H A P T
N P
A O w I
7 V
D L
WA w 4
A O
ot LF G
I l
)
,i D
N E
e G
E L
g1 1
t A
b
=
~
-lgI 1
l i
i
)
(
"M Ep ro l!.<
4 ren F
e A E R G P R r
S A
e H
s
%C r
o$
o-l 2O s
\\
l ll,
D TL o
/
t S
E
./
j I
I RO
\\
f.
T X
A y
j iLw R
E 0
i
-4 L
8 R.
A o
A C
i L
s S
C
,i.l
_!A-YI e
IL e
U It-l N
A R
l jf s
o -
D YH ill E
/
\\
A D
r.
A s
N A
o*~
e C
i i
RN v
E i
T t
S r
E o
W 3
l
1 q
Figure 12 j
l w
'c.('
,a
- ^
,...j
- _ lt' i
l
%V O l
A.
Surface flow patterns produced by Train A only operating at low water level.
I
..{-
U v-I
%r,.
i s
3:,
b 7
~~
~ ~ ~
l I
B.
Surface flow patterns produced by Train B only operating at low water level.
i SINGLE TRAIN OPERATION - ADOPTED MODIFICATIONS 1
- - - - - - - - - - - - - - - - - - - - - - -