ML20140C866
| ML20140C866 | |
| Person / Time | |
|---|---|
| Site: | Turkey Point  |
| Issue date: | 01/23/1986 |
| From: | Flanders H FLORIDA POWER & LIGHT CO., WESTINGHOUSE ELECTRIC COMPANY, DIV OF CBS CORP. |
| To: | |
| Shared Package | |
| ML20140C819 | List: |
| References | |
| OLA-2, NUDOCS 8601290013 | |
| Download: ML20140C866 (15) | |
Text
-T -
.s (-
UNITED STATES OF AMERICA D g
. NUCLEAR REGULATORY COMMISSION
/g ' ;
BEFORE THE ATOMIC SAFETY AND LICENSING BOARD T . "' x ,(
h[ < .. . [
\g;p . ,
/
iin the Matter of ) Docket Nos. 50-250-OLA-2 h ,
) 50-251-OLA-2 FLORIDA POWER AND LIGHT COMPANY )
)
(Turkey Point Nuclear Generating ) (Spent Fuel Pool Expansion)
Units 3 & 4) )
AFFIDAVIT OF HARRY E. FLANDERS, JR.
ON CONTENTION NUMBER S
- 1. My name is Harry E. Flanders, Jr. I am Manager of Commercial Product Engineering for the Nuclear Components Division of Westinghouse Electric Corporation. My business address is Westinghouse Electric Corporation, Scenic Highway (State Route 90), Pensacola, Florida, 32504. A sunenary of my professional qualifications is attached as Exhibit A and is incorporated herein by reference.
- 2. The purpose of my affidavit is to address Contention 5, as limited to the structural integrity of the Turkey Point Units 3 and 4 Spent Fuel Pool Storage Racks. The affidavit of Leonard T. Gesinski will address the structural integrity of the fuel assemblies to be stored in the racks.
Contention 5 and the bases for the contention are as follows:
8601290013 060123 PDR ADOCK 0500 0
1 0
Contention 5 That the main safety function of the spent fuel pool, which is to maintain the spent fuel assemblies in a safe configuration through all environmental and abnormal loadings, may not be met as a result of a recently brought to light unreviewed safety question involved in the current rerack design that allows racks whose outer rows overhang the support pads in the spent fuel pool.
Thus, the amendments should be revoked.
Bases for Contention In a February 1, 1985 letter from Williams, FPL, to Varga, NRC, which describes the potential for rack lift-off under seismic event conditions [ sic). This is clearly an unreviewed safety question that demands a safety analysis of all seismic and hurricane conditions and their potential impact on the racks in question before the license amendments are issued, because of the potential to increase the possibility of an accident previously evaluate [ sic), or to create the possibility of a new or different kind of accident caused by loss of structural integrity. If integrity is lost, the damaged fuel rods could cause a criticality accident.
(Hurricane loads were rejected as a basis for this contention in the Licensing Board's Memorandum and Order of September 16,1985)
- 1. Description of the Spent Fuel Racks
- 3. The Turkey Point Spent Fuel Pools have two storage regions.
These regions are shown by the overall pool layout in Figure 1.
- 4. The Region 1 storage racks, shown in Figure 2, consist of three major sections, which are the leveling pad assembly, the upper and lower grid assemblies, and individual storage cells made of stainless steel. The cells within a rack are interconnected by grid assemblies to form an integral structure. Each rack is provided with leveling pads connected to the lower grid assembly, which contact the floor of the spent fuel pool and are remotely adjustable from above to level the racks during installation. The racks are free-standing and are not anchored to the floor or braced to the pool walls.
Suppart pads for the new racks are located on the existing floor embedment plates. Some of the outer storage locations of the new racks overhang (extend beyond) the pads, as shown in Figure 3.
a i
e 1
N- __
i S i E -- !
s , +
n
- N*
s = l e#, #
I 12 E IW E l
. x2 3 x2 .
8 .,. C
-w'-'
l C
ly E lh k' *l I
{
[ lh h I
l .
. m 4's +
. 'e' I_S. T@ $ =1 eo i
h l
l hl'
--. m a
7
_2 -2 e
M3 M3 e 90 . G0
-y l , CASK AREA 3
l e es +<
+ wv n
+
l N u
V I c. a c.
I =E e =9 :
. ca n c_m 0
- e ,
a s l
-e ei l + !
- 1. : Si
=
. =- -= == r. ---ep l
+, ,
i s*
i -
SI i =
- [4
~ y ,
i Cr O =s
-s -s l' == l ee : sea <
3
= '
c2 o Rw
.=
o e+ + e e i
i
,i 1
l 1
Figure 1 i
Pool Layout i
=
10.6 0 =
f i
n d DE g i PLATE 9 f'f b ADJACENT CELL
- 8.75 SOU AR E *
,g ASSEMBLY
~U \ UPPER GRIO ASSEMBLY
.075 b --
[
-C I f PolSON M ATERI AL 165.61 [sh[usty I i
C
- L __
t p -9 - __
l
,J 'I GLOWER GRIO SIDE (i -
- ASSEMBLY PLATE I-h '
$usina vi: uni,Fl//////y.
' i J RT PL ATE
> LEVELING SCREW
\\\\NN\\\\\'i *f
. POOL LINER h.'
- h. ih EMBEOMENT-,
~' CONCRETE M(n;., {& PLATE Figure 2 Region 1 Storage Rack
FUEL ,
EMPTY STORAGE ASSEMBLY , LOCATIONS i M ; N ;;i I
l ,
l l2 l
i i f
I
!i ll il'
, 0 l
OUTER OVERHANGING I I I E I I I I D
STORAGE LOCATION
@_ M
\\\\\\\\\bsNs\\\\\\\\T\\\\\\\\T\\\\\\\ \\\\\\\\\
" SUPPORT POINT SUPPORT P01hT FIGURE 3 CROSS- SECTION OF FUEL RACK WITH FUEL LOADED IN OVERHAND LOCATION
- 5. The Region 2 storage racks, shown in Figure 4, consist of two major sections, which are the leveling pad base support assembly and stainless steel cells. The cells are assembled in a checkerboard pattern, producing a honeycomb type of structure. The cells are welded to a base support assembly !
and to one another to form an integral structure, without the use of grids of the type employed for the Region 1 racks. This design is also prcvided with the leveling pads connected to the base support assembly, which contact the pool floor /embedment plates, and which are remotely adjustable from above to level the rack during installation. The racks are free-standing and are not anchored to the floor or braced to the pool walls. Some of the Region 2 storage locations also overhang their support pads, as described for the Region 1 racks.
i II. Seismic Analysis of the Spent Fuel Racks A. Applicable codes. standards, and NRC criteria
- 6. The Nuclear Regulatory Commission (NRC) staff has identified criteria which it will accept for the performance of seismic analysis of spent fuel storage racks. These criteria are primarily contained in Section 9.1.2 of the Standard Review Plan (SRP) (Ref. 1), entitled " Spent Fuel Storage," and in the "0T Position for Review and Acceptance of Spent Fuel Storage and Handling Applications" (NRC Position Paper) (Ref. 2). These criteria are discussed in more detail below.
- 7. SRP Section 9.1.2 paragraph III.3.a. states that the spent fuel storage racks should be classified and designed to seismic Category I requirements (i.e., able to withstand the effects of the Safe Shutdown Earthquake (SSE) and remain functional).
- 8.Section IV of the NRC Position Paper identifies criteria for performing evaluations of the mechanical and structural integrity of spent fuel pools and racks.Section IV (2) of the NRC Position Paper identifies either of two industry codes.Section III of the ASME Code (Ref. 3) or the
- - _ ~ . _ _ - _ - - -
9.00 l
N CELL -
s, ASSEMBL
~r -
[
J t u A L 0 C
~
U POISON !
%,- /, - 4 I
..A#~ -
]
I
- I '
iss.s e q
- l [
/
l l
]
(
WRAPPER f r, i
a M I BASE PLATE l
l VHHHHA vmmsmnni vnwn SUPPORT PL AT E
( /! LEVELING SCPEW
_ NNNNNNNNN W __
EMBEDMENT CONCRETE _k E l
l Figure 4 Region 2 Storage Rack l
1 l -
1 <
AISC Code, as being acceptable for deriving allowable stresses in spent fuel racks.Section IV (3) of the NRC Position Paper identifies acceptable methods !
for calculating seismic loads,Section IV (5) states that SRP Section 3.8.4 provides acceptable procedures for modeling and analyzing the seismic responses of the spent fuel storage racks, and Section IV (6) identifies the structural acceptance criteria for the spent fuel storage racks. In particular,Section IV (6) of the NRC Position Paper states, among other things, that the design of a storage rack is acceptable if "the amplitudes of sliding motion are minimal, and impact between adjacent rack modules or between a rack module and the pool walls is prevented provided that the factors of safety against tilting are within the values permitted by Section 3.8.5.11.5 of the Standard Review Plan."
- 9. Finally,Section III of the NRC Position Paper identifies criteria for performing criticality analyses for spent fuel pools, including criticality analyses for postulated accident conditions. In particular, Section 111.1.2 of the NRC Position Paper states that "[r]ealistic initial conditions (e.a., the presence of soluble baron) may be assumed for the fuel pool and fuel assemblies" during postulated accident conditions, including the "effect of . . . earthquake on the deformation and relative position of the fuel racks."
- 10. The criteria discussed above are widely used in the nuclear industry for performing seismic analyses of spent fuel racks, and they are recognized as being conservative.
B. Description of the Analyses Performed for the Turkey Point Spent Fuel Storage Racks (1) Conformance with NRC Criteria
- 11. The new spent fuel storage racks for Turkey Point were designed in accordance with seismic Category I requirements. Thus, the design conforms with SRP Section 9.1.2.
i l
_4
- 12. As discussed in the Affidavit of William A. Boyd on Contention 10, the presence of the soluble boron in Turkey Point spent fuel pool water will maintain the stored spent fuel assemblies subcritical, even under postulated accident conditions involving changes in the mechanical or geometric configuration of the fuel assemblies or the storage racks resulting from earthquake. Thus, the design conforms with Section 111.1.2 of the NRC Position Paper.
- 13. lhe mechanical and structural analysis of the Turkey Point spent fuel storage racks was performed in accordance with Section IV of the NRC Position Paper. In particular, the structural analysis of the storage racks was based on the allowable stresses of the ASME Code, as recommended by Section IV (2) of the NRC Position Paper. Further details regarding the assumptions and methods used in the mechanical and structural analysis of the Turkey Point spent fuel storage racks are provided below.
(2) Details of the Mechanical and Structural Analysis
- 14. During a seismic event, a force is imposed upon a structure as a result of ground acceleration. The maximum seismic acceleration used in the analysis of the Turkey Point spent fuel storage racks was the design basis SSE acceleration for Turkey Point identified in Section 2.11 and Appendix 5A of the Updated Final Safety Analysis Report for Turkey Point Plant Units 3 and 4 (Updated FSAR). The design basis SSE acceleration for Turkey Point is 0.159 horizontal ground acceleration (where g is the acceleration of the Earth's gravity). Horizontal response spectra applicable for the spent fuel pool structure are applicable for both orthogonal horizontal directions, and, as specified by Appendix 5 of the FSAR, the vertical component of acceleration was taken as two-thirds of the horizontal ground acceleration.
- 15. 1he resistance to seismic displacement of a given structural element is determined by its mass, by the damping (energy loss) mechanisms inherent in the element, and by its method of connection to the other elements of the structure. Energy loss mechanisms due to damping are present in
. materials even when the applied stresses are within elastic limits for the material. For example, when a tuning fork is excited by striking it, the damping inherent in the material of the fork causes its vibrations to die away rather quickly. The appropriate damping value to employ for seismic analysis of a given structure depends upon the nature of the structure. Damping values are generally expressed as a percent of the critical damping value. Critical damping provides an amount of energy loss that prevents or quickly eliminates vibratory motion of the structure and its elements. Therefore the lower the damping value in percent, the more susceptible the structure is to vibratory motion in a seismic event. The seismic analyses of the Turkey Point spent fuel racks used a conservative structural damping value of 2%, which is consistent with the value specified in Appendix A of the Updated FSAR, and conservative compared to the value of 4% recomended by NRC Regulatory Guide 1.61 for welded steel f rame structures. Damping provided by the water between the fuel assemblies, storage cells, and rack assemblies was conservatively neglected. However, the mass of the water was taken into account in the analyses.
- 16. The close proximity of adjacent racks and storage cells, as well as their size relative to the size of the gaps between them, is such that the mass of the water in the gap provides large hydrodynamic forces which oppose rack or storage cell motions that are out of phase with the motions of its neighbors. Since the maximum deflections, loads, and stresses occur when adjacent storage cells and racks respond in phase, the Turkey Point racks were analyzed as if they were hydrodynamically coupled (move in Phase).
- 17. In addition to the seismic loads, the effects of acceleration-induced motion of the water (sloshing) were considered. However
! no sloshing loads were assumed to be imposed on the rack structures, which occupy roughly the lower one third of the depth of water in the pool, because the sloshing movement of the water would occur in the upper elevations of the pool above the top of the racks.
l l
i 1
- 18. Since the spent fuel racks for Turkey Point are free-standing, the racks would be held in place during a seismic event by gravity and by frictional forces between the rack and pool floor embedments. The frictional forces act to both excite the racks and to restrain them from sliding during a postulated seismic event. Tests by Rabinowicz (Reference 4.) have shown that static and dynamic friction coefficients in the range of 0.2 to 0.8 are appropriate for use in the analysis of the conditions that apply during a seismic event for the Turkey Point spent fuel racks. Thus, the range of friction coefficients (0.2 to 0.8) was used in the analyses. The use of a low dynamic coefficient of friction (0.2) produces maximum rack horizontal displacement or sliding, while use of a high static coefficient (0.8) produces maximum rack horizontal overturning force.
- 19. The dynamic response of the fuel rack assembly during a seismic event determines the loads and stresses on the structure. The dynamic response and internal stresses and loads were obtained from a seismic analysis which was performed in two phases. The first phase employed a two-dimensional model of an individual storage cell and the fuel assembly it contains. The results obtained from the two-dimensional model of the first phase analysis were then employed as input to the second phase analysis, which modeled the complete storage rack in three dimensions. Each of these phases is discussed in more detail below.
- 20. The models used in both phases of the analysis employed the finite element method. The finite element analysis method is widely used in the nuclear power and other industries. It is accepted by the NRC for the seismic analysis of a variety of nuclear power plant structures, from the containment building to the instrument panel racks in the main control room.
In applying the method, the structure to be analyzed is broken up into a finite number of sections or elements which interact at nodal points. A computer program is then used to evaluate the stresses produced on the elements by the seismic accelerations applied.
- 21. The first phase of the analysis employed a two-dimensional
, nonlinear model of an individual rack storage cell and its fuel assembly. Two dimensions, one vertical and one horizontal, are an appropriate choice for the first phase model because the fuel assembly and storage cell are structurally symmetric (identical) about either the x or y horizontal axis. The model is designated as nonlinear because it is used to calculate the fuel assembly to cell impact loads, the amount of levelling pad lift-off (if any), and the amount of rack sliding, parameters which are nonlinear functions of the gap between the wall of the storage cell and the fuel assembly, energy losses at the support locations of the fuel assembly within the cell, and the boundary conditions at the fuel rack support pads. To represent the seismic event, this model used a time history input of horizontal and vertical accelerations imposed on the rack by the spent fuel pool floor. The analytical model of the fuel assembly used in this model was verified by comparison to fuel assembly test data.
- 22. The second phase analyses used a three-dimensional linear model to calculate the response (loads and stresses) in a complete fuel rack assembly. It is important to understand that the two-dimensional nonlinear model accurately predicts the response of a single storage cell and its fuel assembly, appropriately taking into account the nonlinear effects described, while the three-dimensional linear model accurately predicts the internal stress distribution of a complete rack consisting of an array of such fuel storage cells and fuel assemblies. The values used in the loads and stress analyses which are subjected to nonlinear effects are taken from the maximum loaded sections of the three-dimensional linear model and are corrected to agree with the maximum loads and stresses predicted by the two-dimensional nonlinear model. Thus the second phase analyses employing the three-dimensional linear model conservatively account for the maximum values of the nonlinear loads and stresses.
- 23. Fuel rack seismic analyses were performed for two cases involving different assumptions regarding the loading pattern of fuel assemblies in the overhanging locations.
[
These cases were as follows:
o Case 1 provides the present basis for the spent fuel pool expansion license amendments. The analysis assumes that administrative controls are in place to prevent loading of fuel assemblies into the overhanging locations first, i.e., before assemblies are loaded into the other storage locatioas.
O Case 2 is an analysis performed at the request of Florida Power and Light Company, after NRC approval of the license amendments, to determine the potential effect of loading fuel assemblies into overhanging locations. The analysis assumes that fuel assemblies are loaded into the overhanging locations first, before the remaining locations are loaded, as shown in Figure 3.
C. Results of the Analysis Performed for the Turkey Point Spent Fuel Storage Racks (1) Results of the Analysis for Case 1
- 24. The analysis of the behavior of the racks for Case 1 considered full fuel loading (fuel assemblies in all storage locations) and various partial loading conditions. The results of this analysis are summarized as follows:
l o The fuel rack support points did not lift off of or lose contact with the floor of the spent fuel pool when subjected to the specified seismic ground accelerations. The factor of safety i against overturning was much greater than the 1.5 value specified by Section 3.8.5.11.5 of the Standard Review Plan.
o The maximum displacement of a fuel rack was calculated to be 0.256 inches. The gap between adjacent fuel racks is 1.11 inches and the gap between a fuel rack and the spent fuel pool l
walls is even larger. Thus, impact between adjacent rack modules or between a rack module and the pool wall is prevented and the leveling screws will not slide off the embedment plates.
o The fuel rack stresses are within the ACAE Code allowable limits, i.e., the minimum ratio of allowable stress divided by applied stress is greater than one. The minimum ratios of allowable stress divided by applied stress for the leveling pads, grid assemblies, and cell assemblies are 1.27, 1.15, and 1.11, respectively. It should be noted that allowable stresses do not represent the point of material failure, but are values which include conservatisms inherent in the ASME Code.
Thus, the results of the Case 1 analysis conform with the acceptance criteria in the NRC Position Paper and demonstrate that the spent fuel storage racks will be maintained in a safe configuration s' ring
/ postulated seismic events.
(2) Results of the Analysis for Case 2
- 25. In Case 2, the models were adjusted to account for the overhanging fuel mass shown in Figure 3, and the analysis was conducted for
, various partial fuel loading conditions with the appropriate seismic ground acceleration inputs. The results of this analysis are summarized as follows:
o Support point lift off occurred. The maximum lift off of 0.18 inches was produced by loading three outboard rows on the side of the rack with the overhanging storage locations. Lift off of support points is not uncommon for free-standirg racks under seismic conditions and the structural members of the racks are designed to accommodate the stresses produced by the lift off.
The lift off distance was used in an overturn stability calculation, and it was shown that the rack is stable and will not overturn and that the minimum factor of safety against overturn is 8 (which is substantially greater than the 1.5 factor of safety against overturning recommended by Section 3.8.5.11.5 of the SRP).
o The maximum displacement of a fuel rack is 0.709 inches. This is less than the gap between adjacent fuel racks and between the fuel racks and the spent fuel pool walls. Thus, impact between adjacent rack modules or between a rack module and the pool wall is prevented and the leveling screws will not slide off the embedment plates.
o Structural loads and stresses are enveloped by the condition of a fully loaded rack. Thus, the maximum stresses produced by the partially loaded racks in Case 2 are less than the maximum stresses calculated in Case 1. Therefore, the applied stresses in Case 2 are also within the ASME Code allowable stresses.
Thus, the results of the Case 2 analyses conform with the acceptance criteria in the NRC Position Paper and demonstrate that the spent fuel storage racks will be maintained in a safe configuration during postulated seismic events.
(3) Seismic Loads Experienced by the Fuel Assemblies
- 26. The analysis performed for Case 1 and Case 2 conditions show that the maximum acceleration imposed on a fuel assembly is 1.69 The ability of the fuel assemblies to safely accommodate a 1.69 acceleration is discussed in the Affidavit of Leonard T. Gesinski on Contention No. 5.
III. Sumnery and Conclusions
- 27. The Turkey Point spent fuel racks were analyzed employing methods approved by the NRC and by applicable industry standards. The results of those analyses show that the fuel racks are designed so that the stresses l
I produced under SSE seismic conditions are within the allowable limits of the
, ASMF. Code, the fuel racks will not contact each other or the spent fuel pool l walls during seismic events, and the lift off produced for Case 2 seismic conditions meets the stability requirements of the NRC Position Paper and will not result in overturning of the racks.
FURTHER AFFIANT SAYETH NOT Thd foregoing is true and correct to the best of my knowledge, information and belief.
Harry E. Flanders, Jr.
STATE OF FLORIDA )
COUNTY OF ESCAMBIA )
Subscribed and sworn to before me this day of ,
1986. My commission expires * .
M .
-r y /
References (1) U.S. Nuclear Regulatory Commission Standard Review Plan, NUREG-0800 (July 1981).
(2) "0T Position for Review and Acceptance of Spent Fuel Storage and Handling Applications," attached to letter dated April 14, 1978, from Brian K. Grimes (NRC) to All Power Reactor Licenses, as amended by letter dated January 18, 1979.
(3) American Society of Mechanical Engineers (ASME) Boiler and Pressure Vessel Code,Section III, Subsection NF, 1980 Edition, Summer 1982 Addenduh..
(4) Rabinowicz, E., " Friction Coefficients of Water-Lubricated Stainless Steels for a Spent Fuel Rack Facility," Report prepared for Boston Edison Company, Q 23.13, November 5,1976.
l l
l l
l
.~ . =. , _
I EXHIBIT A Professional Qualifications and Experience of Harry E. Flanders, Jr.
My name is Harry E. Flanders, Jr. and my business address is Westinghouse Eltctric Corporation, Scenic Highway, Pensacola, Florida, 32504. I am employed by Westinghouse Electric Corporation (" Westinghouse") as Manager of Commercial Products Engineering in the Nuclear Components Division.
I g aduated from Georgia Institute of Technology, with a Bachelors Degree in Mechanical Engineering in August 1963. While employed by the Lockheed-Georgia Company, I graduated from the Georgia Institute of Technology with a Master of Science degree in Mechanical Engineering in June 1970. I am currently a Registered Professional Engineer in the state of Georgia (Certificate Number
, 7061), and am a member of the American Society of Mechanical Engineers (ASME).
In April, 1974, I joined Westinghouse in the Nuclear Components Division of the Water Reactor Divisions as a Senior Engineer A. My duties in the Advanced Engineering Analysis Department included the seismic and structural analysis of pressurized water reactor internal components. These analyses included the determinatica of reactor internal component stresses to insure margin against the allowable stresses of the ASME Boiler and Pressure Vessel Code, and other i safety criteria. The result of various postulated accidents and normal operating conditions were analyzed to demonstrate that the stresses met the required limits. I was also responsible for preparing related documentation for submittal to regulatory authorities.
l Since that time I have had assignments of increasing responsibility in seismic and structural analysis and was promoted to the position of Principal Engineer in the Commercial Products Engineering (CPE) Department in February 1980. My duties in the CPE Department included the preparation and supervision of preparation of seismic and structural analyses of spent fuel storage racks and 4
i me ,...m----- - - .
-, .,y. .___i--,------ -- - -,r- - - -
, I associated spent fuel storage equipment. These analyses include the
, determination of fuel rack stresses to insure margin against the allowable stresses of the ASME B&PV Code and other safety criteria of the NRC. The results of various postulated accidents, seismic events, and normal operating conditions were analyzed to demonstrate that the stresses met the required limits. I was also responsible for preparing related documents for submittal to regulatory aethorities (NRC). I was responsible for the analyses of many Westinghouse supplied spent fuel racks including the Turkey Point spent fuel racks. In October 1985, I was promoted to my current position of Manager, Commercial Products Engineering, with responsibility for the efforts of several engineers and technicians in the design - analysis of spent fuel racks and associated spent fuel equipment.
Previous to the time I joined Westinghouse I was employed by Burlington Industries, Lockheed-Georgia Company, and Newport News Shipbuilding and Dry Dock Company.
From 1970 to 1974 I was employed by Burlington Industries, Corporate Research and Development Division, Greensboro, North Carolina, as a Research and Development Engineer. My responsibilities included the design, development, and dynamic analyses of high speed textile equipment.
From 1966 to 1970 I was employed by Lockheed-Georgia Company, Structures integrity Department, Marietta, Georgia, as a Senior Aircraft Structural Engineer. My responsibilities included stress analyses of landing gear and associated components of the C5-A aircraft.
From 1964 to 1966 I was employed by Newport News Shipbuilding and Dry Dock Company, Engineering Technical Department, Newport News, Virginia, as a Design Engineer. My responsibilities included stress analysis and dynamic shock analyses of main machinery and associated deck machinery of surface and subsurface ships.
i i
i I
f l _ __