Accident Analysis Service
Don Webb, Director
Accident Analysis Service
PurposeThe purpose of this paper is two fold: 1) To familiarize lawyers involved in motor vehicle accident litigation with the vehicle Airbag Control Module (ACM). 2) To provide a list of other state trial court and appellate court cases (with some commentary) involving the issue of admissibility of Airbag Control Module evidence. Overview A new tool is now available to accident investigators to help them determine cause/fault issues in motor vehicle accidents. This tool is the Vetronix Crash Data Retrieval system, know as the CDR. The CDR is a device that downloads recorded data from the Airbag Control Module (ACM) that determines when or if air bags should be deployed, deploys them and records certain diagnostic and crash-related data. Some have mistakenly compared the CDR to an FDR (Flight Data Recorder), a commercial airliner’s so called black box (actually orange). A FDR continuously records flight data e.g., air speed, yaw, pitch, role, angle of attack, heading and numerous other data points that are used to reconstruct and determine the cause of commercial airline accidents. However,
the CDR interfaces with the Airbag Control Module and downloads crash related data from the ACM. It is not part of the vehicle. “AMC” is a generic designation of what General Motors calls their Sensor & Diagnostic Module (SDM) and Ford calls their Restrain Control Module (RCM). They are all the same thing.
The ACM is the automobile’s FDR. However, the Flight Data Recorder continuously records data from engine start to engine shut down (or plane crash). The automobile ACM starts recording only when it wakes-up due to a collision event and its algorithm determines whether to deploy air bags – a deployment event. AMCs record approximately 5 seconds of pre and 5 seconds of post deployment data. The scope of the data depends on the year, manufacturer, make and model of the vehicle. This data is permanently written (locked) to the EEPROM chip and cannot be erased or over-written.
Accident data is also available for a non-deployment event. This occurs when the ACM wakes up due to a collision event, but the collision is not severe enough to cause the air bags to deploy. If the rate of deceleration, known as “jerk,” does not reach the deployment threshold, the air bags are not deployed. The same data is written to EEPROM chip, but the data is not locked. It remains in memory until it is overwritten by another non-deployment event with a greater jerk.
Type of Data
The type of data that can be downloaded from the ACM depends on the make, model and year of the vehicle. Currently, data from Ford, GM and Isuzu (1990-2005) is available with other manufactures expected to come on-line soon.
ACM data available:
- Crash data
X axis delta-V (change in velocity) & delta-t (impulse time)
Post collision speed
Principal Direction of Force (PDOF)
- Pre-Crash/pre airbag event data
Percent of throttle
Brake switch status (brake light on-off pre-collision)
- Other systems status parameters
Driver seatbelt circuit status (was the seatbelt on?)
Passenger(s) seatbelt circuit status
Other “system status” parameters.
Diagnostic Link Connector (DLC) Download
The most common way to download data is through the Diagnostic Link Connector (DLC). The CDR connects to the DLC through a cable for that vehicle year, make and model. The CDR software downloads the data on to a laptop. Pre-programmed reports, tables and charts are then available for analysis. The raw data contains 64 lines of hexadecimal code. For security, the CDR software embeds a security code within the downloaded data. If anyone alters the data, the CDR software can detect it and lock out the data for processing.
(Partial Hexadecimal Report)
(Direct Connect to ACM)
When the DLC is damaged and cable connection cannot be made, cable connection can be made directly to the ACM. The ACM can be removed from the vehicle and downloaded at a more convenient location. In cases that are expected to go to trial, the ACM is removed for custody/chain of evidence. This is true even if one can connect directly to the DLC. However, it is important to attempt to download data before removal of the ACM just in case the removal process destroys the data.
Training required for CDR certification
The current training courses offered by Collision Safety Institute (CSI) are the Crash Data Retrieval Technician Course (1 day) and the Crash Data Retrieval Analyst Course (3-days - includes the 1 day Technician course). The courses were designed by CSI staff, project engineers from GM and Ford, Vetronix staff and National Highway Traffic Safety Administration staff. Although one can purchase the CDR system without the training, it would be a mistake. In all of the cases reviewed, training in the proper download procedure was crucial to the courts allowing ACM evidence.
How is the ACM data accessed?
The vehicle in question must be covered under the manufacture’s agreement with Vetronix, the manufacturer of the CDR system. A current list of covered vehicles is on the Vetronix web site listed at the end of this paper.
Data can be retrieved by a technician or an analyst who also has specialized training in forensic accident reconstructionist. Because this is new technology, there are very few who are qualified in Oregon/Washington.
Who is qualified to interpret ACM data?
There is a difference between what a CDR technician can provide and what the analyst (forensic reconstructionist) can provide. The technician downloads data and provides the reports generated by the software. However, technicians do not have the background in forensic accident reconstruction and therefore are not qualified to interpret the data (reports) and render opinions.
The CDR is another tool used by the forensic accident reconstructionist to corroborate other evidence and facts of the accident. Very seldom will just canned reports be useful. One has to have the understanding speed, delta-V, Principle Direction of Force, etc. to be able to interpret the data in relationship between the data and the facts of the case.
Who owns the data?
There is very little case law on ACMs and data ownership and admissibility of AMC downloaded crash data. However, most jurisdictions recognize the registered owner of the vehicle as the owner of the data. California restricts the use of the data and who can access it:
Section 9951 of the California Vehicle Code-Data stored in a “recording device” can only be downloaded by the registered owner of the motor vehicle, except under one of the following circumstances:
- The registered owner gives consent
- Where there is a court order
- For the purpose of improving motor safety
- By a licensed dealer or technician for diagnostics/repair.
Note: See the Michigan Senate Bill 902 link at the end of this paper for proposed legislation for on-board recording device in Michigan. Currently, Oregon’s Revised Statues (Motor Vehicle Code) does not address ACM data ownership or use.
ACM data is described in GM’s owner’s manual, “…Your vehicle may record information about the condition of the vehicle and how it was operated, such as engine speed, brake applications, throttle position, vehicle speed, safety belt usage, air bag performance, and the severity of the collision…..”ACM/CDR Cases
The following is a list of court cases that have admitted ACM evidence:
Regina vs. Gauthier (Canada) (2003). This was one of the earliest ACM right to privacy cases. The defense argued that the ACM download was private information. The court held that the “SDM data is not private information since it does not identify the individual, rather information about the vehicle.” The court stated, “SDM evidence is real evidence. It exists without the accused.”
Harris v. General Motors Corp., FED App. 0039P (6th Cir.) (2000).
Anderson-Barahona v. General Motors Corp., No. 99A19714, GA, Cobb County Cir. Ct., (2000)
Bachman, et al, v. General Motors Corp., et al, Illinois App. Ct., 4th Dist., No. 4-01-0237, Appeal from Circuit Court of Woodford County, No. 98L21 (2002). The defense offered the ACM data from her vehicle that was presented by two engineers that designed and developed the device. The court applied the Frye Test to determine admissibility and found the downloading of the data from the ACM had gained general acceptance within the scientific community and admitted the ACM data.
Colorado v. Cain, 1st Judicial District Court, Division 3, Jefferson County, Case No. 01 CR 967 (2002).
Pennsylvania v. Walter Thomas Rhoads, Montgomery County, Court of Common Pleas, Criminal Division, Docket No. 746701 (2002).
California v. Michael Beeler, San Diego Superior Court, Case No. SCD158974 (2002).
People v. Slade,
State of New York (2003). The defendant, Soukup and his friend Slade, were
friends racing, in separate cars, on Route 106 through Old Brookville, N.Y., in
northeastern Nassau County. As they approached the Muttontown Road intersection,
Soukup, driving a 2002 Chevrolet Corvette at nearly 130 miles per hour, slammed
into a 1993 Jeep Cherokee, tearing it in half. A split second later, Slade,
driving a 2002 Mercedes, rammed into the front end of the Jeep, knocking it 300
yards up the road.
The Court ruled that the black box data were admissible, even though the police had obtained it before applying for a search warrant. Turning back the defense argument that his client had a reasonable expectation of privacy in the car and its contents, the judge ruled that Soukup's operation of the vehicle on a public highway knowingly exposed his behavior to the public. Observed by at least three witnesses, his velocity was not a private matter, the judge said. He noted that the police searched not only the engine compartment containing the black box, but also the car's passenger compartment, in which Soukup may have had a reasonable privacy expectation. Nevertheless, the judge found that because the detectives obtained a warrant based not upon information obtained from the car, but from eyewitnesses and observations at the crash scene, the search was covered by the "independent source rule." The judge wrote that the rule preserves the admissibility of otherwise tainted proof if it was "obtained independently from lawful activities untainted by the initial illegality."
Illinois v. William R. Barham, Illinois App. Ct, 5th Dist., No. 5-02-0047, Appeal from the Circuit Court of Johnson County, No. 00-CF-90 (2003).
Florida v. John Walker, 20th Judicial Circuit, Lee County, Case No. 00-002866CF RTC (2003).
Florida v. Edwin Matos, 17th Judicial Circuit, Broward County, Case No. 02015762 CF 10A (2003).
People of the State of California v. Allen R. French, Court of Appeal, California 2nd Appellate Dist, Div 3 (2003).
People of the State of California v. Vincent H. Sanchez, (2003).
People of the State of Michigan v. Stephan Wood, (2003).
South Carolina v. Andrew Cassels, Beaufort County, General Session Indictment No. 2002 GF 070372 (2003)
Florida v. Suzette Ubals, 17th Judicial Circuit, Broward County, Case No. 01017144 CF 10A (2003)
Gina Meadows v. Worsham Sprinkler, FSS, et al., Georgia State Court of Fulton County (2004).
People of the State of Michigan v. Timothy Schubert, Michigan (2005).
People of the State of Arizona v. Bishop O'Brien, Maricopa County, Phoenix. (2004)
State of Missouri v. Scott Bragg, Jackson County, Missouri (2005).Padilla v. Price Toyota, New Jersey-Federal (2005). In this personal injury case, plaintiff Kristy Padilla a minor by her guardian ad Litem, Edwin Padilla, et al. filed a complaint against the defendant Price Toyota, et al. in connection with an automobile accident that occurred. Plaintiffs alleged that a defect in the design or manufacture of the passenger side airbag system of the vehicle caused or enhanced Kristy Padilla's injuries. Plaintiffs filed a motion to compel discovery of information from the vehicle's black box following the parties' attempt to resolve this issue. In support of this motion, plaintiffs asserted that there was relevant information, which pertained to the time leading up to the deployment of the passenger side airbag, stored within the Random Access Memory (RAM) or Electronic Erasable and Programmable Read-Only Memory (EEPROM) of the vehicle's black box. The defendants' objection was that the information in the black box was either not reliable, or could not be retrieved by reliable means, and therefore not able to lead to admissible evidence. Plaintiffs' expert R. Scott King, M.E asserted that information in the black box might lead to information concerning the crash event. Further, defendants' expert Motoki Shibata acknowledged that a rough approximation of the crash pulse might be obtained, which had the potential to shed light on the circumstances under which the airbag deployed during the accident. The court found that the defendants' objections were akin to the issues raised in a motion challenging the reliability of expert testimony pursuant to Rule 702. The court decided that the court should not, in the discovery context, preclude plaintiffs from obtaining information based on the reliability arguments raised by the defendants. The court concluded that unreliability might very well support a Daubert motion in the future, although the court in this order made no finding in that regard. The court granted plaintiffs' motion to compel discovery of information from the vehicle's black box.
Motas v. State of Florida, District Court of Appeal of the State of Florida, 4th District, on appeal from, the Circuit Court for the Seventh Judicial Circuit, Broward County (2005). Motas was convicted for two counts of manslaughter resulting from a motor vehicle accident. Speed evidence obtained from an ACM download indicated that Motas speed just 4 seconds before the crash was 114 mph and 103 within one second of the crash. An SAE paper was presented to the court, entitled, “Accuracy of Pre-Crash Speed Captured by the Event Data Recorder, involving accuracy studies of the data concluded that the ACM data overestimated vehicle speeds by a mere 1 mph at low speeds and 3.5 mph at high speeds. Another paper was presented to the court, “Recording Automotive Crash Event Data,” that included case studies on real life crashes, which calculated an accuracy of plus or minus 4% for vehicle speed.
The defense also argued that the trial court erred in allowing ACM evidence as proof speed because the ACM did not meet the requirements under Florida law for mechanisms used by law enforcement officers to measure speed. Several examples of case law were presented that discussed the admissibility of read speed-measuring devices.
The court upheld the lower courts admission of the ACM under Frye and that the law referring to Florida Statute regarding mechanisms used by law enforcement officers to measure speed, the court stated that the “the legislature never intended this statute to apply to a device internal to the vehicle and installed by the manufacture. The ACM recorded the defendant’s speed at the time of the accident automatically without any intervention by a police officer. As the state points out, the ACM was not used by a police officer to determine the speed of the defendant’s vehicle while the officer was engaged in the enforcement of the motor vehicle laws of the state. It was only after a warrant had been issued and executed that the data from the ACM was obtained. We can discern no legislative intent to apply section 416.1905(1) to the ACM”
State Of Ohio V Larry Wilson II, Ohio Court of Appeals Case No. 05-Ca-5, Appeal from the Perry County Court of Common Pleas Case No. 04-CR-0034. The appellant was convicted in Perry County Court of two counts of Aggravated Vehicular Homicide. The appellant and his friend, Jason Baron, who was operating another vehicle in the same direction on State Route 13. The State charged that both vehicles were traveling at a speed greater than the posted 55 MPH limit. At the intersection of State Route 13 and Township Road 312, a Neon automobile started across the intersection operated by Jena Snider with Henna Mumford as a passenger. The Neon was struck by both vehicles killing both Snider and Mumford.
During the ensuing trial, the state introduced evidence from as ACM. In the Assignment of Error Pleadings, the defendant claimed:
(1) The court below erred when it permitted the State to introduce evidence from Sensing and Diagnostic Modules over the defendant’s objection and without sufficient foundation to indicate the reliability of the scientific testimony, in violation of rule 702, Ohio rules of evidence.
(2) The court below erred when it permitted expert testimony as to opportunity for the Dodge Neon to clear the intersection prior to the accident as that testimony was not shown to be reliable, in violation of rule 702, Ohio rules of evidence.
The Ohio Appeals court upheld trial court in both issues.
Most of ACM cases are criminal or product liability cases. ACM data will to be used in other civil cases when trial lawyers become more aware of the how the data can assist them in motor vehicle accident litigation. As of this date, there are no ACM cases in Oregon.ACM data admissibility: Frye, Daubert, Kumho
The trial court’s role as gatekeeper seeks is to prevent junk science evidence from being introduced by an “expert witness.” The US Supreme Court, in Daubert and Kumho, prescribed what has become known as the ”flexible multi-factor test.” This test is designed to assist the court in its role as gatekeeper. The test include seven questions:
(1) What are the expert’s qualifications and stature?It is vital that the expert have training as a CDR Data Analysis where the curriculum and instructional staff are recognized by the National Highway Traffic Safety Administration. CDR Data Technician training is not enough to provide expert opinion evidence. While the Technician may download the data and be in the chain of evidence, the Data Analyst can interpret the data and offer expert opinion in context with the other accident evidence. Of course, the expert witness must have the background, education, training, publication and research in the field of forensic accident reconstruction that will allow him to be recognized by the court as an “expert witness!”
(2) Is the Airbag Control Module “new and novel?”
In 1974, NHTSA began equipping a Disc Recorder system on 1,000 vehicles in several fleets that totaled 26 million miles. Twenty-six-crashes were analyzed; measuring changes in velocity (delta-v) up to 20 mph. General Motors introduced the first regular production driver/passenger air bag systems in selected vehicles. These units contained a data-recording feature for deploying air bags in severe crashes.The equipment and data acquisition procedure has been developed and refined over the years by vehicle manufacture and ACM manufacture engineers, CSI and NHTSA staff. This process, in one form or another, has been on going for over 30 years.
(3)Is ACM data downloading procedure tested or testable?
The procedure has been tested and retested. Crash tests have been compared to ACM data to determine error factor and if ACM data matches real world crash data. Any qualified person using the accepted download procedure will get the same results.
(4)Has then ACM data download procedure been published or peer reviewed?
There are volumes of research papers published by the Society of Automotive Engineers, Ford Motor Company, General Motors, the Nation Highway Traffic Safety Association, the Accident Reconstruction Journal, plus by numerous other accident reconstruction organizations throughout the county.
(5)Is there a reasonable error rate associated with ACM crash data recording and data recovery?
As is the case with anything man made, there is an error factor associated with the CDR. However, the error rate was determined through crash tests and other means, to be reasonable.It is commonly accepted within the scientific community that there is an approximate error rate of plus or minus 4% (speed – delta-V – PDOF). However, for other data points i.e., systems status, there is no error. Having said that, let us examine a case where the ACM data indicates seatbelt use, but in fact, the seatbelt was not worn at all. The driver had connected the seatbelt to the seatbelt receiver but was “sitting” on the seatbelt at the time of the accident. The AMC recorded correctly that the seatbelt was fasten, but had no way of knowing that the driver was sitting on it. This is an example of why just relying on the data alone is a mistake. An analyst would examine the seatbelt system, injuries sustained, seatbelt webbing, seatbelt D-ring, secondary impact within the vehicle, etc. to evaluate the ACM data for consistency. 6)Are there standards for application of ACM crash data recording and data recovery technique? There are standard procedures for year, make and model of the vehicles covered under the agreement with Vetronix. These are published and followed by trained technicians and analysts. (7)Has there been acceptance of ACM crash data recording and data recovery in the scientific community? ACM data capture (recording) and data downloading has been developed by the scientific community beginning as far back as 1974. The techniques used today are the results of testing and development involving the scientific community, law enforcement, motor vehicle manufacturers, and ACM manufactures with oversight of the National Highway Traffic Safety Association. ACM downloads have been accepted in several state and federal courts that have recognized the acceptance within the scientific community.
Insurance fraudInsurance fraud involving the replacement of ACMs is an increasing problem for the insurance industry. Generally, there are two type of fraud. One involves set-up crashes between two vehicles where both drives are involved in a scheme so that one can claim faked injuries. This usually involves a conspiracy with both medical and legal personnel in on the scam. The ACM data records the delta-V, Principal Direction of Force and speed. These data can be used to dispute such fraudulent claims.
Another insurance fraud scam involves fraudulent air bag – ACM replacement. It cost about $1000 to replace the air bag and ACM. Once an air bag is deployed, both the air bag and ACM must be replaced. The shady repair shop fraudulently claims that an air bag was deployed in an accident. They remove a perfectly good air bag and ACM and replace it with a deployed air bag and ACM. If the ACM is downloaded, the data will show air bag deployment, albeit from another vehicle. Once they get paid for the replacements, the original components are put back in the vehicle. However, if the insurance company suspects fraud, they can download the ACM and prove that the allegedly new is not new at all. A new SDM will not have a high number if ignition cycles (number of time the ignition has been turned on since the SDM has been installed).A similar fraud situation involves repair shops removing perfectly good air bags from vehicles and filling the air bag receptacle with paper, rags or anything to fill in the hole so that the victim does not notice that it is missing. The Supplemental Indicator Restraint (SIR), light on the instrument panel, will illuminate during the SDM diagnostics routine when the ignition is turned on. It will remain illuminated if it no air bag is detected. To counter this, these repair shops will disconnect the SIR. Most drivers would not notice if the SIR did not come on during the ignition cycle. This fraud goes undetected until the victim is involved in an accident and the air bag is not deployed.
ACM/CDR use in Oregon/WashingtonCurrently, in Oregon and Washington the use of ACM data is not as prominent as in other areas of the country in civil and criminal cases involving motor vehicle accidents. As more trained CDR technicians and analyst are trained, and the insurance companies and the legal community are aware that this data is available, the use of the ACM data will become routine.
A perfect case for ACM data is the multi-car collision where everyone claims that they were stopped and the other guy rear-ended him. The ACM can determine if the vehicle was in motion or stopped.
Aftermarket Add-onsOn-board diagnostics (OBD) modules similar to the Black Box used by the airline industry to record flight data are now available for purchase by the private sector. Parents of teenage drivers are the biggest customers. CarChip company manufactures such device
Reduced Insurance Premiums Progressive Insurance Casualty Company, under its TripSense program, provides an OBD module that plugs into the DLC connector. It continuously collects and stores data from the vehicle’s computer control systems. Progressive provides the module at no cost to the users (insured). Users mail in the recorded data and receive a 5% discount on their insurance and as much as 25% discount at renewal depending on insured recorded driving habits.These add-on modules record speed, time, distance in daily travel logs, trip detail logs and overall usage pattern logs. As of this date, these modules and data downloads have not been used as evidence in court. They have not accepted by the scientific community and have not gone through the peer review process that the ACM has. However, if a vehicle equipped with one of these modules is involved in a criminal case involving speed, be sure that its data will be introduced as evidence. It will be interesting to see the outcome.
ACM downloads maybe useful in determining the following crash related issues::
- Speed of a vehicle before impact.
- Speed of a vehicle at impact.
- Principle Direction of Force (PDOF)
- Speed lose due to impact (Delta-V)
- Were brakes applied before impact?
- Was a vehicle stationary or moving?
- Do injuries match Delta-V
- Were the seatbelts in use (driver-passengers)?
- Did air bag deploy or not deploy as designed?
- Did the seat-belt pre-tensioners work as designed?
So far ACM downloads, have an excellent track record in the courts. However, it was not the data alone, but the data and supporting forensic reconstruction. ACM data must be part of a comprehensive forensic reconstruction to determine the proper contextual relationship of the data and the accident.
For additional information, contact Don Webb at 503-931-0670 or firstname.lastname@example.org
Third Report to Congress Effectiveness of Occupant Protection Systems and Their Use, National Highway Traffic Safety Administration, U. S. Department of Transportation, Washington DC, December 1996.
Fatality Reduction by Air Bags: Analyses of Accident Data through early 1996, NHTSA Report Number DOT 808 470, July 1998.
The Accuracy of Crash Data from Ford Restraint Control Modules Interpreted with Revised Vetronix Software, by Jonathan M. Lawrence and Craig C. Wilkinson, SAE Technical Paper No. 2005-01-1206.
Using Event Data Recorders in Collision Reconstruction, by Richard Fay, Ric Robinette, John Scott and Derrell Deering, SAE Technical Paper No. 2002-01-0535.
ACM related links
Author’s web site:
SAE web site:
CSI web site:
NHTSA EDR History web site:
NHTSA EDR publications web sit:
Vetronix CDR web site & vehicle list:
Davis “CarChip” web site:
Progressive Insurance CASDite:
Regarding on-board recording devices has been introduced and is now in the Senate Transportation Committee. The full text of the bill: