Ignition systems spark new interest in ICE’s.

Facebooktwittergoogle_plusredditpinterestlinkedinmail

The new generation of downsized turbocharged engines require more than a spark generated at the right time to initiate combustion: Some mixtures exceed stoichiometric ratios and in theory should not even combust.

The modern ignition system has evolved from a purely mechanical switch used to charge the coil, to engine-management controlled, electronic switching circuits. A few modern ignition systems are now full closed-loop systems that offer improved mileage and better performance.

Current ignition system technology.

In most of the cars in production today, the ignition system is an open-loop single-spark system. The module may have features such as coil current limiting, overvoltage or overtemperature protection. Some systems also provide coil primary-current diagnostics and feedback to the engine control unit (ECU), enabling evaluation of the quality of the spark.

These conventional spark-advance systems control combustion by referring to detailed calibration tables that take into account: The engine’s revolutions per minute (RPM), battery voltage, coil-charging times, engine-knock sensing, amongst other system parameters. But these spark-advance systems are still primarily open-loop systems controlling the engine combustion based on the expected performance.

Facing ever leaner mixtures, constraints on component size and rising Break Mean Effective Pressures, ignition systems are faced with challenges that require innovative solutions.

Emerging technologies are smarter.

Currently, new systems are emerging that have seen a radical shift in the technology applied to ignition systems.
In the ion sensing system, the ion current stream developed due to combustion can be deduced by monitoring the secondary current of the ignition coil. Using the value and wave shape of the current, after the actual spark event, the quality of the actual combustion process is determined, thus allowing the engine control unit to optimize the timing of the spark for the best engine performance.

In conventional spark controllers, during normal driving conditions, the actual spark ignition occurs about 15 to 30 degrees before TDC (top dead center). This “dwell angle” compensates for the delay from the start of the spark until the actual combustion of the fuel and development of the pressure wave in the cylinder. However, the optimal timing for the actual combustion to develop the most energy out of the firing is approximately 15 degrees after TDC.

Calibrating this event precisely using several open loop sensors would be a real challenge. But using an ion sensing ignition system, the pressures over time in the cylinders, as well as the quality of the actual combustion can be derived in order to optimize the spark timing.

Furthermore, regulations on engine controllers require the detection and signaling of engine misfires, which this system does by directly measuring the combustion. By so doing pre-ignition or knock can be detected as well as a “miss fire” or a “non-firing plug”; and all of this can be done without any additional sensors, by merely using the existing spark plug in an innovative way.

On the other hand, Multi-Spark or Multi-Charge Ignitions generate a series of sparks rather than just firing one spark. Using several spark events, a leaner air-fuel mixture can be ignited resulting in better fuel consumption whilst improving efficiency and performance.

This concept uses a standard spark ignition and simply pulses the ignition system to generate a stream of sparks. With a control circuit in the coil, the engine controller can set the levels of charge and discharge for each spark ensuring sufficient energy in the coil to achieve ignition.

Up until now, the technology developed under totally different boundary conditions. In the past, the charge was homogenous, stoichiometric, or possibly mildly lean-burn. Under these conditions, the restriction to a small ignition source was not a problem.

With aggressive charge stratification, and much higher lambda (λ) this restriction is hindering the advances required. While the issue of higher Break Mean Effective Pressure and a slightly greater λ could still be overcome by increasing the breakdown voltage and arc duration, glaring problems remain: If it takes longer to ignite the charge, the point at which 50%-mass fraction burning occurs will also be delayed. In addition, longer arc duration, coupled with high energy, will increase plug wear. More importantly, however, nothing can be done about the restrictive location of the spark.

To overcome these inherent weaknesses several advanced technologies are being developed.

The next generation ignition system technology.

One such system is the “Advanced Corona Ignition System”.

Federal Mogul Corona ignition

Image credit: Federal Mogul Corporation

Whereas conventional spark ignition only creates a small arc in the gap between the electrodes of the spark plug, ACIS uses a high-energy, high-frequency electrical

field to produce repeatable, controlled ionization. This creates multiple streams of ions which ignite the fuel mixture throughout the combustion chamber.

Optimized for ease of implementation, the “two-piece” igniter architecture allows engine manufacturers to replace traditional coil and plug systems with no adverse impact on engine design or assembly.

An early leader in this technology, Federal-Mogul Corporation, has already achieved up to 10% fuel efficiency improvement over standard spark ignition in development testing.

This advanced technology offers the following benefits:

• Advanced combustion strategies leading to fuel economy/CO2 improvements in excess of 10%, over a wide speed-load range
• Larger ignition source: Multiple 25 mm long streamers vs. single 1 mm arcs
• Higher dilution (EGR) tolerance: 35% vs. 25%
• Extended lean limit combustion stability λ = 1.8 vs. 1.5
• Robust combustion initiated in as little as 30 μs
• Faster burn/less delay: 5° less ignition advance than conventional systems
• Enhances Homogeneous-charge compression-ignition (HCCI) capability

Equally impressive results have been achieved by MAHLE Powertrain Ltd, using a totally different technology: Pre-chamber initiated combustion.

Mahle pre-chamber initiated ignition

Image credit: MAHLE Powertrain 

Known as ‘Turbulent Jet Ignition’ (TJI), the system utilises a spark-initiated pre-chamber combustion process in an otherwise conventional gasoline engine to achieve fuel economy improvements of up 20%. Engine-out NOx emissions are also virtually reduced to zero levels, negating the need for lean NOx after-treatment.

MAHLE’s TJI system is characterised by auxiliary pre-chamber fuelling, small orifices connecting the main and pre-chamber combustion cavities and a very small pre-chamber volume. The smaller orifice size causes turbulence in the hot gas jets which then penetrate deeper into the main combustion chamber and cause a distributed ignition effect. This process allows extension of knock limits and increased compression ratios (up to 14:1) combined with lower combustion temperatures and reduced throttling / pumping losses, achieving thermal efficiencies in the region of 45%.

With the current European trend of downsizing and supercharging, these systems have the added benefits of promoting power density, while still meeting future consumption and emission regulations.

Although more futuristic, the ignition of a fuel/air mixture by means of laser plasma is essentially nothing new and was already presented at an SAE conference in 1978 by Dr. Dale.

Since then extensive research by several institutions and companies, such as GEJenbacher GmbH and Japan’s National Institutes of Natural Sciences (NINS), have brought the system to the point where it could well enter series production within the next few years.

Laser ignition

Image credit: ASME Internal Combustion Engine Division

Laser ignition addresses all of the previously mentioned short-comings of conventional ignition systems but, being an untried technology in this application, faces several challenges.

In order to create the desired combustion, a laser would have to be able to focus light to approximately 100 gigawatts per square centimeter with short pulses of more than 10 millijoules each. Previously, this sort of performance could only be achieved by large, inefficient, relatively unstable lasers.

The Japanese researchers have however created a small, robust and efficient laser that can do the job. This was achieved by heating ceramic powders, fusing them into optically-transparent solids, then embedding them with metal ions in order to refine their properties.

Made from two bonded yttrium-aluminum-gallium segments, the laser igniter is just 9 millimeters wide and 11 millimeters long. It has two beams, which can produce a faster, more uniform ignition by igniting the air-fuel column in two locations at once. While it cannot propagate combustion with just one pulse, it can do so using several 800-picosecond-long pulses.

A challenge faced by all of these emerging technologies is that of simulation and testing.

Current Computational Fluid Dynamic (CFD) simulation methods focus on modeling the engine behavior during ignition and combustion of the fuel. But the real value of a simulation lies in the possibility to predict relevant events during the combustion cycle. Using suitable software engine designers can test designs before they have to use real hardware for their tests; by so doing saving time and money.

Recently a company, Reaction Design, released a simulation product which utilises Chemkin Pro technology to generate more accurate results in one tenth of the time previously taken. The software utilizes multi-component fuel models which embrace hundreds of substances included in today’s most popular gasoline types. Based on these parameters, it can precisely compute the combustion process and predict ignition timing and propagation within the combustion chamber.

Caught in the vice of ever tightening emissions and lower fuel consumption; and consumers’ expectations of increased power, the heart of the engine remains the combustion process. While a truly viable replacement for the IC engine appears to be some way away, we can expect to see some exciting ICE developments in the next decade.
As far as the strides taken in ignition systems thus far – Homogeneous-charge compression-ignition and stratified-charge compression-ignition engines may just dictate the future of spark ignition.

, , ,

No comments yet.

Leave a Reply

UA-72741985-1