Daniel Schrag

Coating diagnostics for thermal mass flowmeters

Thermal mass flowmeters are based on the measurement of the heat transfer from the sensor into the moving fluid, which is given by Kings law. A coating of the sensor element changes the heat transfer from the sensor to the fluid and leads to an error in the flow measurement. We have developed a possible diagnostics algorithm to detect a sensor coating in the μm range. For this the effect of a small time dependent variation of the heating power is analyzed. The measurement approach is confirmed by a comparison of experimental results and FEM simulations of the quasistationary heat conduction. In addition an analytical model was developed, which can be used to determine the coating thickness and correct the flow value if the thermal properties of the coating material are known. As the relevant frequencies are below 1Hz an adaptive online algorithm was developed, which is able to detect the coating without disturbing the flow measurement.

Impedance spectroscopy for diagnostics of magnetic flowmeter

We have developed and implemented advanced diagnostics features for magnetic flowmeters to measure the medium conductivity as an important process parameter and coating on electrodes or liner, which disturbs the flow measurement. The electrode-electrolyte interface (Helmholtz or double layer) is a major obstacle to determine the conductivity directly from the impedances between electrodes at low frequencies. The impedance matrix of the multi-electrode transducer was measured using impedance spectroscopy and different electrochemical components could be identified. The measurements were verified in numerical field simulations. The results have been used to derive a lumped model of the transducer. Coating is found to have an effect on the electrode-electrolyte interface, whereas conductivity can be determined from the bulk resistances. This served as the basis for a successful implementation of the diagnostics function in an embedded microcontroller system.

Statistical signal processing for the diagnostics of gas bubbles

An advanced diagnostics feature for magnetic flowmeter is presented. Gas bubbles in the liquid do not only falsify the measurement of the flow rate but are an important indicator for the process quality. Based on electric field simulations and measurements in the flow laboratory a physical device model has been developed. Different signal processing algorithms for bubbles or solid particle detection have been designed. They have been analyzed and verified based on statistical signal processing in the laboratory and in the field. This served as the basis for a successful implementation of the diagnostics function in an embedded microcontroller based flowmeter platform.

On increasing the power available to an intrinsically safe wireless HART adapter

A wireless HART adapter is a self-powered device that harvests power from the 4 ‥ 20mA current loop. It wirelessly transmits the digital information present on the loop. Such adapters can be retrofitted to in-place current loops if the voltage loss induced by harvesting is small enough. This paper presents two techniques that more than double the power available to the adapter for a given induced voltage loss. Conversely, these techniques can be used to minimise the voltage loss for a given required power. More precisely, the first technique proposes a new topology for the coupling between HART modulation (AC) and energy harvesting (DC); the power efficiency is increased by 30% compared to a previously used topology. The second technique proposes and power-efficient transistor-based reverse-current blocking barrier. The proposed barrier performs much better in terms of power efficiency (up to 50%) than the conventional diode-based barrier; it is thus a valuable substitute in any low-voltage and low-power electronics that involves compliance with intrinsic safety requirements.

Diagnostics of gas bubbles using wavelet transform

Detection of gas bubbles is an important indicator of the process quality in the industrial applications. Gas bubbles in the liquid not only falsify the measurement of the flow rate, but indicate problems like leakage, cavitations, reactions, boiling, etc. Based on electric field simulations and measurement in the flow laboratory, a physical device model has been developed. This paper presents the wavelet transform-based detection of gas bubbles. Application results based on the acquired signals with and without bubbles in liquid flow are presented.