Advanced Automotive Ignition System

G.J. Rohwein, S.R. Babcock and M.T. Buttram
Sandia National Laboratories
P. 0. Box 5800 - MS-1153
Albuquerque, NM 87185-1153

L.S. Camilli
HDI Systems, Inc.
7424 2nd St., NW
Albuquerque, NM 87007

ABSTRACT

Spark ignition engines of practically all types and sizes utilize ignition systems that can develop up to about 30 Kv for breakdown of the spark plug gap and typically deliver current levels in the range of 30 to 100 ma. Measurement and analysis of these systems show that the energy transfer efficiency is very low, typically in the order of one percent (1%) or less. Once the spark is formed, most of the energy is delivered to resistances in the transformer, spark plug wires and spark plugs. This operating level is adequate (but not necessarily optimum) for older and most modern engines, but will not meet the needs of future lean burn and some alternate fuel engines that require higher energy discharges to effectively ignite the air/fuel mixtures^1,2,3,4. To meet the requirements for increased ignition power and energy, ignition systems will have to be designed to operate at higher transfer efficiencies.

There are two basic approaches to increasing the electrical efficiency of ignition systems. The first is to utilize peaking capacitors across the spark plug⁵. The second is to design and utilize a more efficient discharge circuit in combination with a low resistance spark coil⁶. A third alternative would be to utilize a combination of both. By the first method, transfer efficiency can be increased to nearly fifty percent (50%). By the third, transfer efficiency can be seventy-five percent (75%) or more.

This paper summarizes the results of recent high power ignition experiments and related analyses, for so-called "breakdown ignition" conditions. The paper includes descriptions of both conventional system up-grades and a new higher energy system that features multiple drivers and an energy recovery circuit. It does not include analyses or experiments for arc or glow discharge conditions.

Introduction

It is well known that power-enhanced ignitions extend the lean burn limits and transient response of spark ignition engines^4,7. These enhanced ignitions have been primarily of two types; plasma jet and double discharge. Experiments with such ignitions date to at least the early 1970's. More recently, experiments with certain low inductance peaking capacitors across the spark plugs have also demonstrated similar improvements plus lower hydrocarbon emissions and fuel consumption, particularly with older engines. Additional benefits include better starting and cold running performance and lower combustion pressure variability⁷. In contrast to other methods of ignition enhancement, the high voltage peaking capacitors dramatically increase the electrical-to-plasma conversion efficiency by delivering high current during the resistive phase of the spark discharge^8,9.

Low inductance, low loss peaking capacitors have been under development by DirectHits^TM for a number of years^5,10 with assistance in testing and electrical characterization provided by Sandia National Laboratories under a Technical Assistance Agreement. Adapting them to an engine involves only installation of non-resistor spark plugs with the peaking capacitor placed directly over the spark plug insulator. This technique represents the simplest and most economical method of ignition enhancement.

An alternate method of ignition enhancement involves utilizing a high efficiency capacitor discharge circuit to power a low resistance high voltage transformer. A prototype system of this type has been built and bench tested. Initial results have shown the highest transfer efficiency and the capability of delivering higher peak power than any other system under comparable operating conditions.

Discussion

A. Ignition Circuit Analyses

It is useful to model the circuits and compare the energy transferred to the spark with varying amounts of resistance in the high voltage section of the circuit. Figure 1 shows a schematic of a typical capacitor discharge ignition system. Representative parameters for conventional and three different power enhanced ignitions are shown in TABLE 1. Note that Conventional ignition circuits have high resistance in the secondary side, often as much as 25 K ohms, which limits transfer efficiency to the spark to around 0.2%.

By replacing the 5K ohm resistor plug with a non-resistor type, the transfer efficiency increases to 10% due to the 10 pF capacitance of the plug discharging directly into the spark. This simple exchange of components illustrates the function of a peaking capacitor. By increasing the capacitance to 80 pF the transfer efficiency increases to nearly fifty percent (50%).

If the circuit analysis is repeated using a low resistance transformer, the transfer efficiency is consistently higher than with a standard spark coil. A comparison of transfer efficiencies using standard ignition coils and low resistance transformers and various circuit configuration is shown in Table 1. The highest transfer efficiency for any system results when there is virtually no resistance between the secondary capacitance and the spark and when low resistance rf suppression wire (50ohms/ft) is used.

These experiments and analyses were for breakdown conditions where the ignition current was greater than or equal to 100 ma and a fully developed spark discharge occurred with a duration of nanoseconds to multi-microseconds. For fast discharge conditions, the resistive phase time and resistance change as a function of time may be estimated from Ristic and Sorenson⁹ formulas below:

for

where the units are as noted in the equations.

Using these relations for ignition condition where

These relations show that the gap resistance drops very rapidly and with R² <<4 L/C, the peaking capacitor circuit will ring down into the relatively low residual spark resistance. For efficient energy transfer the peaking capacitor must obviously have very low ESR and dielectric losses.

With conventional ignitions where the pulse duration is comparatively long (tens to hundreds of microseconds), the resistive phase relations of Ristic and Sorenson are not directly applicable. For these conditions, low current arc voltages in the range of 50 to 100 ohms are a better approximation. Knowing the fixed resistances of the secondary circuit, reasonable transfer efficiency estimates can be made from a simple ratio of spark resistance to total secondary circuit resistance n=R_spk/R_tot. From this relation it is apparent that rather wide variations in arc resistance do not affect efficiency much.

Since most automotive engines operate with resistor spark plugs, acceptable ignition of stoichemetric or richer air/fuel mixtures is typically achieved with very low spark energies. So long as the circuit resistance does not increase appreciably, the engines maintain high performance levels. However, aging of spark plug wires often produces resistance increases which gradually degrades mileage with a corresponding increase in hydrocarbon and carbon monoxide emissions. For late model engines with ignitions in good condition, the effect of high power ignition is often small. For older engines (pre-1980), however, high power ignitions can reduce hydrocarbon and carbon monoxide to ranges below 100 ppm and 0.5 percent, respectively, without a catalytic converter. Mileage increases are often ten percent (10%) or greater.

B. Capacitor and Emissions Tests

Prototype peaking capacitors provided by DirectHits^TM were tested both on the bench for their electrical characteristics and on a variety of automobiles. Automobile tests included chassis dynamometer tests at various load and speed points, road tests over a standard course and operational fleet tests conducted in a normal usage mode. A third type of test was a standard EPA 505 conducted on three late model vehicles at Southwest Research Institute (SWRI)¹¹.

Bench tests primarily included high voltage insulation endurance tests, dielectric constant and dielectric loss measurements, all at temperatures to 300 degrees C. Figure 2 is a cross section of a DirectHits^TM capacitor attached to a spark plug. Figure 3 shows a typical current wave form from an 80pF capacitor charged to 18kV where the peak current was 1020A and the ring-down frequency 100 MHz.

Emissions tests on a variety of vehicles were run at high and low-idle points of 900 and 2500 rpm followed by steady load tests on a chassis dynamometer at 1 hp. 40 mph, 5 hp. 40 mph, 10 hp. 40 mph, and 9 hp. 15 mph. Figure 4 is a plot of unburned hydrocarbon reduction relative to stock measured from tests on a 1987 Chevrolet S-10 pickup. It is typical of emission reductions for vehicles greater than four to five years old. Three vehicles less than one year old were tested by SWRI in accordance with EPA 505 test procedures¹¹. Results showed minimal change in emissions with the peaking capacitors.

Fleet tests conducted on several groups of vehicles showed a net mileage gain in all cases. Typical of those were 10 State-owned vehicles tested by the New Mexico Transportation Department which, over a 6-month period, showed an average mileage gain of 12.8%¹². The lowest gain from fleet tests was a group operated by Coca Cola which resulted in a net gain of approximately 5%¹³.

C. High Power, Low Resistance Transformer System

A higher energy, low resistance pulse transformer system that has a four-step energy output control and delivers up to 22A of current to a non-resistor spark plug has been built and is currently being tested. The system was initially designed as a research device to study the effect of varying energy input to the ignition process. However, the basic transformer and drive circuit is readily adaptable to permanent installations.

The energy increments are 85, 170, 255, and 340 mJ. Maximum output current at these steps is 11, 15.6, 19 and 22 A, respectively. Figure 5 shows four overlaid current traces, one for each energy level. Figure 6 is a schematic of a five stage version of the system. Each of the drive stages has an energy recovery circuit which reverses the polarity on its respective capacitor following SCR commutation at negative polarity. Figure 7 shows a full discharge cycle from anode voltage fall to thyristor commutation, energy recovery and final recharge. With a low resistance transformer and energy recovery, the multiple drive system represents the most efficient ignition source available, having a transfer efficiency greater than 75% when used with a circuit-matched peaking capacitor.

Conclusions

Enhanced power ignitions that deliver higher power and more energy than conventional ignition systems, have been shown to improve fuel economy and lower hydrocarbon emissions in many vehicles, particularly those more than five years old. Mileage improvements are typically in the range of 10% and HC emissions reductions around 50% within maximums and minimums well outside these ranges. Because the lean burn limits are extended, high power ignitions also improve starting, cold running and transient throttle response.

Of the options available, high voltage peaking capacitors deliver the highest power and are the most economical for upgrading a conventional spark ignition system. Transfer efficiencies over 40% are possible.

As an alternate to peaking capacitors or in combination with them, a high voltage pulse transformer with low resistive losses can deliver between 100 and 200 times the current (and power) to a spark that a conventional ignition system can. The high power transformer system with selectable energy output should provide researchers with a useful tool in studying the wide range of conditions necessary to optimize the ignition process.

REFERENCES

G.F.W. Ziegler, R.R. Maley, and E.P. Wagner, "Effect of Ignition System Design on Flammability Requirements On Ultra-Lean Turbulent Mixtures", I.Mech.E., paper no. C47/83, 1983.

R. Maly, B. Saggau, E. Wagrner, and G. Ziegler, "Prospects of Ignition Enhancement", SAE Paper 830478, Feb., 1983.

R.W. Anderson, "The Effect of Ignition System Power On Fast-Burn Engine Combustion", SAE Paper 870549, Feb., 1987.

R.W. Anderson and J.R. Asik, "Lean Air-Fuel Ignition System Comparison In a Fast-Burn Engine", SAE Paper 850076, Feb., 1985.

R.P. Corio, J.E. Griswold, R.E. Hensley, R.C. Pate, Glen B. Smith, "Hand Discharge Ignition (HDI)", Hensley Co. Report, Albuquerque, NM, 1984.

M.A. Ward, "A New Spark Ignition System For Lean Mixtures Based on a New Approach to Spark Ignition", SAE Paper 890475, Feb., 1989.

R. Maly, "Spark Ignition: Its Physics and Effects on the Internal Combustion Process", published in: Fuel Economy in Road Vehicles Powered By Spark Ignition Engines, Hilliard, J.C. and Springer, G.S. (Ed.), Plenum, New York, 1984.

J.C. Martin, "Duration of the Resistive Phase and Inductance of Spark Channels", AFWL Switching Note 9, 1965.

T. P. Sorensen and V.M. Ristic, "Rise Time and Time-Dependent Spark-Gap Resistance in Nitrogen and Helium", J. Appl. Phys., V. 48, No. 1, pp. 114-117, 1977.

R.C. Pate, R.E. Hensley, "Combustion Initiation System Employing Hard Discharge Ignition", U.S. Patent No. 4,589,398, May 20, 1986.

M.S. Newkirk, Exhaust Emission Testing of Vehicles Using a Hard Discharge Ignition System, SwRI 08-6295, April, 1994.

Letter, R. A. Leonard, New Mexico Highway and Transportation Department, to L. Camilli, Combustion Technology Products Corp, Subject: Results of Swaser Ignition Fleet Tests Conducted by New Mexico State Highway and Transportation Department dated April 5, 1993.

Private communication, L. Camilli to G.J. Rohwein, Subject: Results of Coca Cola Fleet Tests, October, 1994.

Index of Charts and Tables

Capacitor Discharge Ignition System

Circuit Configuration Parameters

Peaking Capacitor Ringdown

Hydrocarbon Emission From 1987 Chevrolet S-10 Pickup

Overlaid current traces from four stage multi-drive ignition

5-stage multi-drive ignition

Discharge cycle with energy recovery




	Advanced Automotive Ignition System G.J. Rohwein, S.R. Babcock and M.T. Buttram Sandia National Laboratories P. 0. Box 5800 - MS-1153 Albuquerque, NM 87185-1153 L.S. Camilli HDI Systems, Inc. 7424 2nd St., NW Albuquerque, NM 87007 ABSTRACT Spark ignition engines of practically all types and sizes utilize ignition systems that can develop up to about 30 Kv for breakdown of the spark plug gap and typically deliver current levels in the range of 30 to 100 ma. Measurement and analysis of these systems show that the energy transfer efficiency is very low, typically in the order of one percent (1%) or less. Once the spark is formed, most of the energy is delivered to resistances in the transformer, spark plug wires and spark plugs. This operating level is adequate (but not necessarily optimum) for older and most modern engines, but will not meet the needs of future lean burn and some alternate fuel engines that require higher energy discharges to effectively ignite the air/fuel mixtures^1,2,3,4. To meet the requirements for increased ignition power and energy, ignition systems will have to be designed to operate at higher transfer efficiencies. There are two basic approaches to increasing the electrical efficiency of ignition systems. The first is to utilize peaking capacitors across the spark plug⁵. The second is to design and utilize a more efficient discharge circuit in combination with a low resistance spark coil⁶. A third alternative would be to utilize a combination of both. By the first method, transfer efficiency can be increased to nearly fifty percent (50%). By the third, transfer efficiency can be seventy-five percent (75%) or more. This paper summarizes the results of recent high power ignition experiments and related analyses, for so-called "breakdown ignition" conditions. The paper includes descriptions of both conventional system up-grades and a new higher energy system that features multiple drivers and an energy recovery circuit. It does not include analyses or experiments for arc or glow discharge conditions. Introduction It is well known that power-enhanced ignitions extend the lean burn limits and transient response of spark ignition engines^4,7. These enhanced ignitions have been primarily of two types; plasma jet and double discharge. Experiments with such ignitions date to at least the early 1970's. More recently, experiments with certain low inductance peaking capacitors across the spark plugs have also demonstrated similar improvements plus lower hydrocarbon emissions and fuel consumption, particularly with older engines. Additional benefits include better starting and cold running performance and lower combustion pressure variability⁷. In contrast to other methods of ignition enhancement, the high voltage peaking capacitors dramatically increase the electrical-to-plasma conversion efficiency by delivering high current during the resistive phase of the spark discharge^8,9. Low inductance, low loss peaking capacitors have been under development by DirectHits^TM for a number of years^5,10 with assistance in testing and electrical characterization provided by Sandia National Laboratories under a Technical Assistance Agreement. Adapting them to an engine involves only installation of non-resistor spark plugs with the peaking capacitor placed directly over the spark plug insulator. This technique represents the simplest and most economical method of ignition enhancement. An alternate method of ignition enhancement involves utilizing a high efficiency capacitor discharge circuit to power a low resistance high voltage transformer. A prototype system of this type has been built and bench tested. Initial results have shown the highest transfer efficiency and the capability of delivering higher peak power than any other system under comparable operating conditions. Discussion A. Ignition Circuit Analyses It is useful to model the circuits and compare the energy transferred to the spark with varying amounts of resistance in the high voltage section of the circuit. Figure 1 shows a schematic of a typical capacitor discharge ignition system. Representative parameters for conventional and three different power enhanced ignitions are shown in TABLE 1. Note that Conventional ignition circuits have high resistance in the secondary side, often as much as 25 K ohms, which limits transfer efficiency to the spark to around 0.2%. By replacing the 5K ohm resistor plug with a non-resistor type, the transfer efficiency increases to 10% due to the 10 pF capacitance of the plug discharging directly into the spark. This simple exchange of components illustrates the function of a peaking capacitor. By increasing the capacitance to 80 pF the transfer efficiency increases to nearly fifty percent (50%). If the circuit analysis is repeated using a low resistance transformer, the transfer efficiency is consistently higher than with a standard spark coil. A comparison of transfer efficiencies using standard ignition coils and low resistance transformers and various circuit configuration is shown in Table 1. The highest transfer efficiency for any system results when there is virtually no resistance between the secondary capacitance and the spark and when low resistance rf suppression wire (50ohms/ft) is used. These experiments and analyses were for breakdown conditions where the ignition current was greater than or equal to 100 ma and a fully developed spark discharge occurred with a duration of nanoseconds to multi-microseconds. For fast discharge conditions, the resistive phase time and resistance change as a function of time may be estimated from Ristic and Sorenson⁹ formulas below: for where the units are as noted in the equations. Using these relations for ignition condition where These relations show that the gap resistance drops very rapidly and with R² <<4 L/C, the peaking capacitor circuit will ring down into the relatively low residual spark resistance. For efficient energy transfer the peaking capacitor must obviously have very low ESR and dielectric losses. With conventional ignitions where the pulse duration is comparatively long (tens to hundreds of microseconds), the resistive phase relations of Ristic and Sorenson are not directly applicable. For these conditions, low current arc voltages in the range of 50 to 100 ohms are a better approximation. Knowing the fixed resistances of the secondary circuit, reasonable transfer efficiency estimates can be made from a simple ratio of spark resistance to total secondary circuit resistance n=R_spk/R_tot. From this relation it is apparent that rather wide variations in arc resistance do not affect efficiency much. Since most automotive engines operate with resistor spark plugs, acceptable ignition of stoichemetric or richer air/fuel mixtures is typically achieved with very low spark energies. So long as the circuit resistance does not increase appreciably, the engines maintain high performance levels. However, aging of spark plug wires often produces resistance increases which gradually degrades mileage with a corresponding increase in hydrocarbon and carbon monoxide emissions. For late model engines with ignitions in good condition, the effect of high power ignition is often small. For older engines (pre-1980), however, high power ignitions can reduce hydrocarbon and carbon monoxide to ranges below 100 ppm and 0.5 percent, respectively, without a catalytic converter. Mileage increases are often ten percent (10%) or greater. B. Capacitor and Emissions Tests Prototype peaking capacitors provided by DirectHits^TM were tested both on the bench for their electrical characteristics and on a variety of automobiles. Automobile tests included chassis dynamometer tests at various load and speed points, road tests over a standard course and operational fleet tests conducted in a normal usage mode. A third type of test was a standard EPA 505 conducted on three late model vehicles at Southwest Research Institute (SWRI)¹¹. Bench tests primarily included high voltage insulation endurance tests, dielectric constant and dielectric loss measurements, all at temperatures to 300 degrees C. Figure 2 is a cross section of a DirectHits^TM capacitor attached to a spark plug. Figure 3 shows a typical current wave form from an 80pF capacitor charged to 18kV where the peak current was 1020A and the ring-down frequency 100 MHz. Emissions tests on a variety of vehicles were run at high and low-idle points of 900 and 2500 rpm followed by steady load tests on a chassis dynamometer at 1 hp. 40 mph, 5 hp. 40 mph, 10 hp. 40 mph, and 9 hp. 15 mph. Figure 4 is a plot of unburned hydrocarbon reduction relative to stock measured from tests on a 1987 Chevrolet S-10 pickup. It is typical of emission reductions for vehicles greater than four to five years old. Three vehicles less than one year old were tested by SWRI in accordance with EPA 505 test procedures¹¹. Results showed minimal change in emissions with the peaking capacitors. Fleet tests conducted on several groups of vehicles showed a net mileage gain in all cases. Typical of those were 10 State-owned vehicles tested by the New Mexico Transportation Department which, over a 6-month period, showed an average mileage gain of 12.8%¹². The lowest gain from fleet tests was a group operated by Coca Cola which resulted in a net gain of approximately 5%¹³. C. High Power, Low Resistance Transformer System A higher energy, low resistance pulse transformer system that has a four-step energy output control and delivers up to 22A of current to a non-resistor spark plug has been built and is currently being tested. The system was initially designed as a research device to study the effect of varying energy input to the ignition process. However, the basic transformer and drive circuit is readily adaptable to permanent installations. The energy increments are 85, 170, 255, and 340 mJ. Maximum output current at these steps is 11, 15.6, 19 and 22 A, respectively. Figure 5 shows four overlaid current traces, one for each energy level. Figure 6 is a schematic of a five stage version of the system. Each of the drive stages has an energy recovery circuit which reverses the polarity on its respective capacitor following SCR commutation at negative polarity. Figure 7 shows a full discharge cycle from anode voltage fall to thyristor commutation, energy recovery and final recharge. With a low resistance transformer and energy recovery, the multiple drive system represents the most efficient ignition source available, having a transfer efficiency greater than 75% when used with a circuit-matched peaking capacitor. Conclusions Enhanced power ignitions that deliver higher power and more energy than conventional ignition systems, have been shown to improve fuel economy and lower hydrocarbon emissions in many vehicles, particularly those more than five years old. Mileage improvements are typically in the range of 10% and HC emissions reductions around 50% within maximums and minimums well outside these ranges. Because the lean burn limits are extended, high power ignitions also improve starting, cold running and transient throttle response. Of the options available, high voltage peaking capacitors deliver the highest power and are the most economical for upgrading a conventional spark ignition system. Transfer efficiencies over 40% are possible. As an alternate to peaking capacitors or in combination with them, a high voltage pulse transformer with low resistive losses can deliver between 100 and 200 times the current (and power) to a spark that a conventional ignition system can. The high power transformer system with selectable energy output should provide researchers with a useful tool in studying the wide range of conditions necessary to optimize the ignition process. REFERENCES G.F.W. Ziegler, R.R. Maley, and E.P. Wagner, "Effect of Ignition System Design on Flammability Requirements On Ultra-Lean Turbulent Mixtures", I.Mech.E., paper no. C47/83, 1983. R. Maly, B. Saggau, E. Wagrner, and G. Ziegler, "Prospects of Ignition Enhancement", SAE Paper 830478, Feb., 1983. R.W. Anderson, "The Effect of Ignition System Power On Fast-Burn Engine Combustion", SAE Paper 870549, Feb., 1987. R.W. Anderson and J.R. Asik, "Lean Air-Fuel Ignition System Comparison In a Fast-Burn Engine", SAE Paper 850076, Feb., 1985. R.P. Corio, J.E. Griswold, R.E. Hensley, R.C. Pate, Glen B. Smith, "Hand Discharge Ignition (HDI)", Hensley Co. Report, Albuquerque, NM, 1984. M.A. Ward, "A New Spark Ignition System For Lean Mixtures Based on a New Approach to Spark Ignition", SAE Paper 890475, Feb., 1989. R. Maly, "Spark Ignition: Its Physics and Effects on the Internal Combustion Process", published in: Fuel Economy in Road Vehicles Powered By Spark Ignition Engines, Hilliard, J.C. and Springer, G.S. (Ed.), Plenum, New York, 1984. J.C. Martin, "Duration of the Resistive Phase and Inductance of Spark Channels", AFWL Switching Note 9, 1965. T. P. Sorensen and V.M. Ristic, "Rise Time and Time-Dependent Spark-Gap Resistance in Nitrogen and Helium", J. Appl. Phys., V. 48, No. 1, pp. 114-117, 1977. R.C. Pate, R.E. Hensley, "Combustion Initiation System Employing Hard Discharge Ignition", U.S. Patent No. 4,589,398, May 20, 1986. M.S. Newkirk, Exhaust Emission Testing of Vehicles Using a Hard Discharge Ignition System, SwRI 08-6295, April, 1994. Letter, R. A. Leonard, New Mexico Highway and Transportation Department, to L. Camilli, Combustion Technology Products Corp, Subject: Results of Swaser Ignition Fleet Tests Conducted by New Mexico State Highway and Transportation Department dated April 5, 1993. Private communication, L. Camilli to G.J. Rohwein, Subject: Results of Coca Cola Fleet Tests, October, 1994. Index of Charts and Tables Capacitor Discharge Ignition System Circuit Configuration Parameters Peaking Capacitor Ringdown Hydrocarbon Emission From 1987 Chevrolet S-10 Pickup Overlaid current traces from four stage multi-drive ignition 5-stage multi-drive ignition Discharge cycle with energy recovery ©1998 & 1999 by Combusiton Technology Products Corp

Advanced Automotive Ignition System

G.J. Rohwein, S.R. Babcock and M.T. Buttram Sandia National Laboratories P. 0. Box 5800 - MS-1153 Albuquerque, NM 87185-1153

L.S. Camilli HDI Systems, Inc. 7424 2nd St., NW Albuquerque, NM 87007

ABSTRACT

Introduction

Discussion

Conclusions

Index of Charts and Tables

G.J. Rohwein, S.R. Babcock and M.T. Buttram
Sandia National Laboratories
P. 0. Box 5800 - MS-1153
Albuquerque, NM 87185-1153

L.S. Camilli
HDI Systems, Inc.
7424 2nd St., NW
Albuquerque, NM 87007