Light Vehicle Dynamic Rollover Propensity Phases … Vehicle Dynamic Rollover Propensity Phases IV, V, and VI ... Phase I-A – Spring 1997 – ... – Escape = 8.0 % – Blazer =

1

Vehicle Researchand Test Center

14 Mar 02, page 1

Light Vehicle Dynamic Rollover Propensity Phases IV, V, and VI

Research Activities

Garrick J. Forkenbrock

W. Riley Garrott

NHTSA / VRTC

Vehicle Research and Test Center

Overview of NHTSA Rollover Research Phases

2

� Phase I-A – Spring 1997

– Exploratory in nature

– Emphasized maneuver selection and procedure development

� Phase I-B – Fall 1997

– Evaluation of test driver variability

– Introduction of the programmable steering machine

� Phase II – Spring 1998

– Evaluation of 12 vehicles using maneuvers researched in Phase I

� Phase IV

– Spring 2001

– Response to TREAD Act

– Consideration of many maneuvers

� Phase V

– Spring 2002

– Research factors that may affect dynamic rollover propensity tests

– Rollover and handling rating development

� Phase VI

– Evaluation of 25 vehicles using Phase IV recommendations

� Phase III-A – Spring 2000

– Introduction of “Roll Rate Feedback”

� Phase III-B – Summer 2000

– Pulse brake automation

Discussed in this presentation

14 Mar 02, page 2


Phase IV Background

TREAD Act / Congressional Requirements:

– Develop dynamic rollover propensity tests to facilitate a consumer information program

– Consumer Information methodology released by November 2002

– National Academy of Sciences Report

314 Mar 02, page 3


Additional Background

In their assessment of NHTSA’s existing rollover resistance rating system (January, 2002) the National Academy of Sciences recently recommended:

“NHTSA should vigorously pursue the development of dynamic testing to supplement the information provided by SSF.”

414 Mar 02, page 4


Additional Background

� NHTSA is presently providing Rollover Resistance Rating

� Based on vehicle measurements and real world crash data

� Vehicle measurement is Static Stability Factor

� 5 Star ratings are similar to NCAP Crash Ratings

514 Mar 02, page 5


6

T/2

Hcg

Weight W

Pull

Impending Rollover W(T/2)=P(Hcg) Pull/W = (T/2)/ Hcg

Pull/W=SSFWeight W

No Force

14 Mar 02, page 6

7


14 Mar 02, page 7

0%

10%

20%

30%

40%

50%

1 1.05 1.1 1.15 1.2 1.25 1.3 1.35 1.4 1.45 1.5 1.55 1.6

Static Stability Factor

Pro

babi

lity

of R

ollo

ver

per

Sin

gle

Veh

icle

Cra

sh


Maneuver Recommendations

� Alliance of Automobile � MTS Systems Manufacturers Corporation

� Consumers Union � Nissan Motors

� Ford Motor Company � Toyota Motor Company

� Heitz Automotive, Inc. � UMTRI

� ISO 3888 Part 2 Consortium– VW, BMW, Daimler Chrysler

– Porsche, Mitsubishi

814 Mar 02, page 8

9


14 Mar 02, page 9

Phase IV Test Conditions


Test Vehicles

� 2001 Chevrolet Blazer 4x2 � 2001 Ford Escape 4x4

– One star static rollover rating – Three star static rollover rating

– High sales volume – Smaller, car-like SUV

� 1999 Mercedes ML320 4x4 � 2001 Toyota 4Runner 4x4

– “Less aggressive” stability – “Aggressive” stability control control intervention intervention

– Two star static rollover rating – Two star static rollover rating

– First SUV with available – Relatively high sales volume stability control (ESP)

1014 Mar 02, page 10


Vehicle Configurations

� Instrumented

� Fully fueled

� Front and rear mounted aluminum outriggers

� Performed with and without stability control

� Multiple configurations

– Nominal vehicle

– Reduced rollover resistance

1114 Mar 02, page 11


Reduced Rollover Resistance

� Roof-mounted ballast

� Designed to reduce SSF by 0.05

� Increased roll inertia from Nominal condition

– Escape = 8.0 %

– Blazer = 11.5%

� Longitudinal C.G. preserved

� Maneuver sensitivity check Up to 180 lbs

1214 Mar 02, page 12

��

��

��

�

��

��

��

��


Reduced Rollover Resistance(measurements taken without instrumentation)

� 4Runner � Escape

– 180 lbs ballast – 120 lbs ballast

– C.G. raised 1.3” – C.G. raised 1.0”

– SSFNOMINAL = 1.11 (��) – SSFNOMINAL = 1.26 (�� )

– SSFRRR = 1.06 (��) – SSFRRR = 1.21 (��)

� Blazer � ML320

– 180 lbs ballast – 180 lbs ballast

– C.G. raised 1.3” – C.G. raised 1.2”

– SSFNOMINAL = 1.04 (��) – SSFNOMINAL = 1.14 (��)

– SSFRRR = 0.99 (�) – SSFRRR = 1.09 (��)

Note: Nominal SSF differ from those measured without outriggers. 1314 Mar 02, page 13


Test Vehicle SSF Summary

0.90

1.00

1.10

1.20

1.30

Blazer 4Runner ML320 Escape

Sta

tic S

tab

ility F

acto

r (S

SF

)

Baseline

Nominal (no instrumentation)

Nominal (with instrumentation) RRR (no instrumentation)

SSFnom,i = 1.05

SSFnom,i = 1.12

SSFnom,i = 1.18

SSFnom,i = 1.27

1414 Mar 02, page 14


Tires

� OEM specification (as installed on vehicle when delivered) – Make

– Model

– DOT Code

– Inflation pressure

� Frequent tire changes

� Innertubes used during some maneuvers to prevent debeading

Test surface damage due to debeading

� Maneuver speed iterations selected to minimize tire wear within a given test series

15 14 Mar 02, page 15


Test Surface

� All tests performed on TRC’s VDA (a dry, high-mu asphalt surface)

� Tests performed 05/01 to 11/01, 02/02

� Stable friction coefficients

– Peak mu: 0.94 to 0.98

– Slide mu: 0.81 to 0.88

1614 Mar 02, page 16

17


14 Mar 02, page 17

Phase IV Maneuver Review


Characterization Maneuvers

� Used to define NHTSA’s dynamic rollover propensity maneuvers

– Constant Speed, Slowly Increasing Steer

� Used to characterize transient response

– Pulse Steer

– Sinusoidal Sweep

– J-Turn Response Time Tests

1814 Mar 02, page 18


Dynamic Rollover Propensity Maneuvers

� Driver-based Steering

– ISO 3888 Part 2

– CU Short Course

� Automated Steering – J-Turns

– Fixed Timing Fishhook

– Roll Rate Feedback Fishhook

– Nissan Fishhook

– Open-Loop Pseudo-Double Lane Change

� Driver-based Steering, Computer Corrected – Ford PCL LC

1914 Mar 02, page 19

20


14 Mar 02, page 20

NHTSA J-Turn and Fishhooks

� Steering magnitude based on vehicle response

1. Determine the handwheel angle at 0.3 g from Slowly Increasing Steer results

2. Multiply by a scalar (derived with Phase II data)

� Steering rate based on successful Phase II testing

– J-Turn = 1000 deg/sec

– Fishhook = 720 deg/sec

0 500 1000 1500 2000 2500

0

50

100

150

200

250

300

Han

dwhe

el A

ngle

(de

gree

s)

Count Number

0 500 1000 1500 2000 2500-0.1

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

Lat

eral

Acc

eler

atio

n (g

)

Count Number

R2 = 0.99281

actuallinear fit


NHTSA J-Turn

310ML320

3544Runner

287Escape

401Blazer

Handwheel Input

(degrees) Vehicle

2114 Mar 02, page 21


NHTSA Fixed Timing Fishhook(Symmetric)

252ML320

2874Runner

233Escape

326Blazer

Handwheel Input

(degrees) Vehicle

2214 Mar 02, page 22


NHTSA Roll Rate Feedback Fishhook(Symmetric)

23

252ML320

2874Runner

233Escape

326Blazer

Handwheel Input

(degrees) Vehicle

14 Mar 02, page 23

24


14 Mar 02, page 24

Nissan Fishhook

� Adjusts timing to maximize roll motion

� 270 degree initial steer

� Vehicle-dependent reversal magnitude (for fishhooks)– Blazer = 570 degrees

– Escape = 505 degrees

� All rates = 1080 deg/sec

� Response-dependent dwell times– Iterative determination

0 1 2 3 4 5 6 7 8-300

-200

-100

0

Han

dw

hee

l (d

eg)

Time

0 1 2 3 4 5 6 7 8-20

0

20

40

Ro

ll R

ate

(deg

/sec

)

Time

0 1 2 3 4 5 6 7 8

-200

0

200

400

600

Han

dw

hee

l (d

eg)

Time

0 1 2 3 4 5 6 7 8-1

-0.5

0

0.5

1A

Y (

g)

Time


Closed-loop, Path-Following Lane Changes

Consumers Union Short Course

ISO 3888 Part 2

25 14 Mar 02, page 25


Ford PCL LC

� Comprised of a suite of closed-loop paths (double lane changes)

� Data is processed to remove driver effects and facilitate comparison at a constant severity – All vehicles taken to follow the same path

– All vehicles subject to the same lateral acceleration demands

� Test output is an overall dynamic weight transfer metric

2614 Mar 02, page 26

27


14 Mar 02, page 27

Ford PCL LC

Full Lane Lane and a HalfHalf Lane

28


14 Mar 02, page 28

Comments Based on Phase IV Rollover

Resistance Maneuvers


NHTSA J-Turn

� Lowest speed of two-wheel lift is metric

� Uses Programmable Steering Controller

� Simple step-steer (one cycle)

� Handwheel magnitude dependent on vehicle response

2914 Mar 02, page 29


J-Turn with Pulse Braking


� Uses Programmable Braking and Steering Controller

� Addition of Braking Controller makes maneuver substantially harder to perform

� Timing of brake pulse dependent on vehicle response (Roll Rate Feedback)

� Results significantly influenced by whether vehicle has working ABS

3014 Mar 02, page 30


Fixed Timing Fishhook


� Dwell time independent of vehicle response

� Handwheel magnitudes dependent on vehicle response

� Handwheel inputs within ranges established during ISO and CU double lane change testing

� Timing may be better for one vehicle than another

3114 Mar 02, page 31


Roll Rate Feedback Fishhook


� Handwheel magnitudes dependent on vehicle response

� Handwheel inputs within ranges established during ISO and CU double lane change testing

� Dwell time also dependent on vehicle response

� Timing should no longer favor one vehicle over another

3214 Mar 02, page 32


Nissan Fishhook


� Iterative procedure requires additional testing time

� Large number of tests required many tire changes (to reduce tire wear concerns)

� Reversals are harsh; increases steering machine wear

3314 Mar 02, page 33


Ford Path Corrected Limit Lane Change (PCL LC)

3414 Mar 02, page 34


Ford PCL LC� Metric Dynamic Weight Transfer at 0.7 g based

on one of four standard paths (DWTM)

� Method removes driver dependence by normalizing data

� Extra instrumentation needed to run

� Extra tire testing required (tire measurements)

� Concerns about 0.40 second window used for metric calculation (mitigates dynamic weight transfer observed)

� Metric now measured during tests performed with a driving robot

3514 Mar 02, page 35

�


ISO 3888 Part 2 Double Lane Change

� Suggested rating metric is maximum achievable “clean” run speed

– “Clean” run � no cones struck/bypassed

� Test driver generated steering inputs

� Not as repeatable as programmable steering controller inputs

� Tests are straightforward to perform

� Course adapts to vehicle width

3614 Mar 02, page 36

�


Consumers Union Short CourseDouble Lane Change

� Suggested rating metric is maximum achievable “clean” run speed – “Clean” run � no cones struck/bypassed

� Test driver generated steering inputs

� Not as repeatable as programmable steering controller inputs

� Tests are straightforward to perform

� Course does not adapt to vehicle size

3714 Mar 02, page 37


Open-Loop Pseudo-Double Lane Change

� Uses programmable steering controller

� Having three major steering moves slightlydegrades repeatability

� Straight-forward to perform

� Uses programmable steering controller

� Additional development required

3814 Mar 02, page 38


Reporting of Phase IV Findings

Draft of Phase IV NHTSA Technical Report has been written

– Reviews in progress

– Anticipated release late Spring ‘02

3914 Mar 02, page 39

40


14 Mar 02, page 40

Phase V Research


Phase V Overview

� Investigate potential use of a centrifuge

� Improved test equipment– Alternative outrigger development

– Quantification of two-wheel lift

� Resolution of existing matters– Cold and hot weather testing

– Surface effects testing

� Finalize methodology for Phase VI– Loading

4114 Mar 02, page 41


Centrifuge

� Metric could be lateral acceleration at wheel lift or weight transfer

� Quasi-static test

� May be demonstrated by NHTSA using a NASA Facility

4214 Mar 02, page 42


Outrigger Development

� Reduce effects of � Compare three designs

outrigger installation – Existing VRTC Design

without compromising � Aluminum driver safety � 78 lbs per outrigger

� Use wheel load – New VTRC Design

transducers to evaluate � Titanium

dynamic load transfer and � 68 lbs per outrigger

cornering forces – Carr Engineering

� Carbon fiber

� 58 lbs per outrigger

� Testing complete

4314 Mar 02, page 43


Carbon Fiber

� Manufactured by Carr Engineering

� Light weight (58 lbs)

� Strong

� Expensive ($25k / set)

4414 Mar 02, page 44


Titanium

� Designed at VRTC using finite element analysis

� Light weight (68 lbs)

� Less roll inertia than aluminum or carbon fiber

� Strong

� 1/3 cost of carbon fiber

� 6Al-4V a common Ti alloy

� Low-mu hemispherical skid pads replace heavier casters

4514 Mar 02, page 45


Quantification of Two-Wheel Lift

� Objective methodology required

� Laser-based height sensors on each wheel– Eliminates video data analysis subjectivity

4614 Mar 02, page 46


Cold and Hot Weather Testing

� Will research the effects of temperature extremes on dynamic rollover propensity

� All testing to be performed at TRC

� Cold weather tests performed during January ‘02

� Hot weather tests to be performed Summer ‘02

4714 Mar 02, page 47


Surface Effects Testing

� Intended to research the effects of different test surfaces on dynamic rollover propensity

� Testing presently underway in Arizona

– DaimlerChrysler Arizona Proving Grounds (APG)

– GM Desert Proving Grounds

– Performed with the Blazer and 4Runner

� Results from Arizona will be compared with those produced at TRC

4814 Mar 02, page 48

49


14 Mar 02, page 49

Phase VI


Phase VI Overview

� Maneuvers based on Phase IV findings

� Two load conditions are anticipated

� Titanium outriggers

� 25 Vehicles

� Will include a wide range of make/models for which state rollover rate data is available

5014 Mar 02, page 50

Light Vehicle Dynamic Rollover Propensity Phases … Vehicle Dynamic Rollover Propensity Phases IV, V, and VI ... Phase I-A – Spring 1997 – ... – Escape = 8.0 % – Blazer =

Documents