Event Data Recorders (EDR) and the importance of evaluating the data with the available physical evidence
This expert has retired from practice
TASA ID:
Event Data Recorders (EDRs) are part of Airbag Control Modules, PowerTrain Control Module, Electronic Control Modules or after-market products designed to capture crash- related data from non-commercial passenger vehicles and commercial motor vehicles traveling on the highways today. We all have heard the term "black box" from airplane crashes, and some refer to these non-commercial and commercial motor vehicle EDRs as "black boxes," which would be incorrect as the data captured from motor vehicle EDRs is very limited as compared to an airplane EDR.
Regarding passenger vehicles, in 1992, driver side airbags were mandated in passenger vehicles; however, there were several passenger vehicle manufacturers that were already installing them in their vehicles, and the progression of airbag technology continued to make them safer. In order to make them safer, vehicle manufacturers had to determine how effectively the airbags functioned during a collision, so as a secondary function of the Airbag Control Module (ACM), an additional function was established to record information about the crash. This recording function was called an Event Data Recorder (EDR), and that information could be retrieved only by certain employees within the vehicle manufacturer's employment or contractors retained by the vehicle manufacturer. In 1990, General Motors selected Vetronix to develop the first EDR retrieval unit and later was awarded an exclusive contract to develop the Crash Data Retrieval (CDR) System for GM and the aftermarket. March 2000, Vetronix, which is now owned by Bosch, made the CDR system available to the public. Since its public availability, the CDR system has evolved to include a large percentage of non-commercial passenger vehicles on the highway today, and Bosch now provides training for individuals wishing to become CDR Technicians and Data Analysts.
In August 2006, National Highway Traffic Safety Administration (NHTSA) established regulation NHTSA 49 CFR Part 563. This regulation in part specified the uniform requirements for the accuracy, collection, storage, survivability, and retrievability of onboard motor vehicle crash event data in passenger vehicles equipped with EDR manufactured on or after September 1, 2012. It also listed common data elements which needed to be captured and recorded, as in: speed, throttle position, service brake (on/off) for 5 seconds before the wake-up event and change in velocity (Δ V) for a time duration between 150 to 300 ms (milliseconds) during the impact. A current proposal to regulation 49 CFR Part 563 will establish a new safety standard mandating the installation of EDRs in most light vehicles manufactured on or after September 1, 2014. This means that nearly all non-commercial passenger vehicles manufactured after September 1, 2014, will be equipped with an EDR.
A motor vehicle EDR is designed to continuously capture data snapshots from different vehicle systems as the vehicle is driven. When the ACM's algorithm (an algorithm is expressed as a finite list of well-defined instructions for calculating a function) is triggered or awakened due to a collision, the EDR holds onto the data, and if the airbags are deployed and there is no interruption in the power supply, then the data is generally locked and cannot be re-written. However, if the airbags are not deployed or there was a power interruption during the airbag deployment, the EDR data is unlocked where it can be written over after 250 ignition cycles or from another more significant event. The CDR system is designed to image the information or take a picture of the data recorded by the EDR without the ability to alter the data.
This EDR data, as I tell jurors, is only as good as the paper it is printed on. The EDR data cannot tell how a collision took place or if the speed recorded by the EDR was the actual speed of the vehicle. Was the vehicle sliding on ice or was it rolling over? The collision reconstructionist must be critical of the EDR data, and if the data is accurate, then use the EDR data as supporting evidence to the collision reconstruction analysis and opinions.
I recently testified in a trial and a deposition as an EDR expert for two different cases involving EDRs. In the first case, the plaintiff expert testified that he only reviewed the two EDR reports of the two involved vehicles and no other data, i.e. crash scene photographs, photographs of the vehicles or the police investigation report. That is faulty analysis and I will explain why.
The background of the collision was as follows: It was a late night collision with no street-lighting on a two- lane, same direction, level roadway. The right lane was barricaded and closed to vehicular traffic due to road construction. Construction vehicle #1 was perpendicular to the roadway with part of its equipment extending a few feet into the left lane and approximately 4 feet above the roadway. With Vehicle #2 leading and Vehicle #3 following in the left lane approaching Vehicle #1, Vehicle #2's right headlight area impacted Vehicle #1's equipment. The piece of equipment tore through the fender and hood of Vehicle #2 and stopped intruding into Vehicle #2 when it struck Vehicle #2's firewall. At that time, Vehicle #1 had begun to rotate clockwise as Vehicle #2 still proceeded forward. At some time, Vehicle #3 impacted the rear of Vehicle #2, and all the vehicles came to rest. Vehicle #1 rotated approximately 90 degrees clockwise from its original position, Vehicle #2 rotated clockwise a little over 90 degrees from its original positioning, and Vehicle #3 rotated approximately 20 degrees counterclockwise from its original heading.
The investigating officer imaged the EDR data from both Vehicles #2 & 3 as part of the fatality investigation of the collision. (Click here for Figure #1) Figure #1 was Vehicle #2 EDR Report, and Figure #2 (Click here for Figure #2) was Vehicle #3 EDR Report. I later was retained by one of the attorneys to conduct an analysis of the CDR data. I requested the crash scene photographs, which included photographs of the involved vehicles and the police investigation/reconstruction report which were received and reviewed.
A review of Vehicle #2 EDR data shown in figure 3 (Click here for figure #3) revealed several issues. 1) The airbags did not deploy during this collision. 2) Two associated events were not recorded, and they occurred before the recorded event. 3) The recorded event was a positive acceleration event. 4) Was the pre-crash data associated with this collision? (Figure 3)5
The review of Vehicle #3 EDR data shown in figure #4 depicted the following data. 1) Driver's airbag was deployed. 2) There was pre-crash data. 3) The deployment event was recorded as a negative acceleration event. (Click here for Figure 4)
The plaintiff expert testified that he did not review the police investigation file or photographs and that the event recorded by Vehicle #2 EDR was triggered by the impact with Vehicle #1. Then 20 ms later, it recorded the rear impact from Vehicle #3. There are two major inconsistencies with the analysis. 1) As Figure 3 depicts, there was no acceleration in the negative direction during the 20 ms before the rear impact, which there would have been from this impact with Vehicle #1. 2) There were two events not recorded associated with this collision that took place before the recorded event. According to plaintiff expert, the event recorded was the impact with Vehicle #1, which all by itself was a faulty opinion. The collision between Vehicle #1 and #2 was a "soft" or "long duration" in the sense as an ACM would see it. Only sheet metal and plastic were compromised during the first 30-60 ms of this collision, which the ACM's algorithm pre-determined as "not severe"; therefore no airbag deployment. An analysis of the photographs revealed that the 1st event was Vehicle #2's first contact with Vehicle #1, and then the 2nd event was when Vehicle #1's equipment struck the firewall of Vehicle #2, which was substantial enough to cause the clockwise rotation of Vehicle #1.
Plaintiff expert also testified that the distance between Vehicle #2 and Vehicle #3 was at ½ foot when Vehicle #2 collided with Vehicle #1. As Figures # 3 and #4 show, both vehicles were traveling at nearly the same speed as they approached the crash location. Vehicle #2's operator, who does see Vehicle #1's equipment in the left travel lane, does not slow or apply the brakes before the collision with Vehicle #1. Vehicle #3's operator does remove his/her foot from the accelerator approximately 1 second before impact and applies the brake in 0.5 second, which caused a speed reduction of 6 miles per hour before impact with the rear of Vehicle #2. So how's does Vehicle #3 collide with Vehicle #2 from behind when a)Vehicle #2 is going at least 5 miles per hour faster at the time that Vehicle #2 collides with Vehicle #1, b) all the while the operator of Vehicle #3 was reacting approximately 1 second before the impact with Vehicle #2 c) according to plaintiff expert, only 20 ms of time lapse existed between Vehicle #2 impact with Vehicle # 1 and then Vehicle #3 into Vehicle #2.
Commercial Motor Vehicles (CMVs) are used in the transportation of freight and people throughout the world, and diesel fuel in the 1960s thru the 1980s was an inexpensive way to power the heavy vehicles. Mechanical Engine Control Modules (ECM) were used to control the flow of fuel to the engine. Since the price of diesel fuel has matched and sometimes surpassed the price of gasoline and EPA regulations of clean air, diesel motor manufacturers have developed ways to increase fuel mileage and decrease exhaust emissions through constant monitoring and adjusting the diesel engine parameters by using Electronic Control Modules (ECMs). These diesel engine manufacturers, in order to assist in warranty and repair issues with their engines, started to record engine fault codes and what the engine and/or tractor was doing when the engine problems started, by the use of the secondary function of the ECM, which is also called an EDR.
In 1998, Detroit Diesel began using their DDEC IV ECM with their Series 60 engines. These engines were primarily used in over-the-road tractors. The DDEC IV ECM programming was divided into two sections called Diagnostic Links and Reports. Diagnostic Links allows the operating parameters of the engine to be monitored and changed. The Report section was reporting data recorded by the EDR and contains a lithium battery to power the ECM clock. This clock allowed the events recorded to be date and time stamped along with the odometer mileage reading. However, EDR clock drift does occur and must be rectified and adjusted to establish the time and date the triggered events took place. Since 1998, other diesel engine manufactures have followed suit and have added an EDR function within their ECMs. Aftermarket products that have EDR capabilities have been developed and can be utilized in establishing CMV activity a minute or more before the triggering event.
Detroit Diesel DDEC ECMs /EDRs can record two triggered events at a time, and 3 events can be imaged during an extraction of the EDR data.. The events are "Last Stop," "Hard Brake #1," and "Hard Brake #2." To trigger a recording of a Last Stop, the tractor must be traveling less than 1.5 mph and the ignition is turned in the "off" position. A Last Stop record is extremely volatile, meaning, if the tractor key is not turned off and the tractor is moved, that information is gone forever. A Hard Brake event is triggered when the speed sensor at the output shaft of the transmission recognizes the tractor decelerating 7 mph or more in one second. Hard Brake #1 will be the most current hard braking event triggered unless a power interruption prevents a recording.
The forensic examiner/collision reconstructionist must be critical of all types of EDR data. This data can be erroneous and must be properly critiqued as it relates to the collision in question. If the EDR data is accurate and consistent with the collision facts and analysis, then it is proper to use the EDR data to supplement the reconstruction of the collision. The extended EDR data stored before the collision or triggering event allows a forensic expert to analyze further back from the collision as to the CMV's operation before the collision.
The second collision I would like to discuss involved 3 tractor trailers during early morning (pre-dawn) hours on a four lane interstate with two lanes traveling northeast, a wide grass median, and two lane traveling southwest. The police investigation/reconstruction indicated that Unit #1 (tractor trailer) was traveling northeast, drifted left, and while traveling over 800 feet through the grass median, struck a concrete culvert and came to rest across both southwest lanes of the interstate. According to the police investigation, the operator of Unit #1 exited the cab and was standing on the shoulder of the interstate when Unit #2 (tractor trailer), which was traveling southwest, impacted the right side of Unit #1 at the drive wheels, and the force of impact separated the trailer and tractor of Unit #1. Subsequently, Unit #2 impacted the operator of Unit #1 on the shoulder of the interstate. Several minutes passed, and Unit #3 (tractor trailer) traveling southwest observed the mayhem, braked, and struck cargo debris while traveling through the crash area between Unit #1 and Unit #2. All three of the tractor trailers were not operational after the crash and were towed from the scene. The police did not image the ECMs of the crash-involved tractor trailers; however, the ECMs of Unit #1 and #3 were imaged by an opposing expert, and the reports from the EDRs became available to me for further analysis.
A lawsuit was filed alleging that Unit #2 was responsible for the death of Unit #1's operator, who was a pedestrian at the time, and on face value, one would agree. This case had the benefit of Detroit Diesel EDR data from Unit #1 and Unit #3. Unit #2's EDR was not set to record data, and due to the severity of the collision, a power interruption prevented any engine fault codes from being recorded.
The "Last Stop" record of Unit #1 (Click here for Figure 5) indicated that the last stop took place 86.1 miles and approximately 1.5 hrs before the Hard Brake #1 (Click here for Figure 6). Figure 6 was the Hard Brake 1 record of Unit #1. You can see the deceleration of Unit #1 as it is crossing the grass median beginning at T-10 where there was no throttle input by the operator. Unit #1 struck the concrete culvert at T-01 and T-00 and lost tire rotation in excess of 7 mph in one second. The impact with the culvert also caused the tractor's transmission to shift to neutral. This was noted by the high engine rpms after the impact with the culvert, with throttle at 100 % and the tractor's speed still decreasing to final rest. At T+07, the tractor's speed was recorded at zero mph; the engine rpms are at 1350. At T+08, the tractor's speed was recorded at zero mph and the engine rpms are at 933. At T+09, the tractor's speed was zero mph and the engine rpms are zero. In my years of reviewing EDR data and imaging the EDR data, the only time that a diesel engine's rpms dropped from nearly 1000 to 0 in one second was due to incomplete recording from a power interruption. This would be consistent with the last stop not being recorded when Unit #1 came to a stop across the southwest lanes of the interstate.
So, this power interruption took place approximately 2 seconds after Unit #1 came to a stop across the interstate. The power interruption could only be caused by the impact from Unit #2, since Unit #1 was stopped. Would this give enough time for the operator of Unit #1 to exit the cab and walk to the shoulder? We will come back to this.
Unit #1 EDR clock drift was established and corrected to a time of 3:47:12 AM for the hard brake #1 event.
Unit #3 EDR data of the "Last Stop" and "Hard Brake #1" nearly mirror each other, which is usually the case with a towed, non-power interruption ECM/EDR recording. Unit #3 EDR clock drift was established and corrected to a time of 3:48:36 AM for the triggered hard brake event of Unit #3. (Click here for Figure 7).
The time difference between Unit #1's EDR trigger and Unit #3's EDR trigger was 1 minute, 24 seconds. The operator of Unit #1 would not have had severe injuries from the impact with Unit #2 since the passenger compartment of Unit #1 was not damaged from the collision. There was no physical evidence on Unit #2 to indicate it had struck the operator of Unit #1. This time frame of 1 ½ minutes would clearly give enough time for the operator of Unit #1 to exit the vehicle after the collision with Unit #2 and walk onto the highway before Unit #3 arrived at the crash site.
After the physical evidence from a vehicular collision is clearly evaluated, and a concise analysis of the EDR data is conducted, the EDR data can give an accurate objective description of a collision that we have never had in the past. In this last case, the EDR data allowed us to create a time line which contradicted the police investigation as to when the operator of Unit #1 was struck and killed.
The EDR data is a great tool in assisting the trier of fact when determining the responsible party of a vehicular collision; however, the EDR data must be scrutinized before giving it the proper weight as evidence during trial. If you have a vehicle manufactured after 2008, read your vehicle owner's manual; in the first section, you should find additional information regarding your vehicle's EDR system.
This article discusses issues of general interest and does not give any specific legal or business advice pertaining to any specific circumstances. Before acting upon any of its information, you should obtain appropriate advice from a lawyer or other qualified professional.
This article may not be duplicated, altered, distributed, saved, incorporated into another document or website, or otherwise modified without the permission of TASA.