An Experimental Study of Spark Anemometry for In-Cylinder Velocity Measurements

Abstract

It has been demonstrated by previous researchers that an approximate value of the bulk flow velocity through the spark plug gap of a running spark ignition engine may be deduced from the voltage and current wave forms of the spark. The technique has become known as spark anemometry and offers a robust means of velocity sensing for engine combustion chambers and other high temperature environments. This paper describes an experimental study aimed at improving performance of spark anemometry as an engine research tool. Bench tests were conducted using flow provided by a calibrated nozzle apparatus discharging to atmospheric pressure. While earlier studies had relied upon assumptions about the shape of the stretching spark channel to relate the spark voltage to the flow velocity, the actual spark channel shape was documented using high-speed video in the present study. A programmable ignition system was used to generate well-controlled constant current discharges. The spark anemometry apparatus was then tested in a light duty automotive engine. Results from the image analysis of the spark channel shape undertaken in the present study have shown that the spark kernel moves at a velocity of less than that of the freestream gas velocity. A lower velocity threshold exists below, which there is no response from the spark. It is possible to obtain a consistent, nearly linear relationship between the first derivative of the sustaining voltage of a constant current spark and the freestream velocity if the velocity falls within certain limits. The engine tests revealed a great deal of cycle-to-cycle variation in the in-cylinder velocity measurements. Instances where the spark restrikes occur during the cycle must also be recognized in order to avoid false velocity indications.

Voltage rise anemometry in turbulent flows applied to internal combustion engines

“…Furthermore, many researchers could not apply statistically based post-processing and could not extract the assessable sections of the signal (Gardiner et al 2008). Finally, in the past the focus was never on measuring turbulent kinetic energy.…”

“…He compared VRA to hot wire anemometry (HWA). Later some further investigations were performed (Schaudt et al 1985;Gardiner et al 2008;Kim and Anderson 1995;Pashley et al 2000;Paa, 2019). Especially, the research project of Schaudt et al (1985) extended the expertise.…”

Design of a test rig for fundamental investigations of spark characteristics

“…The sensitivity of the electric properties of the arc column (Figure 1d)) to changes in ambient temperatures is expected to be lower than the pressure influence (in contrast to the breakdown properties (Figure 1c))), due to the exponential dependence of electric conductivity in gases on the local temperature. 28 This makes column properties rather insensitive to ambient temperature conditions due to the negligible small changes in electric conductivity caused by ambient temperature changes at the arc periphery. 28 Since the focus is on the interaction of the arc column with the flow at the spark plug (Figure 1d)), high ambient pressures have a higher priority in the experimental investigations than high gas temperatures.…”

“…28 This makes column properties rather insensitive to ambient temperature conditions due to the negligible small changes in electric conductivity caused by ambient temperature changes at the arc periphery. 28 Since the focus is on the interaction of the arc column with the flow at the spark plug (Figure 1d)), high ambient pressures have a higher priority in the experimental investigations than high gas temperatures. Thus, a temperature range of 25 °C–100 °C is envisioned in order to keep the test rig design within reasonable bounds.…”

“…The maximum flow velocity at the spark plug is 15 m/s, the maximum pressure and maximum temperature at ignition timing are 15 bar and 27 °C and the peak firing pressure is 70 bar. Gardiner et al 28 also direct the flow toward the spark plug with a nozzle (nozzle diameters 12 mm and 63.5 mm), yet in contrast to Schneider et al, 27 only under inert conditions. Based on these results, flow velocities up to 15 m/s are achieved; these are superficial (volume averaged) velocities.…”