W3P85-0130, Forwards Rept on License Condition 16 Evaluating Potential Adverse Effects of Failure of Containment Coatings on post-accident Fluid Sys
| ML20113C322 | |
| Person / Time | |
|---|---|
| Site: | Waterford  |
| Issue date: | 01/17/1985 |
| From: | Cook K LOUISIANA POWER & LIGHT CO. |
| To: | Knighton G Office of Nuclear Reactor Regulation |
| References | |
| W3P85-0130, W3P85-130, NUDOCS 8501220292 | |
| Download: ML20113C322 (31) | |
Text
LO UISI AN A / 44 oe-o~o emm . aeax.00e POWER & LIGHT / New onteANS LoUSANA 70174-8000 . (504) 388-=345 M$0l$hY U January 17, 1985 W3P85-0130 3-A1.01.04 A4.05 Director of Nuclear Reactor Regulation Attention: Mr. G.W. Knighton, Chief Licensing Branch No. 3 Division of Licensing U.S. Nuclear Regulatory Commission Washington, D.C. 20555
SUBJECT:
Waterford 3 SES, Docket No. 50-382 LP&L Report on the Evaluation of Containment Coatings
Dear Sir:
The Waterford 3 Operating License NPF-26 incorporated a License Condition requiring Louisiana Power & Light Company to evaluate the potential adverse effects of the failure of coatings inside containment on post accident i fluids systems.
The evaluation specified by the License Condition has been completed, and the evaluation is documented in the attached LP&L report.
Please contact myself or Robert J. Murillo, Safety and Environmental Licensing Unit Coordinator, should you have any questions concerning the subject report.
Yours very truly, K.W. Cook Nuclear Support & Licensing Manager KWC/RJM/pcl Attachment cc: E.L. Blake, W.M. Stevenson, R.D. Martin, D.M. Crutchfield, J. Wilson, G.L. Constable pg124g2;ac o gghggg a.
}Q0 p P ,
t -
bec: R.S. Leddick, R.P. Barkhurst, F.J. Drunnnond, T.F. Cerrets, D.E. Dobson, G.G. Hofer (Ebasco), W. A. Cross (LP&L Bethesda Office) ,
J.M. Veirs (CE), R.M. Nelson, R.A. Savoie, M.J. Meisner, G.E. Wuller, Project Files, Administrative Support (2),
Licensing Library
!~
i C'
E
P LOUISIANA POWER & LIGHT CO.
WATERFORD SES NO. 3 REPORT ON LICENSE CONDITION NO. 16 EVALUATION OF THE POTENTIAL ADVERSE EFFECTS OF THE FAILURE OF CONTAINMENT COATINGS ON POST ACCIDENT FLUID SYSTEMS O
e e
TABLE OF CONTENTS SECTION TITLE PAGE I Introduction 1 II Conclusions 2 III Assumptions Used In Analysis 3 IV Evaluation of Far Field Paint Effects 4 V Evaluation of Near Field Paint Effects 8 VI Evaluation of Secondary Effects of 13 Coating Failure on NSSS Equipment VII Evaluation of Secondary Effects of 16 Coating Failure on B0P Equipment VIII Insulation Inside Containment 19 References 22 TABLE TITLE 1 Containment Coating Materials Assumed to Peel FIGURE TITLE 1 SIS Sump Far Field Effects 2 SIS Sump Chip Motion With Constant Angle k
3 SIS Sump Near Field Effects 4 SIS Sump Plan View l
l 1
I. INTRODUCTION The Waterford 3 Operating License (NPF-26, dated December 18, 1984) has a License Condition (Section 2.C.16) that states the following:
" Prior to January 18, 1985, the licensee shall provide for staff review and approval an evaluation of the potential adverse effects of the failure of coatings inside of containment on post accident fluid systems."
This report provides the evaluation specified by the License Condition.
M II. CONCLUSIONS An evaluation of the potential adverse effects of the failure of coatings inside containment on post accident fluid systems was performed. Highly conservative assumptions were postulated in the evaluation. The evalua-tion determined that the SIS and CSS will remain functional given the most conservative coating failure conditions. The far field and near field effects of coating failure were evaluated. The evaluation deter-mined that pool velocities far from the SIS Sump are insufficient to transport coating particles to the SIS Sump region. Therefore, the SIS Sump screens are not susceptible to any clogging due to the far field effect of coating failure. The evaluation also determined that the near field effect of coating failure, coating falling into an area of the SIS Sump surface between 2.04 ft. and 3.42 ft. from the screen, results in about 36 percent of the vertical SIS screen area remaining unclogged.
This unclogged area is more than sufficient to prevent pump vortexing and to provide for adequate Net Positive Suction Head (NPSH) to ensure proper pump operation.
4 The secondary effects of the injestion of paint chips by post accident fluid systems was also evaluated. The evaluation determined that there are no components in the post accident fluid systems that are susceptible to degradation resulting from paint debris. This part of the evaluation was predicated on the assumption that paint chips could reach the SIS Sump screen, pass through the screen, and enter the post accident fluid systems. The evaluation covered the SIS, CSS, and NSSS.
Finally, the additional potential clogging due to insulation debris was evaluated. The findings indicate that there is no potential for SIS Sump blockage due to insulation debris.
i
III. ASSUMPTIONS USED IN ANALYSIS The coating materials in the Containment at Waterford 3 are expected to withstand the service conditions during normal plant operation and 1
post-LOCA operations. However, it is postulated that all the coating materials except coatinge on the concrete surfaces inside the containment (which was applied to the requirements of ANSI N101.2, N5.12 and N101.4) would fail during a LOCA. This assumption, although unrealistic, provides a conservative groundwork for an analytical evaluation of the effects of the coating failures on plant safety systems.
! For the SIS Sump blockage analysis, another extremely conservative and equally unrealistic worst case assumption was made for the particle size distribution of the failed paint. For the paint debris transport -
analysis, it was assumed that all the paint fails as 0.078" diameter disk shaped particles. This is the smallest particle size which cannot pass through the sump screens and is it more transportable than larger Particles.
The SIS and the CSS will not be placed into recirculation mode operation immediately after a LOCA as the ECCS and the CSS will be initially operated in injection mode using water from the Refueling Water Storage Pool. The SIS and the CSS will be switched to recirculation mode at the completion of the injection mode at which time the Containment floor will be flooded with water. The minimum flooded water level was determined as El. -3.67' and the maximum flooded level at El. +0.5'(Figure 3).
The evaluation of the SIS Sump blockage can be divided into far field effects and near field effects.
Far field effects concern paint and insulation that may be carried from j anywhere in the Containment to the immediate sump region and then clog the SIS Sump screens. Near field effects concern only paint falling in a region adjacent to the SIS Sump screens, from which calculations show that the paint could reach the screens and clog the vertical screen surfaces.
IV. EVALUATION OF FAR FIELD PAINT EFFECTS At the initiation of the recirculation phase of the post-LOCA operation, dislodged paint is subject to a circulating water flow. Fluid velocity, debris density, and debris size were analyzed to determine if debris transport occurs.
Paint Transport Velocity The transport velocity for paint particles was derived following the basic concepts of NUREG/CR-2791 (Reference 4). In addition, the assumptions and derivations for generalized motion and transport of palat particles, that were developed in the Texas Utilities Generating Company (TUGCO) Report (Reference 1), were utilized. As shown in the TUGC0 -
- Report, with a model of forces balanced on a paint particle, the critical water velocity to initiate tumbling of a paint particle can be expressed as
p ,) t (2ge) 0.5
~
VTumble = g,( Pp (1)
Cp p , _
where: g, = static friction coefficient pp = density of paint j
p, = density of water j t = thickness of particle 8e = Newton's constant Cp = drag coefficient Similarly, the critical water velocity to initiate sliding of a paint particle can be shown as:
Valide = W1 (#p - #w) ( rd/4) (2gef 0.5 (2)
(1+ #d) CD #w .
where: d = diameter of particle
- d = dynamic friction coefficient and the remaining parameters were the same as defined with Equation 1.
- In addition, in the TUGC0 Report, a sensitivity study was done on the l
l
Paint particle size, density drag coefficient and friction coefficient.
The following conclusions were drawn from the study:
(a) The thickness of the paint particle has no effect on its transport velocity.
(b) The smaller the paint particle size, the higher the potential for its transport.
(c) The greater the relative density difference between the paint particle and the moving water, the lower the potential for transport.
(d) The higher the drag coefficient between the paint particle and -
the moving water, :he higher the potential for transport.
(e) Variation in the friction coefficient between paint particle and concrete floor of the Containment does not significantly affect t
the transport velocity.
The paint particle size was assumed conservatively to be 0.078 inch, the smallest size that could clog the SIS Sump screens. The critical water velocities to initiate tumbling and sliding of this small paint particle l were determined as:
l V
Tumble
= 0.074 ft/sec V
S11de
= 0.181 ft/sec l
l In the calculation, the following conservative data were used:
t = 0.005 in.
i #p = 82.3 lbe/ft3
- w = 60 lba/ft3 (at 200*F) i
. . . _ _ ~ - . . . . - - - . _ , _ _ , _ ,
0.6
=
g, 3
C " l'1 D
4
- d
=
0.42 The paint thickness and density are based cn data in Table 1 and the SIS Susp water temperature is a conservative estimate at the time of SIS / CSS recirculation. The friction and drag coefficients are the typical values used in the TUGC0 study.
For the purpose of performing a simple, yet still conservative evaluation of the far field effects, it is further ascused that all the failed paint will find its way down to the Containment floor (El. -11') since all the -
upper elevation floors are gratings. By making this assumption it is only necessary to evaluate several critical locations at E1.-11'. To calculate the local water velocity at the selected critical locations, the flow rates shown in FSAR Table 6.3-4 (Reference 2) for the SIS and CSS during long-term recirculating mode were used. Four Regions, A, B, C and D, with potentially higher local velocities were identified in Figure
~
- 1. Regions A and D were considered to see whether any paint debris can j be transported from the First and Second Quadrants to the Third and Fourth Quadrants. Regions B and C were considered to determine whether any paint debris from the Third and Fourth Quadrants can be transported to the near sump regions (for potentially clogging the SIS Sump screens). In calculating the maximum local velocities at the given.
region, the break is always considered near the given region to determine the highest value, while the Containment Spray flow is assumed to be uniformly disl:ibuted inside the Containment. The maximum local velocity l in Regions A and D is calculated as 0.126 ft/sec; Region B is 0.105 ft/sec; and Region C is 0.252 ft/sec. Since all the tumbling debris will eventually orient itself such that the maximum surface area is oriented parallel to the Containment floor (Reference 4), the critical slide l velocity becomes the governing parameter for the debris transport along the Containment floor. By comparing the local water velocity against the critical slide velocity (0.181 ft/sec) and the critical tumble velocity
.(0.074 ft/sec) calculated earlier it can be concluded that although paint l
I
debris in Regions A, B and D can be tumbled by the water flow, it will not be transported to the other regions. Only the paint debris falling into Region C can possibly be transported into the region near the SIS Sump. However, based on the near field effects evaluation, the velocity at 3 feet away from the screens is merely 0.0277 ft/see which would not even tumble the smallest paint debris assumed. Therefore, paint debris being transported from Region C to the region near the SIS Sump will settle at least 3 feet away from the SIS Sump screens and will not be transported further to clog the screens. Even if it is assumed that paint debris in Region C can be transported toward the SIS Sump screens, the TSP baskets (Section V) around the SIS Sump would block the paint debris from reaching the screens. Consequently, it is concluded that the SIS Sump screens will not be clogged by any paint debris falling into the far field regions.
l l
V. EVALUATION OF NEAR FIELD PAINT EFFECTS This section of the report summarizes the results of the analysis done to study the behavior of paint fragments that become dislodged in the event of paint failure and fall to the surface of the pool of water surroundius the SIS Sump during the post-LOCA recirculation mode.
Introduction Motiot of paint fragments through a pool of water is affected by many parameters, including fragment size, shape, density, and water velocity.
For the purpose of this analysis conservative assumptions (which allow for maximum clogging of the sump screens) were made. These include the following:
(1) All paint covering the Containment dome, structural steel, and uninsulated piping is considered capable of peeling. A list is provided in Table 1.
~
(2) The top of the SIS Sump screen is completely clogged. Only the sides of the screen are available.
(3) The peeling paint is uniformly distributed throughout the Containment.
(4) The paint peels as particles shaped like a disk, with a diameter equal to the minimum hole size in the SIS Sump screen at the minimum thickness and minimum density possible.
(5) The maximum drag coefficient possible acts on the paint particles to minimize the settling velocity.
(6) The paint particles are horizontally oriented as they travel through the pool to minimize the settling velocity.
r (7) Paint motion is assumed to be the type of motion with constant angle - this is used to get the paint vertical velocity.
(8) Maximum flooding level exists.
(9) When the paint flakes, the flakes align themselves one next to the other on the SIS Sump screens.
Additionally, the following observations have been made:
, (10) Ductwork above the SIS Sump screen at El. + 6.0' blocks the falling paint from falling onto the pool surface.
(11) I-beams, below the pool surface supporting the grating at El. -
-4.0' collect the paint falling on the water surface above the beam and this prevents these chips from reaching the screen.
Methodology The methodology used in analyzing this problem is the following:
(1) The settling (vertical) velocity of the paint fragment is determined.
(2) The average drift velocity of the water in the pool (moving towards the SIS Sump screens) is determined.
(3) Using (1) and the SIS Sump depth, the time for the paint to reach the bottom of the SIS Sump screen (E1. -11.0') is determined.
(4) Using (2) and (3) the maximum and minimum distance from the SIS Sump screen for which paint can hit the screen are determined.
(5) The amount of paint that falls into the area on the pool surface from which it can then hit the screen is determined.
(6) Tha am:unt of cerain arsa blocksd is daterminsd.
l (7) Geometrical considerations that may shield some of the screen j from being blocked are accounted for.
(8) A determination is made as to whether the area of the screen left unblocked is sufficient for proper pump operation.
Analysis of Motion with Constant Angle This analysis assumed that the paint fragment is a disk which hits the
~
pool surface at any incident angle. Conservatively, small paint fragments are assumed to be momentarily arrested at the water surface, then to start their travel through the water at the angle of impact with zero initial velocity. Any angle of impact is equally probable since tumbling motion would be expected. However, as shown in Reference 1, the -
final paint vertical velocity is a minimum (which is conservative) for an incident ' angle (6) of 0'. Referring to Figure 2, the equation describing the vertical motion of the paint through the water when the pitch angle is assumed constant is the following:
pp Yp du = p V p p g - 9, Vp g -C D (*) #w Aproj sin S W2 dt 2
-CL ($) p,Aproj cos # W2 (3)
T where: W2 = u2 + (yo - y)2 Aproj =
rd sin (4) l 4
l F
l
and: u = vertical component of the fragment velocity, defined -
aspoptivedownward v = horizontal component of the fragment velocity
- p = point density
- w = water density W = fragment relative velocity
=
V, velocity of pool toward the screen C = drag coefficient D
C = lift coefficient L
- = angle from pool surface to the velocity vector V
p
fragment volume t
thickness
= acceleration of gravity a
Since a steady-state velocity is quickly reached and assuming S = 900 ,
Equation 3 reduces to:
0=p p vp g - p, Vpg -C3 (4) av b roj w2 (4)
T Conservatively assuming that v = V,, & = 90* and 8 = 0* and using the conservative maximum CD = 1.9 from Reference 1, Equation 4 reduces to:
0=p p [rd 2 ts - Pw rd2 tg - CD av rd W2 (5) q4 ) (4) 3 4 or ( p p p ,) g = CD #w u2 (6)
'It 3
Using conservative values of a p = 82.3 lba/f 3t , p, = 60 lbs/f t ,
t = 5 mils and CD = 1.9 we have:
u = .072 ft/sec Knowing the maximum pool height it can be calculated that the time to I reach the SIS Sump screen bottom is 159.3 seconds.
. . ~ . , . - . _ . . . . - - ... ..- . - -- . - - . - - - . -
From Reference 2, FSAR Subsection 6.2.2.2.2.1, the water velocity at the screen is .13 ft/sec. Accounting for the open area of the screen and the pool depth, the drift velocity in the pool towards the screen is 0.0277 ft/sec. The maximum horizontal distance travelled by the paint is then 3.62 ft. From geometry, as shown on Figure 3, the minimum distance at the pool surface horizontally from the screen from which paint can reach the screen is 2.04 ft. However, a series of Trisodium Phosphate Dodecahydrate (TSP) baskets surrounding the screens interfere with some of the falling paint. As a result of blocking by the baskets, the maximum distance from which paint can hit the screens is reduced to 3.42 ft. Although the baskets would block paint from reaching the bottom of the screens, gravity would bring down excess paint from the top of the screens to block the bottom also. Of all of the paint that is assumed to peel, a percentage of it would fall into a near field critical area on -
the pool surface from which it could reach the screen. This area surrounds the SIS Susp screens between a distance of 2.04 ft and 3.42 ft. However, a significant portion of this area above the screens is blocked by I-beams and ductwork. The actual situation is shown in Figure 4.
There are I-beams supporting the grating at El. - 4.0' which, in effect, will produce a shadow on the screens by blocking the paint that falls on them. Additionally, there is ductwork at El. +6.0' that shields part of the pool. The net effect of the shielding of the SIS Susp screens by the I-beams and ductwork is that approximately 36 percent of the vertical screens remain open.
Reference 3 (which required that only 10 percent of the vertical screens remain open)shows that this amount of open area is more than sufficient to prevent vortering at the pump intakes and also allows for sufficient Net Positive Suction Head (NPSH) to ensure proper pump operation.
VI. EVALUATION OF SECONDARY EFFECTS OF C0ATING FAILURE ON NSSS EQUIPMENT As secondary effects, coating failure was also evaluated for small paint chips passing through the SIS Sump screen. In this case fine particles (less than 0.078 in) enter the SIS and also the Reactor Coolant System (RCS). The effects of this type of coating failure on NSSS are summarized below.
j Source Term i
The maximum quantity of paint which could reach the SIS Sump, calculated from the near field evaluation, is approximately 115 lbs. To evaluate the potential adverse effects of this material on post-accident fluid i systems, it must be assumed that this paint takes the form of fine -
particulate matter which mixes homogeneously with the recirculating coolant. The design of the SIS Sump screens ensures that particulate matter greater than 0.078 inches in size cannot be ingested into the SIS or the RCS, Identification of Concerns Two conditions of concern have been postulated with this particulate matter suspended in recirculating coolant: (a) detrimental effect on the HPSI, LPSI, and CSS pumps or restriction of flow through the reactor
- core, or (b) consequence of the paint fines settling in the fluid systems.
HPSI Pumps The HPSI pumps are used in the recirculation mode following any LOCA.
The pump vendor has confirmed that particulate matter 0.090 inches in diameter will have no detrimental effect on operation of the HPSI pumps.
LPSI Pumps The LPSI pumps are not required during recirculation. However, for small breaks, it will be possible to enter the shutdown cooling mode. The LPSI
pumps are used in this mode. The pump vendor has confirmed that particulate matter 0.250 inches in diameter will have no detrimental effect on operation of the LPSI pumps.
Reactor Core In the reactor core, the location of the smallest flow area (the potentially limiting flow area) occurs at the fuel spacer grid-fuel rod intersection. By design, particulate matter up to 0.090 inches in size will not become lodged in this area. It should be noted that avoidance of the potential for flow blockage in the reactor core was the principal design basis considered in the sizing of the SIS Sump screen.
Regardless, if it is assumed that some blockage occurs, it follows that such blockage would be localized (i.e., at individual grid / fuel rod _
intersections). Such local subchannel blockage would have no effect on core cooling during recirculation since the open lattice fuel assembly design allows subchannel-to-subchannel as well as assembly-to-assembly crossflow.
Paint Fines in Reactor Vessel The paint fines which are suspended in recirculating coolant will tend to settle in locations of low fluid velocity. The point of lowest velocity within the system is the bottom of the reactor vessel. The potential for paint, which has settled out in the bottom of the vessel, being swept out and into the core in the event that a reactor coolant pump is started, has been considered.
The Emergency Operating Procedure Guidelines recommend starting a Reactor Coolant Pump and maintaining forced circulation flow during plant cooldown following a small break LOCA; however, this requires that the break be small enough that the HPSI pumps are maintaining the RCS at about 1300 psi. At this condition, the SIS will operate in the injection mode from one to two hours before switching to recirculation. Assuming that plant conditions permit starting a pump, and that the decision to
begin cooldown is made two hours into the event (allowing about one hour of recirculation) only a fraction of the 115 lbs or 1.39 f t3 of paint l 1
available for ingestion will actually be present within the RCS, and only i a fraction of that will have settled out.
For the purpose of this discussion, it may be conservatively estimated that 150 in3 (approximately 6%) of the paint fines have settled out.
This quantity of particulate matter, if swept into the core, may cause some temporary local blockage; however, the forces and velocities present will immediately break up such collections. Moreover, as noted earlier, local subchannel blockage would have no effect on core cooling during recirculation, since the open channel fuel assembly design allows subchannel-to-subchannel as well as assembly-to-assembly crossflow.
4 2
1
VII. EVALUATION OF SECONDARY EFFECTS OF C0ATING FAILURE ON B0P EQUIPMENT f
For the purpose of this portion of the analysis it is assumed that paint particles, of sizes less than 0.078 inches pass through the screen and enter the SIS and CSS. The following is an evaluation of the impact of this paint intrusion on components of these systems.
l Following a LOCA or MSLB, fluid escaping from the primary or secondary system break or discharged by the CSS, will accumulat e in the SIS Sump and the lower level of the Containment. This water will contain any failed paint that would be removed from its original location by either the environmental conditions present within the RCB and/or the impingement of fluid. The recirculation mode would transport paint particles that pass through the screens to the HPSI and CSS pumps for -
reinjection into the RCS and Containment, respectively. To ensure a complete evaluation, the LPSI pumps and SCDS have been included since their operation is conceivable during long-tern post-LOCA operations.
1 There are various types of equipment that are utilized in these systems.
~
They are:
valves (gate, globe, check, butterfly and stop-check)
- pumps (vertical and horizontal centrifugal) heat exchangers i -
miscellaneous (orifices / flow elements, spray nozzles, vortex breakers and instrumentation).
Each of the above items have been evaluated for detrimental effects caused by the transported paint. The paint particles are not expected to Present any erosion problems for any of the material in the equipment identified above (for erosion to take place it is necessary to have high flow velocity and some concentration of very hard abrasive particles in the flow. - none of these conditions are present in the system following LOCA).
_. _~. . _ , - _ _ . . . ~ . _ _ . _ . -
,u 4 4 - - a * -- aL-. _- ..m___ . _ - -_#_._A.m. p. a ,,_4 Concerning the potential of blockage due to an accumulation of paint, it i is expected that particle settling will occur only 'ni areas of low
, velocity. Based on this, flow blockage is not expected to occur in the individual pipelines or within any of the valves or pumps. The flow velocity within the tubes of the Shutdown Heat Exchanger is sufficiently high enough to avoid plugging. Paint particles could accumulate in the low velocity section of the heat exchanger's channel heads. However, this would not be detrimental as the buildup would not exceed the height of the tube openings on the tube sheet. Once the paint buildup approached this height, it would leave the low velocity region and thus be exposed to higher flow velocity.
The Containment Spray Nozzle is not susceptible to blockage since the nozzle throat diameter is auch greater than the paint particle (0.375">> _
! 0.078"). The vortex breaker is composed of grating plates of similar size to the sump trash rack. Each opening is approximately 1/2" x 1-1/4" and would not be susceptible to blockage. Instrumentation taps are taken from the sides or top of the piping. Therefore, particle settling would not occur. Also, since the fluid in these lines is stagnant, particle transportation required for blockage conditions in the lines, would not
' occur. All orifices and flow elements are provided with bore sizes that I greatly exceed the particle size that can pass through the screens.
4
- Another potential detrimental effect that can be attributed to failed 4
coating is the loss of heat transfer through the Shutdown Heat Exchanger. In the initial phases of an accident heat from the i Containment is removed by cooling the SIS Sump water before it is redischarged via the CSS. Cooling is accomplished at the Shutdown Heat i
Exchanger where heat from the SIS Sump water is transferred to tha Component Cooling Water Systen via the tube asterial. Any fouling of the tubes would reduce the performance of the heat exchanger. Since the paint is not expected to decompose and the flow through the tubes is relatively high, no significant particulate fouling is expected.
Also considered was the potential for a loss of equipment performance due i
. to paint particle transport. This would include loss of function and i equipment damage. All valves have been reviewed to determine if an 1
accumulation of paint could impair valve operation. The only valve type that could be affected is the gate valve. This valve has an area, known as the crotch, which could be susceptible to sediment accumulation. The crotch is located directly opposite the stem and accepts the outer Portion of the disc as the valve is closed. If this area were significantly blocked, the disc could be prevented from fully seating and result in a partially open valve. This event could only occur if valve was open for an extended duration and then be required to close. A review of the affected systems has shown that there are no gate valves, that are required to close after being open.
All pumps have been examined for possible detrimental effects due to paint transportation. The Containment Spray Pumps were designed to pass Particles of up to 1/4 inch in diameter. Each pump is provided with oil -
lubricated bearings. The mechanical seal water supply line is provided with an abrasive separator to remove particulates from the seal water.
If a failure of the separator is assumed, the seal would either be exposed to abrasive fluid or a loss of sealing fluid. However, considering the design and construction of the seals, complete pump
~
degradation or failure is unlikely.
VIII. INSULATION INSIDE CONTAINMENT Type and Description There are four types of insulation used inside the Containment. They are:
- 1. Metal Reflective,
- 2. Metal Encapsulated,
- 3. Figerblass Insulation Encapsulated with glass : 7th, and
- 4. Radiant Energy Shield (fire wrap).
The metal reflective insulation is built of stainless steel panels. The panels consist of interior and exterior sheets. The exterior sheets are 24 gage austenitic steel, type 304. The interior material is three
- layers per inch of 0.002 inch thick waffled, type 304, stainless steel
! sheets. This insulation is manufactured by Transco. This insulation is attached, using either of two methods. One method is buckle fasteners l (positive lock-quick release buckle fasteners). The other method is by
- using stainless steel, #14, self-tapping screws. It is estimated that the total area of metal reflective insulation is approximately 3,500 square feet. This insulation is used on the Reactor Vessel, Reactor Head and Reactor Coolant Pumps.
The second type of insulation is metal encapsulated. This insulation consists of Owens-Corning inner material and Transco's encapsulating material. Owens-Corning identifies this insulation as TIW Type II, FG (fiberglass) encapsulated. This material conforms to the property l requirements of government specifications: HH-1-5588 (Amendment 3), Form B-Blanket and Felt, Flexible, Type I Blankets, Flexible, Class 7 and 8; MIL-1-24244 Chemical Requirements and USCG 164.009/135/2. Transco provides the encapsulation material for the fiberglass insulation. The fiberglass is totally encapsulated. The method of attachment is buckle P
fasteners (positive lock-quick release buckle fasteners). The
~
construction os the totally encapsulated modules is exactly the same as
- for the reflective-type, except that non-reflective insulation (fiberglass) is used inside. It is estimated that approximately 7,400 linear feet of insulation is supplied for piping while another 15,000 square feet of insulation is supplied for equipment. This insulation is applied to the Steam Generators, Pressurizer and skirt, Regenerative Heat Exchanger and Quench Tank in addition to various piping systems.
The third type of insulation is a fiberglass insulation fully encapsulated with glass cloth. The fiberglass cloth material is referred to as a Temp Mat insulation as manufactured by Alpha Company. This i " donut" or " flexible collar" insulation is tied around the 91 CEDM nozzles and 10 instrumentation nozzles. The collars are held together -
using two (2) levels of stainless steel wire. It is estimated that apprximately 155 linear feet of this type of insulation is used inside Containment.
1 The last type of insulation is the radiant energy shield. This shield is 4 a fire resistant ceramic fiber blanket full encapsulated within high silica content fabric mesh. Final assembly includes an interior liner of fiberglass mat. The encapsulation is accomplished by joining outer layers with a quarts thread and reinforced nylon thread. The radLant energy shield is formed by wrapping the assembly about conduit or electrical cable trays by use of galvanized metal frames and attached with galvanized threaded bolts and screws. It is estimated that approximately 200 linear feet of radiant energy shield is used inside i Containment.
l Debris Generation i
i All four types of insulation are designed to remain intact when exposed to Containment Spray, the only mechanism that would dislodge the insulation is jet impingement from a pipe break.
l l
l I
l l l
- I i
4 All metal materials from the insulation will sink. In addition, the smallest metal part (rivet) has a diameter of 1/8 inch, this is larger than the fine screens (0.078 in).
Although the fiberglass and ceramic sats may be lighter than water, the individual fibers have a specific gravity of at least 2.0 and thus would
- sink. In addition, individual fiberglass and ceramic filaments are approximately 5 mils in diameter. Therefore, although individual fibers could pass through the SIS Samp screens they would have no effect on SIS or CSS equipment.
! Jet Impingement Locations i
A review was conducted of the locations of jet impingement cones inside -
Containment. The review concluded that none of the cones generated insulation debris that could fall near the SIS Sump.
Conclusion Although a pipe break with its resulting jet impingement cone could 2
dislodge insulation, the insulation would sink and thus could not block the SIS Susp. In addition, no jet impingement cone would generate insulation debris near the SIS Sump. Therefore, there is no potential for SIS Susp blockage due to insulation debris.
i O
a J
l 1
REFERENCES
- 1. Report by Gibbs & Hill, Inc. for Texas Utilities Generating Company, Comanche Peak Steam Electric Station " Evaluation and Insulation Debris Effects on Containment Ezergency Sump Performance," Revision 1, October 1984.
- 2. Waterford SES Unit No. 3 Final Safety Analysis Report, Docket No.
50-382.
- 3. Report by Western Canada Hydraulic Laboratories, LTD for Ebasco Services, Inc. , Louisiana Power & Light Company, Waterford Unit No. 3 "Model Testing of the Safety Injection Sump," June 1982. (Submitted to NRC via LP&L Letter W3P82-1755, dated June 28, 1982). -
- 4. NUREG/CR-2791, " Methodology for Evaluation of Insulation Debris Effects," September 1982.
6 0
TABLE 1 CONTAINMENT C0ATING MATERIALS ASSUMED TO PEEL Coated Primer and Thickness Approximate Dry Density Surface Topcoat (mils) area (aq ft) (1ba/ gal)'
Carbon steel exposed Dimetcote F-2 2-7 275,825 16.0 to primary containment Dimetcote 6 2-7 275,825 16.0 atmosphere-uninsulated Amercoat 71 2-18 275,825 11.0 piping, structural main Amercoat 90 5-11 275,825 11.0 equipment Containment vessel done Carbozine 11 2-5 30,788 21.3 -
Phenoline 305 5-7 30,788 11.0 Miscellaneous touch up ZRC cold 3-4 8,000 17.8 on galvanized steel galvanizing compound Page 1 of 1 i
l
- o ',
$O y' *
/fll ,' j; f ',.c i fe 7th o 't S'* G __
3 'l l
\ '\
/ p ( '
c)[edwT \
I, I:, ' s, .!(%
SECCHC QUADRANT ;a
/7 . / // b L*HCQ
,m w
.' '\ e\ ..
u
\. 55: '
\\
h'
- ns.o - g '\
l'
~ ..
\\ ,6 t (RDG 110' ' f ;
7'I. .j
' 'S W 1 ' s o- _
i,me=ctea g; i t :/n, WA r,, \/// p /_ f'///4^//// ' !
j,+
, ,1 q/.
i n
i y
i/
\ \ __."~/
g P,~
.c
/.b.fc.1; .
- 6.
!?qt 7 <
- $ 03 % s CVADRANT ' '; -
s
' f, .
[
//gwouRm '!
, c l , -
'ho 2:3 i ,' b;~*,
- . . cv Acaur .,
-; .4._.
~
r x, e i i . ;,
1 ' e'
~r n p
. s .S-
. v-
'*.r '
- p eI LOUISlANA Figure POWER & LIGHT CO.
Waterford Steam FAR F ELD EFFECTS Electric Station
i YO
,2 v, .
/
\
e -
\
p +
W l
V W V
, LL l
l i
LOUISIANA Figure POWER & LIGHT CO. SIS SUMP Waterford Steom CHIP MOTION WITH CONSTANT ANGLE 2 Electric Station
r- -
l
, 3.42' -
2M
- t. 4~1, WATER SURFACE ,
/ PNNT PATH
- fo.0'
$C6 TSP BASKET I
I
/
/ / JA l/
/ /
-n.o'
.75' l.0' i
i l
l I
l LOUISlANA Figure POWER & LIGHT CO. sis sump Woferford Steam NEAR FIELD EFFECTS 3
- Electric Station I
WALL.
DUCT i ,
~ ._
e, -
/ 5-su o
uJ
/ /. I j $ O
/ <
/' , >
'e -
2 5 3
=
, j/ I 3'-fo s '[ ,l' 0 .
i 15'C , _
71
/' ,
/ / j. - . .
! i
- SY4 WIDE, W ALL ', / [
/ { l ,
/ i
, i CPENING , - '
/
/ i .ico '
. I
- I if
/
/
/i
/ y
..e_& / lY '- ,
I
.o
/
/ -
e l p y I BEAM
- i
- ' i N
[
.._ _. ./
LSUMP SCREE}4 LOUISIANA Figure POWER & LIGHT CO. SIS SUMP Waterford Steam Pt.AN VIEW 4 Electric Station