Problem Definition and Requirements Document:
The Large Scale Vehicle 2 (LSV2) CUTTHROAT is a fully autonomous electric submarine. The main propulsion unit of the LSV2, a 3000HP brushless DC motor, is powered by 1,680 valve regulated lead acid batteries each operating at 2V. Two motor drive cabinets serve to change the DC battery voltage and current into AC voltage and current required by the motor. Please refer to figure A1 on page 1 in the appendix for an electrical layout of the submarine.
Being autonomous, any interruption in required operation will enable an automatic resurface protocol and interrupt the collection of test run data. Under hard acceleration during high speed testing, the current required by the motor and thus supplied by the batteries causes significant voltage drop across the internal resistance of the batteries. As the batteries are used, their effectiveness to store energy is reduced. These two factors cause the battery voltage to drop below the low-voltage threshold of 742 volts set by the submarine’s protection equipment. If this minimum voltage is detected the safety system will interrupt the system which in turn will cause the submarine to resurface. Presently, this voltage drop occurs at the end of the fourth high-speed test run when the batteries have been depleted and can no longer supply the necessary voltage for operation.
The researchers at the Naval Surface Warfare Center, Acoustics Research Detachment have requested that 10 volts from an external power source supplement the existing submarine voltage during its fourth test run. Ultimately no changes can be made to the existing power supply or protection scheme due to the proprietary nature of much of the equipment. Because the equipment is proprietary it is difficult to alter the equipment or find information about it. Mitigating this voltage sag to obtain one more high-speed run (for a total of four runs) at 260 RPM requires supplying the system with energy for 25 seconds at the end of the run. For a standard 260 RPM run, an average of 1300 amps runs through the system. This requires that external power source be capable of operating at 13 kilowatts of power and supplying the system with 325 kilojoules of energy over the 25 seconds. This amount of energy must be supplied while keeping the physical size of the design small enough to fit into a 25 inch diameter hatch, and sized appropriately to fit into the submarine’s extra storage space. Any solution designed must take into account that acoustics testing is being performed, and the solution must be nearly silent. Any power integration into the system must not disrupt the current power system or cause damage to any part of the submarine. The solution must also be reliable and easily maintainable. Any solutions requiring separate charging must charge in 8 hours or less.
The task for team Cutthroat Energy has been stated as to research possible solutions to mitigate this voltage sag. The team will research current technologies, review previous case studies in related fields, and use design techniques to create possible solutions. Of the possible solutions, Cutthroat Energy will choose two for further study. Based on engineering analysis, Cutthroat Energy will recommend a final solution Verification of proof of concept will be accomplished by building a scaled laboratory model roughly representing the electrical characteristics of Cutthroat during a 260 RPM run. The final deliverables for the project consist of a portfolio containing a Problem Description and Requirements Document, a Report of Potential Design Solutions, a Recommended System Selection Report, and finally a Report on Final Configuration.
Report of Potential Design Solutions
In order to weigh the effectiveness of each solution, a design matrix was created using criteria related to the needs and the specifications of the project.
The various criteria that were used to evaluate the solutions, in order from greatest importance to least importance, are as follows:
· Acoustical Noise – The solution needs to be quiet. Cutthroat is one of the quietest submarines in the world, and obviously creating a solution that would create excessive noise would be detrimental. There are methods for quieting the solution to some extent but overall silence is extremely important.
· Cooling – The solution needs to either have minimal heat production or the ability to be externally cooled in order to maintain the internal submarine temperature within the operating range of the existing electronics.
· Device Response/Reliability – The solution needs to be able to consistently supply the energy when it’s needed, in a quick fashion so as to correct the voltage sag in an appropriate amount of time.
· Maintainability – The solution should be of a quality that requires little maintenance from the Cutthroat crew. Other than the necessary preparations prior to use there should be minimal work associated with the solution once it’s installed.
· Complexity – To maintain the quality of the project produced by the team, the solution should be relatively simple.
· Power Integration – The solution needs to be able to integrate with the power infrastructure of the submarine without causing any detrimental effects to existing equipment or submarine operation.
· Control Integration – Another vital requirement is the ability to insert the solution into the system without negatively affecting the existing control architecture.
· Size – The size of the solution is important to an extent. It is important that the solution has the ability to fit through the 25-inch diameter hatch of the submarine. The total solution size can be larger than this as long as it can be split up and fit through the hatch piece by piece and then reassembled in the space available.
· Component Cost – The solutions total component cost, while important, is not of the greatest importance. A solution’s cost will pay for itself very quickly from the money saved on each test run.
· Developmental Cost – An important consideration is the amount of man-hours it will take to properly implement the solution.
· Electrical Noise – This factor has proved to be very unimportant but was still something that was considered in weighting the solutions.
The following comprises all the solutions that have been considered to meet the project. At this point all but two solutions have been ruled out as impractical, unfeasible, or overtly difficult.
One of the first ideas explored was to add energy to the system in the form of a flywheel and a generator allowing the conversion of rotational energy into electrical energy. Initial calculations determined that either too large a flywheel was needed if reasonable rotational speed were desired, or impractical speeds were necessary if the size of the flywheel was to be something that could realistically fit in the submarine. For this reason this solution was nixed on account of the project’ssize specification.
Another solution that was explored was to possibly add some energy into the system on the AC side after the energy has passed through the drive cabinets. While this solution did provide the benefit of reducing the current draw, which would in turn reduce the voltage sag, it introduced numerous complex issues that would need to be addressed. The final ax to this solution came from the project sponsor himself when we learned that we would not have access to the proprietary protected components on the AC side of the system.
After ruling these ideas out the design team focused on the DC side of the system. Multiple solutions were suggested as to this regard. The ideas came in two basic forms, adding a solution in series with the system and adding a solution in parallel with the system, as shown in the figures below.
Figure 1. Series Solution
Figure 2. Parallel Solution
For the parallel solution, the first idea was to add capacitors in parallel with the existing battery supply to help correct the voltage sag by supplying charge to equalize the voltage. The benefit to using capacitors is they have a quick energy response that is essential for our solution. Doing some basic calculations, the necessary capacitance was calculated for the amount of energy that would be needed at the solution’s operating voltage. However it would take approximately 300 ultracapacitors in series to realize the required voltage, which would decrease the capacitance. It would then take multiple strings of 300 capacitors to bring the capacitance back up the level that was needed. In addition, it was realized that a very small portion of the energy stored in the capacitors would actually be used. Therefore, this solution was scrapped because of its large size and component cost and its inefficiency.
The next parallel solution considered was to integrate an energy source, most likely capacitors due to their response, into the existing system using a boost converter. This solution was superior to the parallel capacitor solution because it wasn’t necessary to operate the capacitors at the operating voltage of the batteries as the previous solution required. This solution would alleviate the voltage sag by reducing the current required from the batteries and reduce the resistive voltage drop at the batteries. However, figure A3 in the appendix on page 3 is a plot of the drop in voltage at the drive cabinets as a result of the increase in current. The negative slope for this graph gives the battery resistance, which is roughly 36 mW. If you solve for the current required for 10 volts of correction, the amount of voltage sag due to the batteries, given the 36 mW of battery resistance, it comes to 500 amps. At 750 volts, the solution would need to supply 375 kW of power which is roughly 30 times the specified power of the solution. As a result this solution was thrown out due to the power requirement of the solution.
Looking at these two parallel solutions it was decided that the current and energy requirements needed to realize a parallel solution was not an efficient path for this solution. The focus was then turned to a series solution that would supplement the system with voltage instead of current. Listed below are the series solutions that have been considered.
Series switched capacitors is one series design that was explored and is still being considered. This design will switch an ultracapacitor bank in series with the main battery bank when the battery voltage begins failing by utilizing a switching network of contactors and snubber circuitry. This design is the simplest series design being considered due to its relatively simple topology. This design will also allow for system degradation in being able to correct larger voltage sags with the installation of additional stages of capacitors that can be switched into the system. The developmental costs on this design will be relatively low because it requires only minimal changes to the existing system. In addition, maintenance should be relatively low due to the long lifespan of the ultracapacitors. Research is showing that ultracapacitors are good for up to ten years. An additional charging unit will also have to be added to the system to charge up the capacitor bank. The charging scheme that is used to charge the main battery bank will provide complications in charging to such low voltages so an additional low voltage charger will need to be used. Another consideration is the component cost of the design. Upon initial calculation it is estimated that roughly 80 ultracapacitors are needed which amounts to roughly $8000. There will be additional costs with the contactors and snubber circuitry that are still being looked into.
The other series solutions that have been considered consist of using a DC – DC power converter that would regulate the voltage added to the system. The converter types that have been explored include the buck converter, the forward converter and the flyback converter. Different energy sources for the converters that have been considered include the auxiliary supply or a battery bank.
The first converter setup considered was a two stage buck converter. The first stage will be connected to the auxiliary battery and will charge a capacitor bank at the input of the second stage. This design was chosen because the auxiliary battery is not capable of handling the 1300 A that will be circulating through the system when the solution will be operating. As a result the second stage was added to isolate the auxiliary battery from the main system. Isolation of the two stages would be performed through a contactor and snubber circuit that will disconnect the first stage connected to the auxiliary from the main system after the capacitor bank has been charged up. There will also have to be a contactor and snubber circuit at the end of the second stage to disconnect the whole unit from the main system so that the components of the buck converter are not consuming power over the course of the entire underway. With this type of a solution configuration it will be possible to control the discharge of the capacitor bank and also supply additional energy to the system if need by recharging the capacitor bank during idle times of the underway. While the converter is switched in the system most of the components of the buck converter will have to be rated to handle 1300A of current. There is also a diode utilized in the buck converter design that will consume significant power that will need to be designed so that the heat generated will be dispersed efficiently and not increase the ambient temperature of the submarine. Because of the complexity of all these items, along with the fact that the capacitor bank at the second buck converter would be roughly the same size as the series capacitor solution, this solution was ruled out.
Similar to the previous solution another design proposed was to use batteries as the source for a buck converter. However, this option was ruled out due to the fact that the converter would operate at low voltage and high current. This would mean significant resistive losses in the converter components and switching, so this idea was also scrapped.