[Part 1 of this article discusses the major components of a typical remotely operated vehicle system along with everyday underwater tasks ROVs perform.]

3.2 PRIMARY SUBSYSTEMS
The ability to sense the environment, either visually or through other means, and perform work at the desired location, is the mission of the ROV. The subsystems necessary for this task are discussed in the following sections.

3.2.1 Lighting
This explanation of lighting comes courtesy of Ronan Gray of Deep Sea Power & Light. The need for underwater lighting becomes apparent below a few feet from the surface. Ambient visible light is quickly attenuated by a combination of scattering and absorption, thus requiring artificial lighting to view items underwater with any degree of clarity. We see things in color because objects reflect wavelengths of light that represent the colors of the visible spectrum. Artificial lighting is therefore necessary near the illuminated object to view it in true color with intensity. Underwater lamps provide this capability.

Lamps convert electrical energy into light. The main types or classes of artificial lamps/light sources used in underwater lighting are incandescent, fluorescent, high-intensity gas discharge, and light-emitting diode (LED) " each with its strengths and weaknesses. All types of light are meant to augment the natural light present in the environment. Table 3.1 shows the major types of artificial lighting systems, as well as their respective characteristics.

• Incandescent - The incandescent lamp was the first artificial light bulb invented. Electricity is passed through a thin metal element, heating it to a high enough temperature to glow (thus producing light). It is inefficient as a lighting source with approximately 90 percent of the energy wasted as heat. Halogen bulbs are an improved incandescent. Light energy output is about 15 percent of energy input, instead of 10 percent, allowing them to produce about 50 percent more light from the same amount of electrical power. However, the halogen bulb capsule is under high pressure instead of a vacuum or low-pressure noble gas (as with regular incandescent lamps) and, although much smaller, its hotter filament temperature causes the bulbs to have a very hot surface. This means that such glass bulbs can explode if broken, or if operated with residue (such as fingerprints) on them. The risk of burns or fire is also greater than other bulbs, leading to their prohibition in some underwater applications. Halogen capsules can be put inside regular bulbs or dichroic reflectors, either for aesthetics or for safety. Good halogen bulbs produce a sunshine-like white light, while regular incandescent bulbs produce a light between sunlight and candlelight.

• Fluorescent - A fluorescent lamp is a type of lamp that uses electricity to excite mercury vapor in argon or neon gas, producing short-wave ultraviolet light. This light then causes a phosphor coating on the light tube to fluoresce, producing visible light. Fluorescent bulbs are about 40 percent efficient, meaning that for the same amount of light they use one-fourth the power and produce one-sixth the heat of a regular incandescent. Fluorescents typically do not have the luminescent output capacity per unit volume of other types of lighting, making them (in many underwater applications) a poor choice for underwater artificial light sources.

• High-intensity discharge - High-intensity discharge (HID) lamps include the following types of electrical lights: Mercury vapor, metal halide, high-pressure sodium and, less common, xenon short-arc lamps. The light-producing element of these lamp types is a well-stabilized arc discharge contained within a refractory envelope (arc tube) with wall loading (power intensity per unit area of the arc tube) in excess of 3 W/cm² (19.4 W/in²). Compared to fluorescent and incandescent lamps, HID lamps produce a large quantity of light in a small package, making them well suited for mounting on underwater vehicles. The most common HID lights used in underwater work are of the metal halide type.

• LED - A light-emitting diode (LED) is a semiconductor device that emits incoherent narrow-spectrum light when electrically biased in the forward direction. This effect is a form of electroluminescence. The color of the emitted light depends on the chemical composition of the semiconducting material used, and can be near-ultraviolet, visible, or infrared. LED technology is useful for underwater lighting because of its low power consumption, low heat generation, instantaneous on/off control, continuity of color throughout the life of the diode, extremely long life, and relatively low cost of manufacture. LED lighting is a rapidly evolving technology " look for more usage of LEDs in the underwater lighting field soon.

Most observation-class ROV systems use the smaller lighting systems, including halogen and metal halide HID lighting.

The efficiency metric for lamps is efficacy, which is defined as light output in lumens divided by energy input in watts, with units of lumens per watt (LPW). Lamp efficacy refers to the lamp's rated light output per nominal lamp watts. System efficacy refers to the lamp's rated light output per system watts, which include the ballast losses (if applicable). Efficacy may be expressed as 'initial efficacy', using rated initial lumens at the beginning of lamp life. Alternatively, efficacy may be expressed as 'mean efficacy', using rated mean lumens over the lamp's lifetime; Mean lumens are usually given at 40 percent of the lamp's rated life and indicate the degree of lumen depreciation as the lamp ages.

An efficient reflector will not only maximize the light output that falls on the target, but will also direct heat forward and away from the lamp. The shape of the reflector will be the main determinant in how the light output is directed. Most are parabolic, but ellipsoidal reflectors are often used in underwater applications to focus light through a small opening in a pressure housing. The surface condition of a reflector will determine how the light output will be dispersed and diffused. The majority of reflectors are made of pure, highly polished aluminum that will reflect light back at roughly the same angle to the normal at which it was incident. By adding dimples or peens to the surface, the reflected light is dispersed or spread out. When a plain white surface is used, the reflected light is diffused in all directions.

Cameras, sensors, manipulator and tool pack
3.2.2 Cameras
Currently, most small ROV systems use inexpensive charge-coupled device (CCD) cameras as their main viewing device. These camera systems are mounted on small circuit boards and produce a video signal transmitted in a format sent up the tether to the video capture device on the surface. The actual protocol of the signal emanating from the camera and control box (after transmission through the tether) is manufacturer-specific, but usually falls under either composite or RF (radio frequency) video. The protocol of the video signal will determine the receiving adapter on the viewing device. Refer to the manufacturer's instructions for the specific protocol and/or adapter for the system.

Production of an ROV camera assembly can be accomplished on any electronics bench with rudimentary equipment. A simple chip camera system sold through any surveillance camera manufacturing company is mounted on a block along with a motor and gearing system for panning and/or tilting, plus focusing (if manual focusing is desired). Once the camera is mounted (Figure 3.10), simple wiring and switching will accomplish both control of the individual camera as well as switching between various camera systems aboard the vehicle.

Figure 3.10 Typical arrangement - CCD camera system mounted to ring with tilting mechanism.

Various regions of the world use different video formats. In the USA, as well as a few other countries, NTSC (National Television Standards Committee) is the standard format, while most of Europe, Africa, and Asia use PAL (Phased Array Lines) format. The SECAM format used predominately in France has been declining in recent years and will in all likelihood eventually be eliminated.

Camera technology is evolving rapidly. The High Definition format is quickly being adapted to ROVs as it trickles down to the smaller vehicles due to decreasing size and lower cost structure. Digital still camera technology is also being adapted for high-resolution image capture of underwater items. Look for major improvements in size, functionality, and cost in the near future.

3.2.3 Sensors
As stated earlier, most industrial ROV systems provide the capability to transmit data from the submersible to the surface. This allows the ROV system to deliver a suite of instruments to the work site, powered by the vehicle, with data transmitted through the tether to the surface. Any combination of sensor and instrument (heading/gyro/depth, etc.) is available as payload to the modern ROV system, assuming proper data protocol transmissions and power delivery are available. Figure 3.11 shows (left to right) a sonar, hydrology sensor, and manipulator configured for attachment to an ROV.

Figure 3.11 Various sensors for mounting to ROV.

Major issues regarding the integration of sensors involves the data transmission protocol and the method through which this transmission takes place. The ROV manufacturer must provide a throughput within the ROV system to allow for sensor integration. Some examples of common sensors packages placed aboard ROV systems in industrial applications include:

• Radiation sensors
• CTD (conductivity/temperature/depth) sensors
• Pressure-sensitive depth transducer
• Magnetic flux gate compass module
• Slaved or rate gyro for heading stabilization
• Ultrasonic thickness gauges for measuring metal thickness and quality
• Imaging sonar
• Acoustic positioning
• Digital cameras
• Multi-parameter environmental sensors (e.g. turbidity, chlorophyll, DO, pH, and ORP sensors, which are discussed in Chapter 6).

3.2.4 Manipulator and Tool Pack
Most professional ROV systems allow for a simple A/B power source to provide the locomotion needs of intervention tooling packs. On the larger hydraulic ROV systems, a simple independent A and B connector is provided to power turning or cutting equipment for subsea work. On observation-class systems, a simple 12 or 24 VDC source is provided to run manipulators or other small tools needed for specific jobs. Power can also be redirected from main thruster power for more demanding mechanical work.

A basic single-function manipulator package, common on many small ROV systems, consists of a 24 VDC electric motor running a worm gear to open and close small grabber arms for light intervention duties (Figures 3.12 and 3.13).

Figure 3.12 Typical setup for small ROV manipulator.

Figure 3.13 Small manipulator CAD drawing (courtesy of Inuktun Services Ltd.).

A common problem with small manipulators without a limit switch at the end of the travel of the worm gear is that the ROV operator will continue activating the manipulator. When the worm gear reaches the end of travel, the motor can become stalled at the full close or full open point of the jaw. If the jaw is left in this position (i.e. with the worm gear stalled at the end of its travel and torqued against the end of the worm gear stop) and allowed to sit for a length of time, oxidation can seize the screw/teeth, preventing it from moving the worm gear away from the stop. If this happens, the jaw must be either squeezed/pulled to take the pressure off the worm gear, and the motor then activated, or the mechanism must be disassembled to unfreeze the worm gear.

Electrical considerations
3.3 ELECTRICAL CONSIDERATIONS
The following sections discuss specific issues and relationships regarding the tether, power, data, and the connectors that bring it all together.

3.3.1 The Tether
The tether and the umbilical are essentially the same item. The cable linking the surface to the cage or tether management system (TMS) is termed the 'Umbilical', while the cable from the TMS to the submersible is termed the 'Tether'. Any combination of electrical junctions is possible in order to achieve power transmission and/or data relay. For instance, AC power may be transmitted from the surface through the umbilical to the cage, where it is then changed to DC to power the submersible's thrusters and electronics.

Further, video and data may be transmitted from the surface to the cage via fiber-optics (to lessen the noise due to AC power transmission), then changed to copper for the portion from the cage to the submersible, thus eliminating the AC noise problem. Figure 3.14 is an example of the neutrally buoyant tether for the Outland 1000 observation-class ROV system (courtesy of Outland Technology).

Figure 3.14 Cross-section of neutrally buoyant tether.

The umbilical/tether can be made up of a number of components:

• Conductors for transmitting power from the surface to the submersible
• Control throughput for telemetry (conducting metal or fiber optic)
• Video/data transmission throughput (conducting metal or fiber optic)
• Strength member allowing for higher tensile strength of cable structure
• Lighter-than-water filler that helps the cable assembly achieve neutral buoyancy
• Protective outer jacket for tear and abrasion resistance.

Most observation-class ROV systems use direct current power for transmission along the tether to power the submersible. The tether length is critical in determining the power available for use at the vehicle. The power available to the vehicle must be sufficient to operate all of the electrical equipment on the submersible. The electrical resistance of the conductors within the tether, especially over longer lengths, could reduce the vehicle power sufficiently during high-load conditions to effect operations.

The maximum tether length for a given power requirement is a function of the size of the conductor, the voltage, and the resistance. For example, using a water pipe analogy, there is only a certain amount of water that will flow through a pipeline at a given pressure. The longer the pipe, the higher the internal resistance to movement of the water. As long as the water requirements at the receiving end do not exceed the delivery capacity of the pipe (at a given pressure), the system delivery of water will be adequate. If there were to be a sudden increase in the water requirement (a fire requiring water, everyone watering their lawn simultaneously, etc.), the only way to get adequate water to the delivery end would be to increase the pressure or to decrease the resistance (i.e. shorten the pipe length or increase the diameter) of the pipe. The same holds true in electrical terms between tether length, total power required, voltage, and resistance (Figure 3.15).

Figure 3.15 Diagram depicting the power budget for power through the tether.

Ohm's law deals with the relationship between voltage and current in an ideal conductor. This relationship states that the potential difference (voltage) across an ideal conductor is proportional to the current through it. So, the voltage (V, or universally as E) is equal to the current (I) times the resistance (R). This is stated mathematically as V = IR. Further, power (measured in watts) delivered to a circuit is a product of the voltage and the current.

Thus, based on Ohm's law, the voltage drop over a length of cable can be calculated by using the formula, V = IR, where V is the voltage drop, I is the current draw of the vehicle in amps, and R is the total electrical resistance of the power conductor within the tether in ohms. The current draw of a particular component (light, thruster, camera, etc.) can be calculated if the wattage and voltage of the component are known. The current draw is equal to the component wattage divided by the component voltage (or amps = watts/volts).

For example, referring to the table of electrical resistances for various wire gauges (Table 3.2), the voltage required to operate a 24-volt/300-watt light at 24 volts over 250 feet of 16-gauge cable can be calculated as follows: The current draw, I, of a 24-volt/300-watt lamp operating at 24 volts is 300 watts/24 volts = 12.5 amps. The resistance of 16-gauge wire is approximately 4 ohms/1000 feet (Table 3.2).

Since the total path of the circuit is from the power supply to the light and back to the power supply, the total resistance of the cable is twice the length of the cable times the linear resistance, or for this example, R = (2 × 250 ft) × (4 ohms/1000 ft) = 2.0 ohms. Since V = IR, the voltage drop, V, is equal to 12.5 amps × 2.0 ohms = 25 volts. This means that 25 volts is lost due to resistance, so the power supply will need to provide at least 49 volts (the 24 volts necessary to operate the light plus the additional voltage loss of 25 volts) to power this 24-volt/300-watt light over a 250-foot cable.

Power source, AC vs. DC, and data throughput/transmission
3.3.2 Power Source
The ROV system is made up of a series of compromises. The type of power delivered to the submersible is a trade-off of cost, safety, and needed performance. Direct current (DC) allows for lower cost and weight of tether components; Since inductance noise is minimal, it allows for less shielding of conductors in close proximity to the power line. Alternating current (AC) allows longer transmission distances than that available to DC while using smaller conductors.

Most operators of ROV systems specify a power source independent of the vessel of opportunity. The reason for this separation of supply is that the time the vessel is in most need of its power is normally the time when the submersible is most in need of its power. Submersible systems attempting to escape a hazardous bottom condition have been known to lose power at critical moments while the vessel is making power-draining repositioning thrusts on its engines. This can cause entanglement of the vehicle. With a separate power source, submersible maneuvering power is separated from the power needs of the vessel.

With the advent of the lightweight micro-generators for use with small ROVs, the portability of the ROV system is significantly enhanced. Some operators prefer usage of the battery/inverter combination for systems requiring AC power. Also, some smaller systems use only DC as their power source. Either method should have the power source capable of supplying uninterrupted power to the system at its maximum sustained current draw for the length of the anticipated operation.

3.3.3 AC Versus DC Considerations
Electrical power transmission techniques are an important factor in ROV system design due to their effect upon component weights, electrical noise propagation and safety considerations. The DC method of power transmission predominates the observation-class ROV systems due to the lack of need for shielding of components, weight considerations for portability, and the expense of power transmission devices.

On larger ROV systems, AC power is used for the umbilical due to its long power transmission distances, which are not seen by the smaller systems. AC power in close proximity to video conductors could cause electrical noise to propagate due to EMF (electromotive force) conditions. The shielding necessary to lower this EMF effect could cause the otherwise neutrally buoyant tether to become negatively buoyant, resulting in vehicle control problems. And the heavy and bulky transformers are a nuisance during travel to a job site or as checked baggage aboard aircraft.

Larger work-class systems normally use AC power transmission from the surface down the umbilical to the cage (the umbilical normally uses fiber-optic transmission, lowering the EMF noise through the video) since the umbilical does not require neutral buoyancy. At the cage, the AC power is then rectified to DC to run the submersible through the neutrally buoyant tether that runs between the cage and vehicle.

3.3.4 Data Throughput
The wider the data pipeline from the submersible to the surface, the greater the ability for the vehicle to deliver to the operator the necessary job-specific data as well as sensory feedback needed to properly control the vehicle. With the cost of broadband fiber-optic transmission equipment dropping into the range of most small ROV equipment manufacturers' budgets, more applications and sensors should soon become available to the ROV marketplace.

The ROV is simply a delivery platform for transporting the sensor package to the work location. The only limitation to full sensor feedback to the operator will remain one of lack of funding and imagination. The Human-Robot Interface (the intuitive interaction protocol between the human operator and the robotic vehicle) is still in its infancy; However, sensors are still outstretching the human's ability to interpret this data fast enough to react to the feedback in a timely fashion. This subject is probably the most exciting field of development for the future of robotics and will be of considerable interest to the next generation of ROV pilots.

3.3.5 Data Transmission and Protocol
Most small ROV manufacturers simply provide a spare twisted pair of conductors for hard-wire communication of sensors from the vehicle to the surface. The strength of this method is that the sensor vendor does not need engineering support from the ROV manufacturer in order to design these sensor interfaces. The weakness is that unless the sensor manufacturers collude to form a set of transmission standards, each sensor connected to the system 'hogs' the data transmission line to the detriment of other sensors needed for the task.

A specific example of this problem is the need for concurrent use of an imaging sonar system and an acoustic positioning system. Unless the manufacturers of each sensor package agree upon a transmission protocol to share the single data line, only one instrument may use the line at a time. A few manufacturers have adapted industry standard protocols for such transmissions, including TCP/IP, RS-485, and other standard protocols. The most common protocol, RS-232, while useful and seemingly ubiquitous in the computer industry, is distance limited through conductors, thus causing transmission problems over longer lengths of tether.

The move toward open source PC-based sensor data processing has led to the production of data protocol converters for use in ROV sensor interpretation. Most small ROV sensor manufacturers transmit data with the RS-485 protocol, requiring a converter at the surface to both isolate the signal and to convert it to USB (or RS-232) protocol for easy processing with a standard laptop computer. Standards for these protocol converters are slow in evolving (due to the size of the customer base). Thus, the ROV system integrator must become familiar with the wiring and pin arrangement for these converters to assure data transmission from the sensor, through the vehicle and tether to the software at the surface, is achieved.

Underwater connectors
3.3.6 Underwater Connectors
The underwater connector is said to be the bane of the ROV business. Salt water is highly conductive, causing any exposed electrical component submerged in salt water to short to ground. The result is the 'Ubiquitous ground fault'. The purpose of an underwater connector is to conduct needed electrical currents through the connector while at the same time squeezing the water path and sealing the connection to lower the risk of electrical leakage to ground.

The underwater connector is lined with synthetic rubber that blocks the ingress path of water while allowing a positive electrical connection. Connectors sometimes experience cathodic delamination, causing rubber peeling and flaking from the connector walls. Connector maintenance (Figure 3.16) should include:

• Use small amounts of silicone grease to lubricate the connector, thus allowing easier slide on and off. Using too much grease, a widespread problem, can interfere with sealing.
• Always pull the connector by its body instead of its tail (cable), since the wire splice is located in the connection. Pulling on the tail could part the solder joint and ruin the electrical continuity within the connector.
• Keep the connectors as clean as possible through regularly scheduled maintenance tasks that include cleaning the contacts and lubricating the rubber lining.
• Spray the connector body with silicone spray to keep the housing from drying out, which could result in flaking and rubber degradation.

Figure 3.16 Underwater connectors must be serviced to assure proper electrical connectivity.

Even when the contacts are right and the connector has good design features, the connector must be appropriate for the intended use and environment. The connector materials must be able to withstand the environmental conditions without degradation. For example, extended exposure to sunlight (ultraviolet energy) will cause damage to neoprene, and many steels will corrode in sea water. Check that the connector will fully withstand the environment.

The connector must not adversely affect the application. For example, all ferrous materials (steel, etc.) should be avoided in cases where the connector's magnetic signature might affect the system. In extreme cases, even the nickel used under gold plating could have an effect and should be reviewed.

The physical size of the connector, its weight, ease of use (and appropriateness for the application), durability, submergence (depth) rating, field repairability, etc. should all be assessed. The use of oil-filled cables or connectors should be considered.

Ease of installation and use is especially important, so realistically appraise the technical ability of those personnel who will actually install or use the equipment. If they are inexperienced, a more 'user-friendly' connector may be a better choice. And, if possible, train operators in the basics of proper connector use: Use only a little lubricant, avoid over-tightening, note acceptable cable bending radii, provide grounding wires for steel connectors in aluminum bulkheads, etc.

Splicing and repairing underwater cables and connectors, while quite simple, requires some basic precautions to avoid water ingress into the electrical spaces, thus grounding the connection. Examples are shown in Figures 3.17-3.22.

Figure 3.17 Internal workings of male/female underwater connectors.

Figure 3.18 Slicing is conducted by peeling back the conductors of the tether.

Figure 3.19 Pin out the conductors on the tether to correspond to the connector.

Figure 3.20 The electrical connection is made through standard bench techniques.

Figure 3.21 The connected conductors are laid into a potting mold then sealed with potting compound.

Figure 3.22 Finished connector along with plastic guard.

Coming up in Part 3: The control systems of the ROV.

Printed with permission from Butterworth Heinemann, a division of Elsevier. Copyright 2007. "The ROV Manual: A User Guide for Observation Class Remotely Operated Vehicles" by Robert D Christ and Robert L. Wernli, Sr. For more information about this title and other similar books, please visit www.elsevierdirect.com.

Related links:
Components of an ROV system - Part 1: Mechanical and electromechanical systems
Under the Hood--Extreme: Monitoring Gas Extraction in the Norwegian Sea
Open-Source Robotics and Process Control: Sensor, Actuator and Control Circuit Examples - Part 1 | Part 2: Stepper Motor Controller | Part 3: Speed-controlled DC motor | Part 4: Two-axis attitude sensor | Part 5: RS-422-Compatible Indicator Panel
Building flexible architectures for configurable UAV systems