|
Testing Times - The Complexities of Validating the Safe Operation of Automotive Systems
By Peter Phillips and Dr Anthony Martin, MIRA Ltd
Abstract: The modern road vehicle has evolved in parallel with technology, bringing with it: Anti-lock Braking Systems (ABS), Traction Control Systems (TCS), Adaptive Cruise Control (ACC), smart rain sensors, Electronic Park Brakes (EPB) and Remote Keyless Entry (RKE). Modern day technology has also been implemented in vehicles to ensure that fuel consumption is kept to a minimum using complex powertrain control modules (PCM), and that safety is of paramount importance with a typical Supplementary Restraint System (SRS) comprising in excess of 8 Electro Explosive Devices (EED) used to deploy airbags. The modern road vehicle is becoming increasingly complex and the correct function of the electronic systems built into these vehicles is essential at all times. In recent years, systems that were options only on prestigious vehicles in the past are now becoming standard on basic vehicles, the most recent being the introduction of Anti-Collision Radar (ACR) on Fiat Stilo model vehicles.
To ensure that modern road vehicles operate safely, comprehensive EMC testing is required to prove that all vehicle systems are robust within the electromagnetic (EM) environment in which they must operate. Simulation of real world vehicle modes of operation and real world threats whilst limited to testing vehicles inside an EMC chamber along with the task of accurately quantifying system failure mechanisms and levels are the challenges facing all vehicle manufacturers and EMC test facilities.
Vehicle EMC Testing To prove the basic operation of any product within its intended EM environment is a relatively simple process. When testing the basic functions of a vehicle, the vehicle may be positioned on a dynamometer inside an EMC chamber and radiated with Radio Frequency (RF) energy. This could highlight a large number of events, from engine stalls to airbag deployments. However, whilst basic information is gained about the possible susceptibility of the Powertrain Control Module or Supplementary Restraints System, the real challenge is to determine:
If the PCM and SRS have been tested in a representative mode of operation
If any further modes of operation should be tested to ensure safety under all real world conditions
If any further threats exist such as on-board transmitters (mobile phones, laptops or radio equipment)
Finally, the actual RF levels at which events occur must be quantified in order to give vital information about a systems actual susceptibility. For example, an airbag attached to a vehicles SRS may be susceptible to RF levels above 98 V/m, and may pass an EMC test at 100 V/m, however only a 2% confidence margin can be associated with the system, and in order to know that confidence margin, the level of susceptibility must be assessed. It is to this end that MIRA continually research and implement advanced vehicle test methods to ensure the safe operation of vehicle systems within their intended RF environments and to gain vital information about a systems event thresholds.
This paper presents some of the most common vehicle systems, examining the methods by which these systems are verified during EMC tests. It should be noted that all of the subsequent EMC tests are performed in semi-anechoic chambers where the environment prevents human interaction with the vehicles under test.
Braking Systems One key vehicle safety system is the brakes, now moving away from servo assisted systems to electronic control, the braking system must be reliable at all times. A standard RF immunity test will prove the brakes are not actuated when an RF field is applied, however, ensuring accurate operation of systems such as ABS and TCS during RF illumination is more complex. MIRA have been testing ABS systems for 10 years and the test method used is now refined to ensure accurate ABS operation at each frequency tested.
To perform an ABS test, MIRA disconnect the brake callipers at the wheel and connect the brake pipes to pressure transducers. The vehicle is then driven at speed by the dynamometer with the brake pedal depressed (the disconnected callipers allow the wheels to rotate freely).
The pressure drop on the brake system is monitored throughout the test (see actual ABS test data illustrated in Figure 1).
Figure 1: Actual ABS test data
Other brake system functions can be tested in a similar manner. TCS and electronic brake distribution for stability control are all monitored using the same method (example data is illustrated in Figure 2). For stability programs the steering angle sensor is removed from the steering column and rotated by pneumatic controllers to recreate a steering motion, the yaw rate sensor is also moved by pneumatics to simulate the forces on the vehicle when stability is lost. The braking system applies pressure to various wheels to compensate and assist the driver regain control whilst pressure transducers monitor these variables and report safe operation of the system.
Figure 2: Actual TCS test data
Electronic Park Brake (EPB) systems are partially verified during either Static or Dynamic test modes ensuring that the EPB system is not falsely activated under RF conditions. This test requires no intervention, and the EPB system is inactive during the test period. Finally, the EPB system is monitored during an EPB test mode, to ensure that the brake is activated when commanded under RF conditions. During the EPB test mode, the dynamometer is set to drive the vehicle at 2 km/hour. The rear of the vehicle is raised (using non-conductive blocks) to a height that allows enough friction from the dynamometer to spin the wheels but when the park brake is applied allows the rear tyres to slip on the dynamometer with minimum resistance when the rear wheels are locked. A pneumatic robot controller is used to operate the EPB switch in the vehicle, applying the park brake during RF signal modulation and releasing the brake during the Carrier Wave (CW) dwell. A six second dwell is required to allow full EPB operation, and also to ensure that the park brake motor does not overheat. Monitoring is performed by visually confirming the wheels lock and released during EPB application and release. In addition the cluster warning light is monitored visually and the EPB motor is monitored audibly.
Cruise Control Systems Cruise control testing and forward facing radar for collision avoidance are regularly tested. A standard cruise control test will examine the scenario where, with cruise control engaged, the effect of RF energy can cause speed changes or cruise control dropout. A standard RLS test performed on a vehicle with cruise installed can determine if it is possible to engage cruise control, with adaptive cruise control the challenge is in verifying that the adaptive system will act as intended during RF illumination whilst a moving vehicle is detected in front of the vehicle. Changing the ‘calibrated’ safe distance permits testing of these systems in an EMC chamber. A moving target is positioned in front of the vehicle and is “hidden” by a piece of RF absorber this target is then moved to simulate a vehicle. A test program can be performed that ensures the vehicle slows as a vehicle pulls in front then decelerates as the vehicle in front slows down, and finally verifies the vehicle accelerates when the “car” moves from the front of the vehicle under test. This program can be repeated at each frequency and test signal modulation defined in the EMC test plan.
Supplementary Restraint Systems All modern vehicles are supplied with supplementary restraint systems of varying complexity. However, all restraint systems use the same basic components — an electronic control module connected to an airbag containing an Electro-Explosive Device (EED) via a length of harness. By fitting EMC test vehicles with these systems, it is verified that the supplementary restrains do not unintentionally deploy within their intended EMC environment when the vehicle is under non-crash conditions.
It is known that two main unintentional deployment mechanisms exist for supplementary restraint systems: they will deploy when sufficient RF energy is induced into the wiring harness to which the airbag is connected which will in turn heat the airbag deployment mechanism (EED), or when the restraint control module is falsely triggered and deploys the airbag. Testing the first of these scenarios brings with it a major challenge, as normal tests using live airbags are destructive. Due to the inherent nature of airbags, once the integral EED has deployed, no further information can be gained from the device.
Unintentional Deployment A detailed examination of a vehicles supplementary restraint system reviles a small component within each airbag known as an EED (illustrated in Figure 3). Each EED utilises a small resistive bridgewire that heats when an intended deployment current is applied. Most vehicle airbags work on similar principles, many using common components shared by several vehicle manufacturers. The pyrotechnic unit itself (i.e. the bridgewire housed in the EED) heats the primary explosive to deploy the airbag in question.
Figure 3: Side elevation - typical EED unit
However, when the small EED is connected to a length of harness, the circuit will act as a receptor of RF energy under the appropriate conditions and inadvertent deployment of the pyrotechnic device will be achieved when sufficient AC current flows through the EED bridgewire (without physical connection to a power supply).
It is clear that in order to assess the risk of inadvertent deployment accurate measurements must be made of bridgewire temperature rise due to RF energy using instrumented EED’s with the primary explosive removed. By measuring the temperature rise of EED bridgewires during EMC tests, a level of confidence can be obtained denoting how close to deployment – typically 250mA of bridgewire current - an EED reaches whilst being subjected to RF energy. This type of testing is non destructive, and the instrumented EED probes may be reused.
Historically temperature measurements have been performed with thermocouples but more recently two fibre optic techniques have been devised. Using an advanced fibre optic test method, MIRA monitor the bridgewire current of EED’s using sophisticated fibre optic temperature probes (Figure 4) whilst a vehicle is subjected to its normal immunity test regime.
Figure 4: Instrumented airbag EED
The values extracted during a normal immunity sweep are illustrated in Figure 5. Analysing this data at each frequency tested and using detailed calibration data allows an accurate assessment of bridgwire current values to be obtained that describe:
IM The magnitude of induced current due to modulated signals ICW The magnitude of induced current due to carrier wave signals
Figure 5: Bridgewire temperature measurements
This type of assessment not only allows vehicle manufacturers to obtain crucial information regarding the integrity of the supplementary restraint systems with regards to unintentional deployments due to bridgewire heating, allowing this information to be obtained in a cost effective manner.
Figure 6: EED bridgewire current – vehicle tested at 100 V/m
From the data presented in Figure 6 it is observed that high currents are observed in and around the GSM900 mobile phone band which are close to the no-fire current limit, indicating a potential risk of unintentional deployment if these types of devices are brought into close proximity of the vehicle SRS. Under normal test conditions (without bridgewire current monitoring) this risk would have gone unnoticed, as the airbag would not have deployed.
Assuming that the instrumented EED bridgewires are not damaged during the EMC tests, confidence is also gained as to the immunity of the restraints control module to false triggering due to RF energy.
However, whilst these assessments give confidence that unintentional deployments will not occur, they do not assess if the restrains control module is capable of intentionally deploying an airbag when it is subjected to its intended RF environment.
Correct Operation of the Supplementary Restraints Module We have devoted a significant portion of this section to unintentional deployments, however, there still exists the scenario of a restraint control module not sending an intentional deployment pulse when a crash occurs due to corruption of its internal circuitry under RF conditions. This scenario presents a significant risk to the safe operation of an automotive vehicle, and could result in a catastrophic event.
This type of assessment requires specific diagnostic software to be written for the EMC test period and at present the correct deployment operation of a restraint control module is only tested at component level. This has the effect of reducing the confidence associated with the correct operation of the restraint control module at full vehicle level, as the influence of vehicle wiring harnesses, screening and the resonances of the vehicle cavity will all have an effect.
Rain Sensors Rain sensing is now a common fitment to most automotive vehicles, luxury or standard, and are categorised as safety critical, as they reduce driver distraction by removing the manual operation and speed setting of the wipers. The incorrect operation of this system under RF conditions can cause driver distraction, and in the most severe cases could cause a collision if the wipers were not activated in severe weather conditions.
Rain sensors measure the amount of water hitting the outside surface of the windshield. This is typically accomplished by shining beams of light at approximately 45 degrees to the windshield, this light refracts back to the sensing aperture from the outside surface of the windshield. Under wet conditions, the light pattern refracted back to the sensor is disturbed. The rain-sensing module utilises this information to calculate and set the speed of the wipers.
Two scenarios exist for the incorrect operation of rain sensors under RF conditions, where by corruption of the rain sensing module may falsely trigger the wipers under dry conditions or triggering of the wipers may not occur under wet conditions. The first scenario can be validated during either Idle or RLS immunity tests and false activation of the vehicle wipers can be recorded and thresholds taken. The second scenario however, requires the simulation of wet weather conditions.
As the rain sensing system is activated by the disruption of the refracted light beams returning to the rain-sensing aperture, it is not in this case necessary to use water to activate the sensor. MIRA has researched the operation of typical rain sensors, and has developed a method based on disrupting the rain sensor light beams using a pneumatic controller to repetitively move a cloth over the rain sensing area of the windshield. This recreates similar stimuli to that of heavy rain droplets hitting the windshield. The vehicle may be instrumented with the previously described pneumatic controller, and the wiper control positioned to auto during an ABS immunity mode for example, allowing this system to be tested on the back of other EMC tests.
Remote Keyless Entry The benefits of keyless entry are numerous for vehicle owners. In bad weather conditions vehicles can be entered quickly, the vehicle trunk can be accessed easily when carrying heavy items, and numerous safety features are built into the system – for example all vehicle doors are automatically locked when the key is turned to the ignition on position, keeping unwanted intruders out of the vehicle.
However, if this system fails under RF conditions, it can be a nuisance to the owner and in cases where intended signals are blocked by nearby transmitters, can render a vehicle unusable. It is to this end that MIRA has developed and implemented a test method for validating RKE systems.
The RKE system comprises 2 parts – the Keyless Vehicle Module (KVM) and the hand held key fob. When the door handle is pulled, the KVM transmits a challenge signal to the fob at 125KHz and the key fob responds at 433 MHz. There exists 2 modes of operation that must be validated for correct operation with RKE systems, these are: keyless entry mode and keyless start/stop mode.
Keyless entry mode is used to ensure that doors can be locked and unlocked in the presence of RF energy. Within this test mode, the closed key fob is placed on a 1m high wooden stand, 30 cm away from the drivers door handle. A pneumatic actuator is attached to the driver’s door, and the actuator performs 2 actions as commanded by the test software:
· Pull the door handle to unlock the door · Push the lock button to lock the vehicle
It is important to note that a 433 MHz notch filter is installed between the power amplifier and the radiating antenna to prevent any spurious electromagnetic noise from the amplifiers blocking the RF signals from the key fob during all RKE tests.
The vehicle is secured to the dynamometer, with the engine off and all doors closed. An RF field is applied at every frequency under investigation, and when the field is stable a signal is sent from the equipment controlling the test to the pneumatic equipment
· To lock the doors · To unlock the doors
Figure 7: Relationship between RF field, door actuator and door status
Keyless start/stop mode is used to verify the safe operation of the ignition key functions under RF conditions. The following functions are tested:
Engine can be cranked The steering column lock can be locked and unlocked as commanded The steering column lock does not inadvertently lock or unlock
The vehicle is tested in static mode with all doors locked from inside by operating the internal lock handles. The key fob is located on the centre of the driver seat cushion and a robot arm is installed on the steering wheel column to pull and turn the ‘flyer’ under the command of the test software. The vehicle starter is disconnected, and the starter supply voltage is monitored to indicate engine crank enable, finally, the clutch pedal is depressed with fixed rod.
The vehicle is secured to the dynamometer with the engine off and all doors closed. An RF field is applied at every frequency and signals are sent from the test equipment:
· To push ‘flyer’ to release steering lock and turn to crank position · To turn ‘flyer’ to off position and pull flyer to lock steering
It should be noted that a 10 second RF dwell is required before each ignition on cycle to allow for key re-read by KVM to occur.
Figure 8: Relationship between RF field, key actuator and door status
Challenges for The Future In the future, telematics technologies including radar proximity sensors, vehicle-to-vehicle communications and on-board Doppler sensors for absolute speed measurements will allow a greater capacity of vehicles to use the existing road networks, with vehicles having the ability to travel at set distances from other road users safely. These technologies along with Night Vision Systems (NVS) and advanced crash avoidance systems will increase the complexity of automotive vehicles further.
With an increase in the complexity of the electronic content of vehicles, will also come changes in the shape of future vehicles and the materials with which they are manufactured. These will have the effect of changing the resonant structure of the vehicle shell. New in vehicle communication methods including significant moves towards data buses and alternative harness types will also have an impact not only on the way in which vehicle systems behave under RF conditions, but also the way in which the correct operation of systems are monitored.
In conclusion the future of automotive EMC testing is set to become significantly more complex, and only through continual research, validation and application of advanced test methods can we expect to ensure automotive vehicle safety within the increasing EM spectrum within which they must operate.
Peter Phillips MIRA Ltd, Watling Street, Nuneaton, Warwickshire CV10 0TU, peter.phillips@mira.co.uk
Dr Anthony Martin MIRA Ltd, Watling Street, Nuneaton, Warwickshire CV10 0TU, anthony.martin@mira.co.uk
|
