Predictive Breakdown Modelling for Spark Plug Design 2020-01-0781
The lifetime of a spark plug is often limited by the ability of a spark plug to generate spark as the plug erodes and deforms. Many parameters including the spark plug gap distance, geometric features such as corners, and the gas itself influence the breakdown event. In order to asses the viability of a spark plug design, the most ideal solution is to perform full fidelity plasma simulations for each design configuration. However, these simulations are extremely expensive and take a long time to give meaningful results. Thus, there is a need for a fast spark plug breakdown model which can predict the breakdown effectiveness for a large number of spark plug design configurations over a much shorter period of time, narrowing the range of spark plug design which can then be studied using a higher fidelity plasma simulation tool.
The objective of this work is to describe a fast, predictive breakdown model to simulate the effectiveness of different spark plug configurations. A spark discharge is preceded by a transient filamentary plasma called a streamer discharge. Thus, to evaluate the breakdown effectiveness of a spark plug it is important to predict the formation of transient streamer discharge. Streamer breakdown is characterized by very short time scales (~nanoseconds) and chemical and thermal non-equilibrium. The Raether-Meek criteria relates the probability of the streamer breakdown at a spatial location to the reduced electric field at the location and the net Townsend ionization coefficient at the location. In this work, we apply the Raether-Meeks criteria in a numerical model to evaluate the streamer breakdown probability over the entire spatial region. The Raether-Meeks model output can be interpreted as a parameter that can assess the effectiveness of the spark plug design and identify regions where spark breakdown is likely to occur.
We conducted simulations by varying the spark gaps and aspect ratios of the spark plug design and perform parametric studies to assess the dependence of breakdown of these design parameters. To investigate the practical usability of the model, we will perform simulations on a three-dimensional J-gap configuration for two different cathode shapes namely, a smooth prong cathode and a cathode with grooves.
Ashish Sharma, Douglas Breden, Jay Cress, Laxminarayan Raja
Esgee Technologies, Caterpillar Inc, University of Texas at Austin