A Minimum Cost Design Approach of an ROV for Underwater Inspection Tarek Elsayed *, Amr Hassan *, Yasser Ahmed and Mohamed Darwish * Assistant Professor, department of Mechanical and Marine Engineering Arab Academy for Science and Technology and Maritime Transport AASTMT Alexandria, Egypt tarek@eng.aast.edu Abstract- A minimum cost design approach of an underwater remotely operated vehicle ROV is described in this paper. The Sea Explorer I was designed for general purpose underwater inspection and/or research work with a design depth of 20 meters. The approach is a hand on approach and focuses on minimizing cost. Use of costly commercial watertight motors and thrusters for propulsion was avoided. Conventional low cost DC motors were used instead and were made watertight via manufactured underwater housings, seals and gaskets. An underwater camera system was implemented using a low cost security camera such as the one found in convenience stores. The buoyancy system uses a pneumatic system that fills in a float with air thus providing dynamic buoyancy control from the surface. Stresses on the motor housing resulting from pressure and thermal heating were investigated using the finite element method FEM and the ANSYS software. Underwater hydrodynamic performance of the ROV including resistance, motion and dynamic pressure effects were investigated using the finite volume method and the FLUENT software 1. INTRODUCTION Remotely Operated Vehicles ROVs have been used for the last thirty years all over the world by different organizations to gain access to underwater locations not easily accessible by divers or by other means. Lately ROVs have also been performing tasks that the divers can perform such as hull cleaning due to economical and safety reasons. The improvements in performance and reliability of commercial equipments have increased over the years for the last thirty years. These improvements however come at an increased cost. This paper describes the design, testing and development of an ROV, the Sea Explorer I for underwater inspection, with the objective being to minimize cost. 2. SEA EXPLORER I - DESIGN DETAILS A pipe frame was constructed from Aluminum with a 20 millimeter outer diameter and a 2 millimeter wall thickness. The overall size of the frame is approximately 85cmx55cmx25cm. The two sides of the frame were connected by four transverse ribs. Aluminum was chosen as the frame material mainly because of its lightweight and corrosion resistance characteristics. The advantage of an open frame design lies in its simplicity. Being a standard box frame means that it can be build simply, it provides many points of attachment for ROV components and it allows for future expansion. The electronics box is located in the bottom of the ROV and sits on two of the transverse ribs. A compressed air ballast tank is located on the right side of the ROV and is supported in place by two holders. The main and bow (lateral) thrusters which operate at a voltage of 12V DC and 4 amperes, are also supported in place by holders. The camera housing and light were fitted at the front end of the Aluminum frame. Foam was placed on the top of the frame to provide the necessary buoyancy for the Sea Explorer I as shown in Figure 1. Table 1 provides a summary of the technical details of the Sea Explorer I Dimensions LxBxH: Weight Propulsion ` Mounting frame 85 x 55 x 25 cm 25.5 kg brushless permanent magnet DC electric motors, 12 V, 4 A Aluminum and buoyancy foam Control System - ROV Control 2 control boxes, buttons and control switches - ROV status TV monitor or separate computer display. Power System: Transformer, 15V, 50 Hz, 30 W Standard Equipment: - Camera 200 mw/6 V, 0.2 Lux - Camera swiveling 180, Pan and tilt motor motions, 12V, 0.5 A - Underwater light 120 W,12 V,10 A Umbilical: diameter 15 mm length 20 meters Table 1: Summary of technical details of Sea Explorer I 1
Figure 1: Picture of SEA EXPLORER I Figure 2 shows a solid model of the Sea Explorer I showing overall dimensions, general arrangement and components layout. Figure 2: Sea Explorer I solid model showing general arrangement and components layout 3. STABILITY, BALLASTING AND BUOYANCY CONTROL In order to achieve positive underwater stability, the weight of the ROV with its load and ballast must be nearly equal to the weight of water it will displace when submerged i.e. the buoyancy force [6]. Also the center of gravity of the submerged ROV must be nearly in the same longitudinal position as the center of buoyancy for positive stability. The locations of the centers of gravity and buoyancy of the Sea Explorer I were calculated. Lead weights were added to bring the center of gravity and center of buoyancy to almost the same longitudinal position. Various options were considered for buoyancy control. These include ballast tanks that can be filled with sea water or compressed air [7], thrusters for vertical movement and/or pneumatic systems. Ballast tanks require a carefully designed tank made for expelling water with air or via ballast pumps and watertight motors and therefore additional power consumption and cost. Therefore this option was discarded. Ballasting with air was selected as the best minimal cost alternative. Buoyancy will be provided by inflating a float. Inflating the float would result in increased underwater volume and therefore increased buoyancy. Deflating the float would result in the exact opposite i.e. decreased underwater volume and therefore negative buoyancy. Since this design did not include vertical thrusters or ballast tanks that can be filled with water for vertical motion, the ROV weight was adjusted to be slightly higher than its buoyancy such that it would descend slowly by the force resulting from the difference of its own weight and the buoyancy force. At this condition the float is partially inflated as well to allow control of descent speed if desired. Upon descent to the desired depth inflating the float slowly would result in neutral buoyancy condition and the ROV can then start cruising horizontally. The float is filled with the pneumatic system via activation of a solenoid valve which is electrically controlled from the surface. The float used is a rubber float and was also wrapped with elastic rubber bands to facilitate deflation of the float and air flow through the outlet line upon activation of the solenoid valve for ascending or control of descent speed. The basic components of this system are the compressed air tank used to supply the system and the manifold and solenoid valve assembly used to distribute the compressed air to the system. The tank is a standard compressed air cylinder which supplies air at approximately 10 Bar feeding the pneumatic system. The manifold has two solenoid valves controlling its input and output lines. One solenoid valve is used to open and close the air supply to the buoyancy control float. The other solenoid valve is used to open or close the air outlet line from the buoyancy float. Control of the components of the pneumatic system is provided by the electronics and power supply controlled from the surface. Figure 3 shows a diagram of pneumatic system for buoyancy control. 2
Figure 4: Camera, swiveling motor, camera housing and watertight arrangements 5. ELCTRICAL CONTROL SYSTEM Figure 3: Diagram of pneumatic system for buoyancy control 4. CAMERA SYSTEM AND LIGHTING The camera used is a small color model, 628x582 effective pixels, 0.2 Lux, lowest illuminance, 380 lines resolution, with night-vision capability. The camera operates at 6V DC and consumes 200 mw. The camera was hooked to a swiveling motor to allow pan/tilt motions of the camera. The camera and swivel motor are housed in a container made from Teflon with the top dome made from plexiglass, sealed together with an O-ring and epoxy. The dome provides the viewing port for the camera. The camera housing is 25 cm length x 12cm diameter. When deciding what to use for lighting, the following factors were taken into consideration: costeffectiveness, ease of installation and superior light output. It was decided to use a sealed beam automotive headlight. This lamp was selected because it already has a watertight seal; it is readily available and provides superior light output. A large color television screen is used to display images send to the surface by the camera. Figure 4 shows the camera, swiveling motor, camera housing and watertight arrangements. The control system controls the different functions of the ROV, from controlling the propulsion system to switching on the light/video camera. The control system adopted is a fairly simple relay control system. The ROV is controlled by eight switches, four for forward, reverse and turning, two switches to allow the vehicle to dive and surface and three for the camera and lighting. This is a rather simple design. The tether is a 20 meters marine cable that has eight conductors, (one ground, two for thruster power, two for pneumatic control, two for the camera system and the last to supply power for the light). Figure 5 shows the power supply transformer, the two control boxes, control switches and color television screen is used to display the camera images. Figure 5: Power supply transformer, control boxes, control switches and color TV screen for camera images 3
6. DC ELECTRCIC PROPULSION MOTORS The underwater housing of the electric motors was one of the most important parts of our ROV design. The housing will have to protect the electrical motor from water, humidity and pressure effects. The underwater housing also provides a support for fixing the propeller shaft and propulsion thrusters. The housing would have to withstand the water pressure at the depth of operation of the ROV. The motors will also heat up as a result of continuous underwater operation. A two-compartment Aluminum design was chosen for the motor housing as shown in figure 6. In the event of a water leak, the leak will be confined into the first compartment and still be blocked from reaching the second compartment where the motor is housed. Finite element method FEM [1] was used to predict the thermal stresses acting on the motor housing in addition to stresses resulting from hydrostatic pressure. An inner temperature of 50 C and a surrounding water temperature of 15 C were used for the analysis. Motor housing was made of Aluminum for which the modulus of elasticity E = 73 GPa and coefficient of thermal expansion a=23x10-6 / C. Figure 7 shows the Von Mises stresses acting on the motor housing due to hydrostatic pressure and the Von Mises stresses on motor housing due to combined thermal stresses and pressure effects. At a water depth of 10 meters the Von Mises stresses due to hydrostatic pressure were in the order of 35 N/cm 2 whereas the Von Mises stresses due to combined thermal and hydrostatic stresses were in the order of 10,000 N/cm 2. Figure 6: Cross sectional view in motor housing showing compartments and watertight arrangements Figure 7: Von Mises stresses on motor housing a) Hydrostatic pressure alone. b) Hydrostatic pressure and thermal effects. 7. UNDERWATER HYDRODYNAMIC PERFORMANCE Underwater hydrodynamic performance of the ROV was investigated using computation fluid dynamic techniques CFD and the software FLUENT [3]. The purpose was to estimate the underwater frictional resistance of the ROV and to investigate the dynamic performance and motion of the ROV as a fluid structure interaction FSI problem. The finite volume approach was adopted to predict the viscous underwater resistance of the ROV at various cruising speeds for both horizontal (cruising) and vertical (ascent/descent) motions. In this approach, the governing equations and the fluid domain are subdivided into a finite number of cells and these equations are changed into algebraic form via a discretization process. The convective terms are discretized using first order upwind scheme, and the pressure is interpolated using a standard interpolation scheme. A central difference scheme is then utilized for the diffusion terms. For the pressure-velocity coupling the SIMPLE (semi-implicit methods for pressure-linked equation) is utilized [2, 4, 5]. The algebraic equations obtained from the discretization 4
process are solved iteratively. Solution convergence was monitored by dimensionless residual sum for all variables across the computational points. The minimum residual sum for convergence was set to 1 10-6. The ROV unit and its computational domains were meshed using tetrahedral cells for calculating the viscous resistance. The number of cells used is 192087 in the computational domains of the ROV. Figure 8 shows the coordinate system used, the Sea Explorer I meshed as a moving frame, the fluid domain mesh and wall boundary conditions for underwater horizontal motion. Table 2 summarizes the estimated frictional resistance in Newton of the Sea Explorer I at various cruising speeds for both horizontal and vertical motions. Figure 9 shows the dynamic pressure (Pascal) distribution on the Sea Explorer I at cruising speeds of 0.25 m/sec and 0.55 m/sec respectively. Speed (m/sec) Horizontal Resistance (N) Vertical Resistance (N) 0.14 1.72 0.26 0.25 5.47 0.836 0.5 21.85 3.363 1 87.28 13.47 Table 2: Estimated resistance of ROV at various cruising speeds Maximum dynamic pressure acting on the ROV at a speed of 0.25 m/sec was 55 Pascal, whereas the maximum dynamic pressure at a speed of 0.55 m/sec was 220 Pascal Figure 8: Coordinate system, ROV mesh, fluid domain mesh and wall boundary conditions for underwater horizontal motion 5
5.50e+01 5.23e+01 4.95e+01 4.68e+01 4.40e+01 4.13e+01 3.85e+01 3.58e+01 3.30e+01 3.03e+01 2.75e+01 2.48e+01 2.20e+01 1.93e+01 1.65e+01 1.38e+01 1.10e+01 8.26e+00 5.51e+00 2.76e+00 1.24e-02 Contours of Dynamic Pressure (pascal) 2.20e+02 2.09e+02 1.98e+02 1.87e+02 1.76e+02 1.65e+02 1.54e+02 1.43e+02 1.32e+02 1.21e+02 1.10e+02 9.92e+01 8.82e+01 7.72e+01 6.62e+01 5.51e+01 4.41e+01 3.31e+01 2.21e+01 1.11e+01 5.01e-02 Underwater Intervention Conference 2006 Tampa, FL, January 24-26, 2006 Z Y X Contours of Dynamic Pressure (pascal) Z Y X Figure 9: a) Contours of dynamic pressure (Pascal) cruising speed = 0.25 m/sec b) Contours of dynamic pressure (Pascal) cruising speed = 0.55 m/sec 8. CONCLUSIONS This paper outlined a design approach for an ROV that minimizes cost. Conventional low cost DC motors were used instead of available costly commercial watertight motors and thrusters. An underwater camera system was implemented using a low cost security camera hooked to a swiveling motor. A buoyancy system that uses compressed air to fill in a float was developed. Such system is electrically controlled from the surface and provides real time buoyancy control. Stresses acting on the motor housing were investigated using the finite element method and were found to be within acceptable limits for the material and the design depth. Underwater hydrodynamic performance including resistance, motion and dynamic pressure effects were investigated using the finite volume method. System and subsystem tests were done to evaluate the performance of the Sea Explorer I underwater. The integrated pneumatic system performed successfully and so did the thruster FLUENT 6.2 FLUENT 6.2 housings, camera and lighting. During testing the Sea Explorer I attained the design horizontal cruising speed of 0.25 m/sec as planned. Despite the considerable modeling and meshing effort involved, use of CFD techniques and the FLUENT software has been shown as valuable tool for the evaluation of underwater dynamic performance of the Sea Explorer I and for the estimation of underwater resistance. Estimation of the ROV s frictional underwater resistance at the design speed allows estimation of powering requirements and therefore optimum sizing and selection of propulsion thrusters at the design stage. This analysis can also be used for analyzing the ROVs underwater performance in the presence of currents and/or wave effects Acknowledgments We would like to thank the following people for their support, help and effort in the design and implementation of our ROV Sea Explorer I : Hisham Mokhtar for his help with the calculations of center of gravity, center of buoyancy and solid modeling of the ROV and Mohamed Nagy for his help with the design and implementation of the ROVs control system, switches and associated electronics. REFERENCES [1] ANSYS, User s Manual for Revision 7, Swanson Analysis Systems, Inc., Houston, PA, 1996. [2] B. E. Launder, D. B. Spalding, Lectures in Mathematical Models of Turbulence, Academic Press Inc., 1972. [3] Fluent Europe, 2003, FLUENT 6.1: User s Guide, Fluent Inc. [4] H. K. Versteeg, W. Malalasekera, An Introduction to Computational Fluid Dynamics, The Finite Volume Method, Longman Group Ltd., 1995. [5] Joel Ferziger, Milovan Peric, Computational Methods for Fluid Dynamics, Springer Verlag Berlin Heidelberg, 1999. [6] Lewis, Edward Principles of Naval Architecture, Volume I, Stability and Strength, published by the Society of Naval Architects and Marine Engineers SNAME, 1998. [7] Wasserman, K.S, Dynamic Buoyancy Control of an ROV using a Variable Ballast Tank, Proceedings of Ocean 2003, San Diego, California. 6