Volume 8, Issue 2

R e c Tr fo on us r st t e 15 ru d + ct by Y io e n a is rs t s

2013 ARC-- CSI CRASH CONFerence DVD

C

2013 Vol.8 Iss.2

The International Compendium for Crash Research

Volume 8 Issue 2

Collision: The International Compendium for Crash Research

SIMPLY THE BEST RECONSTRUCTION SOFTWARE

CRASH WITH US...

CALL US NOW TO

BOOK A LIVE DEMONSTRATION

OR TO FIND OUT MORE

1.888.828.0383

Volume 8, Issue 2 - FALL 2013

Diagramming • Reconstruction • Animations • Point Cloud

visualstatement.com/demo

ollision C FALL 2013

Data from the 2013 ARC-CSI Crash Conference Analysis of Event Data Recorder Delta-V Reporting in the IIHS Small Overlap Crash Test Hyundai and Kia Crash Data, the Indispensable Compendium

collisionmagazine.com

Differentiating Potentially Causal Pre-crash Component Damage from Crash Damage

EDCRASH

We have the STIFFNESS data you need for your calculations!!

Now includes a FORCE BALANCE calculation module!!

4N6XPRT StifCalcs®

puts the NHTSA Crash Test database on YOUR 32 or 64 bit MS-Windows computer, automatically calculates A-B-G values Front, Side, and Rear - and Test Specific Crush Factors based on the NHTSA test data, allows the creation of “CLASS” vehicles and determine the “CLASS” stiffness values, PLUS conduct FORCE BALANCE calculations when needed, all of this with no need for you to get on the internet, and all for the LOW price of $650.00*!

Can you do this yourself “for free”?? Only if YOUR TIME = $0.00!!

For more information, program order form, and sample printouts please visit our web site at http://www.4n6xprt.com.

4N6XPRT StifCalcs® is provided by 4N6XPRT Systems

8387 University Avenue - La M esa, CA 91942 - USA Phone: 800-266-9778 / (619) 464-3478 Fax: (619) 464-2206

Individual Vehicle Stiffness Data

Engineering Dynamics Corporation Reconstruction of Accident Speeds on the Highway EDCRASH ™ is an HVE-compatible reconstruction analysis of vehicle collisions. It is the most widely used crash reconstruction computer program in North America. Applications include: • Speed Analysis • Crash Severity EDCRASH is based on the CRASH3 computer program originally developed for the National Highway Traffic Safety Administration (NHTSA), and includes numerous extensions developed

by EDC. The user assigns vehicle impact and rest positions and tire-road friction data acquired during a crash site inspection, and vehicle damage profile data acquired during a post-crash vehicle inspection. The user may also assign wheel lock-up information for damaged wheels. Using this information, EDCRASH calculates impact velocities and velocity changes based on wellestablished linear momentum and damage energy methods. The results are displayed in numerical and graphical reports.

EDCRASH has been extensively validated using numerous staged collisions, the results of which are publicly available in the technical literature. Program results easily pass federal standards for admissibility of scientific evidence.

So, you want Stiffness data for a vehicle, QUICKLY & CONVENIENTLY, but you don’t want to buy a program? Order your data on a vehicle-by-vehicle basis from us at only $40.00 per test!

Year: _________ Make: _____________________ Model:____________________________________ Impact on (circle one): Front - Side - Rear Impact speed (circle one): High - Low YOUR CONTACT INFORMATION

Name: ________________________________________ Company: _____________________________________ Address: ______________________________________ City/State/Zip: _________________________________ Phone: _______________________________________ Fax: __________________________________________ Email: ________________________________________ Visa/MC/AmEx: ________________________________ Exp.:____/____ Case Reference: ___________________ Send Order to:

4N6XPRT Systems

Fax: 619-464-2206

VISUAL REALITY ATM2013

Engineering Dynamics Corporation • 8625 SW Cascade Ave. Suite 200 • Beaverton, OR 97008 USA • 888.768.6216 www.edccorp.com EDCRASH and HVE are trademarks of Engineering Dynamics Corporation. All rights reserved.

Ph: 800-266-9778 8387 University Avenue / La Mesa, CA 91942 / USA E-mail - 4n6@4n6xprt.com

A SOFTWARE SOLUTION FOR EVERY SITUATION:

Total station w/onboard data collection

Robotic total statio n w/Windows CE

Get EVR

□ Reliable □ Compatible See Exhibit A

Evidence Collection Forensic Mapping

Total station or GPS w/data collector

3D Laser Scanner

Evidence Collection Software? w tion n A imacapabilities?

Do we

Yes

No Get Forensic CAD

MapScenes Evidence Recorder • Software for tablets and Windows CE devices • Connect with a total station or GNSS receiver to take measurements and collect evidence • Built for the professional reconstructionist

+ Animation Module

Cloud + Point Module

MapScenes Forensic CAD • A powerful and feature-rich program that provides law enforcement professionals and accident reconstruction specialists with confidence that the evidence they present is precise, compelling, and irrefutable • Produce scaled diagrams for use in cour t, creating a strong visual case for the prosecution or defense. Contains flexible options for point cloud/LiDAR data visualization and anim ation tools • Now with point cloud and animation addon modules • Built specifically for forensic www.MapScenes.com investigators and accident scene reconstructionists 1.800.668.3312

is registered in the U.S. Patent and Trademark Office. All other brand ® MapScenes names, product names or trademarks are property of their respective owners.

Exhibit A

ant

Contents

Fall 2013

Volume 8 Issue 2 8

inside 4

Letter From the Editor

6

NAPARS: Letter From the President

features

38

8

Analysis of Event Data Recorder Delta-V Reporting in the IIHS Small Overlap Crash Test

by Sean Haight and Rusty Haight

26

Differentiating Potentially Causal Precrash Component Damage from Crash Damage By: David Hallman

38

Driver Distraction: Obtaining Mobile Device Digital Evidence

by Mike May and Andrew Russell

46

Evidence Sometimes Overlooked During Vehicle Inspections

by William Brem and Wayne Denham

54

Hyundai and Kia Crash Data: the Indispensable Compendium

by Rusty Haight, Shawn Gyorke and Sean Haight

118

Bombardier Can–Am Spyder Braking and Acceleration Testing

by Dawn Mutis

132

My Turn at the Wheel: "Think of a Number, Then …"

by Erik Carlsson

www.collisionpublishing.com

54

118

Collision Magazine - Volume 8 Issue 2 3

Letter from the editor As we send this second issue of the 8th volume of Collision to press, I am reminded that next February will mark 14 years since the then Vetronix now Bosch Crash Data Retrieval Tool was first offered for public sale. At first, that might not seem like a particularly remarkable observation or landmark (the ten year mark came and went and we’re not quite at some round number like 15) but when you think about it, 14 years - ago at the end of 1999 - we were approaching the dawn of the new millennium and, as we entered that, we also entered a new chapter in crash reconstruction: the advent of publically available crash data. We were also on the eve of comments like: “this ‘black box thing’ will put reconstructionists out of business!” and “this CDR stuff is the ‘next generation’ of police report.”

Hyundai and Kia line vehicles which are accessible using a different system.

Which brings me to one of the central components of this issue of Collision, the multi-part: “Hyundai and Kia Crash Data, the Indispensable Compendium.” It is a first-of-its-kind published look at non-Bosch CDR Tool retrievable crash data.

Covering various aspects of the Hyundai and Kia EDR tool(s), the Compendium includes an examination of the mechanics of the system(s) and resulting reports, a comparison of the retrieved data to control data from crash tests as well as to real world crash analyses and more.

Of course, neither turned out to be quite acAnd, on the subject of “more,” there is more curate. in this issue of Collision. (There always is And just 14 years ago - perhaps as part of a re- more: more original work, more detail, more sponse to the idea that the CDR Tool would value). This issue includes an examination of “end reconstruction as we know it” - the chief data from very small overlap crash tests and complaint was that “well, you can ‘only’ down- how that might effect the application of crash load GM vehicles...” Three years later, when data in similar real world collisions, a look at Ford data access was added, that changed to distracted driving, and two articles on various “well, it’s still ‘just’ two manufacturers and, after aspects of vehicle examination and documentaall, Ford ‘only’ gives you delta-V.” tion including some insight into differentiating between a pre-crash mechanical failure and that How things have changed... Today, of course, which is the product of the crash. And there is there are more than 12 vehicle lines supported still more - like the complete 2013 ARC-CSI in North America as well as other regions of Crash Conference data DVD included with this the world and, by all indications, still more to issue of Collision. come. Moreover, the volume and quality of data has reached a point few probably would have Born, in part, as a product of the CDR User’s anticipated just before the new millennium.

Summit in 2006, it’s appropriate that Collision has emerged as the leading publication of It took 5 more years (after Ford was added) be- its kind related to crash data in particular and fore a third auto manufacturer was added to the crash reconstruction in a general. So one has to list of those supported by the CDR Tool but wonder, where will we be after 14 years? since then, as the number of manufacturers has increased to four times that number, virtually every supported manufacturer’s data set includes not only delta-V but a significant amount of Managing Editor “pre-crash” data as well. And saying there are “12 manufacturers” doesn’t even include those not supported by the Bosch CDR Tool like the

4 Collision Magazine - Volume 8 Issue 2

www.collisionpublishing.com

BACK ISSUES

Collision Publishing Inc PO Box 892885 Temecula, CA 92589 Toll Free: 800-280-7940 Email: editor@collisionpublishing.com Web: www.collisionmagazine.com ISSN: 1934-8681

If you would like purchase back issues of Collision Magazine, please visit www.collisionmagazine.com or call us toll free 800-280-7940.

ADVERTISER INDEX

COLLISION STAFF Scott Baker Sean Haight Tonya Baker Carla McLeod

Owner, Managing Editor Senior Editor Advertising Manager Subscription Services

All rights reserved, © 2006-13 Collision Publishing Inc. The opinions and conclusions expressed in this publication and attached data disk and throughout the articles attributed to specific authors are the opinions and conclusions of the authors noted and not necessarily the editorial staff or anyone else for that matter. While some articles have been reviewed for content, the accuracy of reprinted models or equations cannot be fully guaranteed. It is the responsibility of the reader to apply critical thinking to an individual review of the content and make their own personal judgements as to its value to them, individually.

At the end of the day, facts belong to everybody, any other opinions to us. The distinction is yours to draw...otherwise, the opinions expressed herein are not necessarily those of any employer, not necessarily ours and probably not necessary. Dissenting opinions, discussion or conclusions which may

express an adverse position to those expressed herein by specifically cited authors can be addressed in writing by sending an email to the editor at editor@collisionpublishing.com or sending a “regular mail” letter to us at PO Box 892885, Temecula, CA 92589. By sending correspondence to either our email or snail mail addresses

listed in this publication you are agreeing that: (1) we are by definition, “the intended recipient” (2) all information in the email is ours to do with as we see fit and make such financial profit, political mileage, or good joke as it lends itself to and, (3) this overrides any disclaimer or statement of confidentiality that may be included on your original message. The entire physical universe, including this publication and attached data disk, may one day collapse back into an infinitesimally small space. Should another universe subsequently reemerge, the existence of this publication and data disk in that universe cannot be guaranteed.

SUBMIT AN ARTICLE If you would like to submit an article to be published in Collision please email us your article for review or call us toll free 800-280-7940. Email: editor@collisionpublishing.com

ADVERTISING RATES & INFO If you would like to advertise your products or services in Collision Magazine, rates, availability and other information are available online at: www.collisionmagazine.com

4N6XPRT Systems ARAS360 ARC-CSI Crash Conference ARC Network CDR User's Summit Collision Magazine Back Issues Collision Safety Institute Crash Data Group (CDR) Engineering Dynamics Eos Systems: Photomodeler Laser Technology Inc MapScenes NAPARS Northwestern University REC-TEC VCRware Vericom Computers Visual Statement

PAGE Inside Back, 136 1 130 75 53, 129 131 117 70 Inside Front 44 24 2, 71-74 6 135 25 35 45 36-37, Back Cover

Included with this issue of Collision

is the 2013 ARC-CSI Crash Conference data DVD. The ARC-CSI DVD includes all the photos, video, instrumented crash testing data, scaled diagrams, speaker Power Points, and any other supporting data compiled in a easy to review, menu-driven data DVD.

Next ARC-CSI Crash Conference June 2-5, 2014 www.arccsi.com

www.collisionpublishing.com

Collision Magazine - Volume 8 Issue 2 5

Premier Industry Partner Letter from the President

WREX 2016 NAPARS is working with the rest of the crash reconstruction organizations in determining the feasibility of holding another week long “national” crash reconstruction training conference in 2016. Those of you that remember WREX2000 and F3T2 in 2006 can attest to the usefulness, value, and downright fun of this kind of training event.

The location will be determined based on having a large host city with easy access, multiple conference hotels, and local “boots on the ground” to help with planning. Stay tuned for more on those plans.

NHTSA and you

NAPARS is currently involved in a number of activities to promote crash reconstruction training, as well as working to increase our voice in related industry and governmental policies. In the coming months you should hear more about: NAPARS 2014 Planning is ongoing for the Joint Annual Training Conference in the fall of 2014. In an effort to make training more accessible to different segments of our membership, the location for next year’s conference will be the Portland, Maine area. Although the topics are not yet finalized, we are planning to bring in a number of leading experts in various fields such as Human Factors, Damage Energy, and Memory Creation to pack this conference with valuable training from beginning to end. And perhaps even a lobster dinner or a brewery tour might be in order.

NAPARS, along with the rest of the Traffic Crash Leadership group (TCR) is working to unite our industry’s voice in government and industry decision making with regards to rules and issues that affect our daily work. This includes Notices of Proposed Rulemaking (NPRM) and the issuance of Requests for Interpretation (RFI) that deal with adherence to rules currently in effect. For example, as the end users of vehicle EDR data we should be vocal in dealing with issues that in the past have been (for the most part) discussed in behind closed doors. Requesting transparency in these discussions and negotiations can only lead to better rules for the public, protection of the rights of the manufacturers, and proper access to needed data for our industry.

Again I want to thank you for your support of NAPARS. And as always, if you have any comments, suggestions, or want to get involved send us an email at President@ NAPARS.org.

Chuck Veppert, President, NAPARS

The National Association of Professional Accident Reconstruction Specialists is a non-profit organization whose members have joined together to share the challenge of dealing with the complex problems of accident reconstruction and to upgrade and ultimately professionalize the accident reconstruction field.

6 Collision Magazine - Volume 8 Issue 2

www.collisionpublishing.com

www.napars.org

Industry Partners IAARS is one of the oldest associations dedicated to

expanding the knowledge and training of accident reconstructionists. Formed in 1980, we have members from several different countries as well as many different states across the U.S. We have three levels of membership. Our highest level, Fellow, includes a peer review of a reconstruction project. This peer review feature can be very beneficial to a member seeking expert witness qualification in court. One of the criteria a judge can consider when deciding if a person qualifies as an expert is a peer review of his or her work. We offer opportunities for a member to publish work or research in the field of accident reconstruction. This, also, can be very important when qualifying as an expert witness in court proceedings.

Membership in IAARS has other advantages, including a discount (from IAARS) on a subscription to Collision Magazine,

ASPACI is a non-profit society whose members have

joined to share in the challenge of analysing the complicated issues of road collision investigation.

ASPACI was formed in 1991 by a group of police and collision reconstructionists who felt there was a need for information sharing between police and associated institutions in the Australia, South East Asia, New Zealand and South Pacific area.

Its primary function is to assist in the spreading of knowledge and enhancing expertise in the science. The Association seeks to promote a professional approach to collision investigation and to encourage the highest standard of ethics, integrity and honesty among its members.

The ARC Network

was founded in 1998 and since has grown to be the largest and most respected organization for Accident Reconstruction and Traffic Investigation.

The ARC Network is more than just an organization it is a research and communications Internet portal for everything accident reconstruction. The ARC Network provides research tools, Expert directory, event calendar, industry news feeds, education directlory and a Member's Only area that gives access to various online databases and information.

The ARC Network offers various membership levels to fit your needs and budget. Membership starts at just $79.00 per year.

However, that is where the value of the ARC Network starts.

With your membership you not only get access to everything in the ARC Network, you also receive a, $50 discount to the annual ARC-CSI Crash Conference, a $50 discount to the CDR

member discounts at our training seminars and others listed on our web page, www.iaars.org.

We have annual training opportunities at different locations across the country. Our 2013 conference was a joint seminar with SCARS and SeARS and was held in Mt. Pleasant (Charleston), South Carolina. Staged collisions included two pole impacts, some semi-trailer underride crashes, and one offset head-on crash.

Our 2014 conference will also be a joint conference with SCARS and SeARS and will be held from July 20 to July 25, 2014, at the same location. Check our Web site or www.scarsrecon.org for more information.

Sincerely, Fred Rice, President, IAARS www.iaars.com The early days saw the number of members vary with the base of the association in Victoria, Australia. ASPACI became ASPACI Incorporated in 1993 in Victoria. Now we boast members from all over mainland Australia, New Zealand., United Kingdom, and the United States, as well as parts of Asia.

Membership to ASPACI is open to all persons engaged in collision investigation and its allied fields. The association members include Engineers, Police Officers, Independent Consultants, Scientists, Insurance Corporations, Manufacturers and Safety personnel. ASPACI is not a Trade Association or Political Lobby group, nor is it a Police organisation.

User's Summit and a free copy of Greg Russell's Excel Crush Stiffness spreadsheet program. Extend your savings and sign up as a Police Agency member or Platinum member and receive a complimentary subscription to Collision Magazine!

In addition, as a member you have the choice of being listed in the most-visited Expert Directory on the Internet. Our Expert directory boasts a #1 ranking in Google searches (a feat worth much more than the cost of membership)!

Scott Baker President The ARC Network (800) 280-7940 www.accidentreconstruction.com arc@accidentreconstruction.com

For information on becoming a Collision Industry Partner send an email - editor@collisionpublishing.com www.collisionpublishing.com

Collision Magazine - Volume 8 Issue 2 7

Analysis of Event Data Recorder delta-v reporting in the IIHS Small Overlap Crash Test Sean Haight

Collision Safety Institute

8 Collision Magazine - Volume 8 Issue 2

W. R. "Rusty" Haight

Collision Safety Institute

www.collisionpublishing.com

"Twenty four percent of the frontal crashes that IIHS reviewed in the NASS study were small overlap crashes."

A

bstract A review of small overlap crash test data was conducted during which a seemingly anomalous condition associated with crash severity on the “minor axis” (see note 1) for the vehicles involved in those tests was observed as a function of comparing Event Data Recorder (EDR) and test instrumentation data sets. Where the reconstructionist or researcher might assume that longitudinal and lateral acceleration and/or delta-V would accurately represent the effect on the test vehicle based on past literature in the area, the data reviewed revealed a significant discrepancy between EDR data and the test instrumentation based data. Once the basis for the discrepancy was more fully identified, a corresponding or parallel discrepancy between the EDR data and corresponding values which might be calculated or estimated in a crash analysis or reconstruction was identified.

I

Over the years, there have been numerous studies (see appendix) on the accuracy and performance of Event Data Recorders (EDRs) in instrumented crash tests. However, most have been focused on full frontal impacts and perpendicular side impacts. This study will shift the focus from full frontal barrier crash tests to small overlap crash tests performed by the Insurance Institute for Highway Safety (IIHS) and the resulting effect this type of collision may have on some of the crash severity data recorded in the EDR component of late model vehicles. ntroduction In 2009, IIHS published a “Status Report”1 in which the possibility of creating a new crash test procedure was discussed to analyze vehi-

cle responses to a small overlap crash configuration.

Previous similar studies conducted by IIHS focused on moderate (40%) overlap crash tests. The Institute performed a statistical study of real world crashes to evaluate the applicability of the smaller overlap crash test configuration. Researchers used the 2000-2006 National Automotive Sampling System (NASS) Crashworthiness Data System (CDS) to study frontal crashes on US roads. As a result of the study, IIHS Sr. Vice President of Vehicle Research David Zuby said that “[IIHS] found that a lot of frontal crashes are happening in configurations that aren’t represented in the crash tests being conducted right now, either by us or the National Highway Traffic Safety Administration.”1 Twenty four percent of the crashes that IIHS analyzed in the study1 were small overlap crashes where only about twenty five percent of the vehicle is engaged with the impacted object. Mr. Zuby concluded that “based on results of this study, we think small overlap tests have good potential to improve the crashworthiness of new vehicles.” The report, along with other technical papers on small overlap crashes provided sufficient justification to further analyze this crash configuration in a laboratory setting. In 2012, IIHS announced that they began testing new vehicles in the small overlap configuration.2 By early 2013, IIHS completed testing on eleven mid-sized luxury/near luxury vehicles and eighteen midsized moderately priced vehicles.3 By August of 2013, IIHS completed testing on six small cars in the small overlap crash configuration.4

S

mall Overlap Test Procedure A brief summary of the IIHS small overlap test will identify data relevant to this study. All the details of the small overlap crash test proce5 dure is available at the IIHS website (iihs.org). The vehicle is aligned such that 25% of the vehicle width

www.collisionpublishing.com

Collision Magazine - Volume 8 Issue 2 9

overlaps with a flat faced rigid barrier with 150 mm radius edge, as shown in Figure 1. The test vehicle is accelerated by a propulsion system to 40 mph (64 km/h) upon impact with the barrier. A fully instrumented 50th percentile Hybrid III Dummy is placed in the driver seat and two pressure-sensitive tape switches placed on the front of the vehicle are used to determine the moment of impact. This time (a “time zero”) is synced with the data acquisition and camera systems.

IIHS also records vehicle acceleration at a location along the longitudinal axis of the vehicle in the rear floorpan.

One set of accelerometers, measuring longitudinal, lateral and vertical accelerations are used in each crash test. To ensure the accelerometers stay on axis during the crash, they are mounted to a metal block which is welded to a strong part of the floorpan. Vehicle accelerometers are oriented according to SAE J2116. Figure 2 shows the typical location of the accelerometer array relative to the center of

gravity of the vehicle and a target is positioned on the roof of the vehicle over the location of the accelerometer array which can then be seen in later overhead video.

IIHS also captures valuable crash related data from high speed camera footage. Onboard cameras focus on dummy interaction with the vehicle interior while off-board cameras capture vehicle motion and dynamic deformation. All cameras capture 500 frames per second. For this study, the most relevant camera view is the off-board wide overhead angle. Figure 3 shows series of frames captured from a typical small overlap crash test from this “overhead wide” perspective.

T

est Results A group of vehicles from the IIHS small overlap crash testing program were chosen for analysis (see Table I). These vehicles were selected for this study to include many of the available manufacturers supported

Figure 1: Vehicle alignment for IIHS small overlap crash test.5

Figure 2: Accelerometer location for IIHS small overlap crash test.5 10

Collision Magazine - Volume 8 Issue 2

www.collisionpublishing.com

by the Bosch Crash Data Retrieval (CDR) Tool as well as some Hyundai and Kia vehicles supported by the GIT/ Snap-On EDR tool.

Model Year 2012 2012 2013 2013 2013 2013 2013 2013 2013 2013 2013 2014

Make

Model

Retrieval Tool

Dodge Toyota Buick Chevrolet Dodge Ford Honda Honda Hyundai Kia Volvo Kia

Avenger Camry Encore Cruze Dart Focus Accord Civic Tucson Soul XC60 Forte

Bosch CDR Tool Bosch CDR Tool Bosch CDR Tool Bosch CDR Tool Bosch CDR Tool Bosch CDR Tool Bosch CDR Tool Bosch CDR Tool Hyundai EDR Tool Kia EDR Tool Bosch CDR Tool Kia EDR Tool

Table 1: IIHS crash test matrix

As might be anticipated with the implementation of 49CFR563, many of the data parameters retrieved and reported by the CDR and GIT/Snap-On tools are reported accurately when they are compared to the IIHS test documentation and instrumentation. However, some of the data parameters, such as steering angle and brake application, cannot be fully analyzed because there was no driver inputs to compare the recorded parameters with. Perhaps the most uniquely interesting data parameters, and what we will focus on for this study, are the longitudinal and lateral delta-V. Figures 4 and 5 show typical longitudinal and lateral deltaV data from a CDR Tool generated report compared to instrumentation data from an IIHS small overlap crash test.

The retrieved crash data in these figures were directly transcribed from the CDR report and are offered here without further processing.

While it is clear that there are some minor discrepancies in the longitudinal delta-V, around 175-200ms into the event, what is particularly noticeable, is the difference in lateral delta-V from the CDR report when compared to the test instrumentation lateral delta-V. The example (seen

Figure 3: Series of frames captured from a typical small overlap crash test www.collisionpublishing.com

Collision Magazine - Volume 8 Issue 2 11

Figure 4: Longitudinal detla-V from CDR report and IIHS instrumentation

Figure 5: Lateral detla-V from CDR report and IIHS instrumentation in Figures 4 and 5), which is from a 2013 Honda Civic (test number CEN1301), shows that the instrumentation based delta-V, at 150ms and beyond, is about 10 mph (16 km/h) more than the delta-V found in the CDR report. This figure also shows that the instrumentation and the CDR reported values begin to diverge at about 50ms into the event. Tables 1 and 2 show the comparison of the longitudinal and lateral delta-V data for each vehicle at 12

Collision Magazine - Volume 8 Issue 2

specific time intervals. The difference shown in each table is calculated as the CDR/GIT EDR reported delta-V subtracted from the integrated IIHS acceleration data, or delta-V. All reported data in these tables follow SAE-J211 sign conventions.

As one may see in Table 2, data from all of the vehicles tested in the IIHS small overlap crash test program under-

www.collisionpublishing.com

Longitudinal Delta-V Difference (IIHS – CDR/GIT EDR Tool) (mph) MY Make Model 50ms 100ms 150ms 2012 Dodge Avenger 0.5 1.0 - 0.2 2012 Toyota Camry - 1.0 - 0.6 0.2 2013 Buick Encore 1.0 0.2 0.5 2013 Chevrolet Cruze 0.1 - 1.1 - 1.1 2013 Dodge Dart 2.76 .5 - 0.1 2013 Ford Focus 2.5 4.3 5.3 2013 Honda Accord - 0.1 - 0.1 0.3 2013 Honda Civic 0.6 - 0.9 0.3 2013 Hyundai Tucson - 0.8 0.9 0.9 2013 Kia Soul 2.1 2.1 2.4 2013 Volvo XC60 0.6 1.0 - 0.1 2014 Kia Forte 1.6 2.9 3.2

200ms - 0.5 1.2 1.1 - 0.6 0.2 6.6 1.0 1.6 2.5 3.6 - 0.1 4.7

250ms N/A N/A 2.2 - 0.4 0.5 7.9 2.4 3.4 3.9 3.8 0.1 6.0

300ms N/A N/A 3.0 - 0.2 1.3 N/A N/A N/A N/A N/A N/A N/A

250ms 13.5 N/A 9.9 1.0 7.3 9.4 9.0 10.0 11.9 5.5 4.1 8.5

300ms N/A N/A 10.2 0.5 7.5 N/A N/A N/A N/A N/A N/A N/A

Table 1: Longitudinal delta-V comparison for each vehicle Lateral Delta-V Difference (IIHS – CDR/GIT EDR Tool) (mph) MY Make Model 50ms 100ms 150ms 2012 Dodge Avenger - 2.4 3.3 6.0 2012 Toyota Camry - 2.4 1.4 * N/A 2013 Buick Encore 0.9 6.1 9.2 2013 Chevrolet Cruze - 2.8 0.3 1.0 2013 Dodge Dart - 0.9 2.9 5.5 2013 Ford Focus 0.4 6.4 9.2 2013 Honda Accord - 1.6 5.9 9.0 2013 Honda Civic - 0.2 8.6 9.8 2013 Hyundai Tucson 0.6 7.6 7.4 2013 Kia Soul - 1.1 2.0 4.7 2013 Volvo XC60 ** - 0.1 1.9 3.4 2014 Kia Forte 0.1 5.4 7.6

200ms 5.9 N/A 9.7 1.9 6.1 9.6 9.0 10.0 11.7 5.6 4.1 8.1

* Last data point at 69ms ** Delta-V was calculated manually from acceleration data Table 2: Lateral delta-V comparison for each vehicle

www.collisionpublishing.com

Collision Magazine - Volume 8 Issue 2 13

reported the lateral delta-V. In one case, the EDR lateral delta-V was under-reported by as much as 14 mph (23 km/h) when compared to the test instrumentation data.

Of the 12 vehicles selected for this study, the average difference between the test instrumentation lateral delta-V and that reported by the CDR/GIT Tools (at the last data point) was 8 mph (13 km/h).

Inasmuch as one would expect data from an airbag control module EDR to accurately capture acceleration such that it may be used for deployment decision(s) together with the observation that the EDR reported longitudinal data was very close to that reported by IIHS, the discrepancy between the lateral data sets seems counterintuitive and was seen, initially, as a potential anomaly. At the very least, that there was a near universal underreporting discrepancy in the data suggested it should be more closely examined and explained.

“

"EDR reported longitudinal data was very close to that reported by IIHS, the discrepancy between the lateral data sets seems counterintuitive and was seen, initially, as a potential anomaly."

"

Each vehicle manufacturer specifies the characteristics and type of Airbag Control Module (ACM) used in their vehicles and those modules may be provided by multiple first tier suppliers, even for a single vehicle model. One potential explanation for the difference might be differences between suppliers. Similarly, another potential explanation might be an issue with the translation of retrieved data such that there is an unidentified error with the functionality of the retrieval tool (software) used. However; since the recurring trend is that the lateral delta-V discrepancy 14 Collision Magazine - Volume 8 Issue 2

(for the test series under examination) is largely “universal” across not only multiple supplier’s modules as well as both the very different Bosch and GIT tools, these explanations would be unlikely.

Ultimately, the answer may be found in ISO/TR 12353-37 which, at Appendix C, 2.2.5: “…In crashes with vehicle rotation about the vertical axis, crash pulse recorder signals can easily be misinterpreted. The acceleration signal at the crash sensor depends on the distance from the sensor location to the centre of vehicle rotation ... In crashes with vehicle rotation, the crash sensor may be located far from the vehicle's centre of rotation. In this case, the longitudinal and lateral acceleration at the vehicle's centre of gravity is markedly different from that measured by the crash pulse recorder due to rotational components at the sensor location. In some cases the Delta-V (or acceleration) signals can be reengineered if a sufficient rotational acceleration-time record is available. …” 7 Because reconstructionists would optimally endeavor to apply EDR based longitudinal and lateral acceleration and/or delta-V data to a crash analysis and that recorded data may be compared to a simulation or hand calculation for delta-V, this study attempts to go forward to more fully identify the underlying discrepancies between EDR and instrumentation based delta-V data elements.

V

ideo Analysis It is unusual to have two (apparently) conflicting data sets, of the same physical parameter, from the same crash test. In the group of vehicles being examined here, there are two lateral delta-V data sets from the same crash test that are up to 14 mph (23 km/h) different. Naturally, the next step in the analysis would be to derive a “tie-breaker” or third data set that would potentially identify which data set represents a more accurate value.

For this review, the third data set is based on a digital video analysis using point and vector tracking software.

For this study, Tracker - a publically available video analysis and modeling tool - was used to determine the lateral delta-V during the IIHS small overlap crash test series.

This software allows for both manual and automatic point tracking to enable data collection on position, velocity and acceleration. It will also analyze 2-dimensional vectors to report a specific angle for an object relative to a specific coordinate system. Since this software works in a global-fixed coordinate system, a transformation must take place to compare directly with the vehicle fixed coordinate system defined by SAE J211. The equations for this coordinate transformation are:

www.collisionpublishing.com

Where x’ and y’ are the new, vehicle fixed coordinate system directional vectors, x and y are the global-fixed coordinate system directional vectors and theta is the angle of rotation between the two coordinate systems. Using this relationship, the delta-V from the Tracker software can be adjusted to the vehicle fixed coordinate system.

The “overhead wide” video for each test found in the IIHS crash test database was used to analyze the position of the vehicle at the vehicle’s center of gravity, and the location of the IIHS accelerometer array using the target observed on the roof as previously described. Figure 6 is a group of sample screenshots from the Tracker software being used to measure the velocity of these specific points. While the video frame rate is 500 fps, the figure shown is at equal time increments from time zero to 300ms (at increments of 60ms). The red line shown in Figure 6 represents the incremental locations of the center of gravity, the blue line represents the incremental locations of the IIHS acceler-

Figure 6: Sample screenshots from the Tracker software for the 2013 Honda Civic crash test www.collisionpublishing.com

Collision Magazine - Volume 8 Issue 2 15

ometer array. Again, the 2013 Honda Civic small overlap crash test is offered in Figure 6 used as an example.

pulses recorded in the EDR component for the majority of the vehicles found in this series of tests.

One can easily see in the overhead video footage (Figure 6) that there is a considerable offset distance between the incremental points for the path of the center of gravity and those incremental points for the path of the IIHS accelerometer array. In the 2013 Honda Civic test example, it appears that there is approximately 26 inches offset between those two points. The transformed velocity at each point was extracted from the Tracker software package and plotted against the integrated accelerometer data and the CDR reported lateral delta-V. This plot comparison is seen in Figure 7.

The difference in delta-V between the IIHS accelerometer array, calculated delta-V at the center of gravity and the CDR reported value implies that reported delta-V depends on the position and relative motion of the point (sensor or CG) being evaluated consistent with what is found in ISO/ TR 12353-3. Based on the data presented to this point, one might hypothesize that the location of the airbag control module appears to be typically forward of the center of gravity. This assumption would be based on the observation that there seems to be a positive trend for the relative

“ "

Based on this analysis, it can be seen that the integrated IIHS accelerometer array based delta-V and the delta-V calculated from the video analysis are very similar when one focuses the Tracker software on the path of the accelerometer array. However, when one focuses the program to track the path of the center of gravity the resultant deltaV decreases (relatively). Then, while the CDR reported delta-V is 4 mph (6 km/h) at the end of the event, the delta-V at the center of gravity hovers around 9 mph (14 km/h) for the Honda Civic. The same video analysis has been performed for all of the vehicles listed in Table 3.

Table 3 shows the CDR reported lateral delta-V, the calculated delta-V from the IIHS accelerometer array, and the delta-V at the center of gravity calculated using the video analysis at 200 ms. The 200ms point was selected because it fell within the general range for the duration of the crash

"The difference ... implies that reported delta-V depends on the position and relative motion of the point (sensor or CG) being evaluated."

Figure 7: Lateral delta-V comparison from the IIHS accelerometer, CDR report and video analysis 16 Collision Magazine - Volume 8 Issue 2

www.collisionpublishing.com

Lateral delta-V at 200 ms MY Make 2012 Dodge 2012 Toyota 2013 Buick 2013 Chevrolet 2013 Dodge 2013 Ford 2013 Honda 2013 Honda 2013 Hyundai 2013 Kia 2013 Volvo 2014 Kia

Model Avenger Camry Encore Cruze Dart Focus Accord Civic Tucson Soul XC60 ** Forte

CDR Reported 8.2 mph 5.4 mph * 5.6 mph 9.3 mph 7.5 mph 7.5 mph 4.0 mph 4.0 mph 3.1 mph 7.5 mph 12.5 mph 5.0 mph

Video Analysis (CG) 10.6 mph 6.0 mph 10.0 mph 10.8 mph 11.0 mph 9.4 mph 10.3 mph 9.1 mph 10.3 mph 9.2 mph 12.6 mph 10.3 mph

IIHS Calculated 14.1 mph 6.8 mph * 15.3 mph 11.2 mph 13.6 mph 17.1 mph 13.0 mph 14.0 mph 14.8 mph 13.1 mph 16.6 mph 13.1 mph

Table 2: Lateral delta-V for 2013 Honda Civic small overlap test at 200ms after impact

Figure 8: Graphical representation of Table 2 showing lateral delta-V at 200ms

www.collisionpublishing.com

Collision Magazine - Volume 8 Issue 2 17

Figure 9: Location of Airbag Control Module in 2013 Honda Civic delta-V between location of the ACM and the delta-V at the CG as observed graphically in Figure 8.

In this case, we can utilize the Bosch CDR program to get an idea of the general location of the airbag control module. Figure 9 shows the location of the ACM for the 2013 Honda Civic.

Based on the diagram seen in Figure 9, the “SRS UNIT”, or Airbag Control Module, is located under the dash, under/behind the center stack and near the accelerator pedal.

As previously discussed, the identified location of the subject Honda Civic ACM is forward of both the center of gravity and the IIHS accelerometer array. The relative position of the Honda ACM seen in Figure 9 can be generally identified given landmarks within the vehicle (i.e.: distance from the front bumper, position on the center tunnel, the dash, and/or shifter position). Extrapolating the position from that seen in Figure 9 to a corresponding overhead view which can be used for the video analysis meant that, to a certain degree, the position had to be approximated based on position of relative landmarks visible in diagrams, observations in the cockpit and the overhead videos. Figure 10 shows the location that was selected to represent the best approximate location of the ACM. Figure 11 appends the delta-V data from the video analysis (for the ACM) to the previous delta-V plot (Figure 7). Figure 11 demonstrates that the tracking location that is closer to the ACM results in a more representative delta-V. This analysis supports the finding that in small overlap crash tests which may produce significant rotation, the location of the measuring device – whether the ACM of instrumentation installed by researchers - does have an ef18

Collision Magazine - Volume 8 Issue 2

fect on the calculated delta-Vs relative to the delta-V at the CG of the vehicle. This assumption could also be made for additional crash configurations with similar rotations, for example an eccentric side impact. In the Honda Civic example, the ACM was estimated to be about 16 inches forward of the center of gravity and another 26 inches forward of the IIHS accelerometer array and significant rotation was observed.

The next part of the analysis is to better define this effect and evaluate how a more complete understanding might lead to a more reasonable application of EDR data in crashes like those represented by this test series. Unlike full frontal barrier crash tests, such as those specified for

Figure 10: Approximate tracking location of the ACM for the 2013 Honda Civic

www.collisionpublishing.com

“ "

Figure 11: Location of Airbag Control Module in 2013 Honda Civic compliance with 49CFR563, the IIHS small overlap crash test often causes the vehicle to yaw significantly about its z(vertical) axis. This rotation would cause the accelerometers to report different values given their relative positions along the longitudinal axis. This relationship can be modeled using basic circular motion equations. Figure 12 shows this relationship where “vT” is the tangential velocity, “ω” is the rotational velocity (or how fast the vehicle is rotating) and “r” is the distance from the rotational center to the measurement location.

Figure 12: Rotation of a vehicle about a fixed point This fundamental equation of circular motion indicates that as one increases the distance for a given point from the center of rotation, for a specified increment of time, one would find a higher value of lateral velocity (vT). This

"This analysis supports the finding that in small overlap crash tests which may produce significant rotation, the location of the measuring device does have an effect on the calculated delta-V ..."

www.collisionpublishing.com

Collision Magazine - Volume 8 Issue 2 19

Figure 13: Sample screenshots from the Tracker software for the 2013 Chevrolet Cruze crash test supplements the original analysis in that the IIHS accelerometer array, which is located significantly behind the CG and ACM, produced a higher value for delta-V and, correspondingly, the ACM positioned forward of the CG produced a lower value for delta-V. This rotational effect in the IIHS small overlap crash test can be demonstrated using the overhead video footage from the 2013 Honda Civic and the 2013 Chevrolet Cruze. Figure 13 represents some sample frames from the 2013 Chevrolet Cruze crash tests and the relative movement of the CG (red) and IIHS instrumentation (blue) is shown.

20

Collision Magazine - Volume 8 Issue 2

The footage from the 2013 Chevy Cruze crash test shows that there is a significant difference in the amount of rotation compared to that of the 2013 Honda Civic. Using the Tracker software, vectors can be created and tracked that represent the heading of the vehicle during the test. The angle of that measurement vector can then be exported to show the angle of the vehicle heading as a function of time.

Figure 14 is a plot that shows this rotation angle for both the 2013 Honda Civic and the 2013 Chevrolet Cruze.

Figure 14 shows that the angle of rotation for the 2013 Honda Civic at 200ms is -23 degrees while the 2013 Chevrolet Cruze only undergoes -3 degrees of rotation.

From this analysis, a comparison can be made to the CDR

www.collisionpublishing.com

reported delta-V for each vehicle. Table 2 shows that there was a large discrepancy between the reported and measured delta-Vs for the Civic (9-10 mph, 14-16 km/h), while there was a minimal difference for the Cruze (1-2

mph, 2-3 km/h). This comparison shows how the angle of rotation can significantly affect the reported delta-V. Figures 15 and 16 show the longitudinal and lateral delta-V for the 2013 Chevrolet Cruze, where the rotation of the

Figure 14: Vehicle rotation as a function of time for Civic and Cruze tests

Figure 15: 2013 Chevrolet Cruze longitudinal delta-V

Figure 16: 2013 Chevrolet Cruze lateral delta-V www.collisionpublishing.com

Collision Magazine - Volume 8 Issue 2 21

vehicle was limited to under 15 degrees. These plots can be compared to Figures 4 and 5 which show the same data for the 2013 Honda Civic.

C

onclusions The findings in this study provide some insight into CDR/GIT EDR Tool relative reporting applicibilty in crash configurations similar to the IIHS small overlap crash test. It can be seen that, across all vehicle manufacturers, there is the potential for significant differences between the accelerations and delta-V values from the CDR/GIT EDR report and other instrumentation placed in the vehicle. It is clear that the source of this difference is a combination of rotation and location of the sensor rather than an measurement error in the airbag control module or other instrument.

If a reconstructionist is interested in the change in velocity of the center of gravity of a vehicle in a similar style crash, the CDR/GIT EDR report may over/under-estimate the lateral delta-V because the ACM is typically not colocated with the vehicle’s center of gravity. If there is a large rotation, as can be seen in the 2013 Honda Civic example, the minor axis (here the lateral axis) reported delta-V may be significantly under-reported. In other words, the ACM reported delta-V does not necessarily reflect the velocity change at the center of gravity – it represents what is measured at the location where it is mounted. Reconstructionists must take this fact into account when trying to evaluate the change of velocity at the center of gravity.

It is important to discuss the impact of these findings relative to 49CFR563.8 Table III of the rule specifies the accuracy and resolution of event data recorders if they are recorded. For example, maximum longitudinal delta-V must achieve an accuracy of plus or minus ten percent while maintaining a recording resolution of one kilometer per hour. Additionally, the maximum lateral delta-V must meet the same accuracy and resolution requirements.

Without conducting a proper analysis and simply comparing the crash data, it may seem that these ACMs do not meet the accuracy requirements. However, this comparison is not necessarily fair and equal. In fact, NHTSA has defined which types of test determine whether or not an ACM would pass the Part 563 rule.

The NHTSA laboratory test procedure for Part 563 9 explicitly defines which types of tests can be used to “determine the performance of a vehicle with respect to the requirements of 49CFR563.” The types of tests listed in this procedure are: Frontal Rigid Barrier, Frontal Offset (40%, 25 mph, 40 km/h), and Side Deformable Barrier.

It is important to note that the IIHS small overlap test is 22 Collision Magazine - Volume 8 Issue 2

not listed in this table and is very different from the 40% frontal offset crash test. The offset frontal test described in the NHTSA Part 563 test procedure involves a deformable barrier, while the IIHS small overlap test is a rigid barrier and the speed of the NHTSA offset frontal is slower the IIHS test.

More importantly, the NHTSA test procedure explicitly describes how the reference accelerometer is placed. In section 10.2, the test procedure calls for the documentation of the “installation of reference accelerometer(s) colocated with the EDR primary accelerometer(s).” Furthermore, the procedure states the a “second triaxial accelerometer shall be on the vertical centerline passing through the vehicle’s longitudinal crash sensor, either on the outer cover of the module containing the vehicle crash sensor or a nearby rigid structure.” By physically mounting the reference accelerometer to the outside shell of the ACM, the probability of any measurement differences is greatly reduced.

This action would ensure a fair comparison of data for the purpose of testing EDR accuracy.

In conclusion, it is important for the CDR Analyst to understand where and how the data is being measured. In small overlap collisions, or other crashes with significant rotations, the reported accelerations and delta-V values may not represent the dynamics of the center of gravity, or at the location of an occupant.

R

eferences 1. Status Report. Rep. 2nd ed. Vol. 44. Arlington: Insurance Institute for Highway Safety, March 7, 2009. Print.

2. Insurance Institute for Highway Safety. IIHS News.

New Crash Test Aims to Drive Improvements in Protecting People in Frontal Crashes. News Releases.

N.p., 14 Aug. 2012. Web. 3. Sherwood, Chris. "An Update on the IIHS Small Overlap Ratings Program." 2013 SAE Government Industry Meeting. Washington, DC. 30 Jan. 2013.

Speech.

4. Insurance Institute for Highway Safety. IIHS News.

Performance in Small Overlap Front Test Earns 6 Small Cars Good or Acceptable Ratings and Top Safety Pick+ Award. News Releases. N.p., 8 Aug. 2013.

Web. 5. Small Overlap Frontal Crashworthiness Evaluation Crash Test Protocol (Version II). Rep. Arlington, VA: Insurance Institute for Highway Safety, December 2012. Print.

www.collisionpublishing.com

6. Instrumentation for Impact Test - Part 1 - Electronic Instrumentation. Tech. no. J211/1. Warrendale, PA: Society of Automotive Engineers, 2007. Print. Polarity (sign) notation found in this paper adopts the normal sign conventions found in SAE J211.

7. ISO/TR 12353-3; “Road Vehicles - Traffic Accident Analysis - Part 3: Guidelines for the Interpretation of Recorded Crash Pulse Data to Determine Impact Severity” 8. Federal Motor Vehicle Safety Standards; Event Data Recorders, 49CFR563 (2012). Print.

9. Laboratory Test Procedure for Part 563, Event Data Recorders, U.S. DOT – NHTSA. TP-563-00 (2012).

Print.

Appendix of representitive EDR delta-V performance references: 1. Lawrence, J., Wilkinson, C., King, D., Heinrichs, B. et al., "The Accuracy and Sensitivity of Event Data Recorders in Low-Speed Collisions," SAE Technical Paper 2002-01-0679, 2002, doi:10.4271/2002-010679.

2. Wilkinson, C., Lawrence, J., Nelson, T., and Bowler, J., "The Accuracy and Sensitivity of 2005 to 2008 Toyota Corolla Event Data Recorders in Low-Speed Collisions," SAE Int. J. Trans. Safety 1(2):420-429, 2013, doi:10.4271/2013-01-1268.

3. Tsoi, A., Hinch, J., Ruth, R., and Gabler, H., "Validation of Event Data Recorders in High Severity

Full-Frontal Crash Tests," SAE Int. J. Trans. Safety 1(1):76-99, 2013, doi:10.4271/2013-01-1265.

4. Lawrence, J. and Wilkinson, C., "The Accuracy of Crash Data from Ford Restraint Control Modules Interpreted with Revised Vetronix Software," SAE Technical Paper 2005-01-1206, 2005, doi:10.4271/200501-1206.

5. Wilkinson, C., Lawrence, J., Heinrichs, B., and King, D., "The Accuracy and Sensitivity of 2003 and 2004 General Motors Event Data Recorders in Low-Speed Barrier and Vehicle Collisions," SAE Technical Paper 2005-01-1190, 2005, doi:10.4271/2005-01-1190.

Note 1: A convention adopted for the purposes of this study is the term “minor axis.” While there are some who might incorrectly assume, based on the testing done here

by IIHS, that this term only refers to the lateral (Y) axis as the “minor axis” for all use cases, in the right collision alignment (i.e.: an eccentric side impact), the “minor axis” would be the longitudinal (X) axis.

A

cknowledgements The authors would like to thank the Insurance Institute for Highway Safety, especially Becky Mueller and Raul Arbelaez, for their contributions to this study. IIHS is an independent, nonprofit scientific and educational organization dedicated to reducing the losses — deaths, injuries and property damage — from crashes on the nation's roads. We would also like to thank Jeff Suway, Shawn Gyorke, and Dr. Shaun Kildare for their comments and contributions.

www.collisionpublishing.com

Collision Magazine - Volume 8 Issue 2 23

I AM

CRASH INVESTIGATION It’s not like it looks on TV. Crash scene investigations are never cut and dry. Time is always of the essence.

No two scenes are alike, so I need my mapping equipment to be easy to run and versatile enough to handle any situation I’m up against.

We hear you.

Look into our TruPulse® 200X laser with TruAngle® and QuickMap 3D, an easy to operate and less expensive alternative to total stations.

Watch our recorded Crash Scene Mapping Webcast.

Request a quote or demonstration.

www.lasertech.com/collision 877.696.2584

differentiating potentially causal pre-crash component damage from crash damage david m. hallman, m.s., p.e.

26 Collision Magazine - Volume 8 Issue 2

www.collisionpublishing.com

a

bstract Statistics show that 10 – 15% of crashes are reported to have been triggered or caused by vehicle mechanical failure. Examination of involved vehicles in these cases requires an investigator to have skills in both crash reconstruction and vehicle mechanical systems. Important information is provided by both the post-crash vehicle examination and the original crash scene. This article will focus on some of the methods used to differentiate potentially causal pre-crash component damage from crash damage.

i

ntroduction Crash investigations vary widely depending on the circumstances of the incident being examined. Some of the investigations involve vehicle mechanical component failures and the investigator must determine whether the failure happened prior to or during the crash. In the event that a component is found to have failed pre-crash, the investigator must also determine if the component failure caused or triggered the crash. These determinations can dictate the direction of the investigation. Investigators need to be cautious interpreting all of the factual evidence

www.collisionpublishing.com

Collision Magazine - Volume 8 Issue 2 27

from the vehicles and the scene. In most cases mechanical failures on vehicles can be identified as pre-crash and / or crash induced if the evidence is examined carefully. In some cases it may be necessary to have an experienced mechanic or engineer assist with the evaluation. All evidence must be considered together to determine the events of the crash and ultimately the root cause. Crash investigators come from many different backgrounds and sometimes it is necessary to call in reinforcements to assist with certain aspects of a case. The purpose of this article is to provide some basic information for investigators with varying levels of mechanical ability and understanding.

This article begins by covering some vehicle basics including several of the vehicle systems. The systems covered here are those that are most likely to be involved in a crash where a mechanical cause or trigger is suspected. This does not mean that other systems are not involved, only that they are not typically causal. This discussion will begin at the tires where the vehicle contacts the road and work up. The next section will briefly discuss the “method of inquiry” known as the Scientific Method. This method provides a good basis for all type of investigations and can be a way to guide an investigation. It can also be a good way to define the steps of your investigation to clients and ultimately to a jury. The next section will continue with a discussion of one method for vehicle examination and evaluation as well as evaluation of road evidence prior to and after the vehicle inspection. Information from each examination may lead

back to the other for further evaluation of the evidence. All of the facts and evidence must be considered together in order to get a complete understanding of the crash. The final section will briefly discuss one methodology for separating pre-crash damage from crash damage.

v

ehicle Basics - Tires Based on some research, tires appear to be blamed for many of the crashes that occur on the road today. Research done on the National Highway Transportation Safety Administration’s (NHTSA) National Automotive Sampling System (NASS) database website shows that tires are very often listed as the mechanical failure for crashes. In-depth examination of some of those cases indicates that while tires were listed, actual tire failure was not always the cause. For instance, wheel-off incidents were sometimes coded as tire failures.

A wheel-off could be the failure of a wheel, bearing, lug nuts, wheel studs, an installation failure or some combination of those. It likely would not be caused by a physical tire failure.

Post crash tire examination can be complicated depending on tire conditions. If tire failure is a suspected cause or trigger for a crash, a complete examination of all of the remaining tire fragments should be performed. These fragments can give clues to the failure mode or modes leading to a better understanding of the root cause of the failure. It is important to understand that many tire failures involve

Figure 1: Tire identification 28

Collision Magazine - Volume 8 Issue 2

www.collisionpublishing.com

over-heating of the tire. This can happen for a variety of reasons including but not limited to running the tire low and / or overloading the vehicle (and the tire). Tire failures can also be caused by previous damage, incorrect repairs, tire age or worn tires. Many of these conditions are detectible by an astute investigator at the initial inspection.

A post-crash inspection should begin with a visual inspection of all of the tires on the vehicle. Tires should be inspected for wear and wear patterns, sidewall and tire face damage and tread depth. Tire sidewall information should also be documented (Figure 1). All of these checks are non-destructive and can reveal a great deal of information. Pressure checks may result in changing the conditions present prior to the crash and should be done cautiously. A high quality pressure gage should be used that can provide pressure readings starting at zero pounds per square inch (psi) rather than the typical 10 or 20 psi. A high pressure gage may be required for medium and heavy trucks and busses.

v

ehicle Basics - Wheels One of the main potential issues with wheels, particularly medium to heavy truck wheels can be inappropriate installation. Wheel nut torque is specified for a wheel based on the level of bolt clamp required to react the expected loads applied to the wheel through the tire / road contact. Typical wheel nut torque for a passenger car or light truck is approximately 80 – 90 ft-lbs. Many mechanics will install passenger vehicle wheel nuts with an impact wrench which typically results in them being over-tightened. This over tightening results in high tension in the studs. It is also possible to over tighten the nuts with a hand lug wrench meaning that even wheels installed by the vehicle owner could be over-tightened. This makes automotive wheel installation much more forgiving. Since the lug nuts are very typically over tightened, the tension in the joint is generally adequate to retain the wheel.

Required torque values for medium and heavy trucks however are significantly higher than they are for passenger vehicles, particularly for dual wheels. These values can be ten times higher than passenger car and light truck torque values (up to 900 ft-lbs). The most important factor in proper wheel installation is that wheel nut torque is simply a method to achieve the correct “bolt stretch” or tension in the wheel stud leading to adequate clamp in the wheel attachment joint. The tension in the actual wheel stud is the key factor in wheel retention. As studs and nuts age, corrode and wear it is possible to apply the correct torque to the wheel nut and yet still not achieve correct tension in the studs meaning that the wheel is not tight enough.

Figure 2: Wheel lug nut rust 7 When a wheel is not tight enough, the nuts can back off over time and a wheel-off incident can occur. There are several indications that a wheel has worked loose. One indication is rust around the lug nuts as seen in Figure 2. This is caused by relative movement between the wheel face and the lug nuts that generates metal particles. Over time, those particles rust and flow out from under the washers and leave rust on the wheel surface.

When tightening a nut on a stud or bolt, the applied torque goes into not only clamping the joint together and stretching the stud, but also into overcoming the friction of the joint including the thread surfaces against each other and the face of the nut against the face of the wheel. Thread wear, damage and corrosion introduce metal particles into the interface between the threads. The applied torque then has to go into overcoming the additional friction rather than tensioning the joint. Why use torque as a measure when installing nuts to bolts or wheel studs? Torque is inexpensive to measure using a hand held or other type of torque wrench. Tension can be measured but would require special strain gauged bolts or other expensive measurement equipment. Torque is used because it is simple and inexpensive.

A post-crash wheel inspection, particularly in a wheel-off case, starts with locating the missing wheel or wheels. This is critical in determining the root cause of the failure. If the lug nuts have fallen off, an attempt should be made to find as many as possible. The surface of the wheel and the remaining wheel studs should be carefully examined. In the event that all of the lug nuts are still present, the investigator should evaluate if there has been a bearing / axle failure or other failure resulting in the wheel separation.

www.collisionpublishing.com

Collision Magazine - Volume 8 Issue 2 29

v

ehicle Basics - Brakes Vehicle brakes have changed dramatically over the last 20 - 30 years. Early passenger vehicles and light trucks were typically equipped with drum brakes on both axles. Later vehicles used disc brakes on the front axle and drum brakes on the rear axle. Many modern passenger vehicles and light trucks use disc brakes on both axles. Medium and heavy trucks / busses vary, but typically disc brakes are used on steer axles and drum brakes in other locations including trailer axles.

Modern vehicles also have anti-lock brakes, stability control and “smart” braking systems that can apply the brakes for the driver in the event of an imminent crash based on feedback from a forward facing radar, video, infrared, ultrasonic or other type of sensor.

Brake failure can be a generic claim made in any type of crash or incident. Experience indicates that lack of post collision tire marks does not necessarily mean that the vehicle brakes failed. Braking systems have some level of redundancy and rarely fail completely and without warning.

These claims need to be carefully evaluated and a complete brake system inspection should be performed. Even if the braking system is badly damaged and non-functional post crash, it can still be possible to determine if the brakes failed prior to the crash. The possible cause or causes of that failure can also likely be determined.

A post-crash brake inspection should not begin by pressing the brake pedal. This should be considered a destructive test and can change the conditions that existed pre-crash.

The first step, depending on the damage to the vehicle, should only be a visual examination of the components that are accessible. The master cylinder fluid level, brake pedal and linkage condition, possibly portions of the brake line and brake pad and rotor condition will all likely be visible without removing any components. A simple visual examination of the components may indicate potential issues with the braking system.

If a more in depth inspection is required (Figure 3), it is best to do that in a shop or lab with a vehicle hoist. It is possible, however, to do this inspection outdoors in a vehicle lot if necessary. Particular care should be taken to preserve and protect the vehicle until a proper inspection can be performed. It is important to remember that heavy corrosion can form on brake rotor and drum surfaces very rapidly and that if brake lines were damaged in the crash, the brake fluid may leak out from the system. The condition of the brakes at an inspection performed at a much later date may be substantially changed from the condition at the time of the crash.

Figure 3: Example brake mechanical failures

30

Collision Magazine - Volume 8 Issue 2

www.collisionpublishing.com

v

ehicle Basics – Steering and Suspension The vehicle steering and suspension systems (Figure 4) are two separate yet interdependent systems. This discussion will include both systems.

The vehicle steering system consists of components that allow the vehicle to be steered including the steering gear or rack and pinion, the inner and outer tie rods and other steering linkage parts depending on the type of steering system being utilized in the subject vehicle. Suspension systems vary widely but can include coil or leaf springs, torsion bars, struts, strut assemblies, control arms, ball joints and many other components.

The simplest way to differentiate between the two systems is to understand that the suspension system “suspends” the vehicle. The steering system allows the vehicle to be controlled or steered typically by the front wheels. A small number of passenger vehicles utilize four wheel steering but that will not be included in this discussion. The investigator should have a basic understanding of both systems and the differences between them. This will assist in making a determination of the potential failure modes of each system post-crash. This will also help an investigator differentiate potential pre-crash damage from crash damage.

The suspension system not only suspends the vehicle, it also acts as a damping system to isolate the rest of the vehicle from the road. By using springs and shock absorbers of some sort, the vehicle suspension can isolate the vehicle occupants from the potentially significant forces seen by the tires and wheels while still allowing for some road feel for the driver. A properly designed and maintained suspension system will provide good driver feedback and behave predictably during vehicle operation.

The steering system also provides some damping of road forces. The tires and wheel can see significant forces during pothole strikes and while turning in normal vehicle operation. Through the use of gear reduction and power assisted steering; the forces seen by the driver are minimized allowing the vehicle to be easily controllable during normal operation. A properly designed and maintained steering system will also provide good road feedback and behave predictably.

The initial post-crash inspection of a steering system or suspension system should be a visual inspection. It is also important to document both systems to the extent possible. If the vehicle was not thoroughly documented at the crash scene, it is important to remember that significant damage can occur during the removal and transportation

Figure 4: Steering and suspension examples www.collisionpublishing.com

Collision Magazine - Volume 8 Issue 2 31

of the subject vehicle. The person or persons removing the vehicle from the crash scene may not be concerned with evidence preservation unless they have been instructed to protect certain parts of the vehicle prior to removal. The investigator will not only need to differentiate pre-crash damage from crash damage, but also from potential transportation damage. This requires the investigator to have a basic understanding of removal and transportation methods as well.

If damaged components are discovered during the inspection, it is important to protect those components from further damage. If components are found to be missing, finding them should be a priority. It is possible they were retained and are inside the vehicle but a scene visit may be necessary as components are regularly left behind by recovery personnel..

m

echanical Examination and Evaluation – Scientific Method “It is a HUGE mistake to theorize before one has data. Insensibly one begins to twist facts to suit theories, instead of theories to suit facts.” – Sherlock Homes, A Study in Scarlett.

The Scientific Method (Figure 5) is a systematic method of investigation. It is well explained in chapter 4 of the National Fire Protection Association’s (NFPA’s) Guide for Fire and Explosion Investigations. This document is also referred to as NFPA 921. In that document, it is defined as a seven step process that begins with recognizing the need (identifying the problem) and ends with selecting a final hypothesis.

Step one is recognizing the need or identifying the problem. In the case of a crash investigation, the problem is a crash occurred so there is a need to investigate. The next step is to define the problem. Again, a crash occurred and defining and understanding the root cause of the crash is necessary. The next step is data collection. Data collection begins the moment an investigator begins work on a case.

The investigator responding to the crash scene begins collecting data upon arrival at the scene, prior to getting out of their vehicle. Data collection and analysis are an ongoing process throughout the investigation. The next four steps are iterative meaning that they may occur multiple times depending on what facts are discovered during the investigation. While the investigator may develop an immediate hypothesis as to the cause(s) of the crash, those must be set aside until data can be collected and analyzed.

Otherwise, there is a significant risk of only considering data that support that hypothesis and dismissing that which does not. This is a consideration for even the most experienced investigator.

32

Collision Magazine - Volume 8 Issue 2

Figure 5: Scientific Method During that process the investigator may develop multiple hypotheses about the root cause or causes of the crash. The next steps are critical, choosing and testing the more likely hypotheses developed during the investigation and being able to explain why they are more likely. This can be done through thought experiments as well as physical testing. A thought experiment is consideration of the facts of the case in comparison to the hypotheses of root cause. Simply stated, does the hypothesis make sense based on the factual evidence? Does it explain most or all of the factual evidence? The investigative process must also include careful consideration of those facts that don’t appear to support the final hypothesis. In selecting a final hypothesis, all of the evidence should be considered including vehicle and scene evidence. Any evidence which does not support the final hypothesis should also be explained.

m

echanical Examination and Evaluation – Vehicle Inspection Vehicle inspection is an important part of any crash investigation. Inspection methods vary widely among investigators. A good vehicle inspection covers all aspects of the vehicle, regardless of initial apparent relevance to the crash. A good practice is to either develop or acquire a vehicle inspection form or checklist and use it consistently during all vehicle in-

www.collisionpublishing.com

spections performed. This insures that the inspections are done in a consistent manner and that all vehicle systems are properly documented.

Depending on when the vehicle inspection is taking place, it is a good practice to collect general and specific information about the subject vehicle prior to the actual vehicle exam. A complete vehicle exam is unlikely to occur at the actual crash scene so typically an investigator would have some information regarding vehicle make, model, production year and VIN prior to the vehicle mechanical exam.

If possible it is also helpful to know the subject vehicle’s service history as well. At a minimum, it is important to understand what the undamaged vehicle components on the subject vehicle should look like.

There are good general and vehicle specific references available including:

4. Complete inspection of the vehicle interior including: a. Glove box. b. Door Pockets. c. Under seats. d. Center console. e. Trunk.

ii. Vehicle components.

5. Component inspection including: a. Tire / wheel condition including tire tread depth and air pressure.

1. Haynes or Chilton’s vehicle manuals.

i. Empty or partially filled vehicle fluid containers.

i. Document tire size, DOT numbers, brand, damage.

2. Exemplar vehicles (most car lots will allow you to examine exemplar vehicles on their lots).

b. Steering linkage / suspension condition.

3. Other people.

c. Brake condition including, pedal, pedal linkage, and master cylinder.

4. Subject matter experts.

6. Other mechanical condition.

5. Vehicle mechanics.

a. Engine

6. Internet sites (be cautious with this).

There are good sources for specific vehicle information including:

b. Transmission c. Axles

1. CarFax or AutoCheck for VIN specific information.

d. Drive Shaft(s)

2. Vehicle service manuals.

e. Fluid levels / leaks 7. Under vehicle visual inspection (on hoist).

3. Recall information (NHTSA or Dealer).

4. Service Bulletins (NHTSA or Dealer).

a. Fluid leaks.

5. SAE, ASME research papers.

b. Damage.

6. General internet search with VIN or make and model.

c. Tires and wheels including back sidewalls.

A typical vehicle inspection could consist of the following steps:

8. Identify potential areas for possible destructive inspection as required.

1. Four or eight photos around the vehicle (4 corners, 4 sides).

echanical Examination and Evaluation – Pre-crash v. Crash Damage Separating crash damage from pre-crash damage can appear to be a complicated process but it does not have to be. The crash investigator need not be an expert in all aspects of vehicle inspection. It is more important to have a basic understanding of the systems to determine when further examination by

2. Photos of the vehicle identification number (VIN) in two locations. a. Photos wide to close-up to locate areas of interest.

3. Note general vehicle condition and age.

m

www.collisionpublishing.com

Collision Magazine - Volume 8 Issue 2 33

“

Catastrophic damage can appear visually different from long term damage. Parts that are shiny, clean or appear new may indicate recent vehicle service. If broken parts don’t appear to have been crushed in a direction consistent with the PDOF, further evaluation may be necessary. Depending on the findings, it may be necessary to do further lab evaluation of parts including but not limited to optical or electronic microscopic inspection, metallurgical analysis, or other analytical testing.

The first step in damage evaluation is to identify which parts are not in their original configuration or location on the vehicle.

"

subject matter experts may be warranted. The first step is to have a basic understanding of the systems being evaluated. That will take some effort on the part of an investigator who is relatively unfamiliar with vehicle mechanical systems. Understanding how the system should function when operating properly can give some insight into how it could behave when it fails. It is important to remember that, depending on the circumstances; the pre-crash damage may or may not have contributed to the crash being investigated.

The first step in damage evaluation is to identify which parts are not in their original configuration or location on the vehicle. This may be obvious depending on the vehicle damage. If a subject vehicle has taken a significant direct hit on the left front corner, an investigator would expect to see damage to the vehicle body, possibly wheel and tire damage and maybe bent or broken steering and suspension parts or damaged brake parts in that area. If the driver is claiming some kind of mechanical failure contributed or led to the crash however, the next step should be to determine if the damage in the contact area is consistent with impact damage. This may seem like a daunting task but knowing the approximate principle direction of force (PDOF) and knowing where the parts should have been prior to the crash can assist in determining if the damage is consistent with the crash. Parts will generally deform or break in the direction they are pushed. Parts damaged during a crash will generally fail at the smallest cross section.

34 Collision Magazine - Volume 8 Issue 2

Continuing with the left front corner crash, a complete vehicle inspection would reveal all damage to the vehicle.

Careful consideration should be given to damage in other areas of the vehicle away from the actual crash damage.

Other damage may be crash related such as a tire on the opposite side that has been de-beaded. If the driver in this case claimed the brakes failed and a leaking brake line is found on the right front corner, away from the crash damage, that may indicate a pre-crash failure. The next step would be to examine the other brake lines on the car and the rest of the braking system. Does the line appear to have been leaking long term? Is the master cylinder empty or low?

Finding a failed mechanical part is only the start of determining the root cause of the crash. Even if that part was worn out or even failed prior to the crash, it may or may not have contributed to the incident. This is where a consideration of all of the data collected becomes very important. If, for instance, brake fluid is found all over the back of the right front tire and on the road, the master cylinder is found to be low and the other brake lines on the vehicle are found to be weather cracked or bulged, this may indicate that the lines are all very old and the subject one was leaking pre-crash. If this is the case, the driver may have experienced a partial brake failure. This does not necessarily mean that the failure caused the crash. Even with a brake fluid leak on one corner, modern brake design allows for other brakes on the vehicle to continue to function. Furthermore, the driver may have been aware of the issue and continued driving the car. The driver may have a container of brake fluid somewhere in the car and been topping the fluid off to avoid making the repair. Depending on the driver’s claims, a complete vehicle inspection including documenting partial containers of oil, brake fluid or other vehicle fluids can become important.

s

ummary Crashes where a mechanical failure is the suspected cause or trigger are a relatively small percentage of the crashes that occur. To properly evaluate these cases, an investigator must have a basic understanding of vehicle systems and appropriate mechan-

www.collisionpublishing.com

ical inspection techniques. The discussion here has focused on some of the basics of vehicle mechanical systems, a brief discussion of the scientific method and its relevance to crash investigation and mechanical inspection, one methodology for vehicle mechanical inspections and a very brief overview of how crash damage can differ in appearance from pre-crash damage. Crash investigators are not expected to be experts in all aspects of vehicle mechanical systems. Subject matter experts can be consulted as necessary. A crash investigator should be particularly careful to evaluate all evidence in a crash to determine the root cause, even that which does not support the final hypothesis. Mechanical component damage that is determined to be pre-crash may or may not be causal.

r

eferences 1. Longhurst, C.J. (2013). Carbibles.com.

Retrieved April 2013, from www.carbibles.com

2. National Highway Traffic Safety Administration, (2013). National Automotive Sampling System. Retrieved April 2013, from http://www.nhtsa.gov/NASS

3. Joscelyn, K.B., Treat, J.R. (02/01/1974). The Role of Steering and Suspension Degradations and Failures in Accident Causation, SAE Technical Paper 740150.

4. Pascarella, R.J., Volger, M.M. (04/14/2008). Analysis of Tie Rod Separations in Motor Vehicle Crashes., SAE Technical Paper 2008-01-0177 5. Durisek, N.J., Granat, K.J., Holmes, E.W. (04/20/2009). Analysis of Front Suspension Ball Joint Separations in Motor Vehicle Crashes, SAE Technical Paper 2009-01-0101 6. Dunn, A.L., et al.(04/14/2008). Vehicle Handling and Control Following Front Ball Joint Failure, SAE Technical Paper 2008-01-0171 7. American Trucking Association, Technology & Maintenance Council (2007). Radial Tire and Disc Wheel Service Manual.

8. American Trucking Association, Technology & Maintenance Council (2010). User’s guide to Wheels and Rims.

www.collisionpublishing.com

Collision Magazine - Volume 8 Issue 2 35

CRASH WITH US...

Easy to Use • 2D • 3D • Accurate • Accident & Crime Scene Testimonial When I first became interested in Accident Reconstruction, I researched a lot of programs. I found that Visual Statement was one of the best, if not the best. I was employed as a State Trooper at the time, and the department eventually purchased the program for the acting Accident Reconstructionists. [...] After starting my own business, I tried a couple of different software companies that were okay, but did not have the capabilities or the ease of use as EdgeFX. I finally buckled down and purchased the program and I am extremely pleased. [...] My recommendation is to go with Visual Statement !!!

Dale Donham Owner, Donham Accident Reconstruction, AR

visualstatement.com/demo

SIMPLY THE BEST RECONSTRUCTION SOFTWARE

• Diagramming • Reconstruction • Animations • Point Cloud

CALL US NOW TO

BOOK A LIVE DEMONSTRATION

OR TO FIND OUT MORE

1.888.828.0383

Driver Distraction Obtaining Mobile Device Digital Evidence

A

n array of Electronically Stored Information (ESI) from varying sources exists following a transportation related accident where the driver has been distracted by using their mobile device. Once collected, this data can be extremely valuable in criminal and civil litigations.

Driver distraction is a leading factor in many crashes, and cell phone use and texting are two of the most common distractions. The term cell phone can be used to describe basic feature phones as well as the more popular smart phones. Smart phones offer traditional telephony features as well as broadband data services enabling web browsing, email, social media like Facebook and Twitter, and 3rd party applications, many of which are designed for use while mobile.

At a high level, driver distractions fall into four types: • Visual – looking at something other than the road • Auditory – hearing something not related to driving • Manual – manipulating something other than the wheel • Cognitive – thinking about something other than driving While there are many distractions which may prevent a driver from focusing on the com-

38 Collision Magazine - Volume 8 Issue 2

plex task of driving, the focus here will be on the distractions that have received the most attention: driver use of cell phones, text messaging, and usage of other electronic devices brought into the vehicle.

There have been numerous studies conducted to assess the impact of driver distraction as a contributing cause of motor vehicle accidents. These studies and common sense have contributed to the passage of laws and regulations by state and local jurisdictions to address these unsafe behaviors.

The National Highway Traffic Safety Administration (NHTSA) estimates that 16% of fatal crashes and 20% of injury crashes in 2009 involved at least one distracted driver (NHTSA, 2010a). Similarly, the more detailed investigations in the National Motor Vehicle Crash Causation Study (NMVCCS) found that in those crashes where the critical reason for the crash was attributed to a driver, 18% involved distraction (Ascone et al., 2009).

Another study found that 29% of the passenger vehicle drivers in NMVCCS crashes and 20% of the large truck drivers in Large Truck Crash Causation Study (LTCCS) crashes were distracted or inattentive (Craft and Preslopsky, 2010). These and other studies taken together conclude that drivers were distracted in 15% to 30% of crashes at all levels, minor to fatal. The proportion of distracted drivers may be greater because investigating officers may not detect or record all distractions.

www.collisionpublishing.com

Mike May and Andrew Russell, Esq. The Cannon Group www.collisionpublishing.com

Collision Magazine - Volume 8 Issue 2 39

The Governors Highway Safety Association publication Distracted Driving – What Research Shows and What States Can Do (GHSA, 2011), reported that crash reports may record driver distractions that the investigating officer believes caused or contributed to the crash (NHTSA, 2010a). Crash reports probably under-estimate distractions for two reasons. First, distraction is difficult to detect as drivers may not admit to being distracted before a crash and, second, there may be no physical evidence of a distraction after the fact. This article intends to introduce various sources of digital data and evidence that does remain after the fact. This evidence may support or refute allegations of driver distraction by means of an electronic mobile device.

M

obile Devices The term “mobile device” is far more inclusive than just cell phones or smart phones.

Today’s personal electronic devices also include tablets, notebook PCs, and an array of other connected devices. Connected devices can be defined as devices that are connected to a network, which may be another source of relevant evidence. Forensic examination of cell phone data has been an investigative technique performed by government and civilian examiners for over two decades. There is a wide variety of commercially available digital forensic tools designed specifically for cell phones.

These tools have primarily focused on the data traditionally associated with cell phones to include call logs, contacts, SMS (Short Message Service) or text messages, calendars, browser history, etc.

The advanced capabilities inherent in today’s smart phones, tablets, and notebooks leverage the diverse radios and on-board sensors to support feature rich applications designed and optimized for a mobile environment. Examples range from navigation applications like Google Maps, to location-based social media like Facebook and Twitter, to location-aware services like FourSquare and Yelp. Much like traditional computers, these devices create logs that record device activities, often including network connections, location, and timestamps. Third party applications like those referenced above often create activity logs which are written to a SQLite database that is stored on the device or on media like an SD card. Tools and techniques beyond the mainstream cell phone forensic solutions may be necessary to acquire, parse and examine this contemporary and relevant data.

Contrary to the statement offered in The Governors Highway Safety Association publication Distracted Driving – What Research Shows and What States Can Do that “there may be no physical evidence of a distraction after the fact,"

40

Collision Magazine - Volume 8 Issue 2

it is apparent that digital evidence, often in the form of metadata, does exist within the mobile device. This offers a window into the activities of the operator prior to, at the time of, and after the accident or incident.

While it is always best to properly seize and secure the mobile device as soon as possible following the accident or incident, relevant remnant data will likely exist on the device for some time. In any event, best practices should be followed when seizing the device. If the device is in the control of another, efforts should be made to obtain a digital image of the device for an independent digital examination and record all steps in the chain of custody, of that device.

While traditional cell phone forensic tools will reveal relevant data, additional data is likely to be found through advanced examination techniques that involve acquiring, parsing, decoding, and examining data derived from third party applications, system files and logs.

C

ellular Network Provider Call Detail Records Network Provider Call Detail Records (CDRs) are reports or files that provide activity reports for a given mobile device for a specified period of time.

It’s important to know that CDRs are provided with cell location data and without cell location data. If the court order or other process fails to specifically demand the cell location data, the CDR will not include it. CDRs typically offer details pertinent to three types of activities or services; voice, data and SMS usage.

Voice call records usually include but are not limited to: •

Connection date and connection time

•

Originating and/or terminating number

•

Elapsed time

Data usage records include much of the same data as other reports but unlike calls with origin and destination numbers, data usage records include the number of data bytes transmitted up and received down. These records would indicate that a data network connection was established.

Data connections enable access to web-based services like Internet browsers, social media sites, and other on-line services.

SMS detail records include date and time, originating and terminating numbers, and a description of what is known about the message. CDRs vary from carrier to carrier so it is important that the analyst have a working knowledge of the subtleties between them. As an example on the AT&T network, if you requested call records and the target subscriber’s device is an iPhone with the operating system iOS

www.collisionpublishing.com

5 or above and they send a text message to another device with the iOS 5 or above operating system, text messages between these devices will not appear in the detail records provided. This is because these text messages are delivered via the iMessage service through Apple. AT&T only captures the transaction as a data session.

C

DR - Location Information If location information was requested and included in the CDR, information about cell sites that serviced the mobile event will be included.

Data includes: •

Location Area Code (LAC) - A "location area" is a set of base stations that are grouped together to optimize signaling

•

Cell ID (CID) – A unique number used to identify each cell site or base transceiver station (BTS)

•

Longitude

•

Latitude

•

Azimuth – Center point of the sector that serviced the call

curred along a route that included the location of the accident. A timeline analysis indicated that a call occurred minutes before the accident was reported to 9-1-1. This call had been processed by the cell tower and sector that served the location of the accident.

Cellular providers do not retain Call Detail Records for an indefinite period of time. As such, it is important to identify early, cases that warrant these records. Steps should be taken as soon as practicable to obtain the records or at least notify the carrier of your intent to obtain the records through judicial or other legal process. If location data is desired, it must be specifically requested in the order.

N

etwork, Server and Peripheral Log Record Analysis Log records are a product of today’s communication and networking infrastructure components.

Log records memorialize Internet communications and network connections made with the various devices along the communication path. At each device along the path, logs will be created that record activity and events that provide valuable evidence. These devices and their log records routinely include:

•

Routers and Switches

As most cell sites transmit and receive using three separate antenna arrays, the azimuth data can be useful in establishing an approximate location of the mobile device.

•

Firewall Logs

•

Web Server Access Logs

Location information can be extremely informative when attempting to answer the questions where and when. Location data is most demonstrative when plotted on a map or overhead imagery. CDRs with location data can be plotted to show the location of the serving cell site along with the azimuth or cone of coverage area that serviced the call. A series of events and their associated cones can be used to derive direction of travel.

•

Intrusion Detection Systems

•

Email Servers

•

File Transfer Protocol (FTP) Servers

•

Domain Name System (DNS) Servers

In a recent case, a commercial motor vehicle was leaking diesel fuel along its route of travel which included a highway on-ramp. Shortly after the vehicle leaking fuel passed, a motorcyclist entered the on-ramp and lost control of the vehicle on the slick road surface causing severe injury and property damage. The vehicle leaking the fuel was located a short time later some distance from the scene of the accident. The driver claimed to have never been on that onramp or even in the area of the motorcycle accident. Call detail records were obtained for the driver’s cell phone. The subsequent analysis indicated that the driver had made several phone calls over a period of time that preceded the accident up to the time that the vehicle was located some distance from the accident. Plotting the location data indicated that the phone was used for several calls that oc-

One of the greatest values of log records is that they exist separate from the mobile device. This allows an examiner to collect and analyze activity that took place on a device even if the device is no longer available or has been damaged.

The type of information that may be retained in a log varies depending on the function and features of the network system or device. For example, firewall log records may show what IP addresses accessed the server, what internal systems were accessed, and a timestamp and duration of the connection. Web Server access logs may include metadata such as Web sites visited, IP address, and detailed click-by-click activity. The collection and analysis of these composite records can create a powerful temporal exhibit demonstrating device connection and use over a brief or detailed period of time.

www.collisionpublishing.com

Collision Magazine - Volume 8 Issue 2 41

The retention period for log records is defined by the carrier, Internet Service Provider (ISP), or corporate internal policy. Like call detail records, the data may differ from company to company and system to system. With so many disparate systems involved in any given network connection, a thorough examination and analysis may reveal systems or devices that may retain the relevant data even though it may be unavailable from the specific system being examined.

“

Paramount in these investigations is the timely process of identifying the records and their location, preserving the records with the custodian, and executing the legal process to access or collect the records.

"

Paramount in these investigations is the timely process of identifying the records and their location, preserving the records with the custodian, and executing the legal process to access or collect the records. After identifying the log records, they should be preserved pending further legal action with a preservation letter, or other appropriate legal process within the jurisdiction.

The resulting analysis and report can present conclusions about the activities performed as well as a time line of events that may be useful in establishing that a mobile device was in use at a particular time. The analysis may simply reveal that a device was logged onto a network or the analysis may indicate the click-by-click activity that was 42 Collision Magazine - Volume 8 Issue 2

being performed. Equally important are the log records that provide identifying information about the user of the device.

As an example, a residential real estate agent is driving clients between home showings. The agent uses an iPad to show clients pictures and provide information about the homes. The iPad has 4G cellular connectivity and accesses a subscription service that provides the pictures and information. While driving, the agent turns the iPad on, establishes a network connection via the cellular carrier, launches the application, enters a login and password, and while navigating the server to access the appropriate listing, drifts into a bike lane, striking and severely injuring a child cyclist.

Log records of this activity will reside in a number of locations. First, the iPad itself will archive the network connection made to the cellular provider as well as the launching of the agent’s application. Second, the cellular carrier will have network records that indicate the date and time of the connection as well as the bytes of traffic that were transmitted and received. Lastly, the third party server that hosts the listing service application will have logs that identify the customer via the login and password. This information will support attribution to the driver. Click-by-click activity may also be logged indicating the steps that were performed by the driver/real estate agent while navigating the site. All of the log records apart from the mobile device itself may offer activity-based time stamps, and possibly location data artifacts.

P

lanes, Trains and Automobiles It’s worth noting that driver distraction is not limited to the operation of just automobiles. One of the worst train wrecks in recent history was caused by an engineer who was texting just before the collision and missed critical signals.

In a separate incident a phone conversation contributed to the Hudson midair collision between a tour helicopter and a Piper Lance that killed nine people. The National Transportation Safety Board report stated the tower controller at the Piper’s departure airport was “engaged in a bantering personal phone call about a dead cat while directing traffic.” The NTSB report noted that during the time the controller was on the phone, he made several errors. For example, after he made the handoff of the Piper to Newark tower for flight following, he failed to provide the pilot the new frequency for two minutes. When he finally called with the new frequency, he spoke very quickly. The pilot read back an incorrect frequency to the controller, but the controller did not notice the difference.

www.collisionpublishing.com

C

onclusion Today, many digital forensic tools used to process evidence from cell phones acquire data from specific locations in the data storage space on the phone and associated subscriber identity module (SIM) card.

Essentially, these tools are designed to retrieve data with forensic value in storage areas where it is expected to be located. Issues arise, however, when this methodology may not identify and collect the data that is valuable to a transportation related incident. That relevant data may be located in other file locations not specifically targeted by the forensic software or hardware.

This issue grows more acute with the introduction of more capable and feature rich smart phones, tablets, and other connected personal electronic devices. These mobile devices provide increased data storage capability and, possibly more complex network activity. In addition, relevant data that may be overlooked could be important as inculpatory evidence, or even exculpatory evidence. As a result, it is imperative that consideration be given to alternate examination and discovery techniques to ensure that all efforts to discover relevant data are utilized when investigating incidents that may have been caused by driver distraction.

"This (distracted driving) is becoming the new DUI. It's becoming epidemic," said NTSB member Robert Sumwalt. While the physical evidence of such distraction may not be as evident as skid marks or vehicle damage, digital evidence or electronic records likely exist either at the scene, on the mobile device, or separate and apart from the device at the network provider, and other third party servers. While the collection of the digital evidence may be more complex than measurements or the photographs taken at the scene the relevant data available may weigh heavily on supporting or refuting the allegation of driver distraction through the use of a mobile electronic device.

Should there be a notion that driver distraction was a contributing cause to the accident, whether or not it is cited in the police report, steps should be taken early to identify the source(s) of data to be collected. If the data is possibly located on a network system, it is important to notify the custodian of these records of the intent to obtain the records through judicial or other legal process.

Once the records are obtained it is crucial that the data be examined and analyzed by a professional familiar with geospatial analysis. The end result of the examination and analysis should be the generation of a report and exhibit to substantiate the merits of the case.

The various sources of digital data and evidence that remain after an incident can be of great value in any case.

One of the greatest values is the ability to prove, or disprove, the necessary elements of proof in a criminal or civil proceeding. This evidence may support or refute an allegation that a mobile device was in use at the time of the incident, and that such use was a causative factor supporting a claim of driver distraction.

About Mike May Mike has 24 years of experience conducting technical investigations. He served 23 years in the law enforcement community retiring from the Connecticut State Police where he managed the Electronic Surveillance Unit for 13 years. Mike has been engaged in the development of mobile device data exploitation tools and techniques for over ten years. He provides technical training, operational support, and policy guidance nationally and internationally. Mike served as the President of the National Technical Investigators’ Association and was a member of the board of directors for 15 years. Today, Mike bridges the technologies once developed for government purposes to applications supporting the mobile e-Discovery sector.

About Andrew Russell, Esq. After serving over 24 years in law enforcement, specializing in computer and digital data forensics, Andrew now concentrates on research and development of software solutions and providing technical support for mobile device forensics. In addition, Andrew provides litigation support for civil and criminal matters where mobile device and related data may be relevant to the proceedings. Andrew is a licensed attorney. Andrew has experience in forensic analysis and investigation of computers and mobile devices, drafting policies and procedures for handling digital evidence, and development of innovative software applications for exploitation of data from various mobile device platforms, specifically as it relates to location-based data.

About The Cannon Group, Inc. The Cannon Group (TCG) is a Mobile e-Discovery solution provider. TCG offers products and services focused on the E-Discovery challenges of mobile devices and related Electronically Stored Information (ESI). TCG follows the EDRM methodology and can support your personnel through all or any phase of the process from Identification through Presentation. The Cannon Group’s staff and resources include attorneys, examiners, network professionals, analysts, graphic artists and project managers.

www.collisionpublishing.com

Collision Magazine - Volume 8 Issue 2 43

Measure and Diagram Scenes and Vehicles from Photographs New!

· flexible total station importer las · er scanner integration · 64 bit version

PhotoModeler ®

· Quickly clear and measure scenes with your camera (ground-based or optionally aerial / UAV photography) · Clear road scenes more quickly and with greater safety · Measure crush using photos from your camera, or unknown cameras · Extract key evidence (distance, position, angles) from 3rd party photos · Model vehicles in 3D using a camera · Photo-based measurement for law enforcement, engineers, and reconstructionists since 1993 · Self-train with detailed video tutorials, attend public training held several times each year, or book a private course · Floating concurrent network licenses available for multiple users

PhotoModeler and PhotoModeler Scanner share these common features: · Fully integrated support for digital and film cameras · Use 3rd party photos and unknown cameras with InverseCamera · On-photo dimension lines for basic diagrams · Export to other CAD, diagramming, or visualization packages (15 formats) · Sophisticated integration of control points from total stations or other sources · Map scenes over large distances using Project Merge · Extract photographic texture for realistic 3D models · Support for single photograph projects - great for forensics · Ortho-photo generation for that “birds-eye” view of your scene · Use the Surface Draw function to add 3D points, lines and curves traced on a single photograph

www.photomodeler.com

PhotoModeler: $1145

· Everything you need for scene diagramming, crush measurement, and more · Includes automation via Coded Target printing and recognition

PhotoModeler Scanner: $2595 · Adds textured-surface densepoint-cloud scanning

* Prices in USD and include free shipping by FedEx

®

Shoe print with PhotoModeler Scanner

1-604-732-6658 toll free: 1-866-532-6658 email: sales@photomodeler.com

William Brem ARCCA, Incorporated, ACTAR #1222 Wayne Denham ARCCA, Incorporated, ACTAR #1995

V

ehicle inspections are an important aspect in the reconstruction of motor vehicle accidents. Often, the reconstructionist is one of the first non-involved parties to examine and document a vehicle after a crash. As part of the vehicle inspection and initial documentation, there may be evidence that could be collected, but is sometimes overlooked, that could ultimately become vital to either the reconstruction itself or to other aspects of a case. This article highlights some of the more important aspects to be aware of that are not often addressed. It is not intended to cover all of the basics of a vehicle inspection, and is not all inclusive. Many of these items may not be required to perform a standard reconstruction, but nonetheless may become important ques-

46 Collision Magazine - Volume 8 Issue 2

tions that are asked of the reconstructionist, or of others, later on as the case evolves.

Photography: Photography is one of the best methods of preserving evidence. When beginning the vehicle inspection you should take several overall photos from all four sides of the subject vehicle and oblique views from all four corners to establish the condition of the exterior vehicle as you found it. Photographs of the interior of the vehicle should also be taken through the windows and opened doors to document the “as found” condition of the interior of the vehicle. Take these photographs before you have disturbed anything. In a single frame for each, attempt to capture a photograph of the entire side, front, rear, and opposite side of the vehicle first. In addition, attempt to take several photographs of the top

www.collisionpublishing.com

and bottom of the vehicle. Because of how/where a vehicle is stored, this is not always possible. During the early stage of photographing the vehicle, take photographs of identifying aspects of the vehicle, including the license plate, the VIN plate (usually visible under the bottom left edge of the windshield), build plate (usually on the B-pillar or on the B-pillar edge of the driver’s door) and other indicators of the vehicle’s trim level should be taken. After completing these initial photographs, the investigator should concentrate on closer photographs involving the areas in question. Before ‘zooming in’ on a close view though, it is best to ‘walk’ the photographs sequentially into the close up, because if a particular bolt, head light, dent, etc. is the focus of the close up, a simple close up photograph may not allow someone else (or you) to later identify where or what side of the vehicle the close up is from. By ‘walking the sequential photographs in,’ it will be clear when reviewing the photographs sequentially where the item is located. The addition of simple identifiers is also helpful. For instance if the object is the left front tire, simply putting an “LF” on a small piece of paper on the tire and having it appear in the photographs is of great help. When using videotaping equipment in addition to, or instead of still photography, many of the same procedures should be followed, i.e., once recording begins, slowly walk all around the vehicle before focusing in on a specific area.

S

pecific Inspection Items: Lamp Inspection: Many cases involve a question of head light usage, brake usage, or turn signal usage. In some cases the condition of the filament of the bulb may assist an investigator in determining its usage.

This information should be documented as soon as possible during an investigation. Frequently, after a collision there is scientific evidence that can be collected or photographed from a bulb, or even its remnants, that could help answer the question of whether a lamp was “on or off” at the time of the collision. At the first inspection it is not always known if this will be an issue, so if possible it is a good practice to inspect the bulbs that are accessible. However, unless you have specific permission, you do not want to disassemble any part of a vehicle.

www.collisionpublishing.com

Collision Magazine - Volume 8 Issue 2 47

This article is not attempting to cover all aspects of light bulb filament examination and/or analysis; other references such as Lamp Examination for Traffic Collision Investigators 1 and Insurance Fraud and Vehicle Collisions 2 provide more details. In general though, start by inspecting, documenting, and photographing the condition of the area around the outside of bulb. Note if there is direct contact damage, induced damage, possible pre-existing damage, etc.

If accessible inspect and take close up photographs of the bulb and its filament, as seen in Figure 1. Having a means of identifying the individual light bulbs greatly assists the documenting process.

“

Having a means of identifying the individual light bulbs greatly assists the documenting process.

"

Seatbelts: Seat belt usage is another frequently asked question that is not always known to be a question during the initial inspection. Several specific close-up photographs can prove to be very beneficial in providing additional evidence to assist in determining whether or not the seat belt was in use at the time of the collision, if and when usage becomes an issue. Start by determining if the seat belt is jammed in its stowed position or jammed in an extended position. Photograph and document your findings. Note items such as the condition of the buckle portion of the seat belt, is there shattered glass from the impact now resting in the buckle opening? (Figure 2a) Is the webbing cut, where is it cut? (Figure 2b) Determining if it was cut to get an occupant out or cut while stowed when rescuers cut the roof off of the car to get someone out is a question that might need to be answered. Did the seat belt webbing rupture or was it cut? If there is a webbing separation, photograph the ends of the separation. Photograph the tongue portion of the seat belt latchplate (Figure 2c) the edges of the latch plate (Figure 2d), as well as the top and front edge of the D-ring where the seat belt loads during a crash.

These are the locations where the webbing routes, turns, and loads during a crash. The loads and friction generated during a crash often result in an abrading of the material when the crash is sufficiently severe. This topic is covered in more detail in references such as A Guide to Determining Occupant Seating Positions from Physical Evidence and Injury Pattern, 3 Diagnosis of Seat Belt Usage in Accidents 4 , and Assessment of Timely Retractor Lockup in Automotive Seat Belt Systems. 5

Figure 1: Close up photographs of a bulb and filament 48 Collision Magazine - Volume 8 Issue 2

www.collisionpublishing.com

Figure 2a: Buckle portion of the seat belt

Figure 2b: Cut seat belt webbing

Figure 2c: Tongue portion of the seat belt latchplate

Figure 2d: Edges of the latch plate

Vehicle tires: Depending on the circumstances surrounding the crash event, there may be questions about tire failure or tire defects. Valuable evidence regarding the tires can be collected from careful inspection. Be sure to document each tire including its make, model, size, and DOT number, as well as photographing its tread (and if possible any differences over the circumference of the tread that might indicate a skid or yaw). Tires that are unevenly worn (Figure 3a) or improperly inflated can lead to a possible hydroplaning incident. Perhaps the tires are unevenly

worn indicating a potential alignment issue. Is there obvious tire damage? Is there any way to determine if the tire damage occurred as a result of the collision or does it appear more likely that the tire damage occurred before the crash? It is also an important detail in accident reconstruction to know if the wheel(s) became locked (Figure 3b) or “pinned” due to collision damage, as this will affect the post-impact energy-speed calculations. Other references such as Tire/Rim Failure Analysis 6 provide a greater overview of this topic.

www.collisionpublishing.com

Collision Magazine - Volume 8 Issue 2 49

Figure 3a: Tires that are unevenly worn

Figure 3b: locked or “pinned” wheels due to collision damage

Transfer Marks: Direct contact will often leave an indication of what was contacted and the nature of the contact..

For instance, a tire contact will often leave a black rubber transfer mark. Where and how this mark is left can assist in determining how the interaction occurred. Also, imprints from license plates, emblems, and bumper/brush guards will assist the reconstructionist in determining how the vehicles were aligned and how the contact occurred.

At times the orientation of these marks will also assist in determining the movement of the vehicle and, therefore, will be beneficial in the reconstruction process (Figure 4).

Windshield Damage: Windshield damage can also provide valuable information about the crash. Most importantly, windshield damage may assist with determining how an occupant moved within the vehicle and/or if they were wearing their seat belt. However, there are numerous different causes for ‘spider web’ windshield fractures in

addition to the occupant’s head. These include a deploying airbag (Figure 5a) and items from the exterior such as a pedestrian or motorcyclist (Figure 5b). Because of this, the location and type of windshield fracture needs to be investigated, including the characteristic bulge created by the object impacting the windshield. Is it from the outside going in or from the inside going out? (Figure 5c and 5d).

References such as Vehicle/Pedestrian Collision Investigation Manual 7 will provide a greater in-depth discussion.

Brake inspection: In reference to cars and small trucks there are two general types of brakes; drum brakes and disc brakes. Both types are operated through a hydraulic system.

Late model cars almost exclusively have either a 4 wheel disc brake set-up or 2 disc brakes (Figure 6a) on the front wheels and 2 drum brakes (Figure 6b) located on the rear wheels.

Figure 4: Transfer marks 50

Collision Magazine - Volume 8 Issue 2

www.collisionpublishing.com

Figure 5a: Deployed airbag

Figure 2b: Windshield damage

Figure 5c: Outside-in windshield fracture

Figure 5d: Inside-out windshield fracture

“

Due to the friction caused each time the brakes are applied, brake pads will begin to show wear and will eventually wear down. Heavily worn brake pads eventually need to be replaced; if not replaced, the metal base holding the brake pad will start to come into contact with the inner drum or disc causing it to scorch and become etched. A brake in this condition will be ineffective in stopping the vehicle, especially at higher speeds or in wet conditions.

After a crash, brakes should be inspected for wear including measuring the brake pads (when possible), checking the wheel cylinder and pushrod, and inspecting the master cylinder and brake lines, along with the calipers and discs for any possible malfunctions. Photographs should be taken of all these components when there may be a question concerning brake conditions or the vehicle’s stopping ability. Any broken or leaking brake line should also be documented. If possible, attempts should be made to determine if this break or leak could have been caused by the collision damage or if the failure preceded (and maybe was a causative factor in) the crash event. Other references on this topic and others covered in this article providing a greater overview include: The Traffic Accident Investigation Manual 8 and Traffic Accident Reconstruction. 9

Due to the friction caused each time the brakes are applied, brake pads will begin to show wear and will eventually wear down.

www.collisionpublishing.com

Collision Magazine - Volume 8 Issue 2 51

Figure 6a: Disc brakes

Figure 6b: Drum brakes

Airbag EDR: Many cars now store information in the Airbag Event Data Recorder (EDR) that can be downloaded.

If a download is not being done in conjunction with the inspection, an investigator should note telltale evidence if one has been performed and/or how feasible it would be to perform a download. Because data is not always locked, an initial investigator should not to apply electrical power to the vehicle if there is a chance that data may be lost. Bosch provides the equipment to perform the EDR download on the vast majority of publicly downloadable vehicles. ARCCA provides a free smartphone (iOS and Android) app that allows for a quick search of a complete list of the Bosch downloadable vehicles as well as what information is likely to be contained within the download. The app is free and available in the app markets or at www.arcca.com/mobile-app/

It is possible that the fractured component played a role in causing the crash event.

Broken Component: During the course of an inspection, the investigator may find a broken component. Commonly found broken components include axle shafts, sway bars, shock absorbers and/or their mounts, struts and/or their mounts, lug studs, etc. Do not try and piece these back together, metallurgic analyses after the fact are best done if the fracture surfaces are ‘fresh’ and because the sections fit together so well, the easiest way to mar the fracture surfaces is by trying to “reconnect” them. If found, photograph the ends and edges of each section and photograph them in proximity to each other, but do not allow the surfaces to touch. Similarly, if possible do not let anything else interact with the fracture surfaces. Also document the damage around the fracture area to assist in making a determination as to if the impact caused the fracture or not.

52

Collision Magazine - Volume 8 Issue 2

Child Seats: Much like seat belts, use, or proper use, of a child safety seat (for the purposes of this article a ‘child safety seat’ can be in any of the following forms, a rear facing infant seat, an infant seat with base, an integrated harness type or a booster seat) may become an issue of interest after a crash event. For this reason, when an investigator is examining a vehicle, if there is a child safety seat in the vehicle, photographs of it should be taken. If the child safety seat is still somehow attached to the vehicle, whether the attachment is by the seat belt, by LATCH anchors and/or top tether strap, this should be photographed as well. Do not disturb the condition or attachment of the child safety seat. If it is necessary to disturb its attachment ot complete another part of your investigation, documents its as found condition with additional photographs and notes. Sometimes a child safety seat (or a component of it) is in the car but may no longer be attached. If this is the case, photograph the components found. Do not attempt to connect it back into place or re-secure it to the car. Additional information can be found in references such as Validation of Loading Marks on Child Safety Seats through Testing. 10 This article was intended to assist with the investigator’s awareness of the more common items that may become an issue later in the crash investigation, but that may not be immediately apparent as issues during the initial vehicle inspection process. Several issues were detailed to increase the understanding of what to look for as well as how to properly photograph available evidence and vehicle dam-

www.collisionpublishing.com

age for future analysis. As noted, it is not intended to be a complete ‘how-to’ guide to conducting a vehicle inspection.

R

2.

3.

4.

5.

eferences: 1. Tony L. Becker, Lamp Examination for Traffic Collision Investigators. Institute of Police Technology and Management, 1995 Robert A. DuBois, Insurance Fraud and Vehicle Collisions. Institute of Police Technology and Management, 1993 Joseph N. Cofone, A Guide to Determining Occupant Seating Positions from Physical Evidence and Injury Patterns. Institute of Police Technology and Management. 1997 Charles A. Moffatt, Edward A. Moffatt and Ted R. Weiman, Diagnosis of Seat Belt Usage in Accidents.

Society of Automotive Engineers. SAE number 840396. 1984 Eisentraut, D., Muzzy, W., Cantor, A., D’Aulerio, L., Whitman, G., Brown, K., and Markushewski, M., As-

sessment of Timely Retractor Lockup in Automotive Seat Belt Systems, Society of Automotive Engineers.

SAE number 971515. 1997 6. Morris D. Smith, Tire / Rim Failure Analysis. Institute of Police Technology and Management, 2012 7. Tony L. Becker, Vehicle / Pedestrian Collision Investigation Manual. Institute of Police Technology and Management. 1997 8. J. Stannard Baker and Lynn B. Fricke, The Traffic Accident Investigation Manual. Northwestern University Traffic Institute. 1986 9. Lynn B. Fricke, Traffic Accident Reconstruction, Volume 2 of the Traffic Accident Investigation Manual.

Northwestern University Traffic Institute. 1990 10. Larry A. Sicher, Gary R. Whitman and John Yannaccone, Validation of Loading Marks on Child Safety Seats through Testing. Society of Automotive Engineers, SAE 2006-01-0906, 2006

Hyundai and Kia Crash Data, the Indispensable Compendium W. R. "Rusty" Haight

Shawn Gyorke

Collision Safety Institute

Crash Data Services, LLC

Sean Haight Collision Safety Institute

T

he Event Data Recorder (EDR) data retrieval tool built by Global Information Technology (GIT) which provides access to data stored in Hyundai and Kia vehicles was first offered for sale in 2012 such that vehicles sold by those manufacturers in the United States after September 2012 would be in compliance with 49CFR563. 1 2 The basic history of the GIT system, 49CFR563 and the basics related to the Tool are found in “Hyundai and Kia Crash Data, a Preliminary Overview” from Collision Volume 8, Issue 1 included by reference here. 3

This Compendium is a follow up to the earlier published preliminary review and is the product of a review and examination of crash test data, “real world” crashes, hands-on experiments and analysis of retrieved data. It is separated into five sub-sections: 1. General Functionality of the Hyundai and Kia EDR Tool(s) 2. A Review of Crash Testing involving Hyundai and Kia Vehicles 3. A Review of Hyundai and Kia Crash Data from "Real World" Crashes 4. Coverage Spoofing; Data from Unsupported Hyundai and Kia Vehicles 5. Compendium Conclusions and Summary

KIA 54 Collision Magazine - Volume 8 Issue 2

www.collisionpublishing.com

The first section, “General Functionality of the Hyundai and Kia EDR Tool(s),” is an examination of the installation and operation of the EDR tool and a limited look at its application. Observations related to potential installation considerations (i.e.: folder or directory choices which might be made during the initial installation process) are addressed. The operation of the Tool as it relates to how connection is made and other aspects of the hands-on use of the Tool are examined as the basis for other activity discussed in later sections of the Compendium such as the “coverage spoofing” section.

This sub-section also offers a general look at the basic report layout for Hyundai and Kia EDR reports and what one might expect to find within the various sections of those reports including the section contents and layout of the hexadecimal data as presented.

In “Crash Testing involving Hyundai and Kia Vehicles,” the authors examine a set of crash tests conducted by the Insurance Institute for Highway Safety (IIHS) involving Hyundai and Kia vehicles and compare the data retrieved from the Airbag Control Units (ACUs) to observations and instrumentation from those tests. An associated review of some of that data is also found in the paper “Analysis of Event Data Recorder Delta-V Recording in the IIHS Small Overlap Crash Test” found in this issue of Collision and particularly that information related to Hyundai and Kia vehicles involved in those tests is included by reference here.

The sub-section “Hyundai and Kia Crash Data from ‘Real World’ Crashes” is a look at data retrieved from a small group of those vehicles accessible using the GIT/ Hyundai-Kia EDR Tool(s) involved in “real world” crashes and, particularly taken in light of the data retrieved and analyzed from the IIHS tests, it offers a look at some of the EDR data in a setting other than a crash test.

Referring, in part, back to the section on functionality, the authors examine the potential for retrieving data from unsupported Hyundai an Kia vehicles in the sub-section “Coverage Spoofing; Data from Unsupported Hyundai and Kia Vehicles.”

The authors examine the potential process and variations on that process and illustrate the possible dangers of this approach to data retrieval.

In the previously published “Hyundai and Kia Crash Data, a Preliminary Overview,” the authors briefly discussed what is otherwise identified as “VIN spoofing” (using Bosch CDR Tool related terminology as a frame of reference) and suggested then that there had been some “success,” limited to the notion that “data has been retrieved,” not that it has been successfully translated accurately.

This section looks at that investigation and experimentation in more detail including some analysis of the issues arising from this otherwise ill-advised

HYUNDAI www.collisionpublishing.com

Collision Magazine - Volume 8 Issue 2 55

practice at least in part in an effort to be able to potentially identify when it is undertaken later with unanticipated results.

be made later in the functionality, operation and/or output in the GIT EDR Tool which may make part(s) of this compendium obsolete or at least “dated.”

The last section of the Compendium is the “Conclusions and Summary” in which the authors have made an effort to further connect the other sub-sections and associated observations. Without a doubt, the Hyundai-Kia EDR Tool(s) is a developing and evolving system which will undergo changes in the coming months and years as a function of this and other research. As there have already been changes to, for example the Data Limitations text found in the reports for the Hyundai and Kia EDR reports since publication of the preliminary review, one might reasonably anticipate that other changes or modifications might

Again using a comparison to (and with a “tip of the hat” to) the Bosch CDR Tool, the end users of the information found in this Compendium are urged to use the latest production release of the GIT EDR Tool software appropriately as indicated in the instructions which come with the relevant system when retrieving data from Hyundai or Kia vehicles. The authors believe that this Compendium will further illustrate what is the best way to ensure that retrieved data has been translated using the most current information provided by these manufacturers.

Editors Note:

In Volume 8, Issue 1 of Collision, in “Hyundai and Kia Crash Data, a Preliminary Overview,” the authors of this Compendium provided a “first look” at crash data newly available from Event Data Recorder (EDR) systems found in Hyundai and Kia vehicles retrievable using what will be referred to here as the Hyundai-Kia EDR Tool(s). The Compendium in this issue is the first-of-its-kind published in-depth review of several important aspects of EDR data available from Hyundai and Kia vehicles and a follow-up on that early, preliminary review. Some of the crash test data initially evaluated and discussed briefly in that earlier review is also discussed in more depth in “Analysis of Event Data Recorder Delta-V Recording in the IIHS Small Overlap Crash Test” in this issue of Collision.

We are often asked about “peer review” of Collision articles. While that topic has been addressed time and again in Collision, for this Compendium, each of the sub-sections has been written and reviewed by individuals who have actually used the Hyundai and/or Kia EDR Tool and who have reviewed and compared the retrieved data to other sources of crash data or information. The best information we have been able to develop is that less than 50 Hyundai and Kia EDR Tools have been sold commercially within North America which makes it far less likely that, for example, some later published paper would be able to assert that same claim in the context of finding an individual with experience with the system to review the text.

During the preparation and research for this article, the authors reached out to Hyundai Motor America (Hyundai), Kia Motors Corporation (Kia) and Global Information Technology (GIT, the manufacturer of the Hyundai and Kia EDR Tool(s)). Hyundai responded that they were interested in the EDR test set up “status”

56

Collision Magazine - Volume 8 Issue 2

and were looking for information on what kind of test was performed and how the testing was done as well as data on the differences between the data extracted using EDR tool and the ACU internal data. Despite follow up calls and emails from the authors offering the available information, no further communication has been had between the authors and any of those entities nor did they participate in the preparation of this Compendium.

For readers who might be familiar with the operation and use of the Bosch CDR Tool, many of the HyundaiKia EDR Tool concepts discussed in the Compendium may be easily compared to those concepts or applications associated with the CDR Tool….whether “good” or “bad.” Bosch is the manufacturer of the longest standing EDR data retrieval tool, no one from Bosch was involved in the research for or preparation of this Compendium. The Bosch CDR Tool has been in existence as the Vetronix then later the Bosch CDR Tool for nearly 14 years at this writing and the experience gained by the authors in the use of that tool over that same period in a variety of settings has been applied during the development of the Compendium. Undeniably, many of the readers of Collision will have personal experience with the use and application of the Bosch CDR Tool and, for that reason, some references have been made to the Bosch tool not so much as a “comparison,” but rather to offer a frame of reference.

For a more detailed and comprehensive look at data from these “EDR Tools,” the authors will be presenting on this article and potential developments in the meantime at the CDR User’s Summit in Houston, TX in January, 2014 and that presentation might be later referenced for updates to this text as necessary.

www.collisionpublishing.com

Section 1: Functionality of the Hyundai and Kia EDR Tool(s) Editor’s note: This section of the Compendium is offered to address the basic installation, operation and function of the Hyundai/Kia EDR Tool. For those who haven’t seen the EDR Tool and for those who might have the Bosch Crash Data Retrieval (CDR) Tool, this section is offered as a frame of reference and as background for other sections of this Compendium.

be installed. There are two installation options(Fig. 1.2); one installs the EDR Tool software to the user’s “C” drive and the other splits the program and “files” folder across two drives identified as “C” and “D.” Without a dedicated “D” drive where an end user might have a portable drive on which they store data and that drive might have a drive letter assigned differently by Windows each time the portable drive is attached, the “split installation” choice may result in difficulty saving reports later and/or lost data. For most users, the installation of all the components to a C drive is preferable.

S

oftware Installation The Hyundai and Kia EDR Tool(s) are sold as individual products with separate software applications; however, the underlying components and the functionality of the software identified as being “for Hyundai” as opposed to “for Kia” are essentially identical. Therefore, for the purpose of this compendium, the authors will only distinguish between the two applications when necessary.

As of November, 2013, an annual software subscription to either the Hyundai or Kia EDR Tool software component costs $195 (no doubt subject to change, the price at publication is offered here for information only). Initial installation software is distributed via CD-ROM, similar in nature to how Vetronix then Bosch originally distributed their CDR Tool software. When sold to an individual, the Hyundai and Kia EDR Tools are each assigned a unique user identification (ID) number and password (PW). The user's ID and password are required when installing the software application(s) (Fig. 1.1).

Figure 1.2: Installation screen In the context of typical Windows environment installation, the way the installation process completed was surprising if not disconcerting. When the software installation process completes and the user clicks on Finish, the installation executable file automatically restarts the user’s computer without at that step until the shutdown process actually begins. While there is a dialog box (Fig. 1.3) which appears briefly, the shutdown process is fast (less than 10 seconds), and if the user has other open applications, work could be lost. Therefore, the authors encourage users to shut down all other Windows applications before beginning the install process.

Figure 1.1: "LoginForm" dialog box where the user will enter a unique ID and Password The install processes for the Hyundai and Kia EDR Tools are effectively identical. While relatively problem free and simplistic, the authors identified a few subtle challenges with regard to the installation of the software packages addressed here later. During the install process, the user is prompted to select the directory to which the software will www.collisionpublishing.com

Figure 1.3: Installation screen indicating pending reboot Collision Magazine - Volume 8 Issue 2 57

The authors successfully installed both the Hyundai and Kia EDR Tool software in Windows XP (Professional 32bit), Widows Vista (Home Premium 64-bit), Windows 7 (Home Premium 64-bit), Windows 8 (Professional 64-bit) and Windows 8.1 (Professional 64-bit) operating systems.

The option for installing all the components to the “C” drive was selected for each. Apart from the surprising shutdown, the installation was generally problem-free across a wide variety of Windows operating systems; both 32-bit and 64-bit.

After successfully installing the software applications, the user is provided with two distinct shortcut icons (Fig. 1.4) on the Windows desktop. The desktop shortcuts are identified as “HMA EDR” and “KMA EDR;” associating each with the separate “Hyundai Motors America” and “Kia Motors America” software installations.

Figure 1.4: Hyundai and Kia EDR Tool Shortcuts

Figure 1.5: Power and PC Communication indicators on the VCI

Figure 1.6: Connection via the DLC Port

T

he EDR Tool hardware Similar to the Bosch CDR Tool, the Hyundai and Kia EDR Tool(s) require communication at two locations in order to facilitate access of a Hyundai or Kia Airbag Control Unit (ACU). First, the user must initiate communication with the Vehicle Communication Interface (VCI) (Fig 1.5), analogous in some ways to the Bosch CDR Tool interface module. Second, the user must also ensure communication between the VCI and the ACU either through the Diagnostic Link Connector (DLC or OBD II port) or directly to an ACU (Fig. 1.6 and Fig. 1.7).

While similar in some ways, the VCI functions in a different fashion than what the Bosch CDR Tool users are familiar with. The VCI has an added power (On/Off) button. Also, the VCI cannot be powered directly (has no direct receptacle). By way of explanation, the VCI must be powered by drawing power (12v) from the vehicle through the power and ground pins at the DLC, or through an external 12 volt power source, facilitated through the ACU cable(s).

58

Collision Magazine - Volume 8 Issue 2

Figure 1.7: Connection direct to ACU Another distinct difference is that, in order to open the Hyundai and Kia EDR Tool Software(s), the end user must have the VCI powered up and connected to the PC. Obviously, this requirement presents a distinct use case problem inasmuch as the end user cannot check for vehicle coverage list found in the software absent connecting the VCI, powering it up and then being able to open the software. When a user attempts to open the Hyundai and Kia software(s), tthe program automatically searches for communication (Fig 1.8) with the VCI.

www.collisionpublishing.com

Figure 1.8: "Checking USB connection between PC and VCI"

Figure 1.10: "This program only can be run after login. Program will be terminating"

Absent the VCI being powered and connected to the PC, the user will see an error message dialog box (Fig. 1.9) or, if the use fails to login before attempting connection (Fig 1.10), the software will close.

Therefore, the vehicle coverage list (as well as other aspects of the software) cannot be accessed absent assembling and powering of the hardware associated with the system. The Hyundai and Kia EDR User’s Manual(s) also fails to provide a vehicle coverage list, and the User Manual is not revised as a result of the update process described later in this section.

Instead of associating a direct-to-ACU cable by vehicle make or model, the Hyundai and Kia EDR User’s Manual(s) only identify the ACU cable(s) associated with specific Hyundai and Kia ACU Identifier Codes (Fig. 1.11). An Identifier Code is simply a two-letter value printed on the label of Hyundai and Kia ACUs. These Identifier Codes help identify what cable(s) may access the subject ACU.

Figure 1.9: "In order to run this program, VCI needs to be connected. Please check the connection between PC and VCI"

By way of example, the Hyundai and/or Kia ACUs bearing Identifier Code(s) ‘RB’ or ‘FS’ would generally be accessible via cable G1ZDDPA005. The end user wouldn’t be able to identify this until (a) the ACU is identified by direct observation within a given vehicle and (b) then after the VCI is connected and powered up and the EDR software started.

Figure 1.11: Adapter identifier table example from the User Manual

www.collisionpublishing.com

Collision Magazine - Volume 8 Issue 2 59

As of November 7th, 2013, the Hyundai and Kia EDR Tool coverage list as found in the current respective software versions is shown in Table 1.1: Hyundai Accent Azera Elantra Coupe Elantra GT Elantra Equus Genesis Genesis Coupe Santa Fe Santa Fe Sport Sonata Hybrid Sonata Tucson Veloster Kia Cadenza Forte Optima Optima Hybrid Rio Sedona Sorento Soul Sportage

2013 Available Available Available Available Available Available Available Available Available Available Available Available Available Available 2013 Not Available Available Available Available Available Not Available Available Available Available

2014 Not Available Available Available Not Available Not Available Available Available Not Available Available Available Not Available Available Available Available 2014 Available Available Available Not Available Available Available Available Available Available

Figure 1.12: Landing screen with update tab highlighted

Figure 1.13: E-mail subject for update notice

Table 1.1: Vehicle coverage

S

oftware Update and Software Version The Hyundai and Kia EDR Software(s), as well as the VCI, must be manually updated. The end user performs this function via the Update tab (Fig 1.12) within the software once it’s running.

In addition to having the VCI powered and connected, the end user must also have an active internet connection at their PC. All updates are accomplished via an online update protocol.

Figure 1.14: Software update process GIT America, Inc., maker of the Hyundai and Kia EDR Tool(s), as well as the supplier of software revisions, generally sends users an e-mail reminder (Fig 1.13) for software updates which are then reflected in the update process as

60 Collision Magazine - Volume 8 Issue 2

www.collisionpublishing.com

seen in Fig. 1.14. The initial software version release for Hyundai was E-N-H-01-00-0000, and the initial software version for Kia was E-N-K-01-00-0000. As of this writing (November 2013), the most current version of the Hyundai and Kia software(s) was E-N-H-01-00-0004 and E-N-K-01-00-0002, respectively. Addressed later, the version number detail seen here is not reflected in the output report seen by the end user.

The e-mail reminder from GIT America, Inc. provides the update version, as well as a very brief summary (Fig 1.15) of the additional vehicles added to the official vehicle coverage list.

Dear EDR Users, An internet update for your Kia EDR unit is now available.

In order to update to version E-N-K-01-00-0002, your EDR unit requires an internet connection and the VCI to be connected to the PC as well as to an ACU or to a vehicle (Required to power up the VCI).

There is no software version history built into the software update process nor is there a “What’s New” section (Fig 1.16) of the Hyundai or Kia Software. Therefore, users do not know what has been “fixed” or “updated” as a result of the software update and the end product report does not reflect the actual software revision or version number.

By way of comparison, the Bosch Crash Data Retrieval System offers a “What’s New” section whenever a revision is provided. Bosch Crash Data Retrieval Tool end users are also offered a summary of changes and bug fixes, as well as a summary of those vehicle models(s) and year(s) added (or potentially deleted).

At times, the effect of a software update with the Bosch Crash Data Retrieval Tool can be substantive, such that the revision can affect the nature in which data is displayed (such as graphing), as well as the actual content or inclusion or exclusion of one or more data parameters. By way of example, with Version 12.0 of the Bosch Crash Data Retrieval Tool, a correction for Toyota/Lexus/Scion data was made, wherein graphing and precision reporting was affected. The end user was alerted to this change in the information accompanying the version release.

The EDR unit will search for internet updates automatically whenever the EDR program is started or the "Internet Update" button is pressed.

The EDR internet update will contain the following: • VCI Firmware update (Ver. 2.06) • 2014MY Optima(TF/QF) Diagnostic Program.

• 2014MY Rio(UB) Diagnostic Program.

• 2014MY Soul(PS) Diagnostic Program. (Note: An additional ACU adapter (Serial #: G1ZDDPA012) It is not included in the Basic kit and is sold seperately).

• 2014MY Sportage(SL) Diagnostic Program.

For more information or to purchase the ACU adapters, please contact EQS at 1-888-542-1011 (Toll-Free).

For technical support, please contact GIT America at 1-888-5494977 (Toll-Free).

Figure 1.15: Example update e-mail

Figure 1.16: Example of version change information from Bosch CDR Tool Later within this Compendium, the authors address how the absence of a version history or “what’s new” section and lack of actual version numbering both with incremental updates, may adversely affect the results of an investigation involving the Hyundai and/or Kia EDR Tool(s).

As previously noted, the report generated by the Hyundai and Kia EDR Tool(s) do not display or one might suggest that they do not display adequately the specific or actual version number of the software the data was retrieved by and then translation with. The reports, to date, only provide users with a translation indicating the imaging was performed with “G-EDR Software Version 1.00” regardless of which actual version or release/update of the software(s) the actual retrieval and translation was performed with.

www.collisionpublishing.com

Collision Magazine - Volume 8 Issue 2 61

sion

i Coll

M

m

e Sa

in agaz

port

Re ple

Figure 1.18: Landing screen with vehicle shortcut tab highlighted

Figure 1.17: Two example reports showing the same G-EDR Software Version numbers regardless of which actual version or release/update of the software(s) the actual retrieval and translation was performed with.

Finding the end user’s actual version number is not intuitive or easily identified and it is not completely identified in a printed report (Fig. 1.17). The underlying version of the software the end user is applying when accessing an ACU and generating a report, can only be discerned by opening the software and selecting the “Update” tab.

When the user selects this tab, they can access, and download, a new version of the software, as well as “see” what version of the software is currently installed (see Fig 1.14).

A

CU Access and Report Generation While connection to a Hyundai or Kia ACU is done via DLC or direct-to-ACU connection, the process within the software can be initiated in one of three ways, all originating from the Vehicle icon/tab (Fig. 1.18) within the EDR software interface.

The first method in which the user may initiate an imaging session is via the Auto VIN Search (Fig 1.19). The “Auto VIN Search” feature will only work when imaging is performed through the DLC, the vehicle status is set to Accessory or Run, and there is good (12 volt) in-car power. If all the previous requirements are met, the user may click the Auto VIN Search. The software will attempt to read the VIN from the vehicle. Since this feature is not available in the direct-to-ACU connection configuration, it follows that the software call for a VIN being “searched for” follows a generic OBD II call to collect the VIN from the vehicle’s powertrain/engine control module or ECU. The VCI being associated with the normal Hyundai or Kia diagnostic scan tool, it follows that the scan tool would retrieve the VIN from the powertrain ECU (akin to the same function within the Bosch CDR Tool).

Figure 1.19: Auto VIN Search button In a second use case, the end user may also begin an imaging session by manually entering a VIN. From the Vehicle tab, the user can manually type the VIN into the appropriate field seen in the upper left portion of the screen captures in Figure 1.20). Based on the VIN entered, the software will identify the Year/Model vehicle to be imaged.

The third available option for beginning a download is more generic. Users can forgo the use of a VIN, and simply select the model, year and engine (Fig 1.20) for the vehicle they are interested in imaging.

62 Collision Magazine - Volume 8 Issue 2

www.collisionpublishing.com

model and year as selected by the end user (where no VIN is entered and the vehicle is selected based on that VIN).

As one who has used the Bosch CDR Tool or similar application may imagine, the nature in which the user may initiate an imaging session allows for potential access to data in and generation of a report from vehicle(s) outside the intended vehicle coverage list. For the purpose of this article, the authors will refer to this ill-advised and unsanctioned use of the Hyundai and Kia EDR Tool(s) as “Coverage Spoofing,” as the scope of its application goes well beyond the VIN Spoofing sometimes possible (and similarly ill-advised) when using of the Bosch CDR Tool.

Coverage spoofing is addressed in greater detail later in this compendium.

Regardless of how the technician identifies the vehicle being imaged, the software allows the user to enter limited case detail (Fig 1.21). The user can enter their name, case number and crash date. None of the fields are mandatory to continue.

Figure 1.21: User can enter their name, case number and crash date Figure 1.20: Example of vehicle selection using model, year, and engine It becomes clear then that the potential access to data does not require a specific VIN and the Hyundai and Kia EDR Tool(s) does not require any specific reference to the subject vehicle by VIN to initiate the data retrieval process.

Even if the user chooses to enter a VIN manually, there is no requirement for the user to provide the correct VIN to the Hyundai or Kia EDR Tool(s). Moreover, there is no reference to a VIN in the report generated after data is retrieved. The only identification of the vehicle is by way of a line on the first page of the report indicating the make

After clicking “OK” to finish the end user input screen, one may click on the “retrieve” icon (Fig 1.22). The software suite will then initiate an imaging (Fig 1.23) of the subject vehicle/module (assuming there is a connection through the DLC or direct-to-ACU).

www.collisionpublishing.com

Figure 1.22: Retrieve icon

Collision Magazine - Volume 8 Issue 2 63

Data Retrieval Tool *.cdr or *.cdrx file. The lack of a raw, un-editable and re-translatable version of data from Hyundai and Kia ACUs creates, at least, an “evidentiary annoyance” not easily remedied. The authors have provided an example of this deficiency and its impact on the data application later in the Compendium.

Figure 1.23: Initiating the imaging of the subject vehicle/module

Figure 1.24: EDR data being translated to a report/PDF file Perhaps most disappointing to those familiar with the Bosch CDR Tool report particularly in the context of litigation is the format in which the Hyundai and Kia EDR Tool(s) produce the end report as well as the format in which the software saves the accessed data. If successful (Fig 1.24), the software will display a PDF version (Fig 1.25) of the translated data for the user in a report format on screen. Likewise, the user is provided an option to save the report as a PDF file.

Figure 1.25: The translated data as a PDF report There is; however, no option to save any “raw” data for later reanalysis within the software package. The software does not provide any sort of file similar to the Bosch Crash 64

Collision Magazine - Volume 8 Issue 2

The Hyundai and Kia EDR Tool is, as stated within the Tool documentation, offered such that these manufacturers would be in compliance with 49CFR563. For example in the “Data Limitations” portion of a Hyundai or Kia report where one may find this entry: “The specifications for EDR are designed to be compatible with NHTSA 49 CFR Part 563 rule” (sic) as seen in the sample report found in this Compendium.) That regulation does not require a secure file which can later be re-translated with updated software nor are concerns about data integrity which might be addressed as a function of litigation anticipated in that regulation.

R

eport Content Hyundai and Kia vehicles, officially covered by the EDR Tools, may record up to two events. Generally, the Data Limitations from these reports indicate that the events are “Deployments” and “Non-deployments” where the Deployments are locked and can’t be overwritten but Non-deployments unlocked and overwritten by later occurring events. This suggests that once two Deployments have been recorded or stored, the module would need to be replaced inasmuch as it can’t record additional data.

A recorded event is typically reported in four parts. By way of example, the Data Limitations portion of a given report may read “For the Event#1: Event#1-1 Event#1-2 Event#1-3 Real-Time Data.” Lacking other detail or explanation or comparable heading within the report body, this would seem to translate to: “Event 1, section 1,” “Event 1, section 2,” “Event 1, section 3,” and a section labeled “Real-Time Data.” Where there is a second event, the headers repeat (i.e.: “Event 2, section 1,” and so on). For a single recorded event, the report length varies between 10 and 30 pages including the four sections as translated as well as the hexadecimal data which is also displayed in the same section division layout.

“Event n, section 1,” includes tables and graphs of delta-V data, generally shown in km/h and milliseconds. “Event n, section 2,” includes event data, to include various precrash parameters, as well as deployment decision timing data (where appropriate). “Event n, section 3,” includes acceleration data (displayed in g’s) in both tabular and graphical form, apparently rounded for the purpose of re-

www.collisionpublishing.com

porting to the nearest ½ g. And finally, the fourth part is identified as “Real-Time Data” which, at this writing, is limited to the data parameter "ignition cycle, download.

Pre-crash dynamic data, such as “speed” (in km/h) or “throttle application” (in %) are illustrated graphically, similar to the reports many may be familiar with as seen in the Bosch CDR Tool reports for the earliest GM vehicles which recorded pre-crash data. More specific examples of report parameters are provided later in this Compendium.

The Hyundai-Kia EDR Tool(s) report includes a display of the data recovered in hexadecimal format and, loosely comparable to the Bosch CDR Tool “Engineering Translation report” for certain GM vehicles, the data is separated into blocks corresponding to the sections (i.e.: “Event n, section 1,”).

A sample report, offered as part of this Compendium is provided only as an example such that those not familiar with the report layout or design might better see what the output side of the Hyundai-Kia EDR Tool looks like and would provide a reference for later portions of this Compendium.

R

eferences 1. "Read Me First." Introduction. Kia EDR User's Manual. Tustin, California: Global Information Technology, 2012. 4-5. Print.

2. "Event Data Recorders" 49 CFR 563. 2012.

3. Haight, Gyorke, Haight. “Hyundai and Kia Crash Data, a Preliminary Overview.” Collision: The International Compendium for Crash Research. Volume 8, Issue 1

Sample Report: The pages of this Hyundai-Kia EDR Tool report are offered as an example for use in better understanding parts of the Hyundai-Kia compendium found this issue of Collision. It may contain information which has been created for research or illustrative purposes and may not readily reflect data which would necessarily be found in an "actual" crash in the same way or as may be produced using a later available version of the Hyundai-Kia EDR Tool.

-------------------------------------------------------------------------------------------------------------------------------------

Vehicle Information KM | SOUL(AM) | 2013 | ALL | AIRBAG SYSTEM Additional Information User : Collision Publishing CaseNumber : Volume 8 Issue 2 CrashDate : 4/14/13 Part No. : 95910 - 2L074 Save on : Sunday, October 28, 2013, at 10:13:39 G-EDR Software Version : 1.00

EDR Data Limitations

port

am ine S z a g Ma 1 sion i l Page l o

Re ple

C

Tools for downloading the EDRs in Hyundai vehicles have been developed only for vehicles produced after September 1, 2012. Currently, there is no tool for downloading accurate data from the EDRs in Hyundai vehicles produced prior to this.

The retrieval of his EDR data has authorized by the vehicle's owner, or other legal authority.

The airbag ECU can store up to two events. Deployment events cannot be overwritten or cleared from airbag ECU. Non-deployment events(which did not qualify as deployable events) can be overwritten by subsequent events.

The specifications for EDR are designed to be compatible with NHTSA 49 CFR Part 563 rule. The EDR data recording specifications of airbag ECU are divided into the following four categories.

- For the Event#1: Event#1-1 Event#1-2 Event#1-3 Real-Time Data.

- For the Event#2: Event#2-1 Event#2-2 Event#2-3 Real-Time Data.

The airbag ECU records data for all or some of the following crash(event) But, depending on the installed airbag ECU, data for side crash and/or rollover crash(event) may not be recored.

Ignition cycle counter(download) will increment by 1 every time when the power mode cycles is changed from OFF/Accessory to IGN/RUN or EDR data is downloaded by using the retrieval tool.

-------------------------------------------------------------------------------------------------------------------------------------

www.collisionpublishing.com

Collision Magazine - Volume 8 Issue 2 65

t epor

eR mpl

a ine S z a g Ma 2 sion i l Page l o

t epor

eR mpl

a ine S z a g Ma 4 sion i l Page l o

C

C

port

Re ple

am ine S z a g Ma 3 sion i l Page l o

port

Re ple

am ine S z a g Ma 5 sion i l Page l o

C

C

Sample Report: The pages of this Hyundai-Kia EDR Tool report are offered as an example for use in better understanding parts of the Hyundai-Kia compendium found this issue of Collision. It may contain information which has been created for research or illustrative purposes and may not readily reflect data which would necessarily be found in an "actual" crash in the same way or as may be produced using a later available version of the Hyundai-Kia EDR Tool.

66 Collision Magazine - Volume 8 Issue 2

www.collisionpublishing.com

t epor

eR mpl

a ine S z a g Ma 6 sion i l Page l o

t epor

eR mpl

a ine S z a g Ma 8 sion i l Page l o

C

C

port e Re

e Sa azin g a M 9 sion i l Page l Co

l

p Sam e n i az Mag 7 n o i is age l P l o C

t epor

eR mpl

www.collisionpublishing.com

Collision Magazine - Volume 8 Issue 2 67

le amp

eS azin g a M 0 ge 1 ision a l P l o C

le amp

eS azin g a M 1 ge 1 ision a l P l o C

rt

Repo

le amp

eS azin g a M 2 ge 1 ision a l P l o C

rt

Repo

le amp

eS azin g a M 3 ge 1 ision a l P l o C

rt

Repo

rt

Repo

Sample Report: The pages of this Hyundai-Kia EDR Tool report are offered as an example for use in better understanding parts of the Hyundai-Kia compendium found this issue of Collision. It may contain information which has been created for research or illustrative purposes and may not readily reflect data which would necessarily be found in an "actual" crash in the same way or as may be produced using a later available version of the Hyundai-Kia EDR Tool.

68 Collision Magazine - Volume 8 Issue 2

www.collisionpublishing.com

t epor

eR mpl

a ine S z a g Ma 14 sion i l Page l Co

t epor

eR mpl

a ine S z a g Ma 16 sion i l Page l Co

port

am ine S z a g a ion M Page 15 s i l Col

Re ple

t epor

eR mpl

a ine S z a g Ma 17 sion i l Page l Co

www.collisionpublishing.com

Collision Magazine - Volume 8 Issue 2 69

ACCIDENT RECONSTRUCTION AND FORENSIC MAPPING SOLUTIONS DESKTOP MAPPING & ANIMATION SOFTWARE | EVIDENCE COLLECTION SOFTWARE & HARDWARE | TRAINING

WHEN YOU’RE LOOKING FOR THE BEST ACCIDENT RECONSTRUCTON SOFTWARE, YOU NEED TO LOOK AT

NOW WITH ADD-ON MODULES MapScenes Forensic CAD with your choice of point cloud and animation options to ensure you get precisely what you need.

FORENSIC CAD Law enforcement and accident investigators use MapScenes Forensic CAD. MapScenes Forensic CAD is the most precise desktop forensic mapping software available in the world. This powerful and feature-rich application provides law enforcement professionals and accident reconstruction specialists, like you, with confidence that the evidence they present is precise, compelling, and irrefutable.

Turn the page for more information about our powerful animation and point cloud modules.

ANIMATION FOR THE REAL WORLD

ANIMATION MODULE Full-featured forensic animation software for creating precise and compelling animations.

The animation module gives you incredible flexibility to accurately animate your scenes - look at your data in ways you never thought possible.

Our rendering capabilities give you the tools you need to analyze, edit, and build engaging animations.

Timeline and advanced object movement, animated cameras, real surfacing, vaulting and other physics tools - you will be amazed at the possibilites!

POINT CLOUD MODULE An incredible point cloud visualization tool to assist with mapping and animation.

Imagine being able to walk through a point cloud, in real-time, in order to complete your line work or conceptualize your dramatic re-creation.

With the point cloud module you can draw lines right in the point cloud viewer, which will show up in your CAD drawing. You can also include the point cloud itself in your final animation for added realism at NO extra effort. Mapping through to animating, your efficiency will soar!

LET THE SCENE SPEAK FOR ITSELF

is registered in the U.S. Patent and Trademark Office. All other brand ® MapScenes names, product names or trademarks are property of their respective owners.

www.mapscenes.com

1.800.668.3312

THE REST OF THE

SYSTEM

EVIDENCE COLLECTION SOFTWARE:

EVIDENCE COLLECTION HARDWARE:

CERTIFIED TRAINING:

Reduce errors, save time, and eliminate frustration.

Incredible hardware options and support.

When accuracy and precision are non-negotiable – there is MapScenes.

Evidence Recorder is the world’s best selling evidence collection software built specifically for the professional reconstructionist.

Using a Windows Mobile or tablet device, Evidence Recorder allows you to experience the power, precision and productivity gains associated with being able to visually verify your ongoing field evidence collection – on-the-spot, right in the palm of your hand. Then, for maximum results, it can seamlessly integrate with the MapScenes desktop software.

Evidence Recorder is supported with more total stations, data collectors, and GNSS receivers than ever.

Training is essential in all facets of law enforcement. MapScenes acknowledges its responsibility to the industry by ensuring that all trainers and educational courses are kept up to date through a base educational standard, preserving quality and integrity.

Combine with your existing hardware, or choose from one of our hardware bundles. Evidence Recorder is included on all hardware configurations and offers the absolute best in value.

We have set an industry benchmark by developing a comprehensive framework that provides investigators with the knowledge and skills required to use the MapScenes system easily, efficiently, and expertly.

Choose from numerous options, including: Nautiz X7, Algiz 7, Archer field PC, or Allegro MX.

THE

SYSTEM:

Evidence Collection Software Evidence Collection Hardware Desktop Mapping Software

is registered in the U.S. Patent and Trademark Office. All other brand ® MapScenes names, product names or trademarks are property of their respective owners.

www.mapscenes.com

Training

1.800.668.3312

#1 Expert Witness Directory for Accident Reconstruction

Membership has its benefits Knowledge Members have unlimited, exclusive access to databases including: Vehicle Specs DB Motorcycle Specs DB Crush Stiffness DB Crash Video Clips PDF Article Library Online Equation Solver

Exposure

Membership starts at only $79/year All Members also receive discounts to

The ARC Network is the most visited web resource for the accident reconstruction and traffic investigation industry.

Listing in the #1 Online Expert Directory #1 Result in Google Professional Profile Web Page

Plus much more

www.accidentreconstruction.com n 800.280.7940

port

Re ple

am ine S z a g a ion M Page 18 s i l Col

t epor

eR mpl

e Sa azin g a M e 21 sion g i a l P l Co

t epor

eR mpl

e Sa azin g a M e 19 sion g i a l P l Co

in agaz M n 20 isio Page Coll

e Sa azin g a M e 22 sion g i a l P l Co

port

Re ple

t epor

am ne S i z a ag ion M Page 23 s i l Col

le R

mp e Sa

t epor

eR mpl

port

Re ple

am ne S i z a ag ion M Page 24 s i l Col

76 Collision Magazine - Volume 8 Issue 2

www.collisionpublishing.com

le amp

eS azin g a M 5 ge 2 ision a l P l o C

le amp

eS azin g a M 6 ge 2 ision a l P l o C

rt

am ine S z a g a ion M Page 29 s i l Col

Repo

rt

Repo

port

Re ple

am ine S z a g Ma 27 sion i l Page l Co

port

Re ple

Section 2: Crash Testing involving Hyundai and Kia Vehicles

C

rash Testing Overview Three different types of crash tests have been conducted by the Insurance Institute for Highway Safety (IIHS) using 2012-2014 Hyundai/Kia vehicles supported by the Hyundai-Kia EDR Tool. The test protocols, reports and raw crash data can be used to shed some light on how well the EDR tool is reporting specific data parameters. All of these tests were performed at the IIHS Vehicle Research Center in Ruckersville, Virginia. They were conducted in a 22,000 square foot crash hall with a propulsion system that can accelerate a full size pickup truck up to 50 mph.1 IIHS implements state of the art measurement equipment including high fidelity accelerometers and fully instrumented crash dummies.

One of the many crash tests that the IIHS conducts, one is the Small Overlap Frontal Crashworthiness Evaluation.

This test is conducted at a speed of 40 mph (64.4 km/h) and the test vehicle is aligned so that a rigid barrier is 25% offset from the vehicle centerline.2 Figure 2.1 shows the vehicle orientation at the moment of impact.

t epor

eR mpl

a ine S z a g Ma 28 sion i l Page l Co

Figure 2.1: Diagram showing the small overlap crash test configuration Prior to the test, the gasoline is removed from the vehicle and the fuel lines are replaced with a Stoddard solvent.

The engine is then started to ensure that the Stoddard solvent has filled the fuel lines completely. To measure www.collisionpublishing.com

Collision Magazine - Volume 8 Issue 2 77

vehicle accelerations, a steel plate is welded to the floor of the rear seating area along the centerline of the vehicle and then three Endevco accelerometers are mounted and oriented in the longitudinal, lateral and vertical directions.

To measure the moment of impact, two pressure-sensitive tape switches are applied to the front of the vehicle. The in-car data acquisition system relies on closure of the tape switch to set a time-zero for the crash pulse and other data elements.

Just before the test begins, a 50th percentile male Hybrid II dummy is placed in the driver seat and the driver seat belt is buckled. The vehicle is turned to the on position to power the vehicle electronics but the engine is not started (turning over). To begin the test, the vehicle is accelerated straight forward by the IIHS propulsion system at an average of 0.3 g until just before it reaches the vertical steel plate barrier such that it approached the barrier at a steady state speed at and just before impact.

The second type of crash test in this data set is the Moderate Overlap Frontal Crashworthiness Evaluation. While this crash test is very similar to the Small Overlap Crash Test it differs in two major aspects. While the impact speed and angle are the same, the percent overlap is increased from 25% to 40% 3 and instead of striking a rigid barrier, the moderate overlap test utilizes a rectangular deformable barrier. All other test preparations are similar to that of the small overlap crash test. Figure 2.2 is a diagram that shows the setup for a moderate overlap crash test at IIHS.

Figure 2.2: Diagram showing the moderate overlap crash test configuration The last type of crash test analyzed in this study is the Side Impact Crashworthiness Evaluation. This test is arranged so that a stationary test vehicle is struck on the driver side by a deformable barrier attached to a movable “crash cart” at a 90 degree angle.4 The speed of the barrier at impact is 31.1 mph (50 km/h) and the “crash cart” weighs about 3,300 lbs (1,500 kg). Figure 2.3 is a diagram of the set-up for an IIHS side impact test.

78 Collision Magazine - Volume 8 Issue 2

Figure 2.3: Side impactcrash test configuration Like the other IIHS tests, the fluids are removed from the vehicle and replaced with Stoddard solvent and the fuel lines are cleared of gasoline. Unlike the previously described tests, only the lateral accelerations are measured near the A and B pillars on the non-struck side by the IIHS as a function of the focus of their research. Unlike the other IIHS crash tests, a 5th percentile SID-II dummy is belted and positioned in the driver seat while a second identical dummy is belted and positioned in the left rear seat. As is the case for all of the IIHS crash tests, video footage and still photography are captured and archived.

T

est Matrix Table 2.1 contains the year, make, model, test type and IIHS test number for each ACM that was downloaded using the Hyundai-Kia EDR Tool. For each test, the ACM was removed from the vehicle by IIHS test technicians and downloaded using direct-to-module cables. In each test, except side impact test CES1306, both longitudinal and lateral accelerations were recorded and analyzed.

A

nalysis of Crash Data As previously mentioned, all of the crash tests included in this paper were performed at IIHS according to IIHS test protocol. Moreover, all of the frontal crash tests were offset and caused the test vehicle to undergo significant rotation during the impact. Because the accelerometer was not colocated with the ACM, the accelerations and the calculated delta-V will have some significant differences. This difference is discussed in a separate paper in this issue of Collision.5 In short, the rotation of the vehicle in an offset crash coupled with a significant distance between measurement sensor and the ACM data reveals two different readings for acceleration and delta-V.

www.collisionpublishing.com

Year 2012 2012 2013 2013 2013 2013 2013 2013 2014

Make Hyundai Kia Hyundai Hyundai Hyundai Kia Kia Kia Kia

Model Accent Rio Accent Elantra Tucson Rio Rio Soul Forte

Test Type Moderate Overlap Frontal Moderate Overlap Frontal Small Overlap Frontal Small Overlap Frontal Small Overlap Frontal Small Overlap Frontal Side Impact Small Overlap Frontal Small Overlap Frontal

IIHS Test Number CEF1107 CEF1201 CEN1329 CEN1310 CEN1237 CEN1330 CES1306 CEN1317 CEN1318

Table 2.1: Selected IIHS crash tests Since this effect occurs in all of the frontal offset crash tests, only the side impact acceleration and delta-V data will be compared directly. For brevity, all other crash tests will be highlighted but only unusual or contextually interesting data parameters will be discussed in more detail.

2012 Hyundai Accent (CEF1107) Although the 2012 Hyundai Accent is not a supported vehicle in the Hyundai EDR tool, the authors decided it would be important to test the device on an older (unsupported) models to document and underscore the dangers of “coverage spoofing.” Because the Hyundai EDR software has no 2012 Hyundai Accent to choose from, the 2013 Hyundai Accent vehicle was selected for this download. This test follows the moderate overlap test procedure described above.

It is very clear in this test that coverage spoofing will provide “some data” and it is equally clear that at least some of the parameters do not make physical sense in the context of the test conducted. For example, Figure 2.4 shows the reported longitudinal acceleration for the moderate overlap crash test and Figure 2.5 shows the acceleration measured

by IIHS for the same test. Obviously, is a large discrepancy between the acceleration reported by the Hyundai EDR Tool and the IIHS accelerometers. It is important to note the -70 g and the +70 g values in the EDR report appear to be the confines of the range for the system. One could propose that there seems to be a data translation error in this case and the software may be reading the accelerations incorrectly. This issue appears to be associated with coverage spoofing the 2012 Hyundai Accent because the 2013 Hyundai Accent, discussed later, does not have similar acceleration values and reinforces that coverage spoofing should be avoided.

Figure 2.4: Reported longitudinal acceleration

Figure 2.5: Measured longitudinal acceleration Another significant observation would be that the delta-V for this test does not seem to be integrated directly from the acceleration provided in the report. If one were to integrate the data in Figure 2.4, the maximum delta-V would be closer to -355 km/h. The actual EDR Tool reported maximum longitudinal delta-V is -73 km/h. This latter value is a reasonable delta-V when one considers that the impact speed for this test is 64.4 km/h.

www.collisionpublishing.com

Collision Magazine - Volume 8 Issue 2 79

While one cannot compare the delta-V calculated by the IIHS accelerometer directly because of the rotation and difference in sensor location, the reported IIHS delta-V is -68 km/h. Without measuring the acceleration at same location as the ACM, it is difficult to determine exactly if the reported delta-V data is accurate or not. This same argument can be made for the lateral delta-V which is reported as 6 km/h.

In this report, the vehicle roll angle is listed as a data table and graph, however each time entry is listed as “not supported” and the graph is blank. Vehicle speed, which is 63 km/h at time zero, is consistent with the test protocol. The engine throttle is listed as 8% even though the engine is not turning over during the crash test. The engine throttle percent data parameter is discussed in more in section 4 of this paper. As expected, the service brake is off during the entire test, the diver safety belt status is reported as “on” and the passenger safety belt status is reported as “off.” Also as expected, the engine RPM is reported as zero since the engine is off during the test. ABS activity is reported as OFF for all time points and the stability control is reported as “on” (as opposed to “active”). Item #15 of the Hyundai EDR report is the steering input, however, each sample is listed as “Not supported (or invalid data).” Also reported as “not supported” is lateral acceleration.

While some of the reported data vales do seem to be consistent with the test data, (i.e.: “vehicle speed,” RPM, seat belt use, etc.) there are some values, such as longitudinal acceleration, that are not reported correctly. These incorrectly reported values should again serve to discourage the reconstructionist from coverage spoofing non-supported vehicles with the Hyundai/Kia EDR Tool.

2012 Kia Rio (CEF1201) The 2012 Kia Rio moderate overlap crash test was also analyzed for any inconsistencies in the reported data parameters. Like the 2012 Hyundai Accent, the 2012 Kia Rio is not supported by the Kia EDR Tool and therefore the authors chose to engage in coverage spoofing with the EDR Tool for the purposes of this analysis as previously described. To accomplish this, the 2013 Kia Rio was selected as the test vehicle using the on-screen list.

The maximum longitudinal delta-V for the 2012 Kia Rio is -77 km/h, which is similar to that of the 2012 Hyundai Accent. The lateral delta-V was reported as 7 km/h. The vehicle roll angle was not supported for all time points and the roll angle graph was left blank. The vehicle speed for this test, which is 65 km/h at time zero, is consistent with the test protocol. Like the 2012 Hyundai Accent moderate overlap test, the throttle is listed as 8% and the service brake is off for all time points. Unlike the 2012 Hyundai Accent, all of the steering input values are reported as zero, which is consistent with a vehicle being pulled into a barrier on a straight track.

The longitudinal acceleration and delta-V values for this EDR report reasonably compare with the IIHS crash test data; however, again, it is not appropriate to compare the data correctly because the sensors were not mounted at the same location as the ACM. Like the other tests, the engine throttle is reported throughout the five seconds prior to impact, however the reported value is 14% to 15% compared to the 8% observed on the previous test example.

The engine speed, ABS activity, stability control and the driver safety belt status, passenger safety belt status are all recorded correctly for this test. Vehicle roll angle is listed as “not supported” for all data points and the corresponding graph is empty.

2013 Hyundai Accent (CEN1329) The 2013 Hyundai Accent was the first vehicle discussed in this data set that is currently supported by the Hyundai EDR Tool and the first small overlap crash test in the data set. The reported vehicle speed at the time of impact for this test is 63 km/h which is consistent with the actual speed of the vehicle at impact as reported.

Figure 2.6: Photo of the 2012 Kia Rio moderate overlap crash test (courtesy of the IIHS) 80 Collision Magazine - Volume 8 Issue 2

Like the previously described tests, the vehicle accelerations and delta-V values cannot be directly compared to the IIHS crash test data, they do; however, seem reasonable given the impact speed of 64.4 km/h and the dynamics of a small overlap crash. The maximum longitudinal delta-V for this test was -47 km/h and the maximum lateral deltaV was 13 km/h. The reported longitudinal and lateral ac-

www.collisionpublishing.com

celerations do seem to report the dynamics of ACM and there does not seem to be any inappropriate data points in either tables.

The vehicle roll angle is not supported, the vehicle speed is 63 km/h, the engine throttle is 14% to 15%, the service brake is off, the engine RPM is zero, the ABS activity is reported as off, and stability control is on. All of these available parameters are consistent with the test data. The driver safety belt is reported as “on” and the passenger safety belt is reported as “off,” which is consistent with the test protocol.

software. The table in this figure is truncated for space; all of the values 0.3 seconds are reported as “invalid data.” The Data Limitations found with the Hyundai-Kia EDR Tool report offers no explanation for the “invalid data” indicator. Lacking insight into this indication, one is left to speculate that this may be a function of clipping, other internal or pre-set system limits or an internal failure of some sort.

Steering input, however, introduces some notable differences between two different 2013 Hyundai Accents. The vehicle in this crash test reports that all of the steering input values are “not supported (or invalid data)” while another download of the same vehicle model (2013 Hyundai Accent) from a real world collision does in report values for this parameter. This real world case is more thoroughly investigated and noted in section 3 of this Compendium in this issue of Collision.

sion

i Coll

2013 Hyundai Elantra (CEN1310) The 2013 Hyundai Elantra is the second small overlap crash test and the second Hyundai EDR tool supported vehicle in this paper. The maximum longitudinal delta-V is -64 km/h and the maximum lateral delta-V is 10 km/h which are reasonably consistent with the 64.4 km/h impact speed and crash dynamics of the small overlap test. The vehicle roll angle is “not supported” and the vehicle speed is 63 km/h at about time zero. The engine throttle for this test was 7% for all of the time points and the service brake was correctly listed as off. The safety belt status for the driver, the passenger belt status, the engine RPM, the ABS activity, and the stability control were all reported correctly as compared to the test design and as conducted. The steering input angle is zero degrees, which is also reported correctly considering that the vehicle was being pulled straight on a track. Both the longitudinal and lateral accelerations are reasonably consistent with a small overlap crash test pulses.

2013 Hyundai Tucson (CEN1237) The 2013 Hyundai Tucson was the third small overlap crash test and the first SUV to be analyzed in this review.

The most striking difference in this test report is the existence of roll angle data. Given all of the reports in this crash test matrix, this vehicle is the only one with a populated vehicle roll angle data table. Not only is the existence of this table singularly interesting, but the way it is reported is also unique. Figure 2.7 shows the table and graph for the vehicle roll angle as reported in the Hyundai EDR tool

M

m

e Sa

in agaz

port

Re ple

Figure 2.7: Graph and table of the reported vehicle roll angle for the 2013 Hyundai Tucson crash test As expected, the vehicle roll angle is 0 degrees from -1.0 through time zero. After impact, the vehicle roll angle is reported as -10 degrees. This value stays constant for 300 ms. After 300 ms, all of the data points are reported as “invalid data.” It is not certain whether or not this data is actually “invalid” or if the module stops recording after the frontal event is complete as a function of the requirements found in Table III of 49CFR563. It is possible that for a longer event, like a rollover, the data collection may continue past 300 ms but inasmuch as Table III calls for 300ms of data (except for a rollover) it would seem the most likely explanation is compliance with that recording period rather than a suggestion that the data represents the event completely or that it would be for the same period in a rollover event.

www.collisionpublishing.com

Collision Magazine - Volume 8 Issue 2 81

The maximum reported longitudinal delta-V is reported as -60 km/h and the maximum reported lateral delta-V is 7 km/h. These values are reasonably consistent with the small overlap crash test dynamics. Like other tests, the vehicle speed at impact is listed as 63 km/h in the EDR report which is very close to the actual impact speed. Engine throttle is reported at 4% and the service brake is off for all time points. The driver and passenger safety belt statuses are consistent with the test setup. The engine RPM, the ABS activity and the stability control are all consistent with the small overlap crash test. The steering input is reported as zero degrees for all time intervals which is accurate for this test. Lastly, both the longitudinal and lateral acceleration data seem reasonably appropriate for the small overlap crash configuration.

with the fact that this vehicle is being pulled straight down a track. The steering angle should reasonably be reported as 0 degrees as observed in all the previously mentioned tests.

The accuracy of steering input becomes more significant when one contemplates the requirements of 49CFR563 discussed in a later section of this Compendium.

2013 Kia Rio (CEN1330)

2013 Kia Rio (CES1306)

The 2013 Kia Rio is the fourth small overlap crash test in this data set. Perhaps the most interesting parameter observed in this report is the steering input data. Figure 2.8 shows the steering input graph and table from the 14th page of this EDR report.

The 2013 Kia Rio is the only side impact test of this data set. For this test, the lateral acceleration and delta-V can be directly compared because the barrier impacts the vehicle in the center, perpendicular to the longitudinal axis. This does not cause the same kind of rotation that is observed in the small and moderate overlap crash tests. Since only the lateral acceleration is measured during the test, it will be compared directly to the data provided in the Kia EDR Tool report. Figure 2.9 shows the lateral acceleration and delta-V from the 2013 Kia Rio report.

For this test, the maximum reported longitudinal and lateral delta-V values, as well as the corresponding accelerations, are consistent with the small overlap crash test.

The vehicle roll angle for crash test was reported as “not supported” for all time values. The vehicle speed, engine throttle, brake status, engine RPM, ABS activity, driver and passenger safety belt status and stability control are all consistent with the small overlap crash test.

Figure 2.10 shows the comparison between the Kia EDR Tool reported lateral acceleration and the IIHS instrumentation acceleration. Figure 2.11 shows the comparison between the Kia EDR Tool reported delta-V and the calculated delta-V provided by the IIHS accelerometers.

It is clear that the lateral delta-V reported by the EDR Tool in this side impact test reasonably compares to the deltaV calculated using the IIHS accelerometer. The maximum delta-V calculated from the IIHS accelerometer is about 26.7 km/h (16.6 mph) while the maximum reported deltaV reported by the EDR tool is 25 km/h (15.5 mph). The percent error between the two maximum delta-V values is about -6.4% if one assumes that the IIHS calculated deltaV is the true value. Figure 2.12 shows a ±10% error corridor for the IIHS calculated delta-V value.

Figure 2.8: Reported steering input for the 2013 Kia Rio crash test The steering input values for this Kia EDR Tool report range from +10 to +15 degrees throughout the 5 seconds of pre-crash data. The values in this table are not consistent 82 Collision Magazine - Volume 8 Issue 2

Unlike the delta-V comparison, the acceleration data provided by the Kia EDR Tool does not compare favorably with the IIHS acceleration data, at least at first glance.

What is not clear to the user of the Kia EDR Tool data is the method for filtering or averaging that takes place when (or perhaps before) the acceleration values are recorded or reported. The data processing which clearly must be taking place could have a significant effect on the reported

www.collisionpublishing.com

gaz n Ma

isio

Coll

ine S

le amp

rt

Repo

Figure 2.9: Reported lateral accerlation and delta-V from 2013 Kia Rio side impact test

Figure 2.10: Comparison between the reported and measured lateral acceleration

Figure 2.11: Comparison between the reported and measured lateral delta-V www.collisionpublishing.com

Collision Magazine - Volume 8 Issue 2 83

Figure 2.12: Instrumentation lateral delta-V with a ±10% band acceleration and since it is not clear which method(s) is used, it is very difficult to compare directly with other raw (control) data.

For example, based on the acceleration recorded by IIHS, the maximum acceleration observed is about 22g (when filtered at SAE60). In comparison, the EDR Tool reported a maximum acceleration of 10.5g. This discrepancy is most likely due to the unknown (otherwise undescribed in any supporting literature) processing that is applied within the ACM or by the EDR Tool software. If one were to apply the same processing methods to the IIIHS accelerations, then the accelerations would most likely be better related.

It is important to note that if one mathematically integrates the acceleration reported by the Kia EDR Tool, it will not equal the delta-V from the same report. Figure 2.13 shows the delta-V that was integrated from the EDR reported acceleration. The integrated acceleration data does have a similar maximum delta-V, however, many of the time points to not match the reported delta-V or the IIHS delta-V. On inspection, it seems to be that the integrated EDR acceleration is time shifted by about 18ms.

If the time shift is analyzed, the integrated acceleration and the delta-V reported by the EDR tool match relatively well. The selection of the 18ms shift has been arbitrarily applied here for illustration alone and may not be applicable to other data sets. In other words, that selected shift reasonably aligns the data sets in this example but may not result in a comparable alignment in another data set.

84

Collision Magazine - Volume 8 Issue 2

Not shown here graphically is the longitudinal acceleration and delta-V. Since this is a perpendicular side impact crash test, there is minimal longitudinal acceleration, and therefore minimal longitudinal delta-V. The acceleration found in the EDR report shows that all of the samples are effectively zero, ranging from -0.5g to 1.0g. Similarly, the longitudinal delta-V has a range from 0 to -1 km/h. This is expected for this type of crash test and seems to be correctly measured and reported.

In this report, the vehicle roll angle is “not supported” and the corresponding graph is empty. The vehicle speed, which is zero for all reported values, is correct because the vehicle is stationary prior to impact. The engine throttle is at 8% for all values prior to impact and the service brake is reported as off for all values. The safety belt status for the driver and passenger are recorded correctly as “on” and “off” respectively. Engine RPM is reported as zero for all values and the ABS activity are all listed as "off." The steering input is correctly reported as zero because the vehicle is lined up with the wheels “straight ahead” prior to the test.

2013 Kia Soul (CEN1317) The 2013 Kia Soul is the fifth small overlap test in this data set. Like the other small overlap tests, the delta-V and the accelerations cannot be directly compared due to the fact that there was significant rotation and the reference accelerometer was not colocated with the ACM. Generally, however, the delta-V and accelerations provided in the EDR report do seem to be consistent with a small overlap

www.collisionpublishing.com

Figure 2.13: Instrumentation lateral delta-V with the integrated acceleration from the EDR report crash at 64.4 km/h. The maximum longitudinal delta-V was reported as -45 km/h and the maximum lateral deltaV was reported as 16 km/h.

In this test, the vehicle roll angle was listed as “not supported” for all values and the corresponding graph was left empty. The engine throttle was listed at 8% and the service brake was reported as off for all values. As expected, the steering input for this test was reported as zero degrees for all of the 11 data samples prior to impact. Finally, the safety belt status was reported correctly for the driver and passenger as “on” and “off” respectively.

The vehicle speed the perhaps the most notable parameter found in this report. The actual impact speed for this test, as reported by IIHS instrumentation, is 64.3 km/h. The vehicle speed, as reported by the Kia EDR Tool ranges from 55 km/h at -5.0 seconds to 61 km/h at time zero.

This discrepancy in vehicle speed represents an error of about 5.1%. This relatively small percent error is; however, the largest observed in all of the tests in the data set.

zero. This is slightly less than the measured impact speed of 64.4 km/h.

The engine throttle percentage for all pre-crash time samples was 1% and the service brake was reported as "off" for each sample prior to impact. The safety belt status for the driver and the passenger was reported as "on" and "off" respectively. The engine RPM was reported as zero for all pre-crash time samples and the ABS activity was reported as "off" for each sample. The steering input seems to be measured and recorded correctly as zero degrees for all precrash samples. Lastly, the acceleration (longitudinal and lateral) reports seem to be consistent with a small overlap crash test.

2014 Kia Forte (CEN1318) The 2014 Kia Forte was the last small overlap crash test that was included for analysis in this study. The maximum reported delta-V was reported as -56 km/h and 9 km/h in the longitudinal and lateral directions which is generally consistent with this type of crash test. Vehicle roll angle was reported as "not supported." The vehicle speed ranged from 59 km/h to 62 km/h for the 5 seconds prior to time

Figure 2.14: Photo of the 2014 Kia Forte from the small overlap crash test (courtesy of the IIHS)

www.collisionpublishing.com

Collision Magazine - Volume 8 Issue 2 85

S

ummary In large part, the Hyundai-Kia EDR Tool reported data sets, compared to that collected by the IIHS in this series of crash tests, appears to be reasonably accurate; however, there are notable exceptions. These include:

• • • •

Potential anomalous data, in particular acceleration values, as a product of “coverage spoofing” (see 2012 Hyundai Accent test) Roll angle “missing/invalid data” (see 2013 Hyundai Tucson test) Steering inputs (see 2013 Kia Rio small overlap test) Acceleration data in general due to unknown raw data processing (see 2013 Kia Rio side impact test)

Arguably, some of these apparent discrepancies or differences might be addressed in a more thoroughly developed and detailed “Data Limitations” which would (should) accompany these types of reports. Lacking detailed information on the application, filtering process, anticipated accuracy, range, and resolution it is difficult to say whether or not the data, taken in total, may be used reliably to “… ensure that EDRs record, in a readily usable manner, data valuable for effective crash investigations and for analysis of safety equipment performance (e.g., advanced restraint systems). (where) These data will help provide a better understanding of the circumstances in which crashes and injuries occur and will lead to safer vehicle designs.”6

A

study. IIHS is an independent, nonprofit scientific and educational organization dedicated to reducing the losses — deaths, injuries and property damage — from crashes on the nation's roads.

R

2.

3.

4.

5.

6.

eferences 1. "Inside IIHS." Vehicle Research Center. Web. 17 Oct. 2013. <http://www.iihs.org/iihs/ about-us/vrc>.

Small Overlap Frontal Crashworthiness Evaluation Crash Test Protocol. Rep. Version II. Arlington, VA: Insurance Institute for Highway Safety, 2012. Print.

Moderate Overlap Frontal Crashworthiness Evaluation Crash Test Protocol. Rep. Version XIV. Arlington, VA: Insurance Institute for Highway Safety, 2012.

Print.

Side Impact Crashworthiness Evaluation Crash Test Protocol. Rep. Version VI. Arlington, VA: Insurance Institute for Highway Safety, 2012. Print.

Haight S., Haight W. R., "Analysis of Event Data Recorder Delta-V Reporting in the IIHS Small Overlap Crash Test." Collision: The International Compendium for Crash Research. Volume 8, Issue 2.

74144 Federal Register, Vol. 77, No. 240, Thursday, December 13, 2012, Proposed Rules; National Highway Traffic Safety Administration, 49 CFR Part 571, [Docket No. NHTSA–2012–0177], RIN 2127– AK86, Federal Motor Vehicle Safety Standards; Event Data Recorders

cknowledgements The authors would like to thank the Insurance Institute for Highway Safety, especially Becky Mueller and Raul Arbelaez, for their contributions to this

Section 3: Hyundai and Kia Crash Data from “Real World” Crashes

I

n addition to comparison testing of data by way of a review of instrumented crash tests, the authors of this Compendium sought to evaluate real-world collisions wherein the data conveyed in the Hyundai and/or Kia EDR report(s) from the Hyundai and Kia EDR Tools(s) might be examined against other forms of evidence and common accident investigation techniques.

Based on the fact that the Hyundai and Kia EDR Tool(s) are only intended for use on vehicles produced after September 1st, 2012 (presumptively model year 2013) or newer 1, consistent with compliance with 49CFR563 2, the authors sought only Model Year 2013 (or newer) vehicles for the comparisons.

86

Collision Magazine - Volume 8 Issue 2

Case #1 - 2013 Kia Soul and a 2013 Cadillac XTS The crash occurred as the driver of the 2013 Kia Soul made a left turn in an attempt to leave a major thoroughfare and enter an adjacent shopping mall. The driver of the 2013 Cadillac XTS was traveling in the opposite direction of the Kia Soul and, as the Kia Soul completed its turn, the front of the 2013 Kia Soul struck the driver’s side of the Cadillac XTS. The crash resulted in the deployment of the side and headcurtain airbags as well as the driver seatbelt pre-tensioner in the 2013 Cadillac XTS. There were no deployments noted in the 2013 Kia Soul. Both vehicles were damaged sufficiently to have been towed from the scene.

www.collisionpublishing.com

NOT TO SCALE, ILLUSTRATION ONLY

Figure 3.1: Case #1 - 2013 Kia Soul (red) and 2013 Cadillac XTS (black) The Event Data Recorders (EDRs) were both imaged; the 2013 Cadillac via the Bosch Crash Data Retrieval (CDR) Tool, and the 2013 Kia Soul via the Kia EDR Tool. Then, a comparison was made between the recorded delta-V from the 2013 Cadillac XTS and the 2013 Kia Soul.

Before the actual comparison could be made, vehicle specifications were sought for both the 2013 Cadillac XTS and 2013 Kia Soul. The published weights for said vehicles were approximately 4215 lbs 3 and 2778 lbs 4 respectively.

In order to account for the additional weights of cargo and occupants, 250 lbs was added to the weight of the 2013 Kia Soul, and 350 lbs was added to the weight of the Cadillac XTS 5. Therefore, the weights of the vehicles, when they collided, for this analysis, were assumed to be about 4565 lbs for the 2013 Cadillac XTS and 3028 lbs for 2013 Kia Soul.

In examining the recorded Longitudinal and Lateral deltaV from the 2013 Cadillac XTS, the authors found that the

www.collisionpublishing.com

Collision Magazine - Volume 8 Issue 2 87

Figure 3.2: Photographs of the 2013 Kia Soul

Figure 3.3: Photographs of the 2013 Cadillac XTS vehicle experienced a Maximum Longitudinal delta-V of about -4 MPH, as well as a Maximum Lateral delta-V of about 5 MPH, both occurring around 150ms after time zero.

Based on the availability of both Longitudinal and Lateral delta-V data for the 2013 Cadillac XTS, the authors calculated the resultant delta-V that the vehicle experienced :

Equation 3.1: Resultant delta-V for the 2013 Cadillac XTS Therefore, the resultant delta-V for the Cadillac, based on its internally recorded delta-V, was about 6.4 mph.

88 Collision Magazine - Volume 8 Issue 2

Based on the fact that extensive research has been performed with regard to the accuracy and reliability of General Motors EDR data 6, 7, 8, 9 and the data “made sense” given the crash case facts, the authors set the 2013 Cadillac XTS as the benchmark by which the subsequent comparison was made with the 2013 Kia Soul data. In short, the 2013 Cadillac XTS resultant delta-V was assumed to be reasonably reliable for the purpose of the subsequent comparison.

In examining the recorded Longitudinal and Lateral deltaV from the 2013 Kia Soul, the authors selected a Longitudinal delta-V of -16 km/h (-9.9 mph) and a Lateral delta-V of -1 km/h (0.6 mph), both occurring at 150ms within their respective delta-V sets. These specific samples were selected by the authors because they represented the same approximate time(s) as observed for the 2013 Cadillac XTS Maximum delta-V data points (both Longitudinal and Lateral).

www.collisionpublishing.com

Figure 3.4: Maximum recorded longitudinal and lateral delta-V from 2013 Cadillac XTS (excerpt, Bosch CDR Tool report)

Figure 3.5: Longitudinal and lateral delta-V reported by the Kia EDR Tool for the 2013 Kia Rio Given the availability of both Longitudinal and Lateral delta-V data for the 2013 Kia Soul, we can calculate the resultant delta-V that the vehicle experienced as:

Adopting the assumption that the force acting on the Kia would be equal opposite to that acting on the Cadillac,10 and given the mass (weight) of the 2013 Kia Soul, as well as that of the 2013 Cadillac XTS, we can utilize the previously calculated resultant delta-V of 2013 Cadillac XTS in order to calculate the anticipated resultant delta-V of the 2013 Kia Soul, absent the use of the 2013 Kia Soul’s EDR report.

Equation 3.2: Resultant delta-V for the 2013 Kia Rio

www.collisionpublishing.com

Collision Magazine - Volume 8 Issue 2 89

Equation 3.3: Calculated Delta-V for the Kia Rio When the resultant delta-V of the 2013 Kia Soul is calculated via the reported Longitudinal delta-V and a Lateral delta-V from its EDR report, the resultant delta-V is about 9.96 MPH. When the resultant delta-V is calculated via comparison with the 2013 Cadillac XTS weight, 2013 Kia Soul weight and 2013 Cadillac XTS resultant delta-V, the calculated value is 9.64 MPH. Clearly then, the delta-V data as reported by the 2013 Kia Soul compares well with delta-V data as retrieved from the 2013 Cadillac XTS (9.96 mph vs. 9.64 mph).

While the extent of the data collected during the aforementioned real-world comparison was encouraging, the authors also experienced reports, generated by the Hyundai and Kia EDR Tool(s) that exhibited debatable data translations in some real-world cases.

Case #2 - 2013 Hyundai Accent The subject real world crash occurred as the driver of the 2013 Hyundai Accent approached a stoplight, with traffic slowing and stopped ahead. In all, there were four (4) vehicles involved in the collision; a 2005 Infinity G35X, a 1999 Infinity Q45, a 2010 Jeep Wrangler and the 2013 Hyundai Accent.

All but the 2013 Hyundai Accent stopped for the aforementioned traffic signal. For an unknown reason, the driver of the 2013 Hyundai Accent failed to reduce her speed before striking, and under-riding, the rear of the 2010 Jeep Wrangler. The 2010 Jeep Wrangler was subsequently pushed into the rear of the 1999 Infinity Q45, which was then pushed into the rear of the 2005 Infinity GX35.

The crash did not involve the deployment of any airbags or seatbelt pre-tensioners in the 2013 Hyundai Accent.

However, due to substantial damage, the 2013 Hyundai Accent was towed from the scene.

The Airbag Control Unit (ACU) from the 2013 Hyundai Accent was accessed, and a report generated, via the Hyundai EDR Tool. Then, an examination of the EDR

90 Collision Magazine - Volume 8 Issue 2

www.collisionpublishing.com

Figure 3.6: Photographs of the 2013 Hyundai Accent

NOT TO SCALE, ILLUSTRATION ONLY

Figure 3.7: Illustration of the 2013 Hyundai Accent crash

report was compared against the known circumstances of the collision.

While an independent validation of the data sets was not possible (due to a lack of instrumented testing), the majority of recorded parameters displayed in the 2013 Hyundai Accent report seemed to fit with the circumstances of the collision. In this case, although the Jeep may have had

data which might have been retrieved using the Bosch CDR Tool, such data was unavailable for this analysis but, by way of example, the Maximum Longitudinal delta-V of about -22 KPH, as well as the Time to Max delta-V of about 215ms for the Hyundai generally fit with the shape and duration of significant under-ride type collisions.

Likewise, the majority of the pre-crash data parameters seemed to fit with driver and/or witness statements, to in-

www.collisionpublishing.com

Collision Magazine - Volume 8 Issue 2 91

Figure 3.9: Reported vehicle speed from the Hyundai EDR Tool for the 2013 Hyundai Accent

Figure 3.8: Longitudinal delta-V reported from the Hyundai EDR Tool for the 2013 Hyundai Accent

clude self-approximated speeds, as well as reactionary measures (i.e.: braking just before impact). For this example; however, there remained data which was inconsistent with observations and logic.

Specifically, the report generated by the Hyundai EDR Tool suggested the driver of the 2013 Hyundai Accent was engaged in what would appear to be at least 250 degrees of counter-clockwise steering prior to the collision (-250° assuming SAE J211 polarity).

In contradiction to this observation, the roadway wherein this collision occurred was straight, at least two independent witnesses observed the 2013 Hyundai Accent traveling generally straight prior to the crash and the damage to the front of the Hyundai suggests that it was aligned directly behind the Jeep at impact. Based on the fact that 92

Collision Magazine - Volume 8 Issue 2

Figure 3.10: Reported brake status from the Hyundai EDR Tool for the 2013 Hyundai Accent the Steering Input data was reported at a constant and apparently maximum value, the authors theorize the error was a product of an improper data translation or a failure of the program to indicate the data element(s) a “clipped value” or “invalid” data element.

This theory was further supported through empirical testing with an identical 2013 Hyundai Accent. By way of example, the IIHS tested a 2013 Hyundai Accent as part of the Small Overlap Crash Test Program (discussed elsewhere in this Compendium). When a comparison of the report generated from the IIHS 2013 Hyundai Accent was made with the report generated from the real-world 2013 Hyundai Accent, the authors noted the Steering Input from the IIHS 2013 Hyundai Accent was translated as “Not Supported (or invalid data).”

www.collisionpublishing.com

Figure 3.11 Reported steering input angle from the for the 2013 Hyundai Accent in Case #2

Figure 3.12: Reported steering input angle from a 2013 Hyundai Accent small overlap crash test

One could theorize, the impossible translations as contained in the report generated by the Hyundai EDR Tool in April of 2013 were or may have been, at some point, remedied via a software update, as an identical vehicle, when accessed in August of 2013, with the same Hyundai EDR Tool, produced a report wherein Steering Input was more reasonably translated as “Not Supported (or invalid data).”

In this case example, the real-world 2013 Hyundai Accent data retrieval took place on April 12th, 2013. Unfortunately, just ten (10) days later, on April 23rd, 2013, GIT America released version “E-N-H-01-00-0003.”

This version of the software, having replaced “E-NH-01-00-0002,” remained current until the time of the accessing and generating a report for the IIHS 2013 Hyundai Accent on August 26th, 2013.

One may conclude that the error in the version of software available in April was remedied in this respect by August 2013; however, on the face of the report for each, the identified software version was 1.0. Unfortunately, the authors were unable to return to the real-world 2013 Hyundai Accent in attempt to re-access the vehicle and generate a report with the most current version of the software, as the vehicle was already salvaged by its insurer. While the available evidence could certainly lead one to hypoth-

Figure 3.13: Report header from case #2

Figure 3.14: Report header from IIHS crash test

www.collisionpublishing.com

Collision Magazine - Volume 8 Issue 2 93

esize on the root of the real-world 2013 Hyundai Accent quandary (mistranslation remedied by a software update), only the producers of the software could confirm such a postulation. In any event, parties in a claim or litigated case would be potentially faced with two reports ostensibly prepared using the same version of the software (1.0) but with different results for at least that particular parameter.

The authors specifically exploit this real-world example as an illustration of some of their concerns predicated on functionality of the Hyundai and Kia EDR Tool(s). As previously explained within the confines of this compendium, the Software Version, as displayed in the PDF document report header, does not actually provide the specific software version being utilized to generate a report. Moreover, the Hyundai and Kia EDR Tool(s) lack a means of preserving a raw data file that could simply be re-translated in the most current version of the software.

“

As this case example illustrates, end users of the Hyundai and Kia EDR Tool(s) are faced with a dilemma wherein the retention of a vehicle or vehicle ACU might play a pivotal role in the same end user's ability to provide a report generated in the most current version of the accompanying software. The Hyundai and Kia EDR Tools produce a report in a Portable Document Format (PDF), and, while these files have a wide viewing capability on virtually any operating system or platform, the PDF files produced by the Hyundai and Kia EDR Tools do not employ any of the security features or document protection capabilities offered through popular PDF commercial software applications, such as the Adobe Acrobat family. 11 The authors have no reason to question method in which data is collected within Hyundai and Kia ACUs. Nevertheless, the authors maintain their concern as to the potential challenges that may arise based on the functionality of the Hyundai and Kia EDR Tools as discussed later in the Conclusions section of this Compendium.

... the PDF files R produced by the Hyundai and Kia EDR Tools do not employ any of the security features or document protection capabilities ...

eferences 1. "Read Me First." Introduction. Kia EDR User's Manual. Tustin, California: Global Information Technology, 2012. 4-5. Print.

2. "Event Data Recorders" 49 CFR 563. 2012.

3. "New 2013 Cadillac XTS Premium Sedan Exterior Specs." 2013 Cadillac XTS Premium Sedan. MotorTrend Magazine, Source Interlink Media, n.d.

Web. 18 Oct. 2013. <http://www.motortrend.com/ cars/2013/cadillac/xts/premium_sedan/1175/specifications/exterior.html>.

4. "The 2013 Soul." Kia Cars, SUVs, Crossovers, Minivans, & Future Vehicles. KIA Motors America, n.d.

Web. 18 Oct. 2013. <http://www.kia.com/us/>.

5. Each driver weighed approximate 180lbs (Driver’s License Data), but the 2013 Cadillac was carrying an additional 100lbs of cargo in its trunk.

6. Chidester, Augustus, et al. "Recording automotive crash event data."Transportation Recording: 2000 and Beyond. International Symposium on Transportation Recorders. 1999.

7. Comeau, Jean-Louis, Alan German, and Donald Floyd. "Comparison of Crash Pulse Data from Motor Vehicle Event Data Recorders and Laboratory Instrumentation." Proc. CMRSC-XIV (2004): 27-30.

8. McClafferty, KJ, et al. "A preliminary comparison of real world frontal impacts and staged crash tests."

94 Collision Magazine - Volume 8 Issue 2

www.collisionpublishing.com

Proceedings of the Canadian Multidisciplinary Road Safety Conference XVI.

10. Newton, Isaac. Mathematical Principles of Natural Philosophy. A. Strahan, 1802.

9. Niehoff, Peter, et al. "Evaluation of event data recorders in full systems crash tests." Proceedings of the 19th International Conference on the Enhanced Safety of Vehicles. 2005.

11. "About Adobe PDF." Adobe Portable Document Format. Adobe Systems Incorporated, n.d. Web. 07 Nov. 2013.

Section 4: Coverage Spoofing - Data from Unsupported Hyundai and Kia Vehicles.

C

overage Spoofing Based on the nature in which the Hyundai-Kia EDR Tool(s) function (previously discussed in this Compendium), an exploration of the ability to retrieve data and generate a report from vehicles outside the intended coverage list was explored. Tests were performed via both DLC connection as well as direct to ACU (directto-module imaging). In summary, the tests returned a myriad of outcomes, ranging from apparent proper translation to manifest errors in significant aspects of the dynamic data being reported.

As discussed in the “Functionality” section of the Compendium, Hyundai-Kia EDR Tools do not require a specific VIN to identifying which specific vehicle one is attempting to retrieve data from. Instead, the end user may attempt retrieval by way of simply selecting the vehicle model, year (at this writing, 2013-2014) and engine. In that section of the Compendium, one may find that, even when no VIN is entered and a vehicle is selected from the “supported list,” the Tool will still retrieve data; at least from supported vehicles. For this portion of the research for the Compendium, the authors chose to attempt to re-

Editor’s note: For readers familiar with the operation and use of the Bosch CDR Tool, many of the Hyundai-Kia EDR Tool concepts discussed in the Compendium may be easily compared to those associated with that tool….whether “good” or “bad.” In this section, the authors address the notion of attempting to retrieve data from Hyundai and Kia vehicles which are not supported by the Hyundai-Kia EDR Tool.

In the context of the Bosch CDR Tool, this would be identified as the practice of “VIN spoofing” or attempting to use a VIN for one vehicle to retrieve data from another vehicle (often one not supported), in effect, trying to “fool” the CDR Tool software into retrieving data not currently supported by the Tool. In a research/testing scenario this is often done but then only with sufficient information and the cooperation of the vehicle or module manufacturer or supplier. In the field, this is often done to otherwise improperly circumvent restrictions put in place by an auto manufacturer on a vehicle or module (i.e.: an ACM with EDR potential) for a number of reasons. In the latter application, the net effect is often that data retrieved may be mistranslated by the Tool software and the “results” while appearing to be “meaningful” are otherwise unreliable and essentially useless.

This section of the Compendium addresses, in the context of the Hyundai-Kia EDR Tool, the idea that one may attempt to retrieve data from otherwise unsupported vehicles and then a review of some of the ramifications of that effort observed to date.

www.collisionpublishing.com

Collision Magazine - Volume 8 Issue 2 95

trieve data from Hyundai and Kia vehicles made during the years prior to the supported years (2013-14) by first selecting their supported equivalent model and then attempting retrieval from other models and the “opposite” manufacturer. Where users of the Bosch CDR Tool would recognize this as the equivalent of “VIN Spoofing,” in this context, since no VIN is required, the authors have elected to refer to this as “coverage spoofing.”

In the context of the Bosch CDR Tool, the end user of that system must first identify a vehicle manufacturer (OE) from a screen showing those OEs supported by that Tool and then either read from that vehicle by way of a generic OBD II process or manually enter a specific VIN to initiate data retrieval. In short, a specific VIN is required by the Bosch CDR Tool. The CDR Tool communicates with the selected module (here, as a corollary, the airbag control module) and, as part of a “hand shake” process identifies the module by a specific software or ROM ID. Assuming that the vehicle identified first by VIN is supported by the Tool and then the specific module is supported and not excluded by virtue of the module’s software ID, data is retrieved. Afterward, in the report generation process, the user entered VIN, software ID or both as well as other information available to the Tool may be used as part of the data translation and report generation process.

Where the user is in control of the VIN entered and that VIN is “spoofed” (specifically, not the actual VIN associated with the ACM being imaged) the results may be demonstrably unreliable. In testing conducted by the Collision Safety Institute 1, results have included, for example, the transposition of the data from the longitudinal and lateral axes or a reversal of polarity for one or more axes. Some manufacturers supported by the CDR Tool have instructed Bosch to include a check of the VIN recorded within the ACM against the user entered VIN and, where those don’t match, no translation is completed and no “plain language” report generated.

sion Safety Institute CDR Technician and Data Analyst courses. 1 Nonetheless, VIN spoofing has been the subject of some debate within the community. For example, some purported experts have promoted the practice where there is some suggested level of "exigency," where none truly exists, to “justify” the practice or where it is simply misrepresented as being conducted “for research” even though the data later finds its way into litigation. There is no doubt that, historically, some data retrieved as a function of VIN spoofing may be reliable as has been demonstrated as a function of testing and comparison to control instrumentation and observations; however, given the objectively demonstrated 1 potential for erroneous translation and reporting, the practice should be avoided.

In this section of the Compendium, the authors have selected examples of field work and crash test data to illustrate examples of what will be referred to in the context of the Hyundai-Kia EDT Tool(s) as “coverage spoofing.” In initial versions of the Hyundai-Kia EDR Tool, there was no reference to the practice one way or the other. In versions of the software released nearer November 2013, a single sentence was added to the apparently generic Data Limitations for the reports from both OEs (see the previously included sample report) indicating the system is not designed for models and model years not listed and a dialog box (Fig. 4.1) appears explaining coverage limits and requiring the user acknowledgement of that restriction to continue.

The Bosch CDR Tool addresses VIN spoofing specifically within the software Help file with, for example (the specific warning including the reference to “VIN spoofing”), that: “…Attempts made to retrieve data from vehicles built for markets the CDR Tool does not support by entering a VIN from a market it is built for (VIN spoofing) may result in inaccurate CDR report data translations and possibly incorrect or erroneous retrieval of data from the ECU.” 2

Figure 4.1: The [EDR Disclosure] confirmation dialog box requiring user confirmation of the limits of vehicle coverage.

Other references to this practice generally may be found in the Help file and within course notes from the Colli96

Collision Magazine - Volume 8 Issue 2

www.collisionpublishing.com

In this research, the authors address coverage spoofing in four ways:

suggest “spoofing” might be at least “productive” particularly toward an understanding of its potential.)

1. Selecting a supported make and model but an unsupported year (i.e.: an unsupported 2012 vehicle accessed by selecting that vehicle for model year 2013) 2. Selecting the wrong model without respect to year (i.e.: selecting a vehicle, whether or not for a year supported, other than the one the Tool is actually connected to) 3. Selecting a vehicle from the “other” manufacturer’s list (i.e.: attempting to download data from a Kia using the Hyundai software and selecting a Hyundai model vehicle whether or not from a supported year) 4. Attempting to retrieve data from an unsupported module by the direct-to-module method using the “wrong” cable

The original investigating police officers were met with an account of this incident such that, when they spoke to the driver of the Elantra, she indicated she was gradually slowing her vehicle while preparing to park in a painted parking spot. She reported that, as she reached her intended parking spot, she applied the brake but the engine revved uncontrollably, causing the vehicle to lurch forward, projecting it over the adjacent sidewalk and into the nearby storefront.

The examples that follow are offered to illustrate the process and, where appropriate, the pitfalls of coverage spoofing rather than endorse the practice.

Coverage Spoofing example: Selecting a supported make and model but an unsupported year - 2011 Hyundai Elantra In August 2013, a 78 year old driver crashed through the front widows of a storefront, depositing her 2011 Hyundai Elantra inside the store (Fig 4.2). Despite the otherwise dramatic nature of the collision, no one was hurt in the incident. (As background to this portion of the research, the authors identified that, for example, there were nominal changes made to the Hyundai Elantra’s basic platform between model years 2011 through 2013 3 which would

Based on the nature in which the event was reported, as well as the allegation of a sudden acceleration incident (SAI)4, the investigating agency sought consent from the vehicle owner to access the ACU data via the Hyundai EDR Tool knowing at the time that the 2011 model wasn’t specifically supported and with the understanding that the resulting report/data might be meaningless – if access were successful and if a report were to be generated. At that time, there was no prohibition, in the literature accompanying the Tool that the 2011 model was “excluded,” although all parties involved were aware it wasn’t specifically supported. In this example, the attempt at “coverage spoofing” was simple; the vehicle was accessed by imaging it as though it were a 2013 Hyundai Elantra. In short, the Tool was “fooled” into believing that it was communicating with a 2013 Elantra.

The report suggested, in the pre-crash data section, that the engine throttle was at 100% for the pre-crash -1.0 and -0.5 data samples; again keeping in mind that the driver’s account was that she was attempting to park her vehicle.

Likewise, there was a lack of any brake switch “on” (activa-

Figure 4.2: Photographs of 2011 Hyundai Elantra Crash

www.collisionpublishing.com

Collision Magazine - Volume 8 Issue 2 97

tion). At least in the initial evaluation of the report, the data retrieved apparently refuted the claim of a SAI. Examination of the remainder of the report yielded no outstanding or obvious discrepancies with the remaining data elements; such as Lateral Acceleration graph or Normal Acceleration graph anomalies. On its face then, the data appeared to be useful in the underlying crash evaluation and analysis; however, the investigating officers knew that there was no way to fully support relying on data retrieved through coverage spoofing.

With this example in mind, the authors undertook additional research into the potential for coverage spoofing with other vehicles not supported by the Hyundai-Kia EDR Tool(s), and were similarly successful in retrieving data often producing reports in which, at least on first examination, no clear discrepancies with the data elements appeared 3. However, despite the apparent “success” with coverage spoofing on some unsupported vehicles, additional attempts provided demonstrably defective translations or, in some cases, failed to retrieve data altogether.

Figure 4.3: Photograph of 2011 Hyundai Sonata Crash

Coverage Spoofing example: Selecting a supported make and model but an unsupported year - 2011 Hyundai Sonata As part of research conducted here, the authors found that, similar to the 2011-13 Elantra example, there were few changes made to the Hyundai Sonata’s platform between model years 2011 through 20136; however, when coverage spoofing was attempted on a 2011 Hyundai Sonata, the results of the imaging session displayed distinctly different results as compared to the otherwise seemingly promising Elantra data set previously discussed.

In this particular example, a 2011 Sonata vehicle was involved in a frontal collision, wherein the driver airbag, as well as driver seatbelt pre-tensioner deployed. In the review of the data successfully retrieved by the Tool from the unsupported 2011 Sonata where it was “spoofed” as a 2013 Sonata, there were clear discrepancies related to some data elements, specifically the Longitudinal Acceleration data, Lateral Acceleration data and the Normal Acceleration data. (The authors would note, at this point, that apparent anomalies in the display of acceleration data, regardless of axis, are a common theme among reports for vehicles accessed by way of coverage spoofing attempts7 and are consistent with similar observations made as associated with “VIN spoofing” in the context of the Bosch CDR Tool.)

In the report for the 2011 (as a 2013) Sonata, the Lateral Acceleration data, reported in g’s every 10 milliseconds,

98 Collision Magazine - Volume 8 Issue 2

www.collisionpublishing.com

Figure 4.4: Excerpts from the 2011 Hyundai Sonata EDR Report

appears to alternate between what would seem to be a maximum values of 25g’s and 0g’s for the 250 milliseconds illustrated in the graph and accompanying data table. In light of past experience with “VIN Spoofing” (in the context of the Bosch CDR Tool), one reasonable conclusion would be that the otherwise unrealistic appearance of the data displayed in the Lateral Acceleration chart (Fig 4.4) is a product of a translation error as a direct result of coverage spoofing.

Moreover, the Normal Acceleration data contained in the report generated by the Tool, also reported in g’s every 10 milliseconds, appears to alternate between values of up to 59g’s and 0g’s for the 250 milliseconds illustrated in the graph (Fig 4.4). Here, one might presume that the 59g’s represent a limit to the range of the system.

However, contrary to the data displayed in the Lateral Acceleration graph (Fig 4.4), the Normal Acceleration data points do not alternate between Maximum Values and Zero. Instead, the graph appears to report unclipped data points. One is left to assume the unreasonable appearance of the data displayed in the Normal and Lateral Acceleration charts are a product of a translation error, brought about by the effects of model year coverage spoofing.

Lastly, the Longitudinal Acceleration does not seem to represent the crash dynamics in this case. By inspection, one would notice that the Longitudinal Acceleration graph shows that there are many consecutive samples larger than -20g's. In fact, when the reported acceleration (Fig 4.4) is

Figure 4.5: Data elements of the 2013 Hyundai Accent imaged as a 2013 Hyundai Accent

integrated to calculate the total delta-V for the event, the result would be larger than -350 km/h. This is simply not realistic given the other facts of this case and the condition of the Sonata (Figure 4.3).

Coverage Spoofing example: Selecting an unsupported model without respect to year – 2013 Hyundai Accent Coverage spoofing can also be facilitated through the application of the GIT Tool, where the user initiates retrieval and generates a report through the application of the wrong model, while still selecting the appropriate year. By way of example, the Hyundai GIT Tool provides official coverage for 2013 Hyundai Accent. However, a user of the Hyundai GIT Tool is also able to produce a report as if the vehicle were imaged as a 2013 Hyundai Azera. This type of coverage spoofing also frequently produces what might be described as anomalous results.

Coverage spoofing the model alone produced abnormal acceleration graphs, both Longitudinal and Lateral, similar to a consolidation of those issues previously identified in the data associated with the coverage spoofing of the 2011 Hyundai Sonata (Figure 4.2). When the 2013 Accent is imaged as a 2013 Hyundai Azera, the longitudinal and lateral graphs yielded what appear to be at-or-near-maximum values, minimum values and values reported as “exceed the maximum sensor range” (clipping).

Figure 4.6: Data elements of the 2013 Hyundai Accent imaged as a 2013 Hyundai Azera

www.collisionpublishing.com

Collision Magazine - Volume 8 Issue 2 99

Coverage spoofing by use of the incorrect model also produced what would appear to be errors in a number of the translations pertaining to SRS deployment timing. If one assumes that the correct translation of data from a 2013 Accent is reflected in the report for that data where the 2013 Accent is the model selected, the translation of the 2013 Hyundai Accent spoofed as a 2013 Hyundai Azera produced mistranslations of the “Time to Deploy – Side Airbag Driver (No Deployment),” “Time to Deploy – Side Airbag Passenger (63 ms),” “Time to Deploy – Curtain Airbag Driver (No Deployment),” and “Time to Fire – Pretensioner Driver (No Deployment)” data parameters.

Coverage Spoofing example: Selecting a vehicle from the "other" manufacturer's list – 2013 Hyundai Accent Coverage spoofing can be facilitated through the application of an improper make, meaning the user completes retrieval and generates a report through the application of the Tool for the “other” manufacturer. By way of explanation, in this regard, coverage spoofing can be done by use of the Hyundai EDR Tool on a Kia vehicle, as well as vice versa.

Through the use of the Kia EDR Tool, the 2013 Hyundai Accent referenced in the previous spoofing example was imaged as a 2013 Kia Optima. A translation of the 2013 Hyundai Accent produced inappropriate translations of the acceleration graphs and deployment data, similar in fashion to the errors associated with coverage spoofing by use of the 2013 Hyundai Azera translation of the data from the 2013 Hyundai Accent (as seen in the previous example).

In testing, data was also imaged from the 2013 Hyundai Accent and translated as other Kia vehicles including the

2013 Kia Soul, the likes of which produced additional, yet different, erroneous translations of the data in the subsequent reports. Specifically, the translation of the data from the 2013 Hyundai Accent as a 2013 Kia Soul failed to populate the acceleration data sections of the report, suggesting such data was potentially “Not supported…” in the Kia. The translation of the data from the 2013 Hyundai Accent as a 2013 Kia Soul also failed to populate the Real-time Data section, specifically the Ignition Cycle at Download, suggesting the data parameter was also “Not supported…” as seen in Figure 4.7.

In all, the 2013 Hyundai Accent was successfully spoofed and data – whether or not properly translated – was recovered where the Accent was spoofed as a: • • • • • • • •

2013 Accent 2013 Azera 2013 Equus 2013 Genesis Coupe 2013 Genesis Sedan 2013 Sonata Hybrid 2013 Sonata Sedan 2013 Tucson

2013 Veloster 2013 Kia Forte 2013 Kia Optima Hy 2013 Kia Optima 2013 Kia Sorento 2013 Kia Soul 2013 Kia Sportage

Each of the successful coverage spoofing attempts performed on the 2013 Accent via either the use of the inappropriate make and/or model, produced incorrect translations absent the 2013 Hyundai Tucson and 2013 Hyundai Veloster. In those attempts, the retrieved data translated to a report identical to the report generated by way of imaging of vehicle as a 2013 Hyundai Accent.

Notably, there were also attempts at coverage spoofing which were met with negative results; the selected vehicle failed to retrieve data from the Accent module and returned an error message indicating a failure to communicate. Where coverage spoofing was attempted but failed, an error dialog box (Fig 4.8) appeared on screen indicating the selected vehicle was not supported/data retrieval failed.

In all, the authors were unable to access the data and produce a report when the 2013 Accent was spoofed as a: • • • •

2013 Elantra Coupe 2013 Elantra Sedan 2014 Equus 2013 I30

Figure 4.7: Top: Ignition cycle data when the 2013 Hyundai Accent is imaged properly.

Bottom: Ignition cycle data when the 2013 Accent is imaged as a 2013 Kia Optima.

100 Collision Magazine - Volume 8 Issue 2

• • • • • • •

www.collisionpublishing.com

• • •

2013 SantaFe Sport 2013 SantaFe 2013 Kia Rio

the Hyundai EDR Tool). However, where the direct-tomodule cable would be connected to the module (the port or opening typically referred to as the “ECU header” or “ECU Connector”) was pinned in a fashion that appeared to allow for the user to connect “Cable 9” (part G1ZDDPA009); a cable which is a standard component in both the Hyundai and Kia EDR Tools.

Figure 4.8: Error dialog box when coverage spoofing fails Coverage Spoofing example: Attempting to retrieve data from an unsupported module by the direct-tomodule method using the “wrong” cable – multiple vehicle examples Coverage spoofing experiments were also performed on Hyundai and Kia vehicles where a combination of the improper year and model or ACU cable was utilized to accessing and retrieve data. In researching vehicle information in anticipation of this Compendium, the authors learned that there were substantive changes made to the Hyundai Santa Fe platform between model years 2012 and 2013.8 Therefore, the authors expected similar substantive changes to the supplemental restraint system, to include the ACM/ACU. An ACU was salvaged from a 2012 Hyundai Santa Fe, involved in a real world collision. This particular module was identified by both Delphi and Bosch part numbers but appears to have been assembled by Bosch in Mexico and bears the Hyundai/Kia identifier "CM" and "FL" (identifiers which are found on both Hyundai and Kia modules) as seen in Figure 4.9.

However, "Cable 9" could not be connected to the ACU until keyways (small plastic tabs within the opening of the ECU header) in the ECU header of the ACU were shaved, thereby allowing for the ACU adapter to properly align with the pins. Once the keyways in the ECU header of the ACU were shaved, the ACU adapter cable (“Cable 9”) aligned properly.

Figure 4.10: Figure 4.10: ECU Connector/Header and "Cable 9"

Figure 4.11: ECU Connector/header before and after the keyway was shaved.

Figure 4.9: ACU from a 2012 Hyundai Santa Fe None of the ACU direct-to-module cables from the Hyundai or Kia user’s manual(s) are suggested for access to ACUs with either of the identifier codes CM or FL (and, of course, the 2012 Santa Fe is not listed as supported by

Figure 4.12: The cable now fits in the keyway

www.collisionpublishing.com

Collision Magazine - Volume 8 Issue 2 101

After this modification of the ECU header, in all, the authors were able to establish communication by way of previously described model and make coverage spoofing, retrieve data and generate a report from the 2012 Santa Fe as a: • • • • • • • • •

2013 Accent 2013 Azera 2013 Equus 2013 Genesis Coupe 2013 Genesis Sedan 2013 Sonata Hybrid 2013 Sonata Sedan 2013 Tucson 2013 Veloster

• • • • • • • •

2014 Kia Cadenza 2013 Kia Forte 2013 Kia Optima Hy 2013 Kia Optima 2014 Kia Sedona 2013 Kia Sorento 2013 Kia Soul 2013 Kia Sportage

As in previous examples, there were also coverage spoofing attempts which were met with negative results. In all, the authors were unable to retrieve data from the 2012 Santa Fe module when it was spoofed as a: • • • • •

2013 Elantra Coupe 2013 Elantra Sedan 2014 Equus 2013 I30 2013 SantaFe Sport

• • • •

2013 SantaFe 2014 Kia Forte 2013 Kia Rio 2014 Kia Sorento

Of particular note here, the authors were unable to establish communication by way of coverage spoofing and retrieve data from the 2012 Santa Fe when spoofed as a 2013 Santa Fe, effectively confirming a substantive change in the SRS between 2012 and 2013, or given the Data Limitations text for these systems, that there was a change in the Santa Fe ACU on or after Sept 2012.

Each of the successful coverage spoofing attempts performed on the 2012 Hyundai Santa Fe’s ACU produced numerous incorrect translations. For example, the data from the 2012 Hyundai Santa Fe ACU as a 2013 Hyundai Accent, generated what may be described as impossible translations (where compared to what was known about the 2012 Santa Fe involved in the underlying crash) including such elements as “Safety Belt Status Driver – Not Supported” and “Time To Deploy Frontal Airbags – 1st Stage Driver – 1ms.” (Fig. 41.3).

Similarly, coverage spoofing of the 2012 Santa Fe as a 2013 Hyundai Accent, produced clear discrepancies with the reporting of Longitudinal Acceleration data, similar to the previously described coverage spoofing of the 2011 Hyundai Sonata. The acceleration data, reported in g’s every 10 milliseconds, appears to alternate between maximum and minimum values for the 250 milliseconds illustrated in the graph (Fig 4.14).

102 Collision Magazine - Volume 8 Issue 2

Figure 4.13: Data elements from 2012 Hyundai Santa Fe imaged as a 2013 Hyundai Accent Perhaps even more concerning, was the nature in which the “pre-crash” parameters were translated when coverage spoofing was applied to the 2012 Hyundai Santa Fe. “Precrash data,” such as “Engine RPM,” “ABS activity” (Fig 4.15) “Stability Control” and “Steering Input,” all exhibited some of the reported translations as “Not Supported” or “Invalid Data,” almost randomly across the various models selected again calling into question the translation of the pre-crash fields.

Figure 4.14: Reported Longitudinal Acceleration when the 2012 Hyundai Santa Fe was imaged as a 2013 Hyundai Accent

www.collisionpublishing.com

ment, anticipated there may be similar substantive changes to the supplemental restraint system, to include the ACU. An ACU was salvaged from a 2010 Hyundai Genesis Coupe, involved in a real world collision. Also similar to the Santa Fe module, this module was identified by both Delphi and Bosch part numbers but appears to have been assembled by Bosch in China and bears the Hyundai/Kia identifier "BK."

Figure 4.15: Reported ABS Activity when the 2012 Hyundai Santa Fe was imaged as a 2013 Hyundai Accent

Figure 4.16: Reported Stability Control

Figure 4.17: Reported Steering Input The Hyundai and Kia EDR Tool Owner’s Manuals, would find that module “BK” would be directly accessible using ACU direct-to-module (Adapter) "Cable 9" (G1ZDDPA009). Likewise, if the user were to open the Hyundai software, they would observe that the 2013 Genesis Coupe purports to contain an ACU with the same "BK" (Fig 4.18). This might lead a casual reader to believe that the data contained in the subject ACU may be translated appropriately if the module identifier were to be an indication that the module type was otherwise supported.

Figure 4.16: Reported Engine RPM In an effort to ensure the test results associated with the coverage spoofing of the 2012 Hyundai Santa Fe were not another event anomaly or product of a power interruption, a second experimental test vehicle was sought out.

The authors identified that there were substantive changes made to the Hyundai Genesis’ platform between model years 2012 and 20139 and, based on the Santa Fe experi-

www.collisionpublishing.com

Figure 4.18: Adapter information found in Owner's Manual Collision Magazine - Volume 8 Issue 2 103

However, again the direct-to-module cable couldn’t be used in the ACU as was the case with the previously discussed 2012 Hyundai Santa Fe. Using the same approach as applied to the Santa Fe ACU above, the keyways in the ECU header on the subject ACU were shaved down allowing for ACU direct-to-module "Cable 9" to properly align with the pins.

49CFR563, deployment was selected by the OE as “Time zero (T0)” which is inconsistent with the apparent description or selection of “time zero” as seen in the remaining Hyundai and Kia reports; spoofed or otherwise.

Despite the potential for access to ACUs bearing the "BK" identifier code, the results of the coverage spoofing attempts on the 2010 Hyundai Genesis Coupe ACU were comparable to that of the 2012 Hyundai Santa Fe.10 For example, the translation of the data retrieved from the 2010 Hyundai Genesis Coupe spoofed as a 2013 Hyundai Genesis Coupe, generated what might be, at best, described as “impossible translations,” for example, “Time To Deploy Frontal Airbags – 1st Stage Driver – 0ms” and “Time To Deploy Frontal Airbags – 1st Stage Passenger – 0ms.” Obviously, a deployment in conjunction with “Time Zero” is impossible unless, in the context of

Coverage spoofing applied to the 2010 Hyundai Genesis Coupe as a 2013 Hyundai Genesis Coupe produced again what might be described as anomalies in the data with respect to the Longitudinal and Lateral delta-V data elements. In that report, the Longitudinal delta-V data

Figure 4.20: Reported Longitudinal Delta-V from the Hyundai Genesis

Figure 4.21: Reported Lateral Delta-V from the Hyundai Genesis

104 Collision Magazine - Volume 8 Issue 2

Figure 4.19: Deployment data elements from Hyundai Genesis

www.collisionpublishing.com

points (reported in kph every 10 milliseconds) appear to trend as might be anticipated for such data element until the about the 190ms point; however, at the 200ms data point, the delta-V magnitude abruptly changes returns to -1 km/h and continues on at that magnitude until the end of the graph (at 250ms). This apparent anomaly is observed in the Lateral delta-V trace as well, with a drop off from 10km/h after 130ms to 9km/h from time mark 140 to 210ms and then at 210ms an abrupt drop-off from 9 km/h to 1 km/h to the 250ms point. The “return” to 1km/h in both cases may be with a function of coverage spoofing and the resultant incorrect translation of data or as a function of an offset applied where the data “should” be 0 (zero) but is reported as 1km/h (again potentially as a function of the spoofing and associated translation).

As was the case in the 2012 Hyundai Santa Fe example, the “pre-crash” parameters as translated when coverage spoofing was utilized to access the data from the 2010 Hyundai Genesis yielded some apparently anomalous results.

“Pre-crash” data, such as ABS activity, Stability Control and Steering Input, all exhibited some of the reported translations as “Not Supported,” “Invalid Data" (Figs. 22 and 4.23) or apparent minimum and/or maximum values, many not reasonably plausible, once again calling into question the translation of the pre-crash parameters when spoofing.

Figure 4.23: Steering Input as reported for Hyundai Genesis The authors discovered during the coverage spoofing attempts with the 2010 Hyundai Genesis, that some of the previously mentioned anomalies appeared to be “remedied” by use of an alternate translation. Another way of saying it is that, by trial-and-error other spoofing efforts resulted in more “reasonable” appearing data. For example, retrieval of data from the 2010 Hyundai Genesis Coupe was also possible via the 2013 Hyundai Accent. In that attempt, the Longitudinal Delta-V trace from the coverage spoofing based data translation/report (Fig. 4.24) produced a more reasonable appearing Longitudinal delta-V trace absent the previously described “drop outs.” However, this alternate coverage spoofing 2010 Hyundai Genesis as a 2013 Accent produced translation errors of the same pre-crash data and resulted in additional apparent discrepancies in other event data fields.

Figure 4.22: ABS and Stability Control data as found in the Hyundai Genesis report

www.collisionpublishing.com

Collision Magazine - Volume 8 Issue 2 105

Figure 4.25 Deployment data from Hyundai Genesis when imaged as a Hyundai Accent In all, the authors were able establish communication through coverage spoofing and generate a report from the 2010 Hyundai Genesis Coupe (each with various apparent erroneous or anomalous data elements) as a: • • • • • • • • •

Figure 4.24: Longitudinal Delta-V for Hyundai Genesis when imaged as a Hyundai Accent For example, nearly all the data fields pertaining to the deployment of the supplemental restraint system devices were apparently affected when data was recovered from the 2010 Hyundai Genesis module when spoofed as the 2013 Hyundai Accent. And while not all errors in translation are covered within the confines of this article, some of the most prominent examples when the 2010 Hyundai Genesis was spoofed as a 2013 Hyundai Accent include: “Time To Deploy Side Airbags – Driver – 6,528ms,” “Time To Deploy Side Airbags – Passenger – 19,584ms,” “Time To Deploy Curtain Airbags – Driver – 47,105ms” and “Time To Fire Pretensioner – Driver – 1,793ms” (Fig. 4.25).” It seems most unlikely (at best) that an ACM/ACU would wait to deploy driver head curtain airbags for over 47 seconds into a side event appropriately severe enough to call for that protective measure.

106 Collision Magazine - Volume 8 Issue 2

2013 Accent 2013 Azera 2013 Equus 2013 Genesis Coupe 2013 Genesis Sedan 2013 Sonata Hybrid 2013 Sonata Sedan 2013 Tucson 2013 Veloster

• • • • • • • •

2014 Kia Cadenza 2013 Kia Forte 2013 Kia Optima Hy 2013 Kia Optima 2014 Kia Sedona 2013 Kia Sorento 2013 Kia Soul 2013 Kia Sportage

As in previous examples, there were also coverage spoofing attempts which were met with negative results. In all, the authors were unable to access the data and produce a report from the Genesis module when spoofed as a: • • • • •

2013 Elantra Coupe 2013 Elantra Sedan 2014 Equus 2013 I30 2013 SantaFe Sport

• • • •

2013 SantaFe 2014 Kia Forte 2013 Kia Rio 2014 Kia Sorento

The reader will likely have noticed that the list of vehicles through which access was and was not possible by way of coverage spoofing attempts with the 2012 Hyundai Santa Fe and with the 2010 Hyundai Genesis are identical.

During a preliminary examination of coverage spoofing attempts, the authors identified something of a pattern or groupings of vehicles which allowed for common access and retrieval and report generation. As with other OEs, Hyundai and Kia are demonstrably using ACMs/ACUs from multiple first tier suppliers with several commonalities (as well as, of course, differences).

www.collisionpublishing.com

In another series of tests, the authors also undertook coverage spoofing on a 2014 Kia Forte. In all, the authors were able establish communication through coverage spoofing and generate a report from the 2014 Kia Forte as a: • • • • •

2013 Elantra Coupe 2013 Elantra Sedan 2014 Equus 2013 I30 2013 SantaFe Sport

• • • •

2013 SantaFe 2014 Kia Forte 2013 Kia Rio 2014 Kia Sorento

However, the authors were unable to access the ACU and produce a report from the 2014 Kia Forte when spoofed as a: • • • • • • • • •

2013 Accent 2013 Azera 2013 Equus 2013 Genesis Coupe 2013 Genesis Sedan 2013 Sonata Hybrid 2013 Sonata Sedan 2013 Tucson 2013 Veloster

• • • • • • • •

2014 Kia Cadenza 2013 Kia Forte 2013 Kia Optima Hy 2013 Kia Optima 2014 Kia Sedona 2013 Kia Sorento 2013 Kia Soul 2013 Kia Sportage

The list of vehicles for which access was/was not possible by way of coverage spoofing attempts with the 2014 Kia was opposite that list for the 2012 Hyundai Santa Fe and 2010 Hyundai Genesis. In the final analysis, the authors discovered that coverage spoofing, regardless of the year, make and model of the subject vehicle, were generally possible via one of the two aforementioned groupings without respect to the “reliability” of the resultant data/report.

O

ther Data Parameters As illustrated in this section, a data parameter may appear in a “spoof ” generated report and may, at least initially, appear to contain reasonable data (apart from the otherwise “odd” appearance, for example, of some of the acceleration data). Other data parameters observed in spoofed and non-spoofed reports alike should then be addressed in this context.

device at the time when the data was downloaded since the first use of the EDR.”11 In the output report generated by the EDR Tool(s), this parameter is found in the “Real Time Data” section (at page 27 of the example report in this Compendium). In that same example report, at page 1 in the report “Data Limitations” section we find this: “… Ignition cycle counter(download) will increment by 1 every time when the power mode cycles is changed from OFF/Accessory to IGN/RUN or EDR data is downloaded by using the retrieval tool.” (sic) (emphasis added) The first part of the definition is clearly consistent with that found in sub-part 563.5; however, the second part “... or EDR data is downloaded by using the retrieval tool.” is not only objectively inaccurate but would also not seem to be in keeping with the specified definition (563.5 noted above)11 not to mention potentially raise other issues again related to the research and litigation application of this data (while those areas are admittedly not contemplated within the scope of 49CFR563 as later discussed in the Conclusions section of this Compendium).

During testing, for more than one vehicle and module, the authors applied power (either by way of the ignition in the car during a “DLC” download or by way of direct-tomodule connection) and then initiated a download having properly identified the supported vehicle or as a function of “coverage spoofing.” Once that download was complete and the PDF report document saved, the process was initiated again without cycling the ignition or removing and reapplying power to the ACU. Without cycling power and even without retrieving data to generate a PDF report (where there was an error encountered during spoofing indicating the vehicle wasn’t supported), the ignition cycle counter “at download” still incremented.

Such that the process is clear, as part of testing this function, the authors:

First, within the Functionality section of this Compendium, one finds a discussion of the process of retrieving data from a given vehicle; the connection of components leading to retrieval. Within that section and then as part of the resultant output report, one finds a data parameter called for in sub-part 563.7 (“Data elements.” at Table I): “Ignition cycle, download.” It is specifically defined at “563.5 Definitions.” as:

1. Connected the Hyundai-Kia EDR Tool to a car’s DLC and the laptop, 2. Brought up the EDR Tool software and selected a vehicle (spoofed or supported), 3. “Clicked” on the “retrieve button” (Fig 1.22), then 4. After the retrieval and report process completed, saved the PDF report

“… Ignition cycle download means the number (count) of power cycles applied to the recording

5. (a) Power was not cycled (the key position wasn’t changed on the car for an in-car download and the

The variations in testing after “step 4” were:

www.collisionpublishing.com

Collision Magazine - Volume 8 Issue 2 107

cables weren’t disconnected where the connection was direct-to-module) (b) The “retrieve button” was “clicked” again (after the previous report was saved) (c) Data was either retrieved and a PDF report generated and saved or an error message was observed indicating a failure to collect/retrieve data (Fig. 4.8) 6. Step 5 (a) and the process was repeated In one specific example, a 2012 Hyundai Sonata was connected for imaging through the DLC. Steps 1 through 4 above completed, the Sonata’s ignition was turned to “IGN/RUN” and the data recovered by spoofing the car as a 2013 Sonata. Then, without changing the car’s ignition setting, the Hyundai hardware was left in place, the Kia software opened and the Sonata downloaded subsequently and ignition cycles observed as seen in Table 4.1.

2012 Sonata spoofed as 2013 Sonata 2013 Elantra 2014 Cadenza

Ignition cycle, download 1832 <failed> 1834

2013 Optima 1835 2014 Sedona 1836 2013 Sorrento 1837 2014 Soul “(parameter) not supported” 2013 Sportage 1839 Vehicle ignition cycled to “off” then cycled back to “on,” Hyundai software initiated, Kia hardware attached 2013 Elantra <failed> Ignition cycled to “off” then back “on” 2013 Accent 1842 2013 Tucson 1843 2013 Genesis 1844 Hyundai software closed, Kia software opened (Kia hardware still attached) 2014 Sportage 1845

Table 4.1: Ignition cycles observed Notably, in the process, for the vehicles selected from the Kia and Hyundai list, one attempt resulted in a failure to communicate/retrieve data (the 2013 Elantra) and one (2014 Soul) returned a report; however, instead of an ignition cycle number, the parameter displayed the words “Not Supported…” This might be compared to the spoofing experiment conducted with the 20120 Hyundai Genesis discussed earlier. When downloaded as a 2013 Accent, for example, the parameter displayed the translated result: 108 Collision Magazine - Volume 8 Issue 2

“Invalid data.” In another example, a 2011 Sportage was spoofed as a 2013 Sportage. The ignition cycles at the event (“Ignition cycles, crash”) were reported as 818 and, at download, reported as 4387.

In the latter example, from a car involved in a “real world” crash, it is difficult to draw too many conclusions from the huge difference in ignition cycles (which could potentially have been from actual vehicle use) but an obvious potential culprit remains it is a function of spoofing.

While, in the particular experiment with the Sonata, other (supported but spoofed) vehicles could have been selected during the process, but the point would seem to be clear from the sequence completed: when the Hyundai-Kia EDR Tool communicates with a module – whether or not data is retrieved (as suggested by the Data Limitations) – the current “ignition cycle counter” is incremented by one. When power is actually cycled at the ignition, the current counter is also, of course, incremented by one.

Connecting and disconnecting hardware or opening and closing software doesn’t appear to be associated with the cycle incrementing. It would seem to be clear then that the current counter is actually incremented (1) when vehicle power (at the ignition) is applied and (2) when the Hyundai-Kia EDR Tool communicates with a module in some fashion not limited to actually retrieving data. Similar experiments were conducted with other vehicles and with direct-to-module connection downloads with similar results (not all were conducted switching the hardware or software packages).

A second data parameter which caught the attention of the authors was the parameter “Engine Throttle” (percent).

Inasmuch as there is no objective frame of reference for readily evaluating the parameter in “real world” crashes, the authors looked to IIHS crashes (some of which are also discussed in the IIHS Crash Test Data section of this Compendium). Across the selected group of IIHS crash tests, the observed throttle percentages observed in the EDR reports (some for spoofed vehicles other were not spoofed) are shown in Table 4.2.

Where the car in these tests would not have a driver and no acceleration would be applied through the accelerator to the throttle of the vehicle (under the hood), one would, initially presume that the parameter should reflect “0” (zero) or at least be relatively consistent across a group of vehicles not to mention one vehicle in a single test. As can be seen in Table 4.2, the results varied from a low of 1% throttle for the entire “pre-crash period” to a variable of 13 to 14% and 14 to 15% in two of the tests. Initially, one might suspect this variance is a function of spoofing (which was done on the 2012 Accent and 2012 Rio; how-

www.collisionpublishing.com

ever, those were at least “both 8%” whereas two tests with the two 2013 Rios had very different results.

Vehicle 2012 Accent 2012 Rio 2013 Accent 2013 Elantra 2013 Tucson 2013 Rio 2013 Rio 2013 Soul 2014 Forte

Throttle Percentage 8% 8% 14% and 15% 7% 4% 13% and 14% 8% 8% 1%

3. "Hyundai Elantra Model History." NADAguides.

NADA Services Corporation, n.d. Web. 19 Oct. 2013. <http://www.nadaguides.com/Cars/2013/hyundai/elantra/Model-history>.

4. SAl is defined as unintended, unexpected, high-power accelerations from a stationary position or a very low initial speed accompanied by an apparent loss of braking effectiveness. - Pollard, John, and E. Donald Sussman. An examination of sudden acceleration. National Highway Traffic Safety Administration, 1989.

Table 4.2: Throttle Percentage found in the reports from the IIHS crash tests Since the throttle on a late model vehicle is open (to some degree depending on a variety of factors and OE specifications) as soon as the ignition is “on,”for there to be "some" percent throttle recorded (as a function of the voltage being observed at/from the throttle position sensor) is not abnormal and should not be confused by one examining the data as being a function of spoofing.

S

ummary The authors believe it is important here to remindthe reader that coverage spoofing is not generally recommended. While it is by way of this sort of comparison that “implausible” or “impossible” data may be identified, a single report obtained by way of spoofing might, on its face, appear to yield reasonable data; however, end users would normally have no way of knowing whether or not the translation was actually done properly for the unsupported module (vehicle) and should not generally rely on data retrieved by way of this practice.

Based on the uncertain nature of coverage spoofing, the authors recommend against its application as, more often than not, coverage spoofing has been shown to produce nominal to dramatic errors in the resultant data translation producing errors or anomalies which might not be readily apparent.

R

2. "Help (file)," (from the) Crash Data Retrieval Tool (software), Robert Bosch, LLC. Version 12.x series released October 2013

eferences 1. Collision Safety Institute, CDR Data Analyst Course curriculum notes, 2001-2013. www. collisionsafety.net

5. The vehicles included a 2012 Hyundai Tucson and a 2012 Kia Rio. 6. "Hyundai Sonata Model History." NADAguides.

NADA Services Corporation, n.d. Web. 19 Oct. 2013. <http://www.nadaguides.com/Cars/2012/hyundai/ sonata/Model-history>.

7. Nearly identical anomalies were found in reports generated through model year Coverage Spoofing on both real-world and instrumented tests, to include a 2012 Hyundai Accent and 2011 Kia Sportage.

8. "Hyundai Santa Fe Model History." NADAguides.

NADA Services Corporation, n.d. Web. 19 Oct. 2013. <https://www.nadaguides.com/Cars/2002/Hyundai/ Santa-Fe/Model-History>.

9. "Hyundai Genesis Model History." NADAguides.

NADA Services Corporation, n.d. Web. 19 Oct. 2013. <http://www.nadaguides.com/Cars/2013/hyundai/ genesis/Model-history>.

10. A third Coverage Spoofing attempt utilizing an ACU from a 2012 Kia Forte (Mobis ACU / Identifier Code ‘TD’), produced nearly identical results as that of the 2010 Genesis Coupe and 2012 Hyundai Santa Fe, despite nominal changes in the vehicle platform between 2012 and 2013 – “Kia Forte Model History."

NADAguides. NADA Services Corporation, n.d.

Web. 19 Oct. 2013. <https://www.nadaguides.com/ Cars/2013/Kia/Forte/Model-history>.

11. "Event Data Recorders" 49 CFR 563. 2012.

www.collisionpublishing.com

Collision Magazine - Volume 8 Issue 2 109

Section 5: Compendium Conclusions and Summary

C

onclusions and Summary This Compendium has been assembled based on the most current data available to and latest research conducted by the authors. As this technology and, more specifically, as the Hyundai-Kia EDR Tool(s) may evolve, opinions offered here may become dated if not obsolete. For this section of the Compendium, the authors have attempted to group their observations, opinions, conclusions and recommendations by topic following the assembly of the Compendium by section. Some of the content of this section may be repeated in or summarized in one form or another in the foregoing sections. For example, observations about the report formatting would be associated with the “Functionality” section of the Compendium and would be grouped under that heading within this summary.

What might be made as something of a blanket observation at this point (and as was discussed in more detail in the previously published “Hyundai and Kia Crash Data, a Preliminary Overview”)1 is that it would be reasonable to conclude, based on the information found in the GIT EDR Tool user’s guide and the “Data Limitations” portion of the Hyundai-Kia EDR Tool reports (a sample is found earlier in this Compendium) which, but for the provisions of 49CFR563 (specifically sub-parts 563.3, 563.6 and 563.12)2, there would be no tool, publically available, which might be used to retrieve crash data from Hyundai and/or Kia EDR components of their ACUs. Parenthetically, the same might be said to one degree or another as this relates to other auto manufacturers. At this point, in light of the provisions, definitions, and requirements found in 49CFR563, in a basic sense, it is the opinion of the authors that the GIT Hyundai-Kia EDR Tool appears to largely address the minimum requirements of “the rule.”

F

unctionality With 49CFR563 as something of a guide, we might look to sub-part 563.12 which reads:

“… Data retrieval tools. Each manufacturer of a motor vehicle equipped with an EDR shall ensure by licensing agreement or other means that a tool(s) is commercially available that is capable of accessing and retrieving the data stored in the EDR that are required by this part. The tool(s) shall be commercially available not later than 90 days after the first sale of the motor vehicle for purposes other than resale.”

As discussed previously, this sub-part2 does not specifically call for a report to be generated nor does it call for a digital 110 Collision Magazine - Volume 8 Issue 2

file to be stored or recorded by the retrieval tool which may be available to the researcher, regulator or reconstructionist to retranslate or re-report in a later revision (read: updated or perhaps “corrected” version) of the retrieval and translation software. Taken literally, the Hyundai-Kia EDR Tool accesses the module where EDR data would be recorded in a vehicle and retrieves that data. Beyond those functions, it generates a report such as the sample report found previously in this Compendium and which is typically laid out as discussed in “Section 1, Functionality.”

From the perspective of an effort designed solely to meet that minimum published standard, this would likely seem sufficient; however, other potential perspectives, or perhaps applications, might be considered. Suppose, for example, that 49CFR563 were to become FMVSS405 (as has been proposed)3 and one were to be tasked with reviewing a collection of EDR data from a group of, solely for the sake of example here, Kia Rios and were interested in whether or not a particular data element found across those reports met the accuracy, range and/or resolution requirements of sub-part 563.8 (Table III).

As the situation currently exists, if one were to be working from PDF files made, potentially, over the span of several months and correspondingly by way of potentially several versions and/or revisions of the Kia EDR Tool software, it would be, at best, difficult to affirmatively identify a trend for a given data element or elements. Moreover, since the software or tool version number (as displayed in the report) doesn’t change (undeniably, there are demonstrably changes and updates to the program periodically), it would be nearly impossible – from the PDF reports alone – to evaluate in which report (if any) a given parameter’s translation was correct (or incorrect) or whether or not that (or which) translation was “current” particularly where there might be differences observed across other reports in the set under study.

We might assume that the chronologically “most recently saved (printed)” report was the most current and might then be adopted as the most reliable; however that presumes - at a minimum - that (1) there had been an update or fix to a select data element (if, for example, one has been identified and some change required) before that report date and (2) it was included in a software update distributed prior to that date, that (3) the operator had updated the software prior to the download and report generation, and that (4) one would be aware of when (by date) updates to the program had been made such that report generation date might be used as a marker in place of incrementally

www.collisionpublishing.com

increasing software revision numbers during a comparative analysis or review.

As it exists now, a report generated in, for example, July 2013 and one generated in November 2013 from a pair of 2013 Kia Rios may have differences, between the two reports, for a given data element. We might assume that the November 2013 report is the most current translation of the data; however, we have no way of knowing at that point – if we were to go back to the Rio ACU which was downloaded in July – whether or not we would get the same report as we find in the November download of the second Rio. Unless we were to have maintained custody of the first Rio or, at a minimum, the ACU from that vehicle (a luxury not available to all researchers or reconstructionists and which poses other issues addressed later), based on a review of just those two reports, what changes may or may not have been made. Some differences between the reports, lacking the ability to return to one vehicle/module or the other for some sort of comparative “second” download, could be a function of options or a lack of a system on one vehicle compared to another.

If, on the other hand, the Hyundai-Kia EDR Tool were (1) to clearly indicate the actual version number of the software in the report and were to (2) save a set of raw data (preferably in a security enhanced digital file (by way of comparison as found in the Bosch CDR Tool and its “.CDRx” file structure) which could be reopened in later released versions of the software and the “new” translation of the data would result in an updated report, this potential issue might be effectively resolved: the original retrieved data could be reexamined again later as (potential) revisions to the program are made.

Overall, by comparison, like the Hyundai-Kia EDR Tool(s), the Bosch Crash Data Retrieval Tool similarly produces an original printout, but the Bosch Tool also provides a secure digital copy of the underlying data which can be retested by newer versions of the Bosch Tool software. The secure digital file is, by comparison, a “photo negative” or a “blood sample” that would serve to support that the output report is an actual duplicate of the original (here, the original data).

Editor’s note: This Compendium is the first-of-its-kind published in-depth review of several important aspects of EDR data available from Hyundai and Kia vehicles and a follow-up to the authors’ article “Hyundai and Kia Crash Data, a Preliminary Overview” found in Collision Volume 8, Issue 1. The conclusions offered in this section reflect observations made by the authors through early November 2013 as this issue of Collision goes to press. Given that event/ crash data technology is subject to frequent changes, updates and evolution, some of the conclusions may not later reflect the state of the system(s) being referenced as a function of subsequent changes, updates or other modifications to the Hyundai-Kia EDR Tool.

As mentioned elsewhere in the Compendium, during the preparation and research for this article, the authors reached out to Hyundai Motor America (Hyundai), Kia Motors Corporation (Kia) and Global Information Technology (GIT, the manufacturer of the Hyundai and Kia EDR Tool(s)). They were also provided with a copy of “Hyundai and Kia Crash Data, a Preliminary Overview.” A representative of Hyundai responded only that he was interested in the EDR test set up “status” and was looking to review the author’s information associated with what kind of test was performed and how the testing was done as well as data on the differences between the data extracted using EDR tool and the “ACU internal data.” Despite follow up calls and emails from the authors offering to review the available information, no further communication has been had between the authors and any of those entities nor did they participate in the preparation of this Compendium.

Going forward then, the opinions and conclusions found in this section of the Compendium are offered to the end user of the GIT EDR Tool or those reviewing data retrieved by those EDR Tool(s) mainly as a guideline and discussion of some of the things to be aware of; perhaps “food for thought.” The reader should keep in mind there is the probability that there might well be changes to the systems made after publication of this issue of Collision which might reverse some of the comments made in this section.

While the authors do not profess to offer regulatory or legal advice or conclusions, some observations and opinions found in this section may well be useful to the researcher or practicing reconstructionist attempting to use and apply Hyundai and Kia EDR data, particularly in litigation, whether civil or criminal. Here, the authors have drawn on more than 50 years combined experience as law enforcement officers, academics and expert witnesses in criminal and civil litigation in the preparation of portions of this section. www.collisionpublishing.com

Collision Magazine - Volume 8 Issue 2 111

The authors full well recognize and reiterate that concerns associated with litigation go beyond the scope of 49CFR563. Nonetheless, the reality is that this technology reaches beyond regulation and research; it undeniably has application in both civil and criminal litigation. Ample examples of this can be found from early litigation5 associated with the technology to more recent cases6, among many others. To that end, an inescapable component of this Compendium would be to extend the analysis to contemplate the effect of the observations made here on litigation.

In the context of litigation then, where two experts might be working with two reports – both from the same, for example, 2013 Hyundai Sonata case vehicle – it is conceivable that Expert #1 would have a report which shows one or more data elements one way and Expert #2 would have a report that displays the same data element(s) another way. As it currently exists, both would have reports generated by what appears to be “Version 1.00” of the program (displayed with the suggested precision found in the report to the second decimal place…) even though they may have different dates for the data access and retrieval (indicated by the “Save on” (sic) date found in the report header).

That there might then be differences in the data parameters but where the apparent (reported) software version number remains unchanged may raise another possible issue associated with the application of this data; data integrity not to mention individual integrity.

The Hyundai-Kia EDR Tool generates a report as a PDF document. The report, in that format, is not a secure digital file and can be easily edited; altered. While that observation is not, of course, a suggested path one might take and surely not one recommended or endorsed by the authors, it is mentioned here as an obvious but unfortunate extension of the application. Where the foregoing example were to come to life, it is a virtual certainty that, someone will suggest that one or the other of these experts (1) “must have altered the PDF” or (2) that one is actually working from a download from a different vehicle since, after all, there is no VIN associated with either report and no objective way to associate the report with a specific vehicle or module (coverage spoofing aside for the moment).

Inasmuch as there is no secure, original digital file to return to for comparison, this allegation will almost assuredly cast a shadow on both of these reports (whether actually “edited” or not) and raises the potential that Crash Data Retrieval Technology in general might well be painted by an ignorant broad brush in a legal decision down the road as in an example such as the convoluted and ill-informed decision found in “LaBorde.”7

112 Collision Magazine - Volume 8 Issue 2

From the perspective of a tool designed to meet the minimum requirements of 49CFR563, an original, secure digital file which might be useful in research or litigation isn’t a priority consideration so the approach taken here by GIT with the Hyundai-Kia EDR Tool(s) is “understandable;” however, the authors believe it’s short sighted and ultimately might be more detrimental; not only to users of the Hyundai-Kia EDR Tool(s) but with the potential for a broader impact on other similar tools or applications whether in regulation, research or litigation. So as to be clear on this point, the authors have no reason to question the method in which data is collected within Hyundai and/or Kia ACU EDR components. Nevertheless, the authors maintain their concern as to the potential challenges that may arise based on the functionality of the Hyundai and Kia EDR Tools in a litigation scenario and potentially similar challenges as may be related to research and even regulation.

As discussed in the preliminary review1, there are those who may suggest that “documenting” the module and getting past the potential issues discussed herein will be overcome by taking some photos and/or video of a data retrieval process or of a module against the background of a damaged car. Still others may suggest the simple preservation of an ACU (ACM) will provide sufficient basis and background to associate the data retrieved and reported with a specific vehicle. This is nothing less than a grossly oversimplified approach, allowing for arguments of credibility, as well as corresponding analogies to the bloody glove that was famously photographed, videotaped and “preserved."

Object or evidence collection and preservation, when done properly, can be a complex practice. Investigators - whether in the civil or criminal arena - must concern themselves with the potential for, and ramifications of, “destructive testing” as well as the potential for a claim of spoliation of evidence.7 Part of a proper evidentiary practice requires pre-planning, multipart photographic and/or video documentation, minimally invasive collection or retrieval, as well as validation and authentication procedures to include the maintenance of a chain of custody of a given object where appropriate. While largely similar in a gross sense, each of the phases or components of an object or thing’s preservation may be applied differently in criminal as opposed to civil investigations.

While not addressed in any expanded fashion within this Compendium, the authors have on reason expect the standard of care associated with the collection and preservation of a vehicle and/or Hyundai or Kia ACUs to be any different than the standard by which any other item(s), likely to be involved in criminal or civil litigation, would be held. The authors recognize the existence of a number

www.collisionpublishing.com

of reasonable and informative resources from which the reader may seek guidance when attempting to produce an effective methodology for collection and/or preservation of articles or objects collected by an investigator. Some “standards” have been proposed, putting style over substance, becoming so overly restrictive as to be impractical. While these “standards” ultimately fail in terms of being effective, some others serve as excellent guidelines leaving room for experience and practicality to those individuals who would observe them.

For example, the American Society for Testing and Materials (ASTM) has developed a number of recommended practices with regard to object and/or evidence collection and preservation including “Collection and Preservation of Information and Physical Items by a Technical Investigator”8, “Examining and Preparing Items That Are or May Become Involved in Criminal or Civil Litigation”8, and “Collection of Non-Volatile Memory Data in Evidentiary Vehicle Electronic Control Units”10. The guides – especially the latter in this context – speak specifically to proper evidence memorialization as well as the issues surrounding the collection of electronic evidence from event data recorders, the likes of which cannot be summarized by simply collecting a few photographs or putting an “evidence” sticker on a module or object. In short, labeling or calling something “evidence” doesn’t make it so.

While the authors cannot provide a strict or comprehensive guideline to which all investigators in every jurisdiction must adhere, these ASTM standards have been previously cited and a lack of observance of these standards was noted by the court in a successful Daubert11 challenge of an expert witness’ qualifications. 12 Prior to the release of the Hyundai-Kia EDR Tool(s), the only widely available system for retrieving “EDR/crash” data was, of course, the Bosch CDR Tool. In his article and oral presentation at the CDR User’s Summit, Melkonian 13 pointed out that the airbag control module is an object, it may well be an “exhibit” at trial, but that module and even the Bosch CDR Tool report are ultimately, cautiously considered “exhibits” as opposed to “evidence.” By definition, “evidence” is what is considered by the terrier of fact (whether judge or jury); a decision in a case is supposed to be based on an evaluation of the admitted evidence. At the end of the day, what the trier of fact will consider with respect to this technology is what a qualified expert testifies to not the presence of an object with a label on it or in some sort of sealed bag.

Where an electronic control unit (typically the ACM) has been downloaded using the Bosch CDR Tool, the module isn’t, of course, going to “testify” in front of the jury; at

best it will effectively become a “paper weight” in the jury room. In some jurisdictions, even the output “CDR report” may not be allowed to go before the jury (although a less likely occurrence). What the trier of fact will consider is the expert’s opinion.

But what if the expert’s opinion is based on something that can’t be demonstrated as a product of the best test conducted on an object? Even if the best test is applied, what if the history of or source of the object is open to or subject to question?

The undeniable reality remains that (1) the Hyundai-Kia EDR Tool(s) output is not a secure digital file, (2) neither the Huyndai-Kia EDR Tool(s) report nor does the ACU (preserved as an exhibit or as “evidence”) contain reference to a specific VIN identifying a specific vehicle as the source of the data, (3) as demonstrated in the “Coverage Spoofing” section, a specific VIN is not even required to retrieve data (and “spoofing” can have its own unanticipated results), and (4) the report does not realistically or clearly represent the current software version number making it nearly impossible to determine if the translation and resulting report are conducted as a function of the best test (the latest translation of the data as distributed by the manufacturer(s)).

Another consideration in this respect is that, like the crash data reports available from other manufacturer’s vehicles accessible using the Bosch CDR Tool, there is no date/ time stamp on the data indicating the date/time of the underlying crash14. Instead, the researcher often relies on comparisons to the results of a situationally complete reconstruction or other characteristics of the data and an investigation to associate data with a specific crash. One of those other characteristics is an ignition cycle counter.

Over time, ignition cycle counts have played a role in litigation 15 as well as research and are important enough on the regulatory side to have been included as a required element in 49CFR563 (Table I).

As illustrated in the “Functionality” and “Coverage Spoofing” sections of this Compendium, the reliability of the ignition cycle count in Hyundai and Kia vehicles between an event and the time of the download which might be used to assist the researcher as well as the litigation expert in evaluating the “age” of the recorded data relative to the time of the retrieval (a comparison of the parameters “ignition cycle, crash” and “ignition cycle, download”) is questionable if not simply unreliable (coverage spoofing aside for the moment).

Another issue to be contemplated not only by testifying experts but by researchers and regulators as well would be the depth and quality of the “Data Limitations.” A com-

www.collisionpublishing.com

Collision Magazine - Volume 8 Issue 2 113

parison can be made here to the Bosch CDR Tool in which one finds the first reference to and first use of the term “Data Limitations” in the context of “EDR technology” or crash data reports. In the CDR Tool, the “Data Limitations” have been, since the product was first released, a common report element used to inform the end user about limits or restrictions associated with the data retrieved and translated, specific nuances and identified potential anomalies associated with a module (or vehicle) type and other considerations (i.e.: the more recently added polarity tables). In some cases, manufacturers have elected to include information as detailed as the potential range for a given data element or parameter.

reasonably accurate; however, there are notable exceptions.

These include:

The Data Limitations found in the report generated by the Hyundai-Kia EDR Tool(s) include elements found in the Data Limitations text found in the Bosch CDR Tool generated reports but are not nearly as well developed nor, demonstrably, informative. Perhaps another way of saying this is that it is the authors’ considered opinion that the Data Limitations found in a Hyundai or Kia crash data report (refer to the sample report included with this Compendium, for example) is virtually useless as it might offer any real, meaningful understanding of the data presented.

Arguably, some of these apparent discrepancies or differences are a function of spoofing and others might be addressed in a more thoroughly developed and detailed "Data Limitations" which would (should) accompany these types of reports (as previously noted).

The authors recognize that there is no requirement for an explanation of or inclusion of some sort of “data limitations” reference document or similar information within the provisions 49CFR563; after all, one can argue there isn’t even a requirement for a report. In general, the functionality and application of the Hyundai-Kia EDR Tool(s) may well be considered to address the minimum requirements of 49CFR563 but the end product – the report – should be carefully considered in light of the considerations addressed in this Compendium separate and apart from issues associated with “spoofing.”

In summary, the authors support a cautious approach to the preservation of exhibits and/or “evidence” from Hyundai and Kia vehicles and cannot specifically recommend a definitive approach to navigating the potentially complex legal issues related to evidence collection and/or retention or admissibility in every possible jurisdiction. However, the reader needs little imagination to visualize how the discussed shortcomings of the Hyundai and Kia EDR Tools could apply to the operational demands of Law Enforcement, insurance claims administration, and expert witnesses alike.

R

eview of Crash Test Data In large part, the Hyundai-Kia EDR Tool(s) reported data, compared to that collected by the IIHS in the assembled series of crash tests, appears to be

114 Collision Magazine - Volume 8 Issue 2

• • • •

Potential anomalous data, in particular acceleration values, as a product of "coverage spoofing" (i.e.: 2012 Hyundai Accent test) Roll angle data shorter than anticipated as "missing/ invalid data" (i.e.: 2013 Hyundai Tucson test) Steering input inconsistencies (i.e.: 2013 Kia Rio small overlap test) Acceleration data reliability (and usefulness) generally as a function of unknown raw data processing (i.e.: 2013 Kia Rio side impact test)

For example, lacking detailed information on the application, filtering process, anticipated accuracy, range, and resolution it is difficult to say whether or not the data, taken in total, may be used reliably to "… ensure that EDRs record, in a readily usable manner, data valuable for effective crash investigations and for analysis of safety equipment performance (e.g., advanced restraint systems). (where) These data will help provide a better understanding of the circumstances in which crashes and injuries occur and will lead to safer vehicle designs."2 Where the reported acceleration data is offered without other explanation or information, it is difficult to discern its ultimate usefulness.

The small amount of roll angle data observed may be tied to the minimum 300ms requirement for other dynamic data as called for in 49CFR563 and, for that reason, it may be “missing” rather than “invalid;” again a recording limitation which should reasonably be addressed in a fully developed “Data Limitations.”

Of course, comparing instrumentation based data from a crash test naturally provides an objective standard by which to evaluate the reliability of the data in the context of 49CFR56 Table III with respect to range, accuracy and resolution to that recorded in the ACU of Hyundai and Kia vehicles.

R

eal World Crashes The “real world” examples serve to illustrate that, in a gross sense, the Hyundai-Kia EDR Tool(s) reported data is generally reliable, within limits, as discussed within that section of the Compendium. Each of the selected cases highlights a different concept or approach to evaluating or using the data in a crash analysis.

www.collisionpublishing.com

One underlying theme; however, is that, as the last case example in that section illustrates, end users of the Hyundai and Kia EDR Tool(s) may find themselves in a situation where the retention of a case vehicle or at least the ACU might play a pivotal role in the end user's ability to provide a report generated in the most current version of the corresponding appropriate software. As also discussed earlier in this Conclusions section of the Compendium, an ACU – by itself – is almost impossible to objectively associate with a specific vehicle absent significant effort on the part of the investigator.

Another issue raised in previous sections of this Compendium which bear repeating here are the potential ramifications of “coverage spoofing.”

C

overage Spoofing The authors believe it is important here to remind the reader that coverage spoofing is not recommend.

A single report obtained by way of spoofing might, on its face, appear to yield reasonable data; however, end users would normally have no way of knowing whether or not the translation was actually done properly for the unsupported module (vehicle) and should not rely on data retrieved by way of this practice.

Based on the uncertain nature of coverage spoofing, the authors recommend against its application as, more often than not, coverage spoofing has been shown to produce nominal to dramatic errors if not “inconsistencies” in the resultant data translation and even producing errors or anomalies which might not be readily apparent without other comparisons being possible.

The general objection of the authors to coverage spoofing is not a condemnation of the data contained within Hyundai and Kia vehicle ACUs prior to model year 2013, but rather a warning against a process wherein the potential for improper translation is considerable and reinforces the recently added warning against the practice as found in the Data Limitations for Hyundai and Kia vehicle EDR reports.

The previously published “Hyundai and Kia Crash Data, a Preliminary Overview”1 includes a discussion of what was then identified as “VIN spoofing” or “coverage spoofing” as described in this Compendium where the authors wrote that “…That ‘success’ is, at this writing, limited to the notion that ‘data has been retrieved’ not that it has been successfully translated accurately…” and “… it is clear that at least some of the crash pulse data, for example, is either (1) being translated incorrectly or (2) not supported in the format anticipated by the Hyundai are Kia software for model year 2013 and, therefore, the output is of questionable accuracy and/or reliability.

In that regard, it will be useful later to identify characteristics of “VIN spoofing” (or its equivalent as it’s related to Hyundai and Kia vehicles) as a function of either testing the reliability of the systems or in an effort to identify where an individual may have used “spoofing” to access data not intended for retrieval or translation by the Hyundai or Kia EDR Tools. Lacking a clear VIN-to-module-todata/report association, this may be an important attribute or benefit of such ongoing research by the authors not to mention its potential application in litigation. …” As of this writing, efforts by the authors are ongoing in an effort to be able to identify where a report is a product of what we have come to call, in this narrative, “coverage spoofing.”

As noted earlier in this Compendium, coverage spoofing generally falls into 4 general categories as it relates to the Hyundai-Kia EDR Tool(s). These are: 1. Selecting a supported make and model but an unsupported year (i.e.: an unsupported 2012 vehicle accessed by selecting that vehicle for model year 2013) 2. Selecting the wrong model without respect to year (i.e.: selecting a vehicle, whether or not for a year supported, other than the one the Tool is actually connected to) 3. Selecting a vehicle from the "other" manufacturer's list (i.e.: attempting to download data from a Kia using the Hyundai software and selecting a Hyundai model vehicle whether or not from a supported year) 4. Attempting to retrieve data from an unsupported module by the direct-to-module method using a cable and making certain modifications The initial coverage spoofing experiments reported in Volume 8, Issue 1 of Collision went to press in May 2013.

In October 2013, Hyundai released software version “EN-H-01-00-0004.” With the release of that later software version, the tool supplier (GIT America) added an “EDR Disclosure,” wherein the user is warned against the utilization of the Hyundai EDR Tool to access vehicles produced prior to September 1st, 2012. The Hyundai EDR software goes on to warn that: “…there is no tool for downloading accurate data from the EDRs in Hyundai vehicles produced prior to this date….”

With the release of version “E-N-H-01-00-0004,” an identical warning was added to the Hyundai EDR Data Limitations: “… Tools for downloading the EDRs in Hyundai vehicles have been developed only for vehicles produced after September 1, 2012. Currently, there is no tool for downloading accurate data

www.collisionpublishing.com

Collision Magazine - Volume 8 Issue 2 115

from the EDRs in Hyundai vehicles produced prior to this.”

On October 31st, 2013, Kia released software version “EN-K-01-00-0002.” With this release, Kia also added an EDR Disclosure, equivalent to the warnings contained in the Hyundai Software.

While the authors largely agree with the spirit and intent of the “EDR Disclosure” (essentially, it says: do not utilize the Hyundai and Kia EDR Tools for vehicles prior to model year 2013), a strict literal interpretation cannot be reasonably possible. The authors do not dispute that the Hyundai and Kia EDR Tool(s) were developed for use only with vehicles produced after September 1st, 2012, and, the authors will agree that the application of the Hyundai and Kia EDR tools on vehicles prior to model year 2013 may produce errors in translation.

However, the suggestion that “no tool” exists for the interpretation of “accurate data” is doubtful in a literal sense.

As the authors previously explained, there is no evidence that there was no data contained within Hyundai and Kia ACUs prior to model year 2013 or that, if translated properly, it would be inaccurate. In fact, the opposite is far more likely to be true. The authors can only confirm the likelihood of inaccurate translations when the current release Hyundai and Kia EDR Tool(s) are used to retrieve data from vehicles prior to model year 2013. Furthermore, one might speculate, given, for example cases such as Padilla v Toyota5, that OEs maintain some form of tool for the proper translation of the data contained within their ACUs prior to model year 2013.

A more appropriate warning, true in its literal sense, might be along the lines of: “… Tools for downloading the EDRs in Hyundai vehicles have been developed specifically for vehicles produced after September 1, 2012. The use of this commercially available tool for retrieving data from vehicles manufactured prior to September 1, 2012 may result in inaccurate data translations.”

Again using a comparison to the Bosch CDR Tool, the end users of the information found in this Compendium are urged to use the latest production release of the GIT EDR Tool software appropriately as indicated in the instructions which come with the relevant system when retrieving data from Hyundai or Kia vehicles. As of this writing and publication of this volume of Collision, as it relates to “coverage spoofing,” the official Hyundai and Kia Tool coverage was available for the vehicles found in Table 1.1.

116 Collision Magazine - Volume 8 Issue 2

S

ummary This Compendium, as a first-of-its-kind look at the recently available Hyundai and Kia EDR Tool(s) should be taken together with the information contained in the “Preliminary Review ” as current at the time of publication and as the basis for ongoing future research into this sort of crash data and retrieved data reporting.

Without a doubt there will be changes in the various system(s) over time which may affect some of the opinions and conclusions offered here; however, as noted with respect to the inability to “retest” retrieved data, some issues may never be resolved. The authors suggest that HyundaiKia EDR Tool(s) report data be contemplated with appropriate caution in litigation and, to some extent in research into collision safety.

R

eferences 1.

Collision: The International Compendium for Crash Research, Volume 8, Issue 1 2. "Event Data Recorders" 49 CFR 563. 2012.

3. 74144 Federal Register, Vol. 77, No. 240, Thursday, December 13, 2012, Proposed Rules; National Highway Traffic Safety Administration, 49 CFR Part 571, [Docket No. NHTSA–2012–0177], RIN 2127–AK86, Federal Motor Vehicle Safety Standards; Event Data Recorders 4. Tucker, Robert L. "The Flexible Doctrine of Spoliation of Evidence; Cause of Action, Defense, Evidentiary Presumption and Discovery Sanction." University of Toledo Law Review 27 (1995): 67-68.

5. Bachman v GM, NO. 4-01-0237, Appellate Court, Illinois 4th District 6. Padilla v Toyota, U.S.D.C. District of New Jersey, Law Division, Civil Action No 04-cv-3422 7. LaBorde, et al, v. Shelter Mutual Insurance, et al, (2011) Third Circuit Court of Appeal, CW11-00956 8. American Society for Testing and Materials (ASTM): Standard Practice for Collection and Preservation of Information and Physical Items by a Technical Investigator (E 1188 – 05) 9. American Society for Testing and Materials (ASTM): Standard Practice for Examining and Preparing Items That Are or May Become Involved In Criminal or Civil Litigation (E 860 – 07) 10. American Society for Testing and Materials (ASTM): Standard Practice for Collection of Non-Volatile Memory Data in Evidentiary Vehicle Electronic Control Units (E2493 – 07) 11. Daubert v. Merrill Dow Pharmaceuticals, Inc., 509 U.S. 579 (1993) 12. Coffey v. Dowling Mfg., Inc., 187 F. Supp.2d 958 (M.D. Tenn. 2002), aff’d, 89 Fed. Appx. 927 (6th Cir. 2003).

13. William J. Melkonian. "CDR Evidence Frequently Asked Questions" Collision: The International Compendium for Crash Research. Volume 4. Issue 1.

14. Although some recent model Chrysler line vehicles, for example, include an odometer reading captured at the time of the event and other systems may capture the equivalent of “engine hours” as lifetime or current ignition cycle operating times.

15. AZ v O'Brien, Superior Court of Arizona, Maricopa County, CR 2003-016197-001 DT. (2004)

www.collisionpublishing.com

www.collisionpublishing.com

Collision Magazine - Volume 8 Issue 2 117

Bombardier Can–Am Spyder

Braking And Acceleration Testing Detective Constable Dawn Mutis #1165 Toronto Police Service Collision Reconstruction Squad

T

he Can - Am Spyder A new vehicle, which is beginning to become popular in our city, is the Can-Am Spyder, manufactured by Bombardier Recreational Products (BRP). At the time of this project, The Toronto Police Service Collision Reconstruction Squad had not investigated any collisions involving a Spyder. The Squad thought it would be wise to gather some data in regards to the vehicle so there would be some familiarity should there be a collision involving one.

hicle, which is exactly what the Spyder pretty much is.

Design work started in the first years of the decade and advanced prototypes were being tested extensively by 2004. The new Spyder 'roadster' was unveiled in February 2007, to be marketed under the Can-Am brand. Originally created by Bombardier in the early '70s for a successful range of motocross, enduro and dual-purpose motorcycles, the name Can-Am was revived for BRP's line of ATVs. The first production Spyder rolled off the Canadian asWith the assistance of Mr. Ben Daponte, Dis- sembly line in Valcourt, Quebec, on Septemtrict Sales Manager for Canada, the Squad was ber 14, 2007. able to do some testing with two models of he Manufacturing Process Spyders.

On the highway the Spyder ackground occupies approximately twoThe Spyder concept began life thirds of the traffic lane. The as a snowmobile you could front wheels have a tendency to follow the ride year-round in northern road’s camber. climates, with a wheel in back instead of a track, but everybody had so much fun riding The Spyder is built around a Surrounding the thing around it occurred to BRP it might Spar Technology (SST™) frame, featuring a be able to create an entirely new class of ve- minimalist steel center beam, surrounding the engine. It minimizes weld points for greater

B

118 Collision Magazine - Volume 8 Issue 2

T

www.collisionpublishing.com

structural integrity. This light and slim frame provides a low center of gravity to enhance stability and handling. The Y-shaped frame extends back to a steel double-sided swing arm located underneath its 6.6-gallon fuel cell.

roll-bar. There is an electronically controlled power steering system which also provides variable assist.

The wheels are a pair of 14x5-inch aluminums on the front and a 15x7-inch aluminum The engine is a Can-Am Spyder Roadster Rotax® rim in the rear. The tires are made by Kenda 998cc V-twin engine. This 60-degree V-twin is (165/65R14 front, 225/50R15 rear). counterbalanced, and runs on regular gas. The lectronic Vehicle Stability Sysdrive belt to the rear is carbon-reinforced.The tem specs indicate that it can accelerate from 0-60 Acceleration results from wheels mph (0-100 km/h) - in 4.5 seconds. being acted on by forces created The suspension is comprised of an automotive- from braking, acceleration and/or steering. derived double-A arm set-up with an integrated Friction is generated at the point of the tire

E

www.collisionpublishing.com

Collision Magazine - Volume 8 Issue 2 119

Figure 1: The Can-Am Spyder

Figure 4: The Spyder carbon-reinforced rear drive belt

Figure 2: The Spyder Surrounding Spar Technology (SST™) frame

contacting the ground. Each tire has a limited amount of friction force than can be generated. If the limit is exceeded, traction is lost.

A vehicle’s acceleration can also effect wheel lift and roll over. Each vehicle has a centre of mass and rollover axes.

Each rollover axis is defined by a line connecting the contact patches of adjacent tires on each side of the vehicle.

Rollover is likely to occur when the centre of gravity passes over a rollover axis. This can occur when a vehicle’s lateral acceleration is high enough to lift a tire off the ground.

Graph of Tire Grip and Wheel Lift Thresholds from U.S. Patent US 8,086,382 B2 (based on practical testing done by Bombardier Recreational Products) 210 - tire grip threshold 216 - wheel lift threshold

Figure 3: The Spyder Y-shaped frame

220 - point in which a precursory condition indicative of wheel lift exists 226 and 228 - point in which a precursory condition indicative of exceeding the tire grip threshold exists

120 Collision Magazine - Volume 8 Issue 2

www.collisionpublishing.com

Figure 5: Graph of Tire Grip and Wheel Lift Thresholds from U.S. Patent US 8,086,382 B2 (based on practical testing done by Bombardier Recreational Products)

The letters are indicative of the EVSS changing the longitudinal and or lateral acceleration to bring the vehicle back to a stable ride.

The wheel lift threshold crosses and is below the tire grip threshold in many combinations of lateral and longitudinal accelerations. This means the vehicle can have tire lift and rollover without the tire first losing grip.

The Electronic Vehicle Stability System (EVSS) was developed in partnership with Bosch. It integrates ABS, traction control, and stability control systems.

The stability control assesses your intended direction via the handlebar and determines the vehicle’s appropriate response. The system individually brakes the wheels and/or reduces excess engine torque to keep control.

A

nti-Lock Braking System Braking is handled by three rotors, controlled by twin 4-piston callipers in front and a single-piston calliper at the back. The entire system is linked together hydraulically with an electronic anti-lock system (ABS). It is actuated via a righthand side foot lever. There is also an electronically activated parking brake.

The brake pedal under the right foot engages all three brakes simultaneously as four-piston callipers squeeze into the discs up front and a single-piston, sliding pins calliper does the same out back.

Sensors monitor the rotation of all three wheels. If any of the wheels is at risk of locking, ABS reduces brake pressure on it. This intervention repeats in rapid succession and can be performed independently on each wheel, allowing for maximum braking performance.

www.collisionpublishing.com

Collision Magazine - Volume 8 Issue 2 121

Figure 6: The Spyder brake system

Figure 7: The Spyder Dynamic Power Steering System

T

raction Control System Whenever the Traction Control System detects excessive rear wheel spin (e.g., in slippery conditions), it first reduces engine power, and if that isn’t sufficient, it will apply the brake to any or all of the three wheels until the speed value returns within range. The system is always on whenever the ignition switch is turned on, and, unlike most cars it cannot be manually disabled.

D

ynamic Power Steering The electronically controlled power steering system provides the rider with variable assist when turning the handlebars. The amount of assistance varies with the speed of the vehicle to adjust the steering input effort to a comfortable level.

Steering is by handlebar with a linkage to the front wheels, and adjustable rods for alignment.

Figure 8: The Spyder RT (Touring)

S

emi-Automatic Transmission (optional) The Spyder has an electronic 5-speed gearbox. A paddle shifter is mounted on the left handlebar where a push of the thumb runs up the gears while a flick of the forefinger brings it back down.

The transmission will automatically shift down for you if a rider doesn’t. It allows riders to stay in first gear at a stoplight without clutching in. It has a reverse gear as well.

R

for the rider.

T vs RS RT (Touring)

The touring model has an upright riding position, and extensive fairing protection

122 Collision Magazine - Volume 8 Issue 2

www.collisionpublishing.com

Figure 9: The Spyder RT (Touring) heated passenger grab handles

Figure 10: The Spyder RS (Sport) The engine is calibrated for touring and has the necessary torque and performance. Electronic fuel injection ensures instantaneous throttle response and improved fuel efficiency. The Rotax engine puts out 80 lb.-ft. of torque at just 5,000 rpm.

There are heated passenger grab handles on the RT. There is also an electronically adjustable windscreen.

Upgrades for the Spyder RT received last year included improved suspension, enhanced ventilation and other ergonomic refinements. For 2012, the Spyder RT roadster features a new Dot-matrix display color for improved visual clarity on night excursions.

RS (sport) Gas-charged, anodized-aluminum FOX Racing Shox front shocks on the Spyder RS-S models are adjustable for preload and have improved compression and rebound damping. These shocks provide a 33% weight reduction per shock.

T

he Future - A Hybrid Can-Am Spyder Can-Am is working on the project with Sherbrooke University in Quebec with funding provided by the government of Canada. They have goals of reducing emissions by 50% and increasing efficiency by the same margin without sacrificing power or performance. This will be achieved with a smaller, 600cc Rotax engine mated to a 20 kW electric motor powered by a lithium-ion battery. The new power source will offer a major improvement in fuel efficiency compared to the current iteration of Spyder (Can-Am reps state the Spyder gets an average of 27-32 mpg, depending on riding style, and up to 35 mpg in special cases).

2

012 Can-Am Spyder RS - Canadian Specifications/Technical Details Canada, MSRP: $ 18,499 CDN

Engine • • • • • • • • • •

BRP-Rotax ® Manufacturer Type Rotax 998cc V-twin Bore and stroke 3.82 inches x 2.68 inches (97 mm x 68 mm) Throttle body injection system multi-point EFI with 2.24 inches (57 mm) Power 106 hp (79 kW) @ 8500 RPM 77 lb-ft of torque (104 Nm) at 6250 rpm 5-speed manual transmission (SM5) with reverse or 5-speed semi-automatic (SE5) with reverse Magnet 500 watts Electric Start 12 volt / 21 amp hour, sealed maintenance free

Chassis • • • • • • • • • • • •

Spyder stainless steel frame Front Suspension Dual A-arm with stabilizer bar Front suspension travel 5.7 inches (145 mm) Swingarm with monoshock rear suspension Rear suspension travel of 5.7 inches (145 mm) Type hydraulic brake actuated foot-3-wheel braking system 4-piston front brake calipers with (250 mm x 6 mm) discs Rear brake calipers single piston pins (250 mm x 6 mm) disc Electronic brake force distribution EBD Hand, foot brake active, committed rear caliper Front tires 165/65 R14 Rear tire 225/50 R15

www.collisionpublishing.com

Collision Magazine - Volume 8 Issue 2 123

• •

Alloy 3-spoke front metallic silver, 14 x 5 inches (355 mm x 127 mm) Rear wheel aluminum metallic silver, 15 inches x 7 inches (381 mm x 178 mm)

Dimensions • • • • • • • •

L x W x H 105 x 59.3 x 45.1 inches (2667 x 1506 x 1145 mm) 68 cm Wheelbase (mm 1727) Seat height 29 inches (737 mm) Ground clearance 4.5 inches (115 mm) Dry Weight 699 lbs (317 kg) Before storage capacity 12 U.S. gal. (44 L) The vehicle maximum load of 440 pounds (200 kg) Fuel Capacity 6.6 U.S. gallons. (25 L)

2

012 Can-Am Spyder RT - Canadian Specifications/Technical Details Canada MSRP Price: $24,399 CDN

Engine • • • • • • • • • •

Manufacturer BRP-Rotax® Type 998cc Rotax V-twin engine calibrated for touring Bore & stroke 3.82 in. x 2.68 in. (97 mm x 68 mm) Injection Multi-point EFI system with 2 in. (51 mm) throttle body Power 100 hp (71 KW) @ 7500 RPM Torque 80 lb.-ft. (108 Nm) @ 5000 RPM Transmissions 5-speed manual (SM5) with reverse Magneto 650 watts Starter Electric Battery 12 volts / 21 amps/hour, sealed maintenancefree

Chassis • • • • • • • • •

Frame SST Spyder Front suspension Double A-arm with anti-roll bar Front suspension travel 5.9 in. (151 mm) Rear suspension Swing-arm with pneumatic adjustable preload Rear suspension travel 5.7 in. (145 mm) Brake: Foot-activated, hydraulic 3-wheel brakingFront brakes 4-piston calipers with (250 mm x 6 mm) discs Rear brake Single-piston sliding pins caliper with (250 mm x 6 mm) disc EBD Electronic Brake Distribution Parking brake Electro-mechanical

124 Collision Magazine - Volume 8 Issue 2

• • • •

Front tires 165 / 65 R14 Rear tire 225 / 50 R15 Front wheels 6-spoke Aluminum Metallic Silver 14 x 5 in. (355 mm x 127 mm) Rear wheel Aluminum Metallic Silver 15 x 7 in. (381 mm x 178 mm)

Dimensions • • • • • • • •

L x W x H 105 x 61.9 x 58 in. (2,667 x 1,572 x 1,473 mm) Wheelbase 67.2 in. (1,708 mm) Seat height 30.4 in. (772 mm) Ground clearance 4.5 in. (115 mm) Dry weight 929 lb. (421 kg) Total storage capacity 41 US gal. (155 L) Maximum vehicle load 525 lb. (240 kg) Fuel capacity 6.6 US gal. (25 L)

T

esting Testing was conducted at 207 New Toronto Street, in Toronto Ontario. This was a city owned asphalt lot which was used to store snow that was removed from city streets in the winter.

Tests were conducted at the scene to determine the coefficient of friction, or the drag factor of the lot. A drag factor is a numerical value, which represents the slipperiness between two surfaces that are in contact with each other, namely tires and pavement. This numerical value was obtained by using the force required to move an object, divided by the weight of the object being moved.

From test data collected for the east side of the lot in a southbound direction, the co-efficient of friction or “drag factor” was calculated to be 0.74.

From test data collected for the east side of the lot in a westbound direction, the co-efficient of friction or “drag factor” was calculated to be 0.73.

Vericom Computers manufactures and sells on board vehicle dynamometers. The VC4000 is an electronic accelerometer computer that provides vehicle brake testing and acceleration or performance testing. It measures the G Force (acceleration factor, drag factor) 100 times per second. The VC4000 accurately measures average G Force and precisely calculates speed, time and distance for any type of vehicle.

The friction coefficient of the roadway surface was determined with the use of a Vericom VC4000, and the two Spyders by conducting braking tests.

www.collisionpublishing.com

Run # 1 2 3 4 5 6

Time 1.55 1.7 2.06 1.65 1.63 1.83

Average Accel -0.954 -0.953 -0.922 -0.882 -0.88 -0.872

Speed (kph) 52.218 57.182 67.035 51.361 50.668 56.348

Distance (Meters) 11.422 13.925 19.492 13.003 12.202 15.913

Pk Gx -1.215 -1.193 -1.157 -1.136 -1.131 -1.146

Pk Gx Time 0.58 0.53 1.24 0.8 0.94 1.83

Table 1: Braking on dry asphalt. Vehicle: RT Date: 2012.06.25 Time: 09:43 Run # 1 2 3

Time 2.96 3.01 2.94

Average Accel 0.479 0.471 0.484

Speed (kph) 50.075 50.057 50.221

Distance (Meters) 19.801 20.25 19.887

Average HP SL 20.222 19.872 20.486

Pk Gx 0.642 0.608 0.598

Pk Gx Time 2.75 2.94 2.78

Table 2: Acceleration on dry asphalt. Vehicle: RT Date: 2012.06.25 Time: 09:54

Figure 11: Impending skid marks visible on asphalt

Figure 12: Acceleration of the RT from a stop

www.collisionpublishing.com

Collision Magazine - Volume 8 Issue 2 125

Run # 1 2 3 4

Time 1.68 2.09 2.07 1.8

Average Accel -0.929 -0.778 -0.906 -0.941

Speed (kph) 55.122 57.422 66.212 59.797

Distance (Meters) 12.829 17.639 19.048 15.09

Pk Gx -1.367 -1.161 -1.274 -1.285

Pk Gx Time 0.46 0.82 0.04 0.48

Table 3: Braking on dry asphalt. Vehicle: RS Date: 2012.06.25 Time: 10:06 Run # 1 2 3 4 5 6 7 8 9 10 11 12

Time 2.49 2.51 2.74 2.37 2.54 5.21 5.22 4.73 5.02 4.79 5.11 5.13

Average Accel 0.541 0.541 0.482 0.58 0.542 0.514 0.527 0.577 0.534 0.563 0.528 0.528

Speed (kph) 47.565 47.977 46.645 48.509 48.56 94.463 97.053 96.391 94.717 95.196 95.203 95.681

Distance (Meters) 15.075 15.233 17.424 14.737 15.638 67.708 69.944 64.342 65.465 61.777 71.269 67.939

Average HP SL 21.764 21.914 18.979 23.791 22.153 40.938 43.075 47.156 42.609 45.254 42.713 42.516

Pk Gx 0.79 0.787 0.781 0.784 0.83 0.925 1.077 1.085 0.814 0.823 0.954 0.823

Pk Gx Time 2.01 0.82 2.02 2.37 0.75 0.73 4.33 4.73 4.67 4.77 5.1 3.5

Table 4: Acceleration on dry asphalt. Vehicle: RS Date: 2012.06.25 Time: 10:11

Figure 13: Braking on dry asphalt

126 Collision Magazine - Volume 8 Issue 2

Figure 14: Acceleration of the RS from a stop www.collisionpublishing.com

Run # 1 2 3

Time 1.96 2.07 2.21

Average Accel -0.882 -0.907 -0.831

Speed (kph) 60.999 66.296 64.861

Distance (Meters) 18.018 19.788 21.635

Pk Gx -1.36 -1.615 -1.363

Pk Gx Time 0.84 0.39 1.14

Table 5: 3 wheel lock (achieved by turning off the key and braking).

Vehicle: RS Date: 2012.06.25 Time: 10:57 Run # 1 2 3

Time 2.41 2.15 1.94

Average Accel -0.901 -0.814 -0.928

Speed (kph) 76.633 61.798 63.531

Distance (Meters) 25.683 19.295 16.912

Pk Gx -2.096 -1.134 -1.451

Pk Gx Time 0.16 0.29 0.09

Table 6: Braking on wet asphalt.Vehicle: RS Date: 2012.06.25 Time: 11:07

Figure 15: Skid marks visible on the asphalt surface. www.collisionpublishing.com

Figure 16: Skid patch and stippling visible on the tires.

Collision Magazine - Volume 8 Issue 2 127

Run # 1 2

Time 3.01 2.97

Average Accel 0.393 0.415

Speed (kph) Distance (Meters) Average HP SL 41.744 16.194 13.829 43.508 16.84 15.342

Pk Gx 0.592 0.763

Pk Gx Time 0.44 2.97

Table 7: Acceleration on wet asphalt. Vehicle: RS Date: 2012.06.25 Time: 9:54 Run # 1 2 3 4

Time 1.85 2 2.06 2.28

Average Accel -0.879 -0.945 -0.942 -0.924

Speed (kph) 57.392 66.704 68.496 74.38

Distance (Meters) 15.02 18.665 19.905 23.737

Pk Gx -1.255 -1.161 -1.189 -1.151

Pk Gx Time 0.16 1.48 1.65 1.4

Table 8: Braking on wet asphalt.Vehicle: RT Date: 2012.06.25 Time: 11:43

Figure 17: Braking was very efficient on a wet surface, which this officer was not expecting

Figure 18: Squeegee mark visible 128 Collision Magazine - Volume 8 Issue 2

Figure 19: Braking the RT on wet asphalt www.collisionpublishing.com

R

eferences 1. Ben Daponte, District Sales Manager, Bombardier Recreational Products, Can-Am Division

2. 2012 Can-Am Spyder Can-Am.brp.com

8. U.S. Patent US8,086,382B2 Three-Wheel Vehicle Electronic Stability System Bombardier Recreational Products Inc. 2011 9. Snow City Cycle Acknowledgements

3. 2012 Can-Am Spyder Totalmotorcycle.com 4. Can-Am Spyder Review John Burns 2011 12 22, Cycleworld.com

Assistance of D. C. J. Bassingthwaite # 5297, D.C. A. Putman #9965, D.C. R. Lanigan #3732, D.C. M. Chin #8077 of Traffic Services, Toronto Police Service

5. 2012 Can-Am Spyder Review Marc Lechapelle MSNAutos.com

Can-Am Spyder RT Quebec licence 3501S, operated by Detective Constable Jeff Bassingthwaite # 5297.

6. 2012 Can-Am Spyder Review Bryan Harley 2011 12 16, Motorcycleusa.com

Can-Am Spyder RS Quebec licence 86634A, operated by Detective Constable Jeff Bassingthwaite # 5297

7. Can-Am Spyder About.com Basem Wadef

www.collisionpublishing.com

Collision Magazine - Volume 8 Issue 2 129

www.

arccsi .com

Don’t Miss the Crash Community Event of the Year!

www.CrashTeamBootCamp.com

nal O i t p O ng i n i a tr

Crash Team

Boot Camp Experience

For the past 12 years, the annual ARC-CSI Crash Conference has been the gold standard for excellence in live, fully instrumented practical crash testing. ARC-CSI continually delivers the widest variety of exceptional speakers and unique presentation topics designed to expand your knowledge of collision reconstruction, investigation and traffic safety.

The ARC-CSI Crash Conference sets the bar for valuable insight into how various crash scenarios occur and has become the basis for a wealth of information on vehicle systems and operations.

���� O�� ����� �O �A�� A�O�� ��A� �A����� ���� ��A���� ����� A��� �O ��O� eXAcT�y ��A� �A�����D

This is a unique opportunity to experience, hands-on, what goes into setting up and conducting crash tests. Boot Camp is limited to about 8-10 people and is offered as a first-come, first-served basis.

Nothing compares to an opportunity to see live, full scale crash tests and then, hours later, get your hands on actual data, video, photos and other documentation to take back and improve your crash investigation/reconstruction job performance.

3 Classroom & hands-on-instruction 3 ARC-CSI Conference registration 3 Conduct your own “team” crash test 3 Present & publish your findings 3 and much more!

June 2-5, 2014 Las Vegas, NV www.arccsi.com 800.280.7940

I^bZ

2 &#(.&' hZX id ^beVXi 7BL KZadX^in 2 .(#'**, [eh HX]dda 7jh KZadX^in 2 )*#')&' [eh

I^bZ

2 %#')*. hZX edhi ^beVXi 7BL KZadX^in 2 '(#%-.- [eh HX]dda 7jh KZadX^in 2 (-#()&' [eh

;^cVaan l^i] K^hjVa HiViZbZciÉh ;M( VcY ^ih VhhdX^ViZY egd\gVbh! i]ZgZ ^h cdl V dcZ hide hdjgXZ [dg egd\gVbh igjan YZh^\cZY [dg i]Z 6XX^YZci >ckZhi^\Vidg VcY GZXdchigjXi^dc^hi# I]Z ;M( 869 egd\gVb Vaadlh [dg Z[[dgiaZhh YdlcadVYh d[ IdiVa HiVi^dc YViV l]^aZ egdk^Y^c\ V h^beaZ id deZgViZ 869 [dgbVi [dg egdYjX^c\ hXVaZ Y^V\gVbh d[ k^gijVaan Vcn hXZcZ ndj XVc ^bV\^cZ# 6cY l^i] kZgn a^iiaZ egVXi^XZ! Vcn gZXdchigjXi^dc XVc WZ Wgdj\]i id a^[Z l^i] Vc VXXjgViZ (9 6c^bVi^dc# 6h V iZhi^[n^c\ ZmeZgi! > ]VkZ XdcYjXiZY i]djhVcYh d[ gZXdchigjXi^dch VcY ]VkZ iZhi^ÒZY ]jcYgZYh d[ i^bZh# L^i] i]^h `^cY d[ WVX`\gdjcY! > `cdl l]Vi XVc VcY XVcÉi WZ gZa^ZY dc [dg ig^Va# L]Zi]Zg ndjg XVhZ ^h Xg^b^cVa dg X^k^a ^c cVijgZ! ndj XVc gZan dc K^hjVa HiViZbZciÉh ;M( id egdk^YZ ZkZgni]^c\ ndj cZZY id YZe^Xi i]Z Zk^YZcXZ VcY hjeedgi ndjg de^c^dc ^c V bdhi VXXjgViZ VcY egd[Zhh^dcVa bVccZg#

Curtis �. Fl�nn � AC�AR ��2� � Fl�nn � Associates� �nc. � �l �aso� �e�as

Real�otion 3D Features� 6XXjgVXn id &$&%%%i] hZXdcY

EgZX^h^dc iZggV^c bdYZah [gdb ndjg bZVhjgZbZ

7j^ai ^c e]nh^Xh Zc\^cZ Xdcigdah i^bZ Y^hiVcXZ kZ 8djgi VXXZeiZY ZkZgnl]ZgZ >cXgZY^Wan ZVhn id aZVgc

I^bZ 2 &#(.&' hZXdcYh id ^b 7BL KZadX^in 2 .(#'**, [eh 7BL 9^hiVcXZ 2 &')#-.%& [i# id ^beVX HX]dda 7jh KZadX^in 2 )*#')&' [eh

I^bZ

2 %#+&,+ hZX edhi ^beVXi 7BL KZadX^in 2 &(#*+() [eh HX]dda 7jh KZadX^in 2 &,#'(.' [eh

I^bZ

2 '#&(+, hZX edh 7BL KZadX^in 2 % HX]dda 7jh KZadX^in 2 %

Eaj\ ^c ndjg gZXdchigjXi^dc gZhjaih!

VcY ]^i EA6N# DkZg ((!%%% jhZgh VgdjcY ldgaY gZedgi i]Vi Vc^bVi^dch VgZ YdcZ ^c b^cjiZh#

�ow else can �ou e�plain a comple� event that happens in less than 3 seconds�

cA�� uS For A �I�e demo And We WI�� AnImATe your crASh �����������3�3

lll#k^hjVahiViZbZci#Xdb &#---#-'

Subscribe Today and Save!

2012 ARC-CSI DATA DVD INCLUDED

C

The International Compendium for Crash Research

Volume 7 Issue 2

FALL 2012

Collision: The International Compendium for Crash Research

An Analysis of

RollOver Data dynamics of vehicle to vehicle side impact crash tests

Volume 7, Issue 2 - Fall 2012

Reconstruction Essentials for Tread Separation Accidents Involving Axle Tramp

A Comparison between a Real World Crash Test, HVE Simulation and 3D Scanning

Analysis of the GM Sensing & Diagnostic Module in 360 Degree Linear Momentum Collisions: Real Case Analysis CDR Data Presentation & Validation in Legal Proceedings

Use our m-FX Pocket PC data collector with any total station – see your points recorded in 3D.

Crash Data Use AccelereX on the same Pocket PC to measure deceleration ( drag factor ).

LEA 4130R3: Enhanced hardware reliability for increased productivity. Make accurate, long-range,

Event Calendar prismless measurements through obstacles, and enjoy the large keyboard for easy entry of point names.

Incorporating CDR Services in an Existing Forensic Practice Organization News

For more information on Sokkia’s LEA product line or to find a dealer near you, contact customer service at

800.4.SOKKIA, or visit www.sokkia.com.

Motorcycle Collinear Collisions Involving Motor Vehicles Equipped with EDRs

LEA

Includes CD

Use hawkeye our photogrammetry program to take accurate measurements with your camera.

Law Enforcement Applications

800.4.SOKKIA

AccelereX

m-FX

Accelerometer

Clear the Road in minutes!

DataCollector

To your deSkTop on your pc – We can help you

Total Station download

Motorcycle

Accident Investigation

Use FX3 Reconstruction software to download your measurements points and auto plot your scene diagram.

Use FX3 to generate a powerful 3D scene complete with textured surfaces and any models you need.

Volume 2 Issue 1

To The courT room

In the court room Analyzing the Collision - We can help you

 

  







        



                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                             

“ 2 in - 3 out ” Use our FX3 realmotion 3d to instantly animate your reconstruction results on your 3D scene.

  



Use our SmAc FX to simulate or predict collision results.

Present powerful time distance reports and persuasive graphical output.

Overwhelm them!

All FX3 output is detailed and court acceptable.

The International Compendium for Crash Research

of EDR Information

2007 CDR User’s Conference CD

Analysis of Collisions:

Conservation of Linear Momentum

Presentations & Data

The Accuracy of GM EDRs in NHTSA Crash Tests

Reconstruction & Analysis of Steering-induced, On-road, Untripped SUV Rollover

Use our crash FX momentum tool right on the diagram to generate impact speeds.

Use our crashmath application with 720+ reconstruction formulae to get all the answers.

2006 CDR User’s Conference

www.CollisionPublishing.com

Untripped

Rollovers

Use our pocket crashmath – to do preliminary

calculations of impact speeds etc.

From 2006 ARC-CSI Conference

C ollision Two-Dimensional Analysis

1.888.828.0383

www.sokkia.com

Plus . . .

For a full reconstruction of this 46 second chase sequence, visit www.edccorp.com/bullitt

HVE directly answers the following questions using Newton’s laws of motion:

Speed on Taylor St.

;^cVaan l^i] K^hjVa HiViZbZciÉh ;M( VcY ^ih VhhdX^ViZY egd\gVbh! i]ZgZ ^h cdl V dcZ hide hdjgXZ [dg egd\gVbh igjan YZh^\cZY [dg i]Z 6XX^YZci >ckZhi^\Vidg VcY GZXdchigjXi^dc^hi# I]Z ;M( 869 egd\gVb Vaadlh [dg Z[[dgiaZhh YdlcadVYh d[ IdiVa HiVi^dc YViV l]^aZ egdk^Y^c\ V h^beaZ id deZgViZ 869 [dgbVi [dg egdYjX^c\ hXVaZ Y^V\gVbh d[ k^gijVaan Vcn hXZcZ ndj XVc ^bV\^cZ# 6cY l^i] kZgn a^iiaZ egVXi^XZ! Vcn gZXdchigjXi^dc XVc WZ Wgdj\]i id a^[Z l^i] Vc VXXjgViZ (9 6c^bVi^dc# 6h V iZhi^[n^c\ ZmeZgi! > ]VkZ XdcYjXiZY i]djhVcYh d[ gZXdchigjXi^dch VcY ]VkZ iZhi^ÒZY ]jcYgZYh d[ i^bZh# L^i] i]^h `^cY d[ WVX`\gdjcY! > `cdl l]Vi XVc VcY XVcÉi WZ gZa^ZY dc [dg ig^Va# L]Zi]Zg ndjg XVhZ ^h Xg^b^cVa dg X^k^a ^c cVijgZ! ndj XVc gZan dc K^hjVa HiViZbZciÉh ;M( id egdk^YZ ZkZgni]^c\ ndj cZZY id YZe^Xi i]Z Zk^YZcXZ VcY hjeedgi ndjg de^c^dc ^c V bdhi VXXjgViZ VcY egd[Zhh^dcVa bVccZg#

Curtis �. Fl�nn � AC�AR ��2� � Fl�nn � Associates� �nc. � �l �aso� �e�as

I^bZ

2 '#&(+, hZX edhi ^beVXi 7BL KZadX^in 2 % HX]dda 7jh KZadX^in 2 %

Eaj\ ^c ndjg gZXdchigjXi^dc gZhjaih!

VcY ]^i EA6N# DkZg ((!%%% jhZgh VgdjcY ldgaY gZedgi i]Vi Vc^bVi^dch VgZ YdcZ ^c b^cjiZh#

�ow else can �ou e�plain a comple� event that happens in less than 3 seconds�

Plus: Cable cA�� uS For A �I�e demoOptions And We WI�� AnImATe for theyour “CDRcrASh� System �����������3�3 Handyman”

Forensic Photography

& Crash Reconstruction

MINIMIZE ROAD CLOSURES AND SAVE TIME WITH mFX 2008 Accuracy of Selected

Chrysler ACM EDR’s During Braking

• mFX was built for reconstructionists - handheld data collection not just adapted from other software • Save time with all the features you would expect with none of the complexity. With a simple and easy to use interface, data collection is made quick and easy • Use your own hardware whether it be Windows Mobile or a CE based device • Seamless download by directly downloading from mFX to FX3 instead of wasting time downloading into a third-party program

30 – 37 mph (both vehicles)

Driver Inputs

Mustang: 2-3 lb brake pedal force, 200 deg left steer onto Filbert, then 25-65 %WOT Charger: 3-5 lb brake pedal force, 175 deg left steer onto Filbert, then 70-75 %WOT

Maximum Undercarriage Impact Force

Mustang: Fx= -1730 lb, Fy= -13 lb, Fz= -2120 lb Charger: Fx= -8610 lb, Fy= -120 lb, Fz= -11100 lb

Maximum Suspension Force

Mustang: 4330 lb @ L/F; 3270 lb @ R/R Charger: 4810 lb @ R/F; 3530 lb @ L/R

Hubcaps Lost

Mustang: N/A (American Racing Mags) Charger: Five

Vehicle Speed Sensor Calibration

Basic Integral Calculus

for Crash Reconstruction

Other software may claim to answer these questions. Have you closely evaluated their claims and the basis for their results?

Bus Braking Tests:

To learn more about HVE and all of the available tools for crash

With and Without ABS

investigators, call 503-644-4500 or visit www.edccorp.com. Our also Close-Range Photogrammetry professional sales staff will help you get started using HVE today! & Forensic Photography COL-SP09

CDR Evidence: FAQs

Answers

Knowledge = Power www.CollisionPublishing.com

www.CollisionPublishing.com lll#k^hjVahiViZbZci#Xdb &#---#-'-#%(-(

Spring 2009

ACM Reprogramming

I’ve used many different tools and software to collect crash scene measurements.

But I can say without any hesitation that FX3 and mFX is by far the best program I’ve ever used. I look forward to many years of use out of mFX and Vista FX3 software.

Det. Mark A. Rice #1457 • Division Reconstructionist • Columbus Division of Police

plus 2009 CDR User’s Conference CD

A TRIMBLE COMPANY

Engineering Dynamics Corporation • 8625 SW Cascade Blvd., Suite 200 Beaverton, OR 97008 USA • Phone (503) 644-4500 • www.edccorp.com

TOLL-FREE: 1.888.828.0383

bZci#Xdb &#---#-'-#%(-(

EDC COL Ad 4_9_09.indd 1

An Abbreviated History

WET or FROZEN

ACM Access Considerations

Insurance Applications

For CDR Legal Considerations

Time Development of Delta-V & PDOF

during a collision

Crash Data Collection Guide for GM Airbag Electronic Control Units

Beyond the Drag Factor Accident Reconstruction Testing

Speedometers

Post-Collision Readings & Vehicle Impact Speeds

Filament Analysis of Light Bulbs

Average Ignition Cycles For GM Vehicles

Comparison Test:

Vehicle Speed vs Speed Recorded by an SDM

Validating Speed Analysis Calculations with Crash Test Data

ONE STEP AHEAD

C ollision

For a limited time, back issues of Collision Magazine are on sale at lower than subscription prices! If you are missing any back issues, now is the time to complete your collection.

The International Compendium for Crash Research

Volume 4 Issue 2

FALL 2009

What is 30?

Momentum

Vectoring in a New Approach

Auto Extrication for the Reconstructionist

Commercial Vehicle Dynamics

For Collision Reconstruction

Death Investigations & Their Psychological Effect

Supplies are limited so order your back issues today before they’re gone!

To learn more about Collision Magazine, please contact us:

C ollision The International Compendium for Crash Research

Volume 6 Issue 1

Volume 6, Issue 1 - Spring 2011

A Variable Deceleration Rate Approach

Volume 5, Issue 2 - Fall 2010

Volume 5, Issue 1 - Spring 2010

Rollover Crash Reconstruction

www.cdrsummit.com

D DV !

IT ide M s M ta In SU a R st D CD h Te 11 ras 20 & C

CDR Technology

Fall 2010

www.collisionpublishing.com

Collision: The International Compendium for Crash Research

Effect of Directional Tires on Wet Traction

C

2011 Vol. 6 Iss.1

The International Compendium for Crash Research

!

Common Post-impact Speed Analysis

C ollision Volume 5 Issue 2

Collision: The International Compendium for Crash Research

Collision: The International Compendium for Crash Research

Spring 2010

Fu

C

2010 Vol. 5 Iss.2

The International Compendium for Crash Research

Volume 5 Issue 1

e VD sid I D r In CS ste C- Po ARrash 10 e C 20ll Siz

IT M M !

SU ide R Ins CD CD 10 Data

C ollision

WWW.VISUALSTATEMENT.COM

4/8/09 1:26 PM

&

20

C

2010 Vol. 5 Iss.1

Vol. 4 Iss. 2

Volume 4, Issue 2 - FALL 2009

Seeing Across the Years

I^bZ

2 %#+&,+ hZX edhi ^beVXi 7BL KZadX^in 2 &(#*+() [eh HX]dda 7jh KZadX^in 2 &,#'(.' [eh

C

2009

The International Compendium for Crash Research

Volume 4 Issue 1

Volume 4, Issue 1 - SPRING 2009

I^bZ

2 %#')*. hZX edhi ^beVXi 7BL KZadX^in 2 '(#%-.- [eh HX]dda 7jh KZadX^in 2 (-#()&' [eh

C ollision

Collision: The International Compendium for Crash Research

“Multiple Case Studies”

Questions

Single Point Crush Variation I^bZ

2 &#(.&' hZX id ^beVXi 7BL KZadX^in 2 .(#'**, [eh HX]dda 7jh KZadX^in 2 )*#')&' [eh

Fall 2008

In this issue

The Accuracy of Speed Recorded by a Ford PCM

Reconstruction

Collision: The International Compendium for Crash Research

I^bZ 2 &#(.&' hZXdcYh id ^beVXi 7BL KZadX^in 2 .(#'**, [eh 7BL 9^hiVcXZ 2 &')#-.%& [i# id ^beVXi HX]dda 7jh KZadX^in 2 )*#')&' [eh

D DV

Comparing Different Sources of CDR Data To “Real World” Crashes

nf.

>cXgZY^Wan ZVhn id aZVgc

Co

8djgi VXXZeiZY ZkZgnl]ZgZ

“The Trampoline Effect” in Reconstruction

SI

7j^ai ^c e]nh^Xh Zc\^cZ Xdcigdah i^bZ Y^hiVcXZ kZadX^i^Zh

C-C

A �I�e demo And mATe your crASh� �3

EgZX^h^dc iZggV^c bdYZah [gdb ndjg bZVhjgZbZcih

Volume 3, Issue 2 - FALL 2008

gjXi^dc gZhjaih!

gdjcY ldgaY gZedgi i]Vi ^c b^cjiZh#

�plain a comple� event that n 3 seconds�

6XXjgVXn id &$&%%%i] hZXdcY

The International Compendium for Crash Research

Volume 3 Issue 2

and the Effects of Brake, Yaw and Other Factors Volume 3, Issue 1 - spring 2008

I^bZ

2 '#&(+, hZX edhi ^beVXi 7BL KZadX^in 2 % HX]dda 7jh KZadX^in 2 %

Real�otion 3D Features�

AR

^bZ 2 &#(.&' hZXdcYh id ^beVXi BL KZadX^in 2 .(#'**, [eh BL 9^hiVcXZ 2 &')#-.%& [i# id ^beVXi X]dda 7jh KZadX^in 2 )*#')&' [eh

���� ��A���� ����� A��� �O ��O� eXAcT�y ��A� �A�����D

C ollision

Crash site evidenCe

www.arccsi.com

D

A 1968 Mustang GT, driven by Lt. Frank Bullitt, is chasing a 1968 Dodge Charger R/T, driven by a man known only as “The Killer,” down Taylor Street on Russian Hill in San Francisco.

08

XgZY^Wan ZVhn id aZVgc

Volume 3 Issue 1

Collision Reconstruction Factors

I CS !

C- side ARD In 09 DV 20 ata

20

j^ai ^c e]nh^Xh Zc\^cZ Xdcigdah i^bZ Y^hiVcXZ kZadX^i^Zh

djgi VXXZeiZY ZkZgnl]ZgZ

���� O�� ����� �O �A�� A�O�� ��A� �A�����D The International Compendium for Crash Research

Collision: The International Compendium for Crash Research

XjgVXn id &$&%%%i] hZXdcY

ZX^h^dc iZggV^c bdYZah [gdb ndjg bZVhjgZbZcih

Collision: The International Compendium for Crash Research

eal�otion 3D Features�

C ollision

The International Compendium for Crash Research

Volume 2 Issue 2

nce DVD C r a s h c o n f e r ewww.CollisionPublishing.com PLUS: 2007 ARC-CSI

For serious Reconstructionists see how FX3 can be right for you!

www.visualstatement.com

C ollision Night-Time Pedestrian

Volume 2, Issue 2 - FALL 2007

CDR Conference Proceedings

PDOF

Principal Direction of Force

One integrated workspace.

Volume 1, Issue 2 - Fall 2006

Volume 1, Issue 1 - June 2006

Crash Data Retrieval

LEA 530R3: This long-range, prismless instrument is ideal for making challenging measurements. Use

You never leave one application - FX3

From The Scene

Sokkia Measurement Solutions for Law Enforcement Applications

A Review Various ACMAccuracy, Moduleefficiency and the convenience of single-person operation are all LEA of Remote Mapper: this powerful robotic instrument.

Case Study: Types combined & DatainRecorded it’s extremely narrow, red laser to pinpoint and accurately measure difficult positions.

Implementing Policy forvisible CDR

software package that contains all the tools required for crash and crime reconstructions. Wether you do a linear momentum analysis,

The International Compendium forSimulation CrashorResearch SMAC create captivating 3D Animations;

on the scene - We can help you

Volume 2, Issue 1 - Spring 2007

Premier Issue

FX3 from Visual Statement is a state of the art reconstruction solution that allows you to perform accurate reconstructions ranging from simple sketches to detailed 3D Environments. FX3

Volume 1 was designed specifically for reconstructionists who benefit from a Issue 2

Collision: The International Compendium for Crash Research

Accuracy of Speed Recorded by an SDM & the Effects of Brake & Yaw Events

C ollision

Collision: The International Compendium for Crash Research

The International Compendium for Crash Research

Volume 1 Issue 1

Collision: The International Compendium for Crash Research

Collision: The International Compendium for Crash Research

A Full Product Line-up

C ollision

Every issue is packed with new, invaluable information on a variety of topics. Moreover, included with every issue is a DVD compendium of the proceedings of either the ARC-CSI Crash Conference or CDR User’s Summit including presentations, reference material, crash data, videos, photos, and more.

Missing Back Issues?

collisionmagazine.com

�A� �A�����D

beVXi

Over the course of the last 8 years, Collision Magazine has become the preeminent crash reconstruction print publication providing subscribers with access to timely, reliable and fresh research, thought-provoking reconstruction related studies and meaningful crash test data from multiple sources.

A semi-annual publication, Collision serves crash reconstructionists from the government and private sectors internationally.

C ollision

2012 Vol.7 Iss.2

Spring 2011

www.collisionmagazine.com

VEHICLE & OCCUPANT MOTION in Far-Side Impacts ESTIMATING MAXIMUM MOTORCYCLE Acceleration Rates INSURANCE APPLICATIONS FOR CDR Part Two in a Series THE PERFORMANCE OF SEAT BACKS in High Speed Rear Impacts

Plus Articles from the

800.280.7940

"My

n r u t

e h t at

- Erik Carlsson his article is not about a numbers game youngsters like to play, but about a case that went to court in Chicago. The issue that brought the case to court was the performance (or rather the lack thereof ) of the seatbelt in a car involved in a collision on a wintry Chicago road. The plaintiff, a fragile and not so young lady, who was a front seat passenger in the car, slid off the seat and got wedged between the seat and the dashboard when the car collided with the rear end of an oncoming car. [The driver of the oncoming car lost control on the icy road. His car spun around and entered the lane for oncoming traffic.]

The case was unusual in one particular respect, and that is that the defendant carmaker’s own legal team did not even try to make the jurors believe that the injured lady had not buckled her seatbelt. The deciding point was the rescue crew’s emergency call report, in which they stated that the seatbelt webbing had to be cut because the belt kept the trapped occupant’s upper torso as in a vice, pressed against the front edge of the seat frame.

Particulars about the seat belt. The front seatbelts in the subject car were equipped with what the carmaker called a “comfort feature,” a type of seatbelt the National Traffic Safety Board in a report about seatbelt performance referred to as a “window-shade” type. In a more technical term, the seatbelts were described as having a tension reliever.

132 Collision Magazine - Volume 8 Issue 2

" l e e wh The tension reliever was a very simple device used for a period of time by some automakers as an inexpensive alternative to adjustable upper seatbelt mounts. The purpose of the device was to oppose retraction of the belt webbing. Vehicle occupants, whose body shape and size were such that the non-adjustable shoulder belt did not fit well could simply pull out some length of the webbing, then release it slowly. The belt would then stay in the extracted position and make itself unnoticeable by lying loosely on the wearer’s shoulder. (The NTSB made the point that this type of seatbelt may not be noticeable during an accident either!)

Severity of the collision. Naturally, the speed of the car in which the injured passenger was riding became an issue. An engineer with the defendant carmaker served as the company’s accident reconstruction specialist, and produced a report in which he stated that the severity of the collision was equal to the car impacting a barrier at a speed of 20 mph, with a tolerance of plus or minus 3 mph . 1 Based on this report, the defendant carmaker conducted sled tests at 17 and 23 mph, with the test sled occupied by a dummy the size of the injured lady. Those tests showed that the lady must have had at least seven inches of slack in the seatbelt the defense legal team claimed.

Therefore, she only had herself to blame for her injuries because she had not followed the instructions in the Owner’s Manual for the vehicle model. Those instructions state that the wearer of the seatbelt should not allow more than one inch of slack. Had she followed those instructions, she would have escaped injuries, just as the driver did, who was wearing the seatbelt properly. Hence the collision was not severe enough to cause injuries to a properly restrained occupant, the defense team argued.

Particulars: It is of great interest that the injured lady was a passenger in a car driven by her neighbor, who had offered her a ride to the nearby shopping mall because the lady did not have a car! That the defense team would even mention what it says in the Owner’s Manual for the car model is therefore surprising since the lady could hardly be expected to have read the Owner’s Manual in her neighbor’s car. The car owner himself stated that he had never read the Owner’s Manual, nor did he see any reason why there would be instructions

www.collisionpublishing.com

in that manual about how to use the seatbelt. You just pull it out and buckle it, he said.

Advanced Mathematics? How did the company engineer determine that the equivalent barrier speed at the accident was 20 mph, with a 3 mph tolerance level? Well, his method was rather unorthodox, I will say. In his report, the engineer stated that he had not seen the actual damage to the car because the car had already been repaired when he got involved. However, based on photos taken of the car before it was repaired, he estimated that the damage to its front end corresponded to a [barrier impact] speed of 20 mph. Note: His estimate may be on the high side. Had the damage really corresponded to what could be expected at a 20 mph frontal barrier impact, the insurance company would probably have estimated that the repair cost would exceed the market value of the car. If so, the insurance company would not have paid for the repair, but paid the car owner the market value of the car, and the car most likely would have ended up at a COPART or IAA, where it would have been auctioned off to licensed car body shops as a source of used replacement parts.]

Having determined that the speed to use as a starting point was 20 mph, the company engineer multiplied 20 by 1.47, then squared the result. He next multiplied the new

www.collisionpublishing.com

Collision Magazine - Volume 8 Issue 2 133

result by 3,693, then divided by 2 and by 32.2, and lastly, multiplied by 2. The end result was a number slightly over 99,000, which he rounded up to 100,000 even.

He then added 30%, which brought the number up to 130,000. Next, he did all the prior numerical operations in reverse, i.e., he divided 130,000 by 2, then multiplied by 2 and by 32.2, divided the result by 3,693, and took the square root of the result. Lastly he divided the square root by 1.47, which gave the end result of 23 after rounding to the nearest integer.

Next, he went back to the 100,000 he had established earlier. This time he subtracted 30%, which brought the number down to 70,000. He then repeated the reverse operations a second time, which gave the final number of 17 after rounding. Thus, he stated in his report, the computations showed that the [Equivalent Barrier] speed ranged from a low of 17 mph to a high of 23 mph. This is why the carmaker’s testing facilities conducted the sled tests at those speeds.

What exactly was the company engineer doing? The answer is that he was playing a numbers game! His multiplications and divisions and squaring a number meant absolutely nothing because the numerical operations all cancelled themselves when they were done in reverse. The only operation that did not cancel itself was the “adjustment” of the computed number by plus and minus 30%.

Had he not made this adjustment, he would have arrived back at his starting number.

The reason the adjustment by 30% of his computed number changed the starting number by 3 mph, which is 15%, is simply that the adjustment was made to a number that included the square of a number. The reverse operations included the square root of the increased number, and the percentage change was therefore reduced to about one half.

This also means that the company engineer could have greatly simplified his own computations. Had he simply squared his starting number, he would have arrived at 400, 30% of which is 120. Those numbers add up to 520, the square root of which is 23 (again after rounding to the nearest integer.) Likewise, subtracting 120 from 400 gives the number 280, which has a square root of 17. But had the company engineer made his computations so simple, the company’s legal team would no doubt have realized that there was something fishy about how their engineer managed to establish a speed range of 17 to 23 mph. (The carmaker’s testing staff probably did not consider it their business to question how the given speed range had been determined.)

134 Collision Magazine - Volume 8 Issue 2

Was this a deliberate effort to mislead a court? Probably not. I suspect that the carmaker’s engineer did not realize that he was playing a numbers game, but thought he was following the textbook to the letter. In the old British system, the mass of an object is expressed as its weight in pounds divided by “acceleration due to gravity,” which is usually rounded to 32.2 feet per second squared. Furthermore, the kinetic energy in foot-pounds is expressed as the square of the speed (in feet per second) multiplied by the mass, divided by 2. Hence, the company engineer simply followed the book by calculating the vehicle’s kinetic energy in ft*lbs, based on his speed estimate and the vehicle’s stated curb weight.

It may be noted that the company engineer included in his series of numerical operations a multiplication factor of 2, which is not part of the kinetic energy formula. The reason for this is that he took into account that the rear end of other car must have been damaged as well. For simplicity, he assumed that the damage energy to that car equaled his calculated kinetic energy of the car in which the injured lady was riding. Whether the damage energy to the other car actually equaled the engineer’s calculated kinetic energy is of course not known, but that doesn’t matter anyway since the multiplication factor cancelled itself during the reversed computations. The company engineer could therefore have picked any number (other than zero) for the damage energy to the other car, or simply omitted it in the computations, just as he could have omitted the vehicle’s curb weight in his computations.

A question a reader of this article probably will ask is what reason the carmaker’s engineer had for adjusting his computed energy number up and down by 30%? The answer is just as interesting as his mathematics. He explained at his subsequent deposition that he knew from his experience that his calculated energy, based on his speed estimate, would be within plus or minus 30% of the colliding vehicle’s actual damage energy. Unfortunately, the attorney who took his deposition did not ask how the engineer had been able to establish the actual damage energy of the collision-damaged vehicles he had inspected in the past so as to enable him to determine how close his computed energy numbers, based on his speed estimates, speed, were to the real ones!

Actually, he stated in his report that he estimated that the speed of the car was 20 mph when the vehicle’s collided.

It is obvious, though, that he meant that the Equivalent Barrier Speed was 20 mph since this is the only speed that can be based on damage alone in a two-vehicle collision.

1

www.collisionpublishing.com

For more than 75 years, the Northwestern University Center for Public Safety has provided professional development to law enforcement professionals throughout the United States and abroad. NUCPS offers the world-renowned School of Police Staff and Command, as well as courses and programs in police management, police operations, police motorcycle operator and instructor training, crime scene investigation and forensics, crash investigation, transportation engineering and more.

nucps.northwestern.edu/collision nucps@northwestern.edu • 800-323-4011

Crash Investigation and Reconstruction The concept of scientific crash investigation has been a part of the Northwestern University Center for Public Safety (NUCPS) since its inception in 1936.

Upcoming Courses Unless otherwise noted, all classes will be held at NUCPS’s state-of-the-art facilities on the Northwestern University campus in Evanston, Illinois. For an up-to-date listing of crash investigation and reconstruction courses scheduled on campus and throughout the country, please visit nucps.northwestern.edu/collision.

ONLINE Crash Investigation 1 Online, Self-paced Start any time

Crash Investigation 2 Online, Self-paced Start any time

We literally wrote the book on crash investigation and reconstruction. In addition to writing the books, we continue to be world leaders in providing training in traffic crash investigation and reconstruction through our many courses of study.

We offer the most comprehensive program of its kind, with the course series beginning at basic crash investigation and culminating in advanced traffic crash reconstruction.

Crash Investigation 1 Online, Instructor-led January 13–March 21, 2014

Crash Investigation 2 Online, Instructor-led January 13–March 21, 2014

CDR Data Analyst April 7–11, 2014

Math & Physics Workshop for Crash Reconstruction April 7-11, 2014

Vehicle Dynamics April 14–18, 2014

Traffic Crash Reconstruction 1 April 21–May 2, 2014

Traffic Crash Reconstruction 2 May 5–9, 2014

EVANSTON, ILLINOIS Traffic Crash Reconstruction 3 February 24–28, 2014

Crash Investigation 1 March 10–21, 2014

Crash Investigation 2 March 24–April 4, 2014

CDR Technician Level 1 April 3, 2014

CDR Technician Level 2 April 4, 2014

Heavy Vehicle Crash Reconstruction May 12-16, 2014

Pedestrian Vehicle Crash Reconstruction May 19–21, 2014

Advanced Crash Reconstruction — Utilizing Human Factors Research May 19–23, 2014

Traffic Crash Reconstruction Refresher May 28–30, 2014

Understanding Heavy Vehicle EDRs September 8–12, 2014

NOW AVAILABLE FROM NORTHWESTERN UNIVERSITY CENTER FOR PUBLIC SAFETY

TRAFFIC CRASH RECONSTRUCTION,

2ND EDITION

Purchase your copy today. Spend a total of $500 or more and receive a 5% discount. nucps.northwestern.edu/tcrbook 800-323-4011

Expert AutoStats®

4N6XPRT BioMeknx™

Expert AutoStats® is a computer program containing over 42,000 vehicles cars, vans, utility vehicles, and light pickups. 1945-2012 model years are represented.

This program can be used alone or with our Expert Qwic Calcs® and Expert VIN DeCoder® programs. Also accessible from within REC-TEC.

(Lap top & D esktop W indow s 3 2 & 64 -bit O S C om p atible)

Enter:

M anufacturer and Year

Select:

M odel from pop-up window

Output:

Wheelbase Overall Length Height Wt. Distrib

Curb Weight Width Drive Wheels Front/Rear Track

M id-60's to present output also includes (when available): Front Overhang Rear Overhang Bumper Heights Hood Height Turning circle Wheel radius Bumper-to-Hood Bumper-to-Window

Expert AutoStats® is currently owned by over 700 private individuals and 300 Law Enforcement agencies in the U.S.A., Canada, and South America. Additional information, brochure, and order from can be downloaded from:

www.4n6xprt.com/4n6as.htm

4N6XPRT BioMeknx is designed as a collection of the Biomechanical data of importance to the Accident Investigator into one easily accessible reference location. It is designed as a program to be useful to ALL accident investigators, not just the Biomechanical expert.

The program includes: 1. Over 65 printable illustrations and charts 2. 10 Separate categories or sections of information related to forensic biomechanics 3. Information on walking speeds of people from 17 months to > 65 years 4. Calculation modules for: - segment length from total body height, - segment weights and Centers of M ass (CoM ) based on total body weight, - Predicted M aximum Walking Speed based on total height 5. Information was obtained by review of over 65 scientific texts, several of which are now "out of print"

www.4n6xprt.com/4n6bm.htm

PROGRAM ORDER FORM (Please Print) Contact Name: ________________________________________________ Company: ____________________________________________________ Street: ______________________________________________________ City:____________________ State:____ Zip:__ __ Country: __________ Phone: (_____) _______________ Fax: (_____) ___________________ E-M ail: ______________________________________________________ Expert AutoStats® 4N6XPRT BioM eknx™ Expert Qwic Calcs® 4N6XPRT Ped & Bike Calcs® 4N6XPRT StifCalcs® Expert VIN DeCoder® (pkg) Expert TireStuf® Order Processing

- $ 625.00 ea.

- $ 495.00 ea.

- $ 275.00 ea.

- $ 375.00 ea.

- $ 650.00 ea.

- $ 550.00 ea.

- $ 85.00 ea.

= ________ = ________ = ________ = ________ = ________ = ________ = ________

- Check = $5.00 / Credit Card = $10.00 / G ovt. P.O . = $15.00 = _______ Notarized Affidavit Requirement $25.00 per required Notarized Signature = _______

Q Deliver via electronic download (Email address Required) Q Ship on USB - $35.00 per disk per program California Sales Tax Enclosed is:

- 8.50%

TOTAL (U.S. Dollars)

= 0.00 = ______

= ________

= ________

Visa ___ MasterCard ___ Am. Ex. ___ Exp ____/____ - Check ___ - Gvt PO ___

Card Number: Name on Card: Signature:

_______________________________________________ _______________________________________________ ________________________________________________

Please M ake Checks*/Purchase Orders Payable to:

4N6XPRT Systems®

8387 University Ave - La M esa, CA 91942-9342 - U.S.A. FAX: (619) 464-2206 Phone: (619) 464-3478 Web: http://www.4n6xprt.com C hecks * M U ST * be draw n on a U.S. bank. O rders are shipp ed within 10 working days of receipt.

Prices subject to cha ng e without notice.

M ultip le orders and Package discounts availab le. Please call for details.

EDCRASH

We have the STIFFNESS data you need for your calculations!!

Now includes a FORCE BALANCE calculation module!!

4N6XPRT StifCalcs®

puts the NHTSA Crash Test database on YOUR 32 or 64 bit MS-Windows computer, automatically calculates A-B-G values Front, Side, and Rear - and Test Specific Crush Factors based on the NHTSA test data, allows the creation of “CLASS” vehicles and determine the “CLASS” stiffness values, PLUS conduct FORCE BALANCE calculations when needed, all of this with no need for you to get on the internet, and all for the LOW price of $650.00*!

Can you do this yourself “for free”?? Only if YOUR TIME = $0.00!!

For more information, program order form, and sample printouts please visit our web site at http://www.4n6xprt.com.

4N6XPRT StifCalcs® is provided by 4N6XPRT Systems

8387 University Avenue - La M esa, CA 91942 - USA Phone: 800-266-9778 / (619) 464-3478 Fax: (619) 464-2206

Individual Vehicle Stiffness Data

Engineering Dynamics Corporation Reconstruction of Accident Speeds on the Highway EDCRASH ™ is an HVE-compatible reconstruction analysis of vehicle collisions. It is the most widely used crash reconstruction computer program in North America. Applications include: • Speed Analysis • Crash Severity EDCRASH is based on the CRASH3 computer program originally developed for the National Highway Traffic Safety Administration (NHTSA), and includes numerous extensions developed

by EDC. The user assigns vehicle impact and rest positions and tire-road friction data acquired during a crash site inspection, and vehicle damage profile data acquired during a post-crash vehicle inspection. The user may also assign wheel lock-up information for damaged wheels. Using this information, EDCRASH calculates impact velocities and velocity changes based on wellestablished linear momentum and damage energy methods. The results are displayed in numerical and graphical reports.

EDCRASH has been extensively validated using numerous staged collisions, the results of which are publicly available in the technical literature. Program results easily pass federal standards for admissibility of scientific evidence.

So, you want Stiffness data for a vehicle, QUICKLY & CONVENIENTLY, but you don’t want to buy a program? Order your data on a vehicle-by-vehicle basis from us at only $40.00 per test!

Year: _________ Make: _____________________ Model:____________________________________ Impact on (circle one): Front - Side - Rear Impact speed (circle one): High - Low YOUR CONTACT INFORMATION

Name: ________________________________________ Company: _____________________________________ Address: ______________________________________ City/State/Zip: _________________________________ Phone: _______________________________________ Fax: __________________________________________ Email: ________________________________________ Visa/MC/AmEx: ________________________________ Exp.:____/____ Case Reference: ___________________ Send Order to:

4N6XPRT Systems

Fax: 619-464-2206

VISUAL REALITY ATM2013

Engineering Dynamics Corporation • 8625 SW Cascade Ave. Suite 200 • Beaverton, OR 97008 USA • 888.768.6216 www.edccorp.com EDCRASH and HVE are trademarks of Engineering Dynamics Corporation. All rights reserved.

Ph: 800-266-9778 8387 University Avenue / La Mesa, CA 91942 / USA E-mail - 4n6@4n6xprt.com

R e c Tr fo on us r st t e 15 ru d + ct by Y io e n a is rs t s

2013 ARC-- CSI CRASH CONFerence DVD

C

2013 Vol.8 Iss.2

The International Compendium for Crash Research

Volume 8 Issue 2

Collision: The International Compendium for Crash Research

SIMPLY THE BEST RECONSTRUCTION SOFTWARE

CRASH WITH US...

CALL US NOW TO

BOOK A LIVE DEMONSTRATION

OR TO FIND OUT MORE

1.888.828.0383

Volume 8, Issue 2 - FALL 2013

Diagramming • Reconstruction • Animations • Point Cloud

visualstatement.com/demo

ollision C FALL 2013

Data from the 2013 ARC-CSI Crash Conference Analysis of Event Data Recorder Delta-V Reporting in the IIHS Small Overlap Crash Test Hyundai and Kia Crash Data, the Indispensable Compendium

collisionmagazine.com

Differentiating Potentially Causal Pre-crash Component Damage from Crash Damage