This web page describes my Senior Project done for my
Mechanical Engineering Technology major at Central Washington
University.
Table of Contents 1. Abstract 2. Introduction e. Benchmark 3. Methods a. Approach b. Description c. Analysis 4. Test results 5. Assembly a. Description b. Parts 6. Budget 7. Schedule a. Sponsor c. References 9. Conclusion 10. Acknowledgements |
The goal of the Fuel Cell Educational Project is to demonstrate and promote the fuel cell technology. For this project Central Washington University was provided with Independence 1000 fuel cell power system that uses proton exchange membrane fuel cells. One way to promote the fuel cell technology is to demonstrate fuel cells' superior efficiency. This report discusses a way of measuring the electrical efficiency of fuel cell power system. Appropriate formulas were selected from ASME Performance Test Code 50-2002 that specifically addresses fuel cell power systems. The formula for the electrical efficiency of a fuel cell power system is similar to the basic efficiency formula of any system. The efficiency equals energy output divided by energy input. The energy output equals electric energy output of the fuel cell power system. The total energy input is a sum of the heating value of the fuel consumed and the energy input by the pressure of the fuel. Based on efficiency formulas the variables that needed to be monitored were: fuel pressure, mass flow of fuel, voltage output, and current output. Final system includes a mass flow meter, a clamp current probe, and a pressure gauge. Mass flow meter and a clamp current probe provide analog outputs that can be logged by a data acquisition device. The data acquisition device can monitor voltage directly. The test was performed and a maximum electrical efficiency of 45.8 percent was calculated. The combination of instrumentation tolerances yielded a total percent uncertainty of efficiency measurement of 7.70 percent. The installed instrumentation allows for future use of this setup in engineering labs or demonstrations. 2. Introduction a. Motivation/Scope The Bonneville Power Administration has developed a partnership with Avista Labs and the Combined Heat and Power Consortium to install a polymer electrolyte membrane (PEM) fuel cell system as Central Washington University. The installation was based on a plan developed by a team of students. This team of students was selected based on the individual interest in fuel cell technology and other alternative energy systems. The motivation for the Fuel Cell Educational Project was to demonstrate and promote the fuel cell technology. One way to promote the fuel cell technology is to demonstrate fuel cells' superior efficiency. The required data is collected through fuel cell's internal and external instrumentation. This data is used to demonstrate some technical aspects to the general public. In addition, engineering students can use this data to analyze the fuel cell performance. The scope of this project is to measure the electrical efficiency of the provided Independence 1000 system of fuel cells. The efficiency will be tested under varying loads. However, no internal modifications to the fuel cell will be made. Fuel starvation effects can be observed. However, the manufacturer does not advise intentional fuel restriction. The Independence 1000 will be tested as a complete system without consideration of the various modes of operation and internal circuitry setups. The peripheral fans and other components that draw power will be considered part of the system and will not be excluded from overall efficiency calculations. b. Function Statement Function of Fuel Cell Power System Efficiency Calculation section of the Fuel Cell Educational Project was restricted to:
c. Design Requirements Manufacturer's specifications and safety concerns dictated the following design requirements:
d. Success Criteria The system was evaluated based on the function statements and design requirements:
e. Benchmarks Chevy engine test setup located in the Power Technology Building is an example energy conversion device efficiency measurement setup. That lab setup was designed to be used by engineering and technology students within a 2-hour lab period. Large gauges on the instrumentation panel allow groups of people to read the values during the lab. The components of the engine efficiency measurement setup don't pose great safety hazards. Engine setup does not occupy excessive floor space but is not compact. No uncertainty percentage was calculated for engine lab. 3. Methods
For this electrical efficiency calculation of the fuel cell power system American Society of Mechanical Engineers (ASME) Performance Test Code (PTC) 50-2002 was referenced. This code is titled "FUEL CELL POWER SYSTEMS PERFORMANCE." The testing was not done per PTC 50-2002 as this code was designed for a lengthier test of a fuel cell power system at one operating condition. However, electrical efficiency, total power input, and total power output formulas used in this report are from PTC 50-2002. Appropriate formulas were selected from ASME Performance Test Code 50-2002 that specifically addresses fuel cell power systems. Independence 1000 is a system using sixty individual fuel cells. This system includes control electronics and fans requiring power as well as a power converter that wastes some power. Therefore, most formulas dealing specifically with a fuel cell are not applicable in this case. ASME Performance Test Code 50-2002 can be used to measure electrical efficiency of any type of fuel cell power system. b. Description For analysis of the electrical efficiency of the fuel cell power system a test boundary was determined (Figure 1). All energy crossing this boundary is ether energy input or energy output. Only values underlined on figure 1 are measured. The other outputs carry waste energy of a low quality and are therefore not considered in the electrical efficiency calculation. Other inputs are at ambient conditions and do not carry significant amounts of quality energy.
Figure 1 For this system designed to test electrical efficiency of the fuel cell power system following values were measured at test boundary:
The mass flow was measured using Omega FMA 1726ST mass flow meter (Figure
2). This flow meter is capable of measuring mass flow up to 30 SLM, satisfying
the requirement for measuring normal operation flow and capturing some of the
excessive flow during purging event. FMA
1726ST uses a power supply and outputs a proportional voltage that can be
converted to Standard Liters per Minute (SLM). Fuel supply
pressure is monitored using standard dial pressure gauge (Figure 3). The
pressure is maintained constant and is not continuously monitored. Electrical output voltage is measured directly from the output terminals and can be input to the data acquisition device directly or through a voltage divider. Due to possible large output current a clamp current probe had to be used (Figure 4). In this setup an Omega HHM72 Clamp Current Probe was used because it provides proportional voltage output that can be automatically recorded by the data acquisition device. For data logging Fluke Hydra unit was used to log the data at one scan per second. While calculating electrical efficiency the internal workings of the fuel cell power system are irrelevant. The basic concept of calculating the electrical efficiency of a fuel cell power system is the same as for any power generating system. The basic formula used for percent electrical efficiency of the fuel cell power system was:
The net_electrical_energy_output is found by integrating the output power graph over the duration time of the experiment. The total_energy_input consists of several components. The components applicable to this setup were:
Where:
The total caloric energy of fuel is found by logging the mass flow of the fuel. The mass flow plot is then integrated over the duration time of the experiment to yield total mass of fuel consumed. The total mass of fuel consumed is then converted to moles and multiplied by the caloric energy content of one mole of fuel. The caloric content (heating value) of hydrogen is 266.1 kJ/mol. For the experiments performed, the pressure of the input air is taken as standard atmospheric. The total energy due to pressure of the fuel input is calculated using the provided formula. If any adjustments to the fuel cell input pressure are made, the different energy due to pressure value has to be calculated. The 25-psig value was used because it is the lowest input pressure recommended by the manufacturer. Independence 1000 power system does not use as typical stack of fuel cells and therefore requires smaller input pressures. According to Avista Labs' engineer Ken Hydzik, Independence 1000 has an internal pressure regulator that further reduces the fuel pressure. Therefore, any changes to the fuel supply pressure do not significantly affect the energy output of the fuel cell. However, including higher fuel supply pressures in calculations erroneously reduces the efficiency of this fuel cell power system. For more details on calculating electrical efficiency see Sample Calculations Appendix. d. Testing and Evaluation Independence 1000 fuel cell power system requires warm up time. Manufacturer states in the operating manual that twenty-minute warm up time is required for the fuel cell to reach full operating output. Even after the full output power is reached the maximum power and efficiency keeps changing as the unit warms up. Therefore, the unit was run for half an hour to warm up before the test was performed. This experimental setup was designed to test the fuel cell at varying loads. The corresponding electrical efficiency was calculated for every load. The fuel cell power system was loaded with a constant resistance and data was logged when the fuel cell achieved steady operation. The data was logged for one minute at each load giving 60 scan points. No measurable difference was observed when the test duration was extended to three minutes. Test duration of one minute allows for averaging out of periodic purging events. One minute is significantly shorter than one hour test duration recommended by ASME PTC 50-2002, but this is an educational experiment designed to be performed during a shorter period of time over a range of loads. Longer tests may be considered if affects of fuel cell warm up have to be averaged out, but this is outside the scope of this project. 4. Test Results
The necessity of integrating the instantaneous flow values to acquire total fuel consumed during the experiment is seen from the graph of instantaneous flow vs. time in figure 5. Independence 1000 has a purge event every twenty seconds by design. During the purge event the valve to the purge line is open. This creates a massive loss of hydrogen that is expelled by the pressure of the fuel. The large values of flow during the purge event and the small flow values right after the purge event create erroneous efficiency values if efficiency is calculated using instantaneous reading. Furthermore, if the average value of the flow is taken, relatively large values of the purge flows skew the average flow up. Therefore,
Figure 2 Similarly to the instantaneous flow graph, instantaneous power graph (Figure 6) displays large variations in the power output. During this experiment batteries that smooth out the power output by the fuel cell power system are disconnected and only the direct output of the fuel cell power system is measured. Therefore, integrating the power output over the time of the experiment gives a better energy out value than any one reading during an experiment. Therefore,
Figure 3 The highest electrical efficiency of Independence 1000 Fuel Cell Power System was measured to be 45.8 percent at 553 W output. The electrical efficiency curve based on all the points in the performed experiment peaks at 620 Watts and forty five percent electrical efficiency (Figure 4). The manufacturer reported the efficiency of this fuel cell power system to be forty-three percent at eighty percent of the rated power (800 Watts). The value of the curve-fit function at 800 Watt is forty-two percent. The one-percent difference may account for including or excluding different inputs in calculation of the electrical efficiency.
Figure 4 As seen from figure 7, the maximum efficiency is not reached at rated output. The manufacturer's engineer Ken Hydzik stated that it is an intentional design. The efficiency of the fuel cell is weighted against the initial cost of the fuel cell. According to him, this design was optimized to give the best fuel cost without excessive initial cost of the fuel cell power system. The percent uncertainty of efficiency is 7.70 %. This equals a tolerance of +/- 3.15 % efficiency. Power values have a percent uncertainty of 3.085 %. This equals a tolerance of +/- 22.79 Watts. Large uncertainty in the electrical efficiency of the fuel cell power system is due to reading manual pressure gauge with percent uncertainty of 2.5 %. Clamp current probe, with uncertainty of 1.5 %, gives the largest error to the power values as well as adding error to the efficiency. The resulting curve-fit function is well within all but one error areas of the experimentation points. The experiment can be easily recreated using existing equipment and this report as a guide. The efficiency calculation instrumentation system consists of purchased parts. Most of the installation is not permanently attached and can be easily re-assembled if removed. Figure 5 shows the diagram for the placement of the instruments.
Figure 5 The only permanently installed component of the efficiency instrumentation is the mass flow meter. It is installed in series onto a fuel supply line. Mass flow meter requires a provided power supply and outputs a voltage proportional to the mass flow of the hydrogen. The output of the mass flow meter is six standard liters per minute for each volt of the output. Output voltage is measured by directly connecting the data acquisition device in parallel with the load. Fluke Hydra can read direct voltage up to 150 Volts, which is above the maximum of fifty-four Volt output of the fuel cell. If another data acquisition device is used with voltage measuring capability below sixty Volts, a voltage divider circuit can be constructed on the provided breadboard to provide a proportional output. A data acquisition device can read this proportional output. Standard voltage divider formula can be used to calculate the divider factor of the voltage divider circuit. Following is a list of the main components of the efficiency instrumentation.
Figure 6 All of the above instruments provide a voltage output only. Additionally, a voltage is measured directly or through a voltage divider. A data acquisition device is preferable. Data acquisition device is needed to record a data over a period of time to average out the bleed events and other variations. The data in this report was acquired using Fluke Hydra data logger. Outputs can also be read using a voltmeter function of a multi-meter for verification or manual data acquisition. Fuel cell power system efficiency calculation was a part of larger Fuel Cell Educational Project. The mass flow meter (Fig. 11) was installed by a professional plumber during the construction of the fuel supply line. Therefore, no clear labor and material costs can be attributed specifically to the fuel cell power system efficiency part of the project. However, specific instrumentation used for the fuel cell power system electrical efficiency calculation can be identified. Figure 10 lists the components of the system, supplier and cost of each component.
Figure 7
Figure 8 Schedule displayed in figure 12 addresses the entire Fuel Cell Educational Project. Fuel cell power system electrical efficiency calculation part of the project figured throughout the project, but no events are specific to this part of the overall project.
Figure 12 This project was a part of Fuel Cell Educational Project sponsored by Dr. Walt Kaminski through a grant by Bonneville Power Administration and Northwest Energy Cooperative. The author of this report is a senior Mechanical Engineering Technology student at Central Washington University (See Resume Appendix). Additional input was provided by Charles Harmon and Erick Nyhus, both MET seniors. The electrical specialist on the student team was John Goes, a senior in Electronic Engineering Technology program at Central Washington University. Several people provided their expertise for Fuel Cell Educational Project in general and Fuel Cell Power System Efficiency Calculation part of the project in particular. Dr. Walt Kaminski was the sponsor and advisor for this project. Dr. Kaminski trained the members of the Fuel Cell Educational Project Student Team in fuel cell technology. Central Washington University's Mechanical Engineer, Pat Nahan, provided the student team with his expertise on designing flammable gas lines and safety systems. Fuel cell manufacturer's engineer Ken Hydzik provided the student team with details about the design and function of the tested Independence 1000 fuel cell power system. Dr. Craig Johnson and Carlos Oncina gave the student team input to ensure that this project was with enough technical merit. Following references were used for Fuel Cell Educational Project: American Society of Mechanical Engineers. Fuel Cell Power Systems Performance (ASME PTC 50-2002). New York, 2002. Avallone and Baumeister. Marks' Standard Handbook for Mechanical Engineers. Tenth Edition. New York: McGraw-Hill, 1996. Cengel, Yunus A. Introduction to Thermodynamics and Heat Transfer. Boston: McGraw-Hill, 1997. Christensen, Peter. "Avista Labs Tour." Avista Labs Headquarters. Spokane, WA. 30 November 2003. Kaminski, Walt. "Fuel Cell Basics Lecture." Hogue Technology 215. 16 October 2003. Weston, Kenneth C. Energy Conversion. St. Paul: West Publishing, 1992. Zumdahl, Steven S. Chemistry. Fourth Edition. Boston: Houghton Mifflin, 1997. The function requirements and design criteria of Fuel Cell Power System Efficiency Measurement part of the Fuel Cell Educational Project were met per evaluation criteria. The instrumentation was setup and test was preformed in one hour. The peak electrical efficiency of Independence 1000 fuel cell power system was measured to be 45.8 percent plus or minus 3.15 percent efficiency. The measured electrical efficiency was within tolerance of manufacturer's stated efficiency. The overall percent uncertainty about the value of efficiency was 7.70. This is a large uncertainty that can be reduces with more expensive instrumentation. However, the materials for this project ran over $ 2400 which is a significant expenditure for educational experiment. Fuel Cell Educational Project Student Team would like to thank following individuals for their assistance: Dr. Walt Kaminski – mentor and sponsor. Mira Vowels – BPA Fuel Cell Program Manager. Pat Nahan – CWU mechanical engineer. Ken Hydzik – Avista Labs engineer. Dr. Craig Johnson – Senior Project instructor. Carlos Oncina – Senior Project instructor.
To see this report in Acrobat format go to Fuel Cell Report. |
Hit Counter