Wang X. Vehicle Noise and Vibration Refinement

Подождите немного. Документ загружается.

10 Vehicle noise and vibration reﬁ nement

• Analyse the test results and suggest solutions for the problems to the

relevant engineering party.

• Install the design solutions supplied by the relevant engineering party

into the vehicle and validate them under the same vehicle operating

conditions.

NVH engineers must recognize that there are many trade-offs between

NVH performance, engine power, fuel economy, development time, cost,

weight, etc. A compromise must be reached in the development process as

shown in Fig. 1.4.

Figure 1.5 shows the steeply increasing committed cost as the commit-

ment to tooling is made during the program. Therefore maximum utiliza-

tion of virtual testing to identify potential issues prior to the tooling stage

will signiﬁ cantly reduce development time and cost, although physical

testing of the subsystems and the complete vehicle is still required to ensure

that the ‘as manufactured’ vehicle consistently meets the performance

targets set.

In order to reduce vehicle development cost and time, an improved

vehicle development process is proposed as shown in Fig. 1.6 where a target

setting, cascading, synthesis and conﬁ rmation approach is applied to design

instead of the traditional design–build–test–redesign approach. This

approach facilitates the use of analytical prediction tools early in the

process, reducing the use of expensive physical prototype testing. It also

allows for design efforts to be shared across the automotive manufacturer

and supplier chain. The process can be implemented with a variety of tools

and speciﬁ c applications. The overall requirement is to maintain commu-

nication between the independent tasks as the results from each sub-

process become available. The activities in the arrowed process ﬂ ow are

Packaging space

Mass

Cost

Compromised solution

available

Noise & vibration requirement

1.4 Compromised solutions in the vehicle development process

(copyright RMIT University, 2008, Wang, X.).

Copyrighted Material downloaded from Woodhead Publishing Online

Delivered by http://woodhead.metapress.com

ETH Zuerich (307-97-768)

Sunday, August 28, 2011 12:00:55 AM

IP Address: 129.132.208.2

Rationale and history of vehicle noise and vibration reﬁ nement 11

positioned relative to time, with the initial target setting activity occurring

ﬁ rst. The intention is for the virtual validation to lead the physical validation

and thus accelerate the overall design process. The parallel ‘V’ image

indicates that both validation paths are required. The virtual path is the

optimization path that leads to the release of design elements requiring long

lead times such as component tooling. The physical path is the conﬁ rmation

path that must capture effects that cannot be modelled with certainty, such

as abrasion and wear, joint fatigue and progressive NVH degradation. The

image also shows that the product development process must be advanced

through continuous virtual and physical correlation.

The overall design engineering process begins with the initial vehicle

target setting as the ﬁ rst stage. The end customers and their demands for

speciﬁ c vehicle characteristics drive the basic requirements. This ‘voice of

the customer’ is combined with the existing internal knowledge and bench-

marks of competitive vehicles already in the marketplace, along with cost,

weight and performance targets to deﬁ ne the primary vehicle assumptions.

The initial assumptions create the top-level design goals for the complete

vehicle.

Target setting

Cascading

Synthesis

Design confirmation

Sub-system

validation

Quality

Product validation

Key

Committed costs

Analytical methods

Physical methods

Concept

Design

Tooling

Sub-system

manufacture

Vehicle manufacture

Production

1.5 Integration of physical and analytical methods (copyright Grote,

P. and Sharp, M., ‘Deﬁ ning the vehicle development process’, Keynote

Paper, Symposium on International Automotive Technology, 2001,

SAE).

Copyrighted Material downloaded from Woodhead Publishing Online

Delivered by http://woodhead.metapress.com

ETH Zuerich (307-97-768)

Sunday, August 28, 2011 12:00:55 AM

IP Address: 129.132.208.2

Corporate product

development

requirements

System targets

(e.g. suspension)

Component targets

(e.g. knuckle)

Cascade

design

requirements

Component analysis

System analysis Vehicle analysis

Vehicle tests

System tests

Design loads

Set targets for design

Synthesize

design

performance

Confirm

physical design

performance

Model correlation

Model correlation Model correlation

Vehicle targets

Lifecycle durability

NVH ratings

Structural durability

Stiffness, damping

Road/tyre/brake isolation

Sub-system weight

Structural durability

Attribute degradation

(alignment, NVH, etc.)

Strength

Fatigue life

Weight

Stress

Fatigue life

Weight optimization

Durability

K&C

Weight

Durability

NVH

Ride & handling crash

Loads confirmation

Structural (NVH)

Ride & handling

Crashworthiness

Component tests

Fatigue life

Crash performance

CAFE (weight)

Ride & handling

1.6 The emerging vehicle development model (copyright Grote, P. and Sharp, M., ‘Deﬁ ning the vehicle

development process’, Keynote Paper, Symposium on International Automotive Technology, 2001, SAE).

Copyrighted Material downloaded from Woodhead Publishing Online

Delivered by http://woodhead.metapress.com

ETH Zuerich (307-97-768)

Sunday, August 28, 2011 12:00:55 AM

IP Address: 129.132.208.2

Rationale and history of vehicle noise and vibration reﬁ nement 13

The second stage is target cascading which subdivides the top-level design

goals into system, subsystem and individual component goal levels. This

stage relies heavily on CAE tools to deﬁ ne all of the vehicle components

with individual target loads and constraints in a digital model. There may

be several iterations as the targets are interpreted for each layer of design.

There are many trade-offs to be determined with different criteria for

safety, performance, NVH, durability and other disciplines.

The third stage is synthesis where the designs for individual components

and sub-assemblies are completed using a variety of computer aided design

(CAD) tools. This stage utilizes CAE tools to generate virtual test results

which may require modiﬁ cation to the initial target values and thus addi-

tional trade-offs. These tools typically allow for analysis and prediction of

expected results that can be compared to the target data values previously

established. One goal of this stage is to identify these necessary trade-offs

before committing to speciﬁ c product design, thus accelerating the overall

design process. This stage is therefore referred to as ‘virtual testing’ as

various design levels can be tested with computer simulation methods prior

to the manufacture of any prototype physical parts. The sequence of the

virtual testing process is component, sub-assembly, sub-system and the full

vehicle. The design goals at each level are thus validated with corresponding

virtual tests at each level.

The fourth stage is conﬁ rmation by physical testing where prototype

parts, sub-systems and systems are subsequently evaluated and validated in

a similar sequence to the virtual testing process until the complete vehicle

is ready for ﬁ nal evaluation. Each activity within the physical testing stage

results in additional data that can be used to validate computer models.

The continuous feedback may require additional changes to the target

levels and vehicle design parameters.

The hybrid simulation is also used to set targets for vehicle development

in the ‘V’ approach. The process starts with full-vehicle performance targets

that are cascaded down to requirements for sub-systems (drivetrain, chassis,

suspension, etc.), and ﬁ nally to components (bushings, mounts, struts, etc.).

Hardware is then designed, built and assembled into a prototype vehicle in

the bottom part of the ‘V’ where physical testing usually leads to several

redesign cycles to iron out problems. In this ‘V’ approach, most car compa-

nies use simulation tools such as Finite Element Analysis (FEA) to help

speed the process after CAD has deﬁ ned the geometry of sub-systems,

assemblies and parts. By that time, important design decisions have been

made and considerable time and expense are required for any reconﬁ gura-

tion. This problem can be eliminated with function-driven design that

aims to accurately establish functional performance requirements through

target setting, much earlier in the process before the detailed design has

started. This eliminates the repetitive build–test–redesign cycles later in

Copyrighted Material downloaded from Woodhead Publishing Online

Delivered by http://woodhead.metapress.com

ETH Zuerich (307-97-768)

Sunday, August 28, 2011 12:00:55 AM

IP Address: 129.132.208.2

14 Vehicle noise and vibration reﬁ nement

development by performing analysis earlier with more system-level full-

vehicle simulation during the conceptual target-setting stage. By perform-

ing up-front engineering, the performance targets can be more accurately

established, and thus better strategic decisions on vehicle design can be

made. In this way, time and expense can be saved later in the bottom part

of the ‘V’, as fewer prototype testing cycles are required.

1.5 Vehicle noise and vibration term deﬁ nitions

Noise, vibration and harshness, also known as noise and vibration, abbrevi-

ated to NVH and N&V respectively, is the name given to the ﬁ eld of mea-

suring and modifying the noise and vibration characteristics of vehicles,

particularly cars and trucks. Harshness is somewhat of an historical misno-

mer. Noise and vibration can be measured, but harshness is a more subjec-

tive assessment. There is a psychoacoustic measurement called harshness

but it does not correlate very well with many harshness issues. Interior

NVH is the noise and vibration experienced by the occupants of the vehicle

cabin, while exterior NVH is largely concerned with the noise radiated by

the vehicle, and includes drive-by noise. The noise being generated by ﬂ uid

pressure ﬂ uctuation and passage through the air is called airborne noise.

The noise radiated from a structure’s surface that is vibrating is called

structure-borne noise. Noise is used here to describe audible sound, with

particular attention paid to the frequency range from 30 to 4000 Hz.

Vibration is used to describe tactile vibration, with particular attention paid

to the frequency range from 30 to 200 Hz.

1.6 History of motoring and vehicle reﬁ nement

It is frequently difﬁ cult to trace the earliest examples of automobiles. In

1885, Karl Benz invented a motorized tricycle in which the wheels were

made of timber and steel. Those who rode such a vehicle experienced bad

harshness. In 1888, John Dunlop invented air-ﬁ lled or pneumatic tyres. In

1904 Continental presented the world’s ﬁ rst automobile tyre with a pat-

terned tread. The wheel/road-induced noise and vibration were reduced by

the air-ﬁ lled or pneumatic tyres. Other vibration isolators such as rubber

bushes and engine mounts were also invented and introduced in vehicles

for the reduction of noise vibration harshness. Figure 1.7 shows a 1905

four-cylinder Tarrant with chain-driven rear wheels. In 1909, Henry Ford

launched his mass production method for the Model T which made cars

available to a large section of the public. The hard, tedious, repetitive work

created resentment, resulted in poor workmanship and quality and pro-

duced badly ﬁ nished, unreliable vehicles. Vehicle reﬁ nement became nec-

essary. The Volkswagen ‘Beetle’ had little reﬁ nement but many innovative

Copyrighted Material downloaded from Woodhead Publishing Online

Delivered by http://woodhead.metapress.com

ETH Zuerich (307-97-768)

Sunday, August 28, 2011 12:00:55 AM

IP Address: 129.132.208.2

Rationale and history of vehicle noise and vibration reﬁ nement 15

features. This vehicle was designed by Ferdinand Porsche in the late 1930s

at the behest of Adolf Hitler. By the 1970s, its styling was antiquated,

its air-cooled engine was noisy, yet it sold well throughout the world,

in particular in the USA. In fact, the metallic sound of the air-cooled

engine is not recognized as noise but accepted as one of the brand charac-

teristics of today’s Porsche cars. Today’s VW ‘Beetles’ have been well

reﬁ ned in all aspects, including noise vibration harshness. Vauxhall and

HSV are reﬁ ned vehicle brands for GM passenger cars, while Tick Ford

and FPV are reﬁ ned vehicle brands for Ford in which the driveability and

ride are upgraded. Lexus are reﬁ ned Toyota passenger cars in which noise

and vibration are greatly reduced. High series passenger cars are reﬁ ned

from low series ones by improving noise vibration harshness, driveability

and ride. The term ‘vehicle reﬁ nement’ is placed in the mind of the

customer as being a relevant factor in the decision making process of

buying a car.

The early vehicle noise and vibration reﬁ nement materials were grease,

motor oil, rubber bushes, washers, gaskets, springs, mass dampers, bolts

and nuts, cushions, earplugs and gloves; the early noise and vibration reﬁ ne-

ment tools were stereoscopes, screwdrivers, earphones, tape recorders, dial

gauges, balancers, water bulb levelling meters, hands, eyes and ears, which

are still practical for use and popular today. Engineers subjectively evalu-

ated vehicle noise and vibration performance and solved the NVH prob-

lems by traditional mechanical tools and methods.

1.7 The 14–16 horsepower four-cylinder Tarrant – the Melbourne-built

car that set an Australian 1000-mile (1600-km) speed record in 1905

(copyright Tuckey, B., Australians and Their Cars, Bondi Junction,

NSW: Focus, 2003).

Copyrighted Material downloaded from Woodhead Publishing Online

Delivered by http://woodhead.metapress.com

ETH Zuerich (307-97-768)

Sunday, August 28, 2011 12:00:55 AM

IP Address: 129.132.208.2

16 Vehicle noise and vibration reﬁ nement

In 1876, Emile Berliner, Elisha Gray and Alexander Graham Bell

invented the ﬁ rst microphone used as a telephone voice transmitter. The

microphone associated with the ﬁ rst articulate telephone transmitter was

the liquid transmitter of 1876. In 1938, Hans J. Meier at MIT was the ﬁ rst

person to construct a commercial strain gauge accelerometer. Magnetic

recording was conceived as early as 1877 by Oberlin Smith. The ﬁ rst wire

recorder was the Valdemar Poulsen Telegraphone of the late 1890s. Since

their ﬁ rst introduction, analogue tape recorders have experienced a long

series of progressive developments resulting in increased sound quality,

convenience and versatility, and Fig. 1.8 shows such a recorder from 1945.

Computer-controlled analogue tape recorders were introduced by Oscar

Bonello in Argentina as shown in Fig. 1.9 where the mechanical transport

used three DC motors and introduced two new advances: automated micro-

processor transport control and automatic adjustment of bias and frequency

response. In 30 seconds the recorder adjusted its bias and provided best

1.8 Peirce 55-B dictation wire recorder from 1945 (copyright

http://en.wikipedia.org/wiki/Wire_recorder).

1.9 Solidyne GMS200 tape recorder with computer self-adjustment,

Argentina, 1980–1990 (copyright http://en.wikipedia.org/wiki/

Tape_recorder).

Copyrighted Material downloaded from Woodhead Publishing Online

Delivered by http://woodhead.metapress.com

ETH Zuerich (307-97-768)

Sunday, August 28, 2011 12:00:55 AM

IP Address: 129.132.208.2

Rationale and history of vehicle noise and vibration reﬁ nement 17

frequency response to match the brand and batch of magnetic tape used.

The microprocessor control of transport allowed fast location of any point

on the tape. Around 1820, Charles Xavier Thomas created the ﬁ rst success-

ful, mass-produced mechanical calculator, the Thomas Arithmometer, that

could add, subtract, multiply and divide. It was based mainly on Leibniz’s

work. Mechanical calculators, like the base-10 addiator, the comptometer,

the Monroe, the Curta and the Addo-X, were used throughout the twent-

ieth century (e.g. see Fig. 1.10) right up until the 1970s. The IBM PC AT

286, 386 and 486 appeared in the early 1980s, then came PC Pentium tech-

nology. PC data acquisition and sound card technology have replaced

digital and analogue tape recording technology for recording and analysing

noise and vibration test data.

1.7 References and bibliography

Grote, P. and Sharp, M. (2001), ‘Deﬁ ning the vehicle development process’, Keynote

Paper, Symposium on International. Automotive Technology, SAE.

Happian-Smith, J. (2002), An Introduction to Modern Vehicle Design, SAE

International, Butterworth-Heinemann.

Harrison, M. (2004), Vehicle Reﬁ nement – Controlling Noise and Vibration in Road

Vehicles, SAE International, Elsevier Butterworth-Heinemann.

Tuckey, B. (2003), Australians and Their Cars, Bondi Junction, NSW: Focus.

Wang, X. (2005), Introduction to Motor Vehicle Design, RMIT Publisher.

1.10 A mechanical calculator from 1914. Note the lever used to rotate

the gears (copyright http://en.wikipedia.org/wiki/History_of_

computing_hardware).

Copyrighted Material downloaded from Woodhead Publishing Online

Delivered by http://woodhead.metapress.com

ETH Zuerich (307-97-768)

Sunday, August 28, 2011 12:00:55 AM

IP Address: 129.132.208.2

Target setting and benchmarking for

vehicle noise and vibration reﬁ nement

X. WANG, RMIT University, Australia

Abstract: In order to develop new vehicle products with well-reﬁ ned

noise and vibration performance, noise and vibration targets must be set

up. Market research helps to select the best-in-class competitors’ vehicles

and determine benchmark vehicles to be studied. Benchmark analysis

together with CAE modelling facilitates vehicle noise and vibration

target setting and target cascading. This chapter summarizes objectives,

signiﬁ cance and scope of vehicle noise and vibration target setting and

benchmarking. Examples are given to illustrate how to conduct vehicle

noise and vibration benchmarking, target setting and target cascading.

Key words: target setting, target cascading, benchmarking, interior noise

target, exterior noise target, subjective evaluation, objective testing,

whole-vehicle noise and vibration, components/subsystems noise and

vibration, sound pressure level, sound quality, sound power, articulate

index, statistical energy analysis, transfer path analysis.

2.1 Introduction

As mentioned in Chapter 1, in the ﬁ rst phase of a vehicle development

program, market analysis, benchmark study and target setting are impor-

tant tasks. The object of benchmark study is to determine the best-in-class

competitors. Requirement, design and performance constitute the three

stages of the vehicle development process as shown in Fig. 2.1. The purpose

of target setting is to establish design requirements. Vehicle targets are set

based on the benchmark study, the voice of the customer and business/

industry/government regulation, as shown in Fig. 2.2.

Market analysis determines which group of customers the vehicle is

targeting; customer wants from this group and competitors’ vehicles for

benchmark study are then determined. The competitors’ vehicles are ana-

lysed to determine the competitor best-in-class systems and subsystems;

overall vehicle speciﬁ cations and targets are then determined as shown in

Fig. 2.3.

Copyrighted Material downloaded from Woodhead Publishing Online

Delivered by http://woodhead.metapress.com

ETH Zuerich (307-97-768)

Sunday, August 28, 2011 12:00:57 AM

IP Address: 129.132.208.2

Target setting and benchmarking for vehicle noise 19

Requirement Design Performance

Voice of the

customer

Business/industry

regulation

Balance/constraints

BOM, mass, cost/ﬁnancial

Vehicle

technical req.

Vehicle

design

Vehicle

performance

Customer

acceptance

Subsystem

technical req.

Subsystem

design

Component

technical req.

Component

design

Subsystem

performance

Component

performance

2.1 Vehicle development process.

Customer

wants

Customer, corporate and regulatory goals

lead to mission specifications (targets)

for a vehicle

How do we derive

actionable design

targets from these

top-level

specifications?

Corporate

wants

Regulatory

musts

2.1.1 Objectives and signiﬁ cance of vehicle noise and

vibration target setting and benchmarking

Benchmarking allows for greater understanding of vehicle systems, subsys-

tems and components and their design targets. Apple-to-apple, back-to-

back comparisons give information about what are good designs, and what

are poor designs, and what are realistic design targets. The development

focuses are then better calibrated. The purpose of the benchmark study is

to identify the competitor best-in-class and facilitate noise and vibration

target setting for the vehicle system, subsystems and components.

2.2 Basic inputs for vehicle development target setting [2].

Copyrighted Material downloaded from Woodhead Publishing Online

Delivered by http://woodhead.metapress.com

ETH Zuerich (307-97-768)

Sunday, August 28, 2011 12:00:57 AM

IP Address: 129.132.208.2