SEARCH · Engineering Papers
Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.
Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.
This work was performed to assess actual relationships of phosphoric acid fuel cell voltage responses obtained from subscale cells with air-cooled and full size cells in stacks as a function of operating parameters. Key relationships were verified and found to be in agreement with the Nerst equation. An empirical equation was developed to predict cell voltages as a function of the measureable parameters.
Two 25-cell, 13 inch x 23 inch (4kW) stacks were started up to evaluate the reliability of component and stack technology developed through the end of 1983. Both stacks started up well and are running satisfactorily on hydrogen-air after 1900 hours and 800 hours, respectively. A synthetic-reformat mixing station is nearing completion, and both stacks will be operated on reformate fuel. A stack-protection control system was placed in operation for Stack No. 2, and a similar set-up is in preparation for Stack No. 1. This system serves to change operating conditions or shut the stack down to avoid deleterious effects from nonstack-related upsets. The capability will greatly improve changes of obtaining meaningful long-term test data.
The cell stack can be operated as a fuel cell stack or as an electrolysis cell stack. The stack consists of a series of alternate fuel cell subassemblies with intervening electrolysis cell subassemblies, and interspersed cooling plates. The water produced and consumed in the two modes of operation migrates between adjacent cell subassemblies. The component plates are annular with a central hydrogen plenum and integral internal oxygen manifolds. No fluid pumps are needed to operate the stack in either mode.
The technology development for materials, cells, and reformers for on site integrated energy systems is described. The carbonization of 25 cu cm, 350 cu cm, and 1200 cu cm cell test hardware was accomplished and the performance of 25 cu cm fuel cells was improved. Electrochemical corrosion rates of graphite/phenolic resin composites in phosphoric acid were determined. Three cells (5 in by 15 in stacks) were operated for longer than 7000 hours. Specified endurance stacks completed a total of 4000 hours. An electrically heated reformer was tested and is to provide hydrogen for 23 cell fuel cell stack.
This paper describes performance characterization tests of a 64 cell hydrogen oxygen PEM fuel cell stack at NASA Glenn Research Center in February 2003. The tests were part of NASA's ongoing effort to develop a regenerative fuel cell for aerospace energy storage applications. The purpose of the tests was to verify capability of this stack to operate within a regenerative fuel cell, and to compare performance with earlier test results recorded by the stack developer. Test results obtained include polarization performance of the stack at 50 and 100 psig system pressure, and a steady state endurance run at 100 psig. A maximum power output of 4.8 kWe was observed during polarization runs, and the stack sustained a steady power output of 4.0 kWe during the endurance run. The performance data obtained from these tests compare reasonably close to the stack developer's results although some additional spread between best to worst performing cell voltages was observed. Throughout the tests, the stack demonstrated the consistent performance and repeatable behavior required for regenerative fuel cell operation.
It is shown that fuel cell devices are particularly attractive for the high-efficiency, high-reliability space hardware necessary to support upcoming space missions. These low-temperature hydrogen-oxygen systems necessarily operate with two-phase water. In either PEMFCs (proton exchange membrane fuel cells) or AFCs (alkaline fuel cells), engineering design must be critically focused on both stack temperature control and on the relative humidity control necessary to sustain appropriate conductivity within the ionic conductor. Water must also be removed promptly from the hardware. Present designs for AFC space hardware accomplish thermal management through two coupled cooling loops, both driven by a heat transfer fluid, and involve a recirculation fan to remove water and heat from the stack. There appears to be a certain advantage in using product water for these purposes within PEM hardware, because in that case a single fluid can serve both to control stack temperature, operating simultaneously as a heat transfer medium and through evaporation, and to provide the gas-phase moisture levels necessary to set the ionic conductor at appropriate performance levels. Moreover, the humidification cooling process automatically follows current loads. This design may remove the necessity for recirculation gas fans, thus demonstrating the long-term reliability essential for future space power hardware.
Reciprocating piston piezohydraulic pumps were developed originally under the Smart Wing Phase II program (Lynch) and later under the CHAP program (CSA, Kinetic Ceramics). These pumps focused on 10 cm scale stack actuators operating below resonance and, more recently, at resonance. A survey of commercially available linear actuators indicates that obtaining power density and specific power greater than electromagnetic linear actuators requires driving the stacks at frequencies greater than 1 KHz at high fields. In the case of 10 cm scale actuators the power supply signal conditioning becomes large and heavy and the soft PZT stack actuators generate a lot of heat due to internal losses. Reciprocation frequencies can be increased and material losses significantly decreased through use of millimeter scale single crystal stack actuators. We are presently targeting the design of pumps that utilize stacks at the 1-10 mm length scale and run at reciprocating frequencies of 20kHz or greater. This offers significant advantages over current approaches including eliminating audible noise and significantly increasing the power density and specific power of the system (including electronics). The pump currently under development will comprise an LC resonant drive of a resonant crystal and head mass operating against a resonant fluid column. Each of these resonant systems are high Q and together should produce a single high Q second order system.
As both NASA and the aeronautics industry recognize the need for higher fuel efficiency and lower carbon emissions in both commercial airline and private aviation applications, development of all-electric or hybrid electric aircraft have garnered renewed interest in the aviation community. For the particular example of the hybrid-electric option, the solid oxide fuel cell (SOFC) is an attractive option for the power source, due to its potential to utilize aviation fuels thereby having minimal impact to aviation infrastructure. SOFC stack performance depends upon many factors, one of the most important is the way the oxidant and fuel gases are delivered to the fuel cells. System modeling of various aircraft configurations for FUELEAP (Fostering Ultra-Efficient, Low-Emitting Aviation Power) point to the need to operate SOFC stacks at high current densities. This creates challenges in the thermal profile of the stacks with potential to create large thermal gradients and hot spots. This study investigates two types of commercial solid oxide fuel cell stacks, the cross flow and co-flow gas designs, both convectively cooled with cathode air. High fuel utilization factors were also employed under varying electrical loads expected from the demands of flight. In addition, performance, range of operation and endurance were investigated under conditions of high current loads and thermal cycling. Evaluations include the study of gas kinetic using electrochemical spectroscopy. Testing took place at the facilities of NASA Glenn using a commercial test system (FuelCon AG, Magdeburg Germany). These studies are crucial to the Glenn Research Center's ability to conduct research, evaluation and development of the next-generation SOFC based stacks for cutting-edge energy technologies for aerospace applications. This study supports NASA's Convergent Aeronautics Solutions' (CAS) FUELEAP project.
Development of a PEM fuel cell powerplant (PFCP) for use in the Space Shuttle offers multiple benefits to NASA. A PFCP with a longer design life than is delivered currently from the alkaline fuel will reduce Space Shuttle Program maintenance costs. A PFCP compatible with zero-gravity can be adapted for future NASA transportation and exploration programs. Also, the commercial PEM fuel cell industry ensures a competitive environment for select powerplant components. Conceptual designs of the Space Shuttle PFCP have resulted in identification of key technical areas requiring resolution prior to development of a flight system. Those technical areas include characterization of PEM fuel cell stack durability under operational conditions and water management both within and external to the stack. Resolution of the above issues is necessary to adequately control development, production, and maintenance costs for a PFCP.
Ultrasonic/sonic impacting penetrators (USIPs) are recent additions to the series of apparatuses based on ultrasonic/sonic drill corers (USDCs). A USIP enables a rod probe to penetrate packed soil or another substance of similar consistency, without need to apply a large axial force that could result in buckling of the probe or in damage to some buried objects. USIPs were conceived for use in probing and analyzing soil to depths of tens of centimeters in the vicinity of buried barrels containing toxic waste, without causing rupture of the barrels. USIPs could also be used for other purposes, including, for example, searching for pipes, barrels, or other hard objects buried in soil; and detecting land mines. USDCs and other apparatuses based on USDCs have been described in numerous previous NASA Tech Briefs articles. The ones reported previously were designed, variously, for boring into, and/or acquiring samples of, rock or other hard, brittle materials of geological interest. To recapitulate: A USDC can be characterized as a lightweight, low-power, piezoelectrically driven jackhammer in which ultrasonic and sonic vibrations are generated and coupled to a tool bit. As shown in the figure, a basic USDC includes a piezoelectric stack, a backing and a horn connected to the stack, a free mass (free in the sense that it can slide axially a short distance between the horn and the shoulder of tool bit), and a tool bit, i.e., probe for USIP. The piezoelectric stack is driven at the resonance frequency of the stack/horn/backing assembly to create ultrasonic vibrations that are mechanically amplified by the horn. To prevent fracture during operation, the piezoelectric stack is held in compression by a bolt. The bouncing of the free mass between the horn and the tool bit at sonic frequencies generates hammering actions to the bit that are more effective for drilling than is the microhammering action of ultrasonic vibrations in ordinary ultrasonic drills. The hammering actions are so effective that the axial force needed to make the tool bit advance into the material of interest is much smaller than in ordinary twist drilling, ultrasonic drilling, or ordinary steady pushing.
Two short stacks were pressure tested at 446 kPa (4.4 atm.) and the pressure gains were more than the theoretically predicted gains. Temperature profiles were observed to be independent of operating pressure. The pressure drop was found to be inversely proportional to operating pressure as expected. Continuous pressurized operation of a stack for 1000 hours verified the compatability of the fuel cell component design. A simple pressurization procedure was also developed. Six separate designs, covering two gas cooling schemes (DIGAS and separated) and two cooling channel geometries (straight through and treed), were analysed on the net voltage output basis. Separated cooling with 5 cells per cooler was recognized to be the best among the designs considered.
A 50 cell bipolar nickel-hydrogen battery was assembled to demonstrate the feasibility of constructing a high voltage stack of cells. Various component combinations were tested in this battery. The battery had approximately 1 ampere-hour of capacity and was constructed from components with an active area of 2" X 2". The components were parametrically varied to give a comparison of nickel electrodes, hydrogen electrodes, separators, fill procedures and electrolyte reservoir plate thicknesses. Groups of five cells were constructed using the same components; ten combinations were tested in all. The battery was thoroughly characterized at various change and discharge rates as well as with various pulse patterns and rates. Over a period of 1400 40% DOD LEO cycles some of the groups began to exhibit performance differences. In general, only separator variations had a significant effect on cell performance. It also appears that shunt currents may have been operating within the stack, resulting in electrolyte transfer from one cell to another, thus contributing to cell performance variations.
A 50 cell bipolar nickel-hydrogen battery was assembled to demonstrate the feasibility of constructing a high voltage stack of cells. Various component combinations were tested in this battery. The battery had approximately 1 ampere-hour of capacity and was constructed from components with an active area of 2 x 2 inches. The components were parametrically varied to give a comparison of nickel electrodes, hydrogen electrodes, separators, fill procedures and electrolyte reservoir plate thicknesses. Groups of five cells were constructed using the same components; ten combinations were tested in all. The battery was thoroughly characterized at various change and discharge rates as well as with various pulse patterns and rates. Over a period of 1400 40-percent DOD LEO cycles some of the groups began to exhibit performance differences. In general, only separator variations had a significant effect on cell performance. It also appears that shunt currents may have been operating within the stack, resulting in electrolyte transfer from one cell to another, thus contributing to cell performance variations.
A segmented involute-foil regenerator has been designed, microfabricated and tested in an oscillating-flow rig with excellent results. During the Phase I effort, several approximations of parallel-plate regenerator geometry were chosen as potential candidates for a new microfabrication concept. Potential manufacturers and processes were surveyed. The selected concept consisted of stacked segmented-involute-foil disks (or annular portions of disks), originally to be microfabricated from stainless-steel via the LiGA (lithography, electroplating, and molding) process and EDM (electric discharge machining). During Phase II, re-planning of the effort led to test plans based on nickel disks, microfabricated via the LiGA process, only. A stack of nickel segmented-involute-foil disks was tested in an oscillating-flow test rig. These test results yielded a performance figure of merit (roughly the ratio of heat transfer to pressure drop) of about twice that of the 90% random fiber currently used in small ~ 100 W Stirling space-power convertors in the Reynolds Number range of interest (50-100). A Phase III effort is now underway to fabricate and test a segmented-involute-foil regenerator in a Stirling convertor. Though funding limitations prevent optimization of the Stirling engine geometry for use with this regenerator, the Sage computer code will be used to help evaluate the engine test results. Previous Sage Stirling model projections have indicated that a segmented-involute-foil regenerator is capable of improving the performance of an optimized involute-foil engine by 6-9%; it is also anticipated that such involute-foil geometries will be more reliable and easier to manufacture with tight-tolerance characteristics, than random-fiber or wire-screen regenerators. Beyond the near-term Phase III regenerator fabrication and engine testing, other goals are (1) fabrication from a material suitable for high temperature Stirling operation (up to 850 C for current engines; up to 1200 C for a potential engine-cooler for a Venus mission), and (2) reduction of the cost of the fabrication process to make it more suitable for terrestrial applications of segmented involute foils. Past attempts have been made to use wrapped foils to approximate the large theoretical figures of merit projected for parallel plates. Such metal wrapped foils have never proved very successful, apparently due to the difficulties of fabricating wrapped-foils with uniform gaps and maintaining the gaps under the stress of time-varying temperature gradients during start-up and shut-down, and relatively-steady temperature gradients during normal operation. In contrast, stacks of involute-foil disks, with each disk consisting of multiple involute-foil segments held between concentric circular ribs, have relatively robust structures. The oscillating-flow rig tests of the segmented-involute-foil regenerator have demonstrated a shift in regenerator performance strongly in the direction of the theoretical performance of ideal parallel-plate regenerators.
A segmented involute-foil regenerator has been designed, microfabricated and tested in an oscillating-flow rig with excellent results. During the Phase I effort, several approximations of parallel-plate regenerator geometry were chosen as potential candidates for a new microfabrication concept. Potential manufacturers and processes were surveyed. The selected concept consisted of stacked segmented-involute-foil disks (or annular portions of disks), originally to be microfabricated from stainless-steel via the LiGA (lithography, electroplating, and molding) process and EDM. During Phase II, re-planning of the effort led to test plans based on nickel disks, microfabricated via the LiGA process, only. A stack of nickel segmented-involute-foil disks was tested in an oscillating-flow test rig. These test results yielded a performance figure of merit (roughly the ratio of heat transfer to pressure drop) of about twice that of the 90 percent random fiber currently used in small approx.100 W Stirling space-power convertors-in the Reynolds Number range of interest (50 to 100). A Phase III effort is now underway to fabricate and test a segmented-involute-foil regenerator in a Stirling convertor. Though funding limitations prevent optimization of the Stirling engine geometry for use with this regenerator, the Sage computer code will be used to help evaluate the engine test results. Previous Sage Stirling model projections have indicated that a segmented-involute-foil regenerator is capable of improving the performance of an optimized involute-foil engine by 6 to 9 percent; it is also anticipated that such involute-foil geometries will be more reliable and easier to manufacture with tight-tolerance characteristics, than random-fiber or wire-screen regenerators. Beyond the near-term Phase III regenerator fabrication and engine testing, other goals are (1) fabrication from a material suitable for high temperature Stirling operation (up to 850 C for current engines; up to 1200 C for a potential engine-cooler for a Venus mission), and (2) reduction of the cost of the fabrication process to make it more suitable for terrestrial applications of segmented involute foils. Past attempts have been made to use wrapped foils to approximate the large theoretical figures of merit projected for parallel plates. Such metal wrapped foils have never proved very successful, apparently due to the difficulties of fabricating wrapped-foils with uniform gaps and maintaining the gaps under the stress of time-varying temperature gradients during start-up and shut-down, and relatively-steady temperature gradients during normal operation. In contrast, stacks of involute-foil disks, with each disk consisting of multiple involute-foil segments held between concentric circular ribs, have relatively robust structures. The oscillating-flow rig tests of the segmented-involute-foil regenerator have demonstrated a shift in regenerator performance strongly in the direction of the theoretical performance of ideal parallel-plate regenerators.
A six-man solid electrolyte oxygen regeneration system (SEORS) that will produce 12.5 lbs/day of oxygen has been designed. The SEORS will simultaneously electrolyze both carbon dioxide and water vapor and be suitable for coupling with a carbon dioxide concentration system of either molecular sieve, solid amine or hydrogen depolarized electrochemical type. The total system will occupy approximately 19 cu ft (34.5 in. x .26 in. x 36 in. high) and will weigh approximately 500 pounds. It is estimated that the total electrical power required will be 1783 watts. The system consists of three major components; electrolyzer, hydrogen diffuser, and carbon deposition reactor. There are 108 electrolysis stacks of 12 cells each in the electrolyzer. Only 2/3 of the 108 stacks will be operated at a time; the remainder will be held in reserve. The design calls for 96 palladium membranes for hydrogen removal to give 60 percent redundancy. Four carbon deposition reactors are employed. The iron catalyst tube in each reactor weighs 7.1 lb and 100 percent redundancy is allowed.
With the recent explosion of the Internet and the enormous business opportunities available to communication system providers, great interest has developed in improving the efficiency of data transfer using the Transmission Control Protocol (TCP) of the Internet Protocol (IP) suite. The satellite system providers are interested in solving TCP efficiency problems associated with long delays and error-prone links. Similarly, the terrestrial community is interested in solving TCP problems over high-bandwidth links. Whereas the wireless community is interested in improving TCP performance over bandwidth constrained, error-prone links. NASA realized that solutions had already been proposed for most of the problems associated with efficient data transfer over large bandwidth-delay links (which include satellite links). The solutions are detailed in various Internet Engineering Task Force (IETF) Request for Comments (RFCs). Unfortunately, most of these solutions had not been tested at high-speed (155+ Mbps). Therefore, the NASA's ACTS experiments program initiated a series of TCP experiments to demonstrate scalability of TCP/IP and determine how far the protocol can be optimized over a 622 Mbps satellite link. These experiments were known as the 118i and 118j experiments. During the 118i and 118j experiments, NASA worked closely with SUN Microsystems and FORE Systems to improve the operating system, TCP stacks. and network interface cards and drivers. We were able to obtain instantaneous data throughput rates of greater than 520 Mbps and average throughput rates of 470 Mbps using TCP over Asynchronous Transfer Mode (ATM) over a 622 Mbps Synchronous Optical Network (SONET) OC12 link. Following the success of these experiments and the successful government/industry collaboration, a new series of experiments. the 118x experiments. were developed.
Unitized Regenerative Fuel Cells (URFC) have recently been developed by several fuel cell manufacturers. These manufacturers have concentrated their efforts on the development of the cell stack technology itself, and have not up to this point devoted much effort to the design and development of the balance of plant. A fuel cell technology program at the Glenn Research Center (GRC) that has as its goal the definition and feasibility testing of the URFC system balance of plant. Besides testing the feasibility, the program also intends to minimize the system weight, volume, and parasitic power as its goal. The design concept currently being developed uses no pumps to circulate coolant or reactants, and minimizes the ancillary components to only the oxygen and hydrogen gas storage tanks, a water storage tank, a loop heat pipe to control the temperature and two pressure control devices to control the cell stack pressures during operation. The information contained in this paper describes the design and operational concepts employed in this concept. The paper also describes the NASA Glenn research program to develop this concept and test its feasibility.