In Germany, a high pressure piston prover has been operated as the national volumetric primary standard for high-pressure natural gas for more than 15 years. The reliability and the claimed uncertainty of this system was proven in international key comparisons in which all relevant physical principles of traceability for gas measurements were assed.
One important part of the uncertainty budget for the volume at the location of the meter under test (MuT) are dynamic effects caused by the limited dynamic in the transient response of the MuT in its indication of the flow. This impact of the limited transient response can be corrected if the volume indication (pulses) of the meter is measured with a high time resolution and if an appropriate model of the transient response for the meter is available.
The paper will describe the results of determination of the system model for a turbine meter and the model parameters. The experimental determination of the system parameters were based on measurements of meter indication as well as flow speed by means of LDV-equipment (at atmospheric conditions) and their verification by means of an ultrasonic-meter (at high pressure). The positive effect on the calibration results with the high pressure piston prover will be demonstrated and discussed.

The volumes of a storage tank and a temperature control loop of the high pressure gas flow standard system in KRISS were increased about three times to calibrate large flow meters and to test the flow capacity of control valves. Sonic nozzle bank consisted of 15 sonic nozzles (10, 13, and 19 mm diameters) is used as reference flow meter. The performance of this large air flow calibration system was evaluated.

Flow-field study was conducted concerning vortex shedding from a circular cylinder with a slit. Experiments were carried out in a water channel and a low-speed wind tunnel with emphasis on the impact of the slit width on the quality of the vortex shedding signal measured. It is found that among the cases studied, whose slit widths ranging from 0 to 0.3 d, where d denotes the diameter of the circular cylinder, an optimal range of slit width, 0.1 - 0.15 d, was identified, in which not only the vortex shedding signals show the best quality, but also the relation of the Strouhal number against the Reynolds number appears to be the most linear. By spanwise correlation of hot-wire measurements in the wind tunnel, it is further noted that in this situation the vortex shedding flow structures appear to be almost perfectly two-dimensional in the near wake region. Moreover, it is worthwhile to mention that for the cases of the circular cylinders with slits, the signal quality of vortex shedding reamins well acceptable for the Reynolds number as low as 2.4 × 10³. Thus, the rangeability of flow measurement is increased significantly.

Due to the mechanism of vortex flowmeter, it is susceptible to the random disturbance from the ambient. With the goal of improving the poor sensitivity and noise immunity of conventional vortex flowmeter at low-flowrate, many researchers have done lots of work, focusing on innovation about the bluff, including the geometry or combinations, and the signal processing. This paper presents an adaptive algorithm for a vortex flowmeter with dual bluff body to improve its sensitivity for low flowrate applications. In our study, we get the distance where intension of vortex getting strongest by experiment. Comparative experimental results of vortex with single bluff body, dual bluff body and dual bluff body combing self-adaption FFT verify the improvements of the flow performance at small flowrate.

Center for Measurement Standards (CMS) developed a laminar flow element (LFE) type flow meter as a transfer standard. This LFE consists of a single straight glass capillary or multiple straight glass capillaries connected in parallel. Two gauges and one thermometer measured the inlet/outlet pressure and inlet temperature, respectively, and the differential pressure was restricted from 2 kPa to 100 kPa. The glass capillaries were manufactured by laser machining, resulting in consistent inner diameter and straight flow path. Characteristics of glass material also prevented the capillary from bending during installation. By means of regression, the turndown ratio of this LFE flow meter could be higher than 20 and the residual would be still within 0.11 %. The reproducibility within 0.03 % indicated that this LFE flow meter can be used as a transfer standard. Measurement with dry air demonstrated that these four LFE meter could span the flow rate at (0.8 to 986) µmol/s within the deviation of ±0.15 %. (1 µmol/s could be converted as 1.3 cm³/min at 0 and 101.325 kPa) Additional measurement with nitrogen demonstrated the feasibility of measurement with multiple gases.