The VizSpark software is used to perform a 2-D simulation of a model spark plug channel with cross-flow. Coupling between arc physics and conjugate heat transfer to the electrode was performed. The ability to predict spark restrike events was also demonstrated. Furthermore, an estimate of electrode ablation is also made.

The spark plug has been a reliable source of ignition for well over 100 years.  The ignition source plays a pivotal role in the efficiency of the combustion ignition event, and there is currently a significant focus on improving both the performance and longevity of the spark plug.  Often, there is a compromise between improving performance versus the spark plug lifetime.  Decreasing the electrode tip radius enhances the electric field, but at the cost of degrading the lifetime of the electrode tip due to material ablation.  Consequently, it is desirable to have the capability of simultaneously designing and optimizing spark plug performance (e.g. arc and initial flame kernel formation) as well as ascertaining the lifetime of the spark plug (e.g. how long will the electrodes last before significant surface degradation has occurred).

 

The lifetime of the spark plug, i.e. how long the spark plug will maintain its performance before requiring replacement, is of prime interest. The intense conditions and high thermal loading that occur during spark breakdown and arc formation ablate material from the electrodes, degrading the sharp edges and degrading spark plug performance. To extend life times, modern spark plugs utilize heat resistance metals such as Tungsten, Iridium and Platinum at the electrode tips with high melting temperature.  Even if arc temperatures do not exceed the melting temperature, the vapor pressure of the electrodes can reach a small yet finite values leading to finite amounts of material ablation for each firing of the spark plug.  Over the millions of firings over a spark plugs life time, the net ablation of the electrode is enough to significantly degrade the electrode shape and adversely impact performance.  It is therefore useful to have the ability to predictively model the ablation and erosion of the electrode surfaces.

Electrode lifetimes and erosion is a process that occurs over an operating lifetime that can consist of millions of firings. The approach suggested here is to model a single firing event over a 1-2 ms transient including arc formation as well as conjugate heat transfer to the electrode material.  The material ablated from the electrode surface is determined using an equilibrium vapor model which calculates mass ablation as a function of the surface temperature and the vapor pressure of the metal.  The etch rate is calculated using the ablation rate and the density of the metal.  The etch depth is then calculated by integrating the etch rate on the electrode surface over time.

The gas temperature is set to 700 K and the gas pressure is set to 3 atmospheres.  The electrode temperature is set to 1200 K (935 C) approximating the conditions of an engine that has been in continuous operation.  The working gas is set as air with an inflow velocity of 20 m/s and the electrode material is iron. A constant applied current density of 4 x 10⁵ A/m² (corresponding to a constant current of 50 mA if the electrode radius is taken to be circular) is imposed on the tip surface of the powered (bottom) electrode.

Figure 2 shows the voltage transient associated with arc stretching and re-strikes.  As the arc stretches and elongates, the resistance increases as is seen by the ramp up in the voltage.  When the arc channel is blown off and a re-strike occurs, the voltage sharply drops.  This see-saw behavior is also observed experimentally [1].

The ablation model utilized here considers mass ablation rate  as function of the metal vapor pressure, and the electrode-arc interface temperature .  The interface temperature is solved for self-consistently as a conjugate heat transfer problem between the metal and the arc.

Figure 3 shows temperature and etch rate transients at the electrode tip at the end of the 0.5 ms spark transient.

The net etch depth over the powered electrode top surface is obtained by integrating the etch rate over time and is shown in Figure 4 at different times in the transient.  Over the 2 millisecond transient, the etch depth calculated is approximately 2-2.5×10^-11 meters.  If the lifetime of the spark plug is measured to be on the order of millions to tens of millions of firings, then the total etch depth at the right most tip over the lifetime of the sparkplug estimated by extrapolating the results from the single firing simulation can reach 0.025 to 0.25 millimeters.

The increased etching depth at the corner will lead to a rounding and loss of sharpness of the edge electrode which will negatively impact the performance of the igniter.

VizSpark provides robust capability for the high-fidelity simulation of high-pressure non-equilibrium discharges such as DBD’s.  Simulation of DBD’s is a particularly challenging problem because of the highly disparate time scales involved in the problem and the potentially wide range of length scales imposed by the small region of discharge formation in the context of a larger application domain.  Furthermore, the high pressure results in very restrictive resolution requirements since the thin propagating streamer must be resolved accurately in order to model the DBD discharge.  This kind of modeling is very much the state-of-the-art in plasma computational modeling community. VizSpark is provably fast, robust, and easy-to-use software and currently a leading industrial plasma simulation tool