Emc Fig1

Emissions Measurements for Alternative Powertrain Vehicles

Increasing demand for vehicles with alternative powertrains prompted this in-depth investigation into today�s automotive emissions standards.

Vehicle manufacturers must satisfy legislative EMC requirements that include radiated emissions measurements for the 30-MHz to 1,000-MHz band. The test methods currently used to validate emissions performance were developed for vehicles powered by internal combustion engines. However, increasing demand for vehicles with lower levels of exhaust emissions and reduced reliance on fossil fuels now is driving the development of alternative powertrains.

Alternative schemes are achieved by introducing electric motors and drives into the powertrain, which then can be driven from a variety of primary power sources. The power sources that have been investigated include batteries, fuel cells, and hybrid schemes that use an internal-combustion engine optimized in terms of efficiency and exhaust emissions. Recent work carried out by the Motor Industry Research Association (MIRA) investigated whether the existing automotive emissions standards need further development to accommodate the particular features of these alternative powertrain schemes.

Current Measurement Practices
Current automotive radiated emissions test requirements, as described in CISPR 12, are based on broadband and narrowband emissions measurements in the 30-MHz to 1,000-MHz band using an antenna at a fixed point relative to the vehicle.1 Unlike many other emissions testing standards, such as CISPR 22, there is no requirement for height scanning of the antenna and rotation of the test object to identify peak emissions.2

The most common test configuration is an open area test site (OATS) or a semi-anechoic chamber (SAC) with an antenna mounted 3 m high at a distance of 10 m from the side of the vehicle and aligned with the front axle. Sweep times can be significant for emissions testing, even without the requirement to identify the direction of maximum emission, so a broadband antenna normally is used to minimize test disruptions.

Broadband emissions are classified as signals with bandwidths that exceed the receiver bandwidth and have pulse repetition frequencies smaller than the receiver bandwidth. Narrowband emissions are classified as signals with bandwidths smaller than the receiver bandwidth and pulse repetition frequencies greater than the receiver bandwidth. For automotive applications, both broadband and narrowband measurements are made using a receiver bandwidth of 120 kHz.

Broadband measurements are performed using a quasipeak detector with the engine running at a constant speed of 1,500 rpm primarily to detect emissions from the spark ignition system. Narrowband measurements are carried out using an average detector with the vehicle switched on but without the engine running to detect emissions from onboard electronic modules. The purpose of the legislative limits is to minimize potential disruption to external radio-based services due to vehicle emissions.

The scope of CISPR 12 includes vehicles propelled by electrical means. However, powertrain systems based on electric, hybrid-electric, and fuel-cell power sources are not fully active under conventional (static) emissions test conditions. Furthermore, the static condition will not fully exercise the electric powertrain components of hybrid systems based on internal combustion engines. As a result, successful evaluation of the electromagnetic emissions from alternative powertrain vehicles is likely to require a dynamic test mode to mimic realistic operating conditions.

Vehicle and Test Configurations
A number of representative alternative powertrain vehicles were selected for test, including the following:

� Two different electric vehicles.

� Two different hybrid electric vehicles.

� A fuel-cell van.

� An electric bus.

� An electric step-through motorcycle.

Measurements also were performed on two representative conventional vehicles to compare the results from alternative powertrain systems with more conventional technology.

Most of the measurements were made in a SAC with a working volume of 22 m � 10 m � 9 m. This chamber is equipped with a variable-wheelbase dynamometer that allows the vehicles to be exercised under realistic load conditions without physically moving.

Figure 1. Alternative Powertrain Testing in a SAC

Conventional emissions measurements at a 10-m range are possible in this chamber since the dynamometer is not required. However, for dynamic testing, it was necessary to align the antenna with the vehicle axis to use the dynamometer (Figure 1). In this configuration, the distance to the antenna was measured from the front axle of the vehicle.

The electric step-through motorcycle could not be safely mounted on the dynamometer so it was secured on its stand and operated at a fixed throttle position with the driven wheel raised above the ground. Measurements then were made with the vehicle freewheeling at a constant speed to provide a dynamic test condition.

Measurements also were performed on an OATS equipped with a 40 m � 12 m ground plane.

Measurements Above 30 MHz
Broadband measurements (Figure 2a) and narrowband measurements (Figure 2b) on the alternative powertrain vehicles were made at 10-MHz intervals for frequencies up to 200 MHz. Results from the two conventional vehicles indicated no excessive broadband emissions, although high-level narrowband emissions were detected from these vehicles while switched off.

One of the hybrid vehicles also was tested in the OATS in the conventional manner, except that data also was recorded for the front antenna position as well as for both sides of the vehicle (Figure 3a). A similar exercise using the freewheeling mode was carried out for one of the electric vehicles (Figure 3b). These results provide an indication of the correlation that can be expected between conventional and dynamic tests and among different antenna configurations.

A conventional broadband test of the hybrid vehicle would have indicated excessive emissions only over a very narrow band around 40 MHz. Additional measurements from the front of the vehicle would show further excessive emissions for frequencies in the 80-MHz to 100-MHz band.

However, the dynamic SAC test also revealed high levels of emission around 30 MHz and in the 50-MHz to 60-MHz band. Conventional methods that specify a constant engine-speed condition are not applicable for electric vehicles. However, the dynamic tests show that electric vehicles do generate emissions.

All but one of the alternative powertrain vehicles tested exceeded the statutory limits, primarily for broadband emissions at frequencies up to about 127 MHz. However, some excessive narrowband emissions were observed below 144 MHz. The results were consistently poorer for vertical polarization than for horizontal and for the front antenna position than for the sides (although the latter was only possible for tests carried out on the OATS).

Measurements Below 30 MHz
As very high emissions levels were found in the lower part of the conventional test band, further measurements were made in the 9-kHz to 30-MHz band. These measurements were conducted using the OATS since the absorbing lining of the SAC is not fully effective at these frequencies.

No emissions were discernible below 30 MHz from either of the conventional vehicles. However, all of the alternative powertrain vehicles appeared to show evidence of some slight increases over ambient levels (2 to 3 dB). Problems with high and fluctuating ambient levels made reliable measurement impossible at the 10-m range although 3-m measurements were more successful. Nonetheless, the fact that all of the alternative powertrain vehicles showed this behavior while neither of the conventional vehicles did suggests that the effect is real.

Practical Measurement Issues
The existing approach to measuring narrowband emissions generally is adequate although the more complex modes of operation that may be found in some alternative powertrain vehicles could require additional development. For example, possibly some of the control systems for hybrid electric vehicles are not activated until the powertrain is active.

Broadband measurement practices require more significant rework since the static test that currently is used to verify the acceptability of spark-ignition engines generally is inappropriate for vehicles using electric motors and power electronics. These components are significant sources of electromagnetic emissions but not fully operational under static conditions, even for hybrids based on conventional engines.

Some automotive test chambers are equipped with dynamometers for dynamic testing of vehicles under radiated immunity conditions. Most of the measurements reported in this study have used such a facility since this represents the most satisfactory approach for dynamic measurements.

In many such facilities, however, the geometry of the chamber will preclude the use of the dynamometer with an antenna placed at 10 m from the side of the vehicle. Results obtained from the OATS, where both front and side positions can be accommodated, show that higher levels are measured at the front rather than at the side. For one vehicle, however, a band of emission was present at the side but not at the front.

The energy capacity of the electric vehicles was so limited that the dynamometer loading had to be set at an unrealistically low loading to ensure that the vehicle would get through the test cycle without discharging the traction batteries. These vehicles are designed for typical urban missions, which are much less arduous than the typical 8 hours of continuous use required for emissions tests. This problem also affected the fuel-cell hybrid vehicles.

Freewheeling operation consumes much less energy but is less representative of real-world conditions than the use of a dynamometer. Another problem that arises in the testing of hybrid electric vehicles based on internal combustion engines is the potential for switching between electrical and mechanical power sources during the course of an emissions sweep. This could lead to changes in the emissions profile as the operating mode fluctuates.

Conclusions
Measurements have been made on a representative sample of alternative powertrain vehicles using both conventional methods and dynamic schemes that exercise the powertrain components. These dynamic tests revealed emissions that would not have been detected using standard test methods. Based on this experience of testing a representative range of alternative powertrain vehicles, a number of recommendations for changes to current test standards have been proposed.3

The energy capacity of the electric vehicles and to a lesser extent the hybrid electric vehicles was found to represent a significant practical problem for emissions testing under dynamic conditions. Nonetheless, a dynamic testing approach is essential to make a realistic assessment of the emissions from the electric motors and associated power electronics deployed in alternative powertrain vehicles. Much faster time-domain emissions measurement methods, currently the subject of research, ultimately may offer a more practicable solution.

The high levels of emissions observed around 30 MHz also prompted investigation of emissions down to 9 kHz. At present, there is no legislative requirement to measure electromagnetic emissions from vehicles below 30 MHz. However, if this does become of interest, a 3-m measurement configuration is likely to be necessary to make reliable measurements.

The results of this work have been presented to CISPR 12, and possible amendments to account for the particular features of alternative powertrain vehicles are under discussion. Although changes to the European Automotive EMC Directive 95/54/EC also are under consideration, currently proposed amendments do not specifically address emissions test methods for alternative powertrain vehicles.4

Electric vehicle charging systems also were briefly investigated in connection with this project. It has been agreed by the CISPR Steering Committee that CISPR 11 will become the relevant standard for these systems although this has yet to be reflected in current stand-ards.5 Vehicle charging systems are considered to fall under Class B of CISPR 11. This standard requires height and azimuth scanning unlike automotive emissions measurement methods.

All but one of the alternative powertrain vehicles tested exceeded the statutory limits, with excessive emissions found at frequencies up to 127 MHz for broadband and 144 MHz for narrowband. Nonetheless, the fact that one of the hybrid vehicles met the legal requirements demonstrates that such vehicles do not represent any more of a threat to radio-based services than conventional vehicles with mechanical powertrains.

References
1. CISPR 12, Vehicles, Motorboats and Spark-Ignited Engine Driven Devices. Radio Disturbances Characteristics. Limits and Methods of Measurements, 1997.
2. CISPR 22, Limits and Methods of Measurement of Electromagnetic Disturbance Characteristics of Information Technology Equipment, 1999.
3. Ruddle, A.R., Topham, D.A., and Ward, D.D., �Recommendations for Emissions Testing of Electric, Hybrid Electric, and Fuel Cell Vehicles,� Proceedings of 5th European EMC Conference, Sorrento, Italy, September 2002, vol. 2, pp. 715-720.
4. Commission Directive 95/54/EC, Official Journal of the European Communities, No. L 266, Nov. 8, 1995, pp. 1-66.
5. CISPR 11, Limits and Methods of Measurement of Radio Disturbance Characteristics of Industrial, Scientific, and Medical (ISM) Radio-Frequency Equipment, 1997.

About the Authors
Dr. Alastair Ruddle received a Ph.D. from Loughborough University. Since joining MIRA in 1996, his research has included electromagnetic modeling for vehicle applications, development of electromagnetic measurements, and analysis of issues relating to system and software safety. e-mail: [email protected]

Debra Topham joined MIRA in 1996 after completing an MSc at the University of Hull. During this time, she has carried out a range of research activities but has a particular interest in vehicle telematics and radio systems. She currently is working on a Ph.D. at Birmingham University.

Dr. David Ward received a Ph.D. from Nottingham University before joining MIRA in 1991. Since then, his work has included the development of automotive EMC test methods, electromagnetic modeling for vehicle applications, and the analysis of functional safety issues. He chairs the Motor Industry Software Reliability Association Consortium.

MIRA, Advanced Engineering Department, Electrical Group, Nuneaton, Warks, CV10 0TU, UK, 011 44 2476 355551.

FOR MORE INFORMATION


For the full report on which this article is based, �Investigation of Electromagnetic Emissions From Alternative Powertrain Road Vehicles,� by A.R. Ruddle, MIRA Project Report No. 01-845060,

www.rsleads.com/409ee-215

September 2004

Sponsored Recommendations

Comments

To join the conversation, and become an exclusive member of Electronic Design, create an account today!