An ultrasonic intake airflow meter for engine testbed applications was developed. An automotive intake airflow meter must fulfill a series of requirements differentiating it from typical ultrasonic flow meters. First, the data sampling rate of the device must be as high as possible to be able to measure dynamic phenomena of the combustion engine. It is therefore necessary to use broadband ultrasonic transducers capable of sending short signals without post-pulse oscillations. The state-of-the-art piezoelectric transducers cannot fulfill these requirements. For this reason, the FLOWSONIX

Figure 1. The Capacitive Ultrasonic Sensor.

The CUTs are characterized by an excellent air impedance matching, enabling a very high sampling rate up to 1 kHz. The full dynamic behavior of any gas flow will be measured correctly. Flow measurements for any engine applications, where pulsation occurs, will be done correctly and with a high accuracy.

Measurement series have been made at zero flow because the accuracy requirements are critical primarily at low flow velocities and at the zero point of the device. For this purpose, the sensor pairs have been mounted in the measuring pipe of the flow meter, and the pipe was hermetically closed.

The setup was powered up for several minutes before start of measurements, and the measuring pipe was thermally stabilized in order to minimize thermal convection flows inside. A longterm measurement with duration ~65 hours was made with different pairs of transducers. Single measurements were made every 10 minutes. The accuracy improvement due to the system identification method is more than a factor 10. The flow velocity measured over 60 hours was stable within ±0.01 m/s, which is five times more than the required 0.05 m/s. Thus, the proposed method removes the estimation artifacts and reduces the drift.

Measurement of intake air on engine testbeds is mostly done with hot film devices. With the following application using a two-cylinder engine with displacement volume of 850 cc, the performance of the flow meter has been investigated and compared to a standard hot film measurement device. The two-cylinder engine with extremely high pulsations in the inlet manifold is a particular challenge for the intake air measurement, as well as for the aspiration characteristics and the air mass per combustion cycle.

Steady-state measurements were performed from 1500 rpm up to 6000 rpm, and transient cycles with different ramp times from 2 s up to 10 s and a full load ramp. For comparison purposes, both the FLOWSONIX and the standard Hot Film Anemometer (HFM) device have been connected in serial to the engine via the throttle flange (see Figure 2).

Figure 2. Both the FLOWSONIX Flow Meter and the standard Hot Film Anemometer (HFM) device were connected in serial to the engine via the throttle flange.

The results have been compared in an additional map and show discrepancies, especially in part load areas where pulsations typically occur. Hot film devices are not able to detect these pulsations that cause deviations in the consumption measurement up to >20%.

Conventional methods like the HFM, due to its unidirectional measurement principle or standard ultrasonic technology based upon piezo transducers, are not suitable for such measurement tasks. Even under stationary conditions, pulsations occur that lead to high deviations with conventional air mass flow measurement devices. The high time resolution of the FLOWSONIX additionally gives the possibility to measure and illustrate the dynamic behavior of the intake airflow.

This work was done by Katarzyna Kudlaty, Karl Köck, and Hubert Krenn of AVL List GmbH. The full technical paper on this technology is available for purchase through SAE International at http://papers.sae.org/2013-26-0118 .