|The TFF. Click to enlarge.|
To support the fair sale of gaseous hydrogen as a vehicle fuel, researchers at the National Institute of Standards and Technology (NIST) have developed a prototype Transient Flow Facility (TFF) to test the accuracy of hydrogen fuel dispensers. The TFF generates transient flow, pressure, and temperature conditions similar to those that occur when a hydrogen-powered vehicle is refueled.
In a paper published in the journal Flow Measurement and Instrumentation, the NIST team reports using the TFF to assess the performance of two Coriolis meters (used to measure mass flow). However, they noted, the TFF can test other meter types and protocols, making it ideal for testing prototype field calibration standards for gaseous fuel dispensers.
More generally, they found, the TFF is capable of a wide variety of tests dealing with transient pressure, temperature, and flow conditions, including gaseous refueling processes, blow-downs, and quasi-stable calibrations.
Flow meters are sometimes used to measure unsteady flows, under conditions where the temperature and pressure also vary. For example, dispensing stations for hydrogen-fueled vehicles comprise a set of pressure vessels (a cascade tube bank) filled to different pressures. As a vehicle is refueled, valves are sequentially opened to connect the vehicle’s fuel tank to the cascade tubes in order of increasing pressure. As each tube is opened, surges of flow and pressure occur at the flow meter that totalizes the flow for customer billing. Rapid, large changes in temperature also occur due to flow work and the subsequent adiabatic cooling and heat transfer to the surroundings. Consumers and inspectors expect < 1% accuracy from meters used in gaseous fuel dispensers, but errors greater than 10% have been reported. At natural gas refueling stations, turbine meters subjected to pulsatile flow over-reported totalized flow by as much as 15%.—Pope and Wright (2014)
The TFF has four, 40 L high pressure tanks (HPTs) that serve as a source of nitrogen or helium at an initial pressure of 42 MPa. These high pressure tanks can be sequentially discharged to simulate cascade filling of a vehicle.
As reported in the paper, during simulated cascade fills, the TFF discharged 3 kg of helium in 3 minutes at flows between 10 g/s and 45 g/s through the two Coriolis meters and the TFF’s standard. The TFF’s expanded uncertainty (95% confidence level) for totalized mass during this cascade fill was 0.45%. For the same simulated cascade fill, both Coriolis meters measured the instantaneous flow within the uncertainty of the TFF and measured totalized flow within the International Organization of Legal Metrology Recommendation 139 maximum permissible errors for meters in gaseous fuel dispensers (1.0 %).
Three automakers plan to begin selling hydrogen-fueled vehicles to consumers in 2015. The state of California has opened nine refueling stations and is funding the construction of an additional 28 hydrogen stations during the next few years to service the growing number of hydrogen fuel cell vehicles on its roads.
NIST Handbook 44, the bedrock reference text for weights and measures inspectors, includes specifications, tolerances and other requirements for commercial weighing and measuring equipment ranging from gasoline dispensers to grocery store scales.
Handbook 44, which has been adopted by all states, stipulates that hydrogen will be sold by the kilogram, and according to Juana Williams, a NIST weights and measures expert, hydrogen-dispensing pumps must be accurate to within 2%, or 20 grams, per kilogram.
It’s much more difficult to measure hydrogen gas delivered at 5,000 to 10,000 psi than it is to measure a product that is a liquid at atmospheric temperatures and pressures. While a kilogram of hydrogen has approximately the same energy content as a gallon of gasoline, the allowable error is slightly less stringent than for gasoline.—Juana Williams
Even with the larger allowance, some have suggested these tolerances are too tight and proposed alternatives as high as 10 or 20%. What isn’t clear is whether these claims arise because the meters are unable to perform within the tolerance specified in Handbook 44 or if the equipment and methods used to conduct testing are contributing larger errors to the process.
Regardless, consumers expect to receive the product they pay for and businesses expect to receive fair payment for the product they sell.
We’ve shown that the master meter in our lab is capable of dispensing helium from a simulated hydrogen dispenser with errors of 1 percent or less. So we can extrapolate that it is possible to measure hydrogen with accuracy sufficient for a fair marketplace.—Jodie Pope, designer of the field testing apparatus
The next challenge is to determine what accuracy is achievable in field installations of hydrogen dispensing systems when using NIST traceable standards and well-defined test equipment and test procedures and to then translate this into guidance for use by weights and measures inspectors and industry.
J. Pope and J. Wright (2014) “Performance of Coriolis Meters in Transient Gas Flows,” Flow Measurement and Instrumentation Vol. 37, 42-53 doi: 10.1016/j.flowmeasinst.2014.02.003