The development of a load sensing trolley for frontal offset testing

This article was downloaded by: [University of Birmingham]On: 11 November 2014, At: 10:22Publisher: Taylor & FrancisInforma Ltd Registered in England and Wales Registered Number: 1072954 Registered office: MortimerHouse, 37-41 Mortimer Street, London W1T 3JH, UK

International Journal of CrashworthinessPublication details, including instructions for authors and subscription information:http://www.tandfonline.com/loi/tcrs20

The development of a load sensing trolley for frontaloffset testingTon Versmissen a , Richard Schram a & Steve McEvoy ba TNO Science and Industry, Business Unit Automotive, Safety Department ,Schoemakerstraat 97, Delft, The Netherlandsb First Technology Safety Analysis , 2 Columbus Drive, Summit Avenue, Southwood,Farnborough, GU14 0NZ, United KingdomPublished online: 21 Sep 2007.

To cite this article: Ton Versmissen , Richard Schram & Steve McEvoy (2007) The development of a load sensing trolley forfrontal offset testing, International Journal of Crashworthiness, 12:3, 235-245, DOI: 10.1080/13588260701441100

To link to this article: http://dx.doi.org/10.1080/13588260701441100

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) containedin the publications on our platform. However, Taylor & Francis, our agents, and our licensors make norepresentations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose ofthe Content. Any opinions and views expressed in this publication are the opinions and views of the authors,and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be reliedupon and should be independently verified with primary sources of information. Taylor and Francis shallnot be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and otherliabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to orarising out of the use of the Content.

This article may be used for research, teaching, and private study purposes. Any substantial or systematicreproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in anyform to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://www.tandfonline.com/page/terms-and-conditions

http://www.tandfonline.com/loi/tcrs20

http://www.tandfonline.com/action/showCitFormats?doi=10.1080/13588260701441100

http://dx.doi.org/10.1080/13588260701441100

http://www.tandfonline.com/page/terms-and-conditions

http://www.tandfonline.com/page/terms-and-conditions

The development of a load sensing trolley forfrontal offset testing

doi:10.1080/13588260701441100

Ton Versmissena, Richard Schrama and Steve McEvoyb

aTNO Science and Industry, Business Unit Automotive, Safety Department, Schoemakerstraat 97, Delft,The NetherlandsbFirst Technology Safety Analysis, 2 Columbus Drive, Summit Avenue, Southwood, Farnborough GU14 0NZUnited Kingdom

Abstract: This paper describes the development of a load sensing trolley for frontal offset testing,to be used as a research tool in the field of compatibility assessment. The main specifications of thetrolley, mass, CoG location, and inertia properties, are based on European and U.S. vehicle geometrydatabases. The influence of different mass and inertia properties was studied using a trolley model andvehicle models in numerical simulations. Based on this information and for research purposes, it wasdecided to make the trolley mass adjustable between 1300 kg and 1800 kg, with tunable inertia propertieswithin realistic limits. The trolley was designed to be equipped with the progressive deformable barrier(PDB) as a deformable element. The PDB was chosen for its stability and its ability to allow a barrierface deformation measurement in order to evaluate the aggressiveness of cars. To study the potentialof additional force measurements in MDB tests, the trolley is also equipped with a lightweight highresolution strain gauge loadcell wall in front of the trolley, just behind the deformable element. Intotal 48 strain gauge load cells of 125×125 mm2 are mounted in 6 rows and 8 columns to the frontof the trolley. The LCW is equipped with a built-in data-acquisition system. With the first prototypetrolley, several calibration tests were performed and it was concluded that the developed trolley is agood working research tool. The future work with the trolley will concentrate on the development andevaluation of a Mobile Progressive Deformable Barrier test procedure, based on the current draft PDBprotocol.

Key words: Compatibility, mobile barrier, EEVC WG15, offset test

INTRODUCTION

Frontal compatibility assessment is still a major topic incrash safety research worldwide. With the changing fleetcomposition, the differences between the cars are increas-ing in terms of mass, stiffness, and geometry. Research inthe field of compatibility is ongoing worldwide and a gen-eral objective of the compatibility research is to ensure thatfuture vehicle developments are more balanced in terms ofoccupant protection of both striking and struck vehicle, incase of vehicle-to-vehicle collisions.

Methods to assess frontal compatibility should take intoaccount several aspects:

Corresponding Author:Ton VersmissenTNO Science and IndustryBusiness Automotive, Safety DepartmentSchoemakerstraat 97DelftThe NetherlandsEmail: [email protected]

– Structural interaction between the two vehicles: the ca-pability of vehicles to ensure that there is a possibility toabsorb the energy of the crash,

– Compartment strength: the capability of vehicles to with-stand the maximum forces occurring during the crash inorder to ensure enough survival space for the occupants,

– Frontal stiffness: possibility to balance the crash energyabsorption.

Moreover, the occupant’s self-protection should not becompromised by increasing the level of partner protection.

Currently none of the regulatory and consumer test pro-cedures is able to assess vehicle-to-vehicle compatibilityon these main aspects. The current procedures are forself-protection assessment and restraint system optimisa-tion only, which is not necessarily beneficial for partnerprotection. Furthermore, there is a lack of worldwide har-monisation in the current protocols.

A short-, mid-, and long-term view on compatibilityassessment, also taking into account the desire for world-wide harmonisation acknowledging fleet differences across

Copyright C© Taylor & Francis Group 235 TCRS 2007 Vol. 12 No. 3 pp. 235–245

Dow

nloa

ded

by [

Uni

vers

ity o

f B

irm

ingh

am]

at 1

0:22

11

Nov

embe

r 20

14

T Versmissen, R Schram and S McEvoy

Figure 1 Future outlook for compatibility assessment. AHOD: average height of deformation. ADOD: average depth ofdeformation. AHOF: average height of fore.

the world, are presented in Figure 1. In the long term, amobile deformable barrier test procedure for compatibilitytesting is generally seen as the best achievable compromiseby both Europe and the United States, and therefore opensthe possibility for harmonisation.

To support the foreseen long-term direction of compat-ibility assessment, TNO Automotive decided to developthe required test tool: a load sensing trolley to be used infrontal mobile offset deformable barrier testing. This pa-per describes the development of this trolley, which can beused as a research tool in the field of compatibility assess-ment.

METHODS

Introduction

The long-term approach of a mobile test procedure is basedon the hypothesis that the striking vehicle is an average car.Therefore, the trolley with deformable barrier should berepresentative of a vehicle class in Europe or the UnitedStates. The main specifications of the trolley, such as mass,centre of gravity (CoG) location, and inertia properties, arebased on European and U.S. vehicle geometry databases[1, 2] and current regulations also using a trolley. A nu-merical model of a trolley, according to the design specifi-cations, was developed and the influence of different massand inertia properties on the trolley performance was stud-ied in numerical simulations using various vehicle modelsin MADYMO.

Trolley specifications

General design criteriaFor the trolley design, the following general requirementswere set:

• For durability and lifetime:• The trolley should be able to sustain at least 100 times

the maximum crash pulse without significant serviceactivities.

• Crash-specific design criteria• The trolley should be able to sustain a 75 g deceleration

crash pulse, based on maximum crash pulses occurringin current test procedures.

• The trolley should be capable of running at a speed ofat least 20 m/s (72 km/h).

• The trolley should be equipped with a remote-controlled electro-hydraulic brake system.

• The axles/wheels of the trolley should be adjustable tocontrol trolley alignment.

• Usability for different crash scenarios:• The trolley should be usable for offset frontal, full

frontal, and side impact crash scenarios.• It should be possible to equip the trolley as left- and

right-hand drive.

Deformable barrierThe trolley was designed to be equipped with a deformableelement in front; in this study the progressive deformablebarrier (PDB) was used as a deformable element. The PDBwas chosen for its stability and its ability to allow a barrierface deformation measurement in order to evaluate thepotential aggressiveness of cars.

Dimensions of the trolleyThe trolley dimensions are based on specifications ofEuropean vehicles and are presented in Figure 2 andTable 1. The standard convention for the vehicle coor-dinate system was followed: x-axis in driving direction,y-axis lateral, and z-axis vertical.

236TCRS 2007 Vol. 12 No. 3 Copyright C© Taylor & Francis Group

Dow

nloa

ded

by [

Uni

vers

ity o

f B

irm

ingh

am]

at 1

0:22

11

Nov

embe

r 20

14

The development of a load sensing trolley for frontal offset testing

Table 1 MDB design specifications (default conditions)

Description Average EU vehicle [2] MDB

Total mass (kg) 1200–1700 1500CG location from front, w.r.t. length (m) 0.43–0.47 0.45Vehicle front to front axle distance H (m) 0.720–0.980 0.900Vehicle front to CG distance I (m) 1.700–2.100 1.900Vehicle front to rear axle distance J (m) 3.200–3.700 3.500Overall length K (m) 3.800–4.700 4.250CG height L (m) 0.560–0.640 0.600Axle height M (m) 0.270–0.290 0.280Wheel base (m) 2.450–2.750 2.600Mass front axle (kg) 710–990 900Mass rear axle (kg) 465–735 600

Figure 2 Schematic drawing of the trolley, top view and sideview.

The default mass of the trolley was selected to be1500 kg and for research purposes the trolley mass wasmade adjustable between 1300 and 1800 kg. All main di-mensions of the trolley were selected to fit in the rangefound for an average European passenger car (test weight).Next to this, MADYMO simulations were performed tostudy the effects of Centre of Gravity location and trolleymass on trolley performance.

Inertia properties of the trolley are based on the NHTSAdatabase [1], which provides inertia properties for a largerange of vehicles. The passenger car information was usedfor this study. No relevant European data were available.In Figure 3 the inertia properties versus vehicle mass aregiven. The inertia properties as a function of vehicle masswere estimated using least square fitting.

For some specific masses of the trolley the selected iner-tia properties are given in Table 2. The design tolerancesare presented between brackets. Since the effect of rollon frontal test results is assumed to be low, a higher de-

sign tolerance has been accepted for the Iyz. The defaulttrolley mass and corresponding inertia properties aremarked bold.

Trolley specifications versus regulatory test devicesTo evaluate how the selected specifications of the MDBtrolley compare to the specifications of various trolleysused in current and expected future regulatory testing,this information is presented in Table 3.

The MDB trolley mass is in line with the future pro-posed test procedures for side impact, EA MDB in Europeand the IIHS in the United States. Although the trolleymass is higher than for trolleys currently used in regula-tions, it is assumed to be more in line with the currentvehicle fleet, see Table 1. The inertia properties of theMDB are comparable to the trolleys used in the UnitedStates. It was decided to have a relatively small groundclearance of the deformable barrier face (150 mm) in orderto be able to assess low subframes, if present, and to be inline with the Progressive Deformable Barrier protocol.

Force measurement

To study the potential of additional force measurements inMDB tests, the trolley is also equipped with a light weighthigh resolution strain gauge loadcell wall (HR-LCW) infront of the trolley, just behind the deformable element.In total 48 strain gauge load cells of 125 × 125 mm2 aremounted in 6 rows and 8 columns to the front of the trol-ley. The HR-LCW, developed by Thames Side-Maywood(TS-M), is equipped with a built-in data-acquisitionsystem.

Crash Wall Load CellFor this type of force measurement testing, it is criticalto utilize load cells with high stiffness/rigidity that wouldaccurately record force data, regardless where it was ap-plied to its surface face. Therefore, the load cell elementsselected are machined from one homogeneous block ofa high strength 2014A aluminium alloy and not manufac-tured from multiple components that would reduce a crashsystem’s rigidity and response performance.

To enable quick and easy verification of the calibratedsystem, strain-gauged load cells were selected, enabling the

237Copyright C© Taylor & Francis Group TCRS 2007 Vol. 12 No. 3

Dow

nloa

ded

by [

Uni

vers

ity o

f B

irm

ingh

am]

at 1

0:22

11

Nov

embe

r 20

14


Figure 3 Inertia properties, based on [1].

use of true force quasi-static calibrations to calibrate thecrash wall load cells. A strain gauge is a device whose elec-trical resistance varies in proportion to the amount of strainin the structure. Strain gauges on the crash wall load cellsaccurately measure the small changes in resistance by usinga full bridge configuration with a voltage excitation source.The general Wheatstone bridge, illustrated in Figure 5,

consists of four resistive arms with an excitation voltagethat is applied across the bridge, with the resultant changein resistance measured as an electrical output.

The crash wall load cell, manufactured from one ho-mogeneous block of material into a solid 5-column loadcell, produces a very rigid sensor. The central column hasthe effect of pivoting the force against the opposite corner,

Table 2 Design inertia properties, including tolerances.

Mass (kg) Yaw Ixy (kg/m2) Pitch Ixz (kg/m2] Roll Iyz (kg/m2)

1300 2140 (±10%) 2040 (±10%) 450 (±50%)1500 2650 (±10%) 2550 (±10%) 550 (±50%)1800 3470 (±10%) 3380 (±10%) 710 (±50%)

Table 3 MDB compared to various regulatory trolleys.

MDB ECE/R95 EA MDB IIHS FMVSS 214

Total trolley mass (kg) 1500 950 ± 20 1500 ± 20 1500 ± 5 1368Ground clearance (m) 0.150 0.260 ± 0.005 0.300 ± 0.005(0.350 ±0.005) 0.379 0.279Distance CG-barrier face (m) 1.900 2.000 ± 0.30 2.000 ± 0.30 1.626∗ 1.626∗Distance CG-front axle (m) 1.000 1.000 1.000 1.123 1.123CG height (m) 0.600 0.500 ± 0.030 0.500 ± 0.030 0.566 ± 0.025 0.500Wheel base (m) 2.600 3.000 ± 0.010 3.000 ± 0.010 2.591 2.591Distance fr axle-barrier face (m) 0.900 1.000 1.000 1.143∗ 1.143∗Track width (m) 1.200 1.500 ± 0.010 1.500 ± 0.010 1.880 1.880Overall length (m) 4.250 4.115 4.115Iyz Roll (kg/m2) 550 542∗ 508∗

Ixz Pitch(kg/m2) 2550 2471∗ 2263∗

Ixy Yaw (kg/m2) 2650 2757∗ 2572∗

∗Note: estimated numbers.


Dow

nloa

ded

by [

Uni

vers

ity o

f B

irm

ingh

am]

at 1

0:22

11

Nov

embe

r 20

14


Figure 4 Crash wall load cell.

Figure 5 Strain gauge grid (left) and Wheatstone.

allowing a parallelogram effect to occur. This parallelo-gram effect allows any off-center loading to be mechani-cally compensated for whilst removing bending effects inthe load cell that would not normally be measured. Thismethod of compensation allows the total column strain tobe mechanically compensated to be within < ± 0.2% lin-earity of the applied force, wherever the force is appliedover the load cell 125 mm × 125 mm face. The load cellson the trolley underwent a rigorous proving when theywere manufactured and are capable of withstanding 200%overload in impact conditions.

Mobile Barrier Crash Wall Load Cell – Technical DataThe technical data of the load cells used in the mobilebarrier crash wall are summarized in Table 4.

The crash wall load cells are mounted on to the MDBtrolley via a mounting plate that includes an integrated dataacquisition system (TDAS G5 from Diversified Techni-cal Systems (DTS)). The small size and low mass of theTDAS G5 DAS solution meets the increasing need forlightweight, crash-hardened data acquisition systems thatcan be moved quickly and easily from application to ap-plication. The TDAS G5 system has docking stations thathave been designed into the load cell mounting plate, al-lowing the TDAS G5 modules to be docked directly intothe LCW. Once docked, the TDAS G5 modules allow au-tomatic data analysis and collection. Each position on themounting plate has an individual ID, which means that the

Table 4 Technical data load cells.

Units Axial Load

Rated load kN 300Safe side load %RL 150Ultimate side load %RL 200Safe longitudinal load %RL 150Ultimate longitudinal load %RL 150Operating temperature range ◦C -20 to +80Compensated temperature range ◦C +10 to +40Environmental protection IP 66Typical bridge configurationLongitudinal load output @ 300 kN mV/V 1.300 ± 1%Excitation voltage (recommended) Vdc/ac 10Excitation voltage (maximum) Vdc/ac 15Nonlinearity %RL <±0.20Repeatability %RL <±0.05Effect of off centre axis loading %RL <±0.20Natural frequency (axial direction) Hz >5000Stiffness/rigidity kN/m >3,500,000Zero load output %RL <2.0Input bridge resistance axial Ohms 320±20Output bridge resistance axial Ohms 300±5Temperature coefficient of zero %/◦C <±0.03Temperature coefficient of span %/◦C <±0.03Insulation resistance @100V dc M� >500Mass of load cell kg 3.5

sensors that store the calibration data can be automaticallyidentified and the LCW is automatically calibrated witheach new setup.

The TDAS G5 data acquisition system is a 32 channelDAS designed for a variety of dynamic impact tests. EachTDAS G5 module has 50MB RAM and 128MB SD cardfor storing the data. Before the crash test commences theTDAS software runs a check of the entire system. TheTDAS G5 memory runs on a circular buffer recordingdata continuously until triggered.

The crashworthy TDAS PLUS Mini Distributor inte-grates power, communication, and event signal distribu-tion functions for multiple TDAS systems into a small,compact package.


Dow

nloa

ded

by [

Uni

vers

ity o

f B

irm

ingh

am]

at 1

0:22

11

Nov

embe

r 20

14


Figure 6 Crash wall load cell complete with gasket as used onthe mobile barrier.

The PC initialises the system and then it can be discon-nected and the TDAS will collect data with no additionalsupport. All MPDB DAS components are illustrated inFigure 7.

Calibration testsA calibration test was performed to assess the final designof the trolley and to evaluate the performance of the highresolution loadcell wall. The calibration test was performedwith the PDB as deformable element mounted to the trolleyface and the trolley was driven into a rigid wall with animpact velocity of 45 km/h at perpendicular impact angle.

To further investigate whether the loadcell wallmeasurements are predictive and distinctive, additionalcalibration tests were performed. Two tests were donewith the trolley, equipped with a 400 mm deep deformableelement with a constant stiffness between 0.34 and0.4 MPa. In the first test, the trolley was driven into aflat rigid wall with an impact velocity of 35 km/h, perpen-dicular impact angle. In the second test, the trolley was

driven with the same test velocity and impact angle intoa rigid wall on which a rigid block with a depth of 100mm was mounted; see Figure 8. The block covers the fourcentral load cells of the HR-LCW (250 × 250 mm).

RESULTS

The following section describes the results of the simula-tion studies. Next the final trolley design is presented andfinally, the results of the calibration tests are given.

Simulation studies

A numerical model of the MDB was built in MADYMOand variation studies were performed with the MDB modelagainst different available vehicle models to study theperformance of the trolley and the influence of changesto trolley properties, such as mass and COG height. Theprogressive deformable barrier face (PDB) was used as de-formable element.

Mass variationsThe MDB mass was varied between 800 and 2000 kg andthe MDB was crashed into several vehicle models, amongstwhich was the multibody Chrysler Neon model developedat TNO Automotive [3]. The simulations were performedwith a closing speed of 106 km/h, which has a crash sever-ity comparable to the Neon against the fixed PDB with60 km/h and the Neon against the MPDB of 1500 kg. Re-sults of the sensitivity studies in terms of dissipated energyand deceleration of the MDB are presented in Figure 9.

As expected, the influence of a different MDB mass onthe dissipated energy by the vehicle can be considered aslinear. Further, it is shown that the lower the mass of theMDB, the higher is the deceleration pulse on the MDB.The maximum values differ between 200 and 450 m/s2.

Figure 7 MPDB DAS components.


Dow

nloa

ded

by [

Uni

vers

ity o

f B

irm

ingh

am]

at 1

0:22

11

Nov

embe

r 20

14


Figure 8 Rigid wall with rigid block of 250 × 250 × 100 mmmounted.

Note that the curves are filtered with a cfc60, thus peakvalues can be higher. The results of the simulation studiessupported the decision to make the mass of the trolleyadjustable for research purposes.

Centre of Gravity locationsThe CoG height of the trolley was varied between 500 mmand 750 mm, with steps of 50 mm. The performance of thetrolley with the extreme CoG heights against a multibodyNeon model is shown in Figure 10.

As illustrated in Figure 10, the trolley with a CoG of500 mm tends to “override” the crash opponent (with thisparticular struck vehicle), whereas the trolley with a CoGof 750 mm pitches. Hence the CoG height of the trolleyshould represent the current vehicle fleet as well as possible.

Trolley design

Based on the above mentioned specifications, several de-sign alternatives have been considered. The possibilityto adjust the mass and inertia properties of the trolleyresulted in additional requirements. The following trolleycharacteristics are listed:

– Steel frame with rigid axis and the wheels positioned out-side the frame. The baseline trolley has a mass of about1000 kg. Including the high resolution loadcell wall (de-sign mass about 300 kg), the total trolley weight is about1300 kg.

– Ballast masses of about 50 kg with a number of fixationlocations enable adaptation of the total trolley weight from1300 to 1800 kg with corresponding inertia properties.

– A ballast mass of 60 kg is installed to compensate for themass and inertia differences in case of switching from LHDto RHD testing. The height of this compensation mass isadjustable in steps of 100 mm, to be fixed on the nonusedloadcell wall fixation beam.

– The ground clearance of the mounted barrier face is default150 mm, but can be adjusted in steps of half the load cellsize (62.5 mm). It was decided to have a relatively smallground clearance of the deformable barrier face, 150 mm,in order to be able to assess low subframes and to be in linewith the Progressive Deformable Barrier protocol.

Pictures of the prototype trolley are given in Figure 11.The barrier is presented with the ballast masses installedin the default conditions with a 1500 kg test weight.

The CoG and inertia properties for the default designconditions were calculated using Autocad software to eval-uate the final trolley design. The results are summarizedin Table 5.

It is concluded that the prototype trolley meets the de-fined tolerances in the default configuration (trolley massof 1500 kg). Calculations for the extreme configurations,1300 kg and 1800 kg, show that the trolley is adjustablewithin the design tolerances, with the exception of theCoG height in the 1800 kg condition, which is about24.3 mm too high, whereas 15 mm was set as thetolerance.

Figure 9 Left panel: dissipated energy of the vehicle. Right panel: acceleration of MDB for MPDB to comp. Neon simulationswith trolley mass variation.


Dow

nloa

ded

by [

Uni

vers

ity o

f B

irm

ingh

am]

at 1

0:22

11

Nov

embe

r 20

14


Figure 10 Trolley CoG variation. Top panel: CoGtrolley = 500 mm, bottom panel: CoGtrolley = 750 mm.

Figure 11 Side and top view of the trolley.

Table 5 Target specifications against resulted specifications.

Target Calculation Tolerance Deviation

Mass 1500 kg 1499 kg ±20 kg 1 kgCOG X 1900 mm 1906.8 ±15 mm 6.8 mmCOG Y 0 mm −3 mm ±15 mm −3 mmCOG Z 600 mm 604.4 ±15 mm 4.44 mmRoll Iyz 550 kg/m2 352 kg/m2 ±50% −36.0%Pitch Ixz 2550 kg/m2 2425 kg/m2 ±10% −4.9%Yaw Ixy 2650 kg/m2 2535 kg/m2 ±10% −4.3%


Dow

nloa

ded

by [

Uni

vers

ity o

f B

irm

ingh

am]

at 1

0:22

11

Nov

embe

r 20

14


Figure 12 MPDB calibration test result.

Calibration tests

Calibration test with PDB barrier faceThe test conditions in the MPDB calibration test were asfollows:

• Trolley mass = 1500 kg• Barrier face: PDB V8 in LHS test position• Trolley velocity = 45 km/h

The comparison of the total recorded force, the summa-tion of the 48 load cells and the trolley mass times trolleyacceleration is presented in Figure 12. The trolley acceler-ation was measured in the trolley CoG.

In general, the curves of the acceleration and force mea-surements show a good correlation. The summation of theload cell wall force results in a lower total force, differencessmaller than ∼8%. The slight difference is most probablycaused by yaw and pitch of the trolley during impact. Thetrolley sustained the test without any problems.

Test with the trolley into different obstaclesThe test conditions were as follows:

• Trolley mass = 1500 kg• Barrier face: 400 mm depth honeycomb with a constant

stiffness between 0.34 and 0.40 MPa, backplate of 3 mm• Trolley velocity = 35 km/h• Two tests with different obstacles• 1 test against a flat rigid wall

• 1 test against a flat rigid wall on which a rigid block, 250 ×250 × 100 mm, is mounted

Pictures of the tests are shown in Figure 13.The comparison of the total recorded force by the load-

cell wall with the trolley mass times trolley accelerationis presented in Figure 14. The total recorded force is thesummation of the 48 load cells.

The acceleration and force measurements show in gen-eral a good correlation. Although the deformable face wasmade from material with a constant stiffness, the total forceis increasing after reaching the theoretical force plateau be-tween 5 and 10 ms. This is caused by locking of the barrierdue to air effects. It is noted that air locking or air inclu-sion is strongly related to the selected test conditions (fulloverlap, rigid wall).

Force against time output for all load cells is given inFigure 15, for the two different tests. The frame aroundthe load cells D and E in row 3 and 4 indicates the locationof the rigid block at the rigid wall in the test.

The loadcell wall clearly detects the block at the wall,visible by the different timing. Because of a slight mis-alignment and some spread of loads due to the backplate,also the load cells in the column next to the block observesome loading (e.g., see column F, row 3 and 4). Note thatthe lowest row of load cells was not fully covered withdeformable material and hence equivalent lower loads arerecorded by row 1.

Figure 13 Test set-up without block (left) and test set-up with block (right), just before impact.


Dow

nloa

ded

by [

Uni

vers

ity o

f B

irm

ingh

am]

at 1

0:22

11

Nov

embe

r 20

14


Figure 14 Trolley into the flat rigid wall (left) and trolley into the rigid block (right).

Figure 15 Forces (kN) in time for all loadcells, frame indicates the rigid block at the rigid wall.

DISCUSSION

This paper presented the development and design of a loadsensing trolley to be used in frontal offset testing, for useas a research tool in the field of compatibility assessment.The main specifications of the trolley were chosen to bein line with the current European vehicle fleet. Since norelevant European data was available on current vehicleinertia properties, these were chosen based on U.S. pas-senger car information and could hence differ slightly fromthe typical European situation.

In the calibration results the phenomenon of air lockingwas observed, causing a stiffer barrier face response thanexpected with the selected uniform honeycomb material.It is questioned how much influence this phenomenonwould have in full-scale barrier-to-car tests, since the crash

opponent will never be completely flat thus reducing thechance of locking, but this is outside the scope of the currentstudy.

The value of high resolution load cell wall measurementson a mobile trolley is not proven yet. A working prototypetrolley is now available, and more research is needed onfurther interpretation and feasibility of the high resolutionload cell wall measurements in this type of testing, particu-larly toward the spread of load due to the barrier backplateas observed in Figure 15.

CONCLUSION AND OUTLOOK

This paper presents an MDB trolley to be used infrontal offset testing that was developed with mass and


Dow

nloa

ded

by [

Uni

vers

ity o

f B

irm

ingh

am]

at 1

0:22

11

Nov

embe

r 20

14


inertia properties representative of the current vehi-cle fleet. Additionally, the trolley was equipped witha high resolution load sensing face. With the proto-type trolley, some calibration tests were performed andit is concluded that the developed trolley is a work-ing research tool. The interpretation of the high res-olution load cell wall results needs further investig-ations.

The future work with the trolley will concentrate onthe development and evaluation of a Mobile ProgressiveDeformable Barrier test procedure, based on the currentdraft PDB protocol. In the future work, the mass and inertia

adjustment possibilities of this research trolley could befurther used.

REFERENCES

1. http://www-nrd.nhtsa.dot.gov/database/nrd-11/veh db.html.Accessed 1 March 2006

2. Martin T., ‘Car geometrical/structural database and analysis ofcar to car geometric compatibility’, UTAC VC-CompatWP1-D9v1.1, June 2004.

3. C.D. v.d. Zweep, ‘Numerical fleet studies 2002–2003, Finalreport’, TNO Automotive, Delft, the Netherlands, TNOReport 03.OR.BV.067.1/CVDZ, 10–2003.


Dow

nloa

ded

by [

Uni

vers

ity o

f B

irm

ingh

am]

at 1

0:22

11

Nov

embe

r 20

14


Dow

nloa

ded

by [

Uni

vers

ity o

f B

irm

ingh

am]

at 1

0:22

11

Nov

embe

r 20

14

Documents

The development of a load sensing trolley for frontal offset testing