Automotive Body Measurement

7/31/2019 Automotive Body Measurement

1/30

Automotive Body

Measurement SystemCapability

Examining the impact of

the measurement system

on dimensional evaluation

processes.

Auto/Steel Partnership


2/30

Automotive Body MeasuremeSystem Capability

Auto/Steel Partnership Program


3/30

Auto/Stee l Partnership

AK Steel CorporationBethlehem Steel Corporation

DaimlerChrysler CorporationDofasco Inc.

Ford Motor CompanyGeneral Motors Corporation

Ispat Inland Inc.LTV Steel Company

National Steel Corporation

Rouge Steel CompanyStelco Inc.U. S. Steel Group, a Unit of USX Corporation

WCI Steel, Inc.

Weirton Steel Corporation

This pub lication is for general information only. The material contained herein should not b ewithout first securing c ompetent advice with respect to its suitability for any g iven app licatio

pub lication is not intended as a representation or warranty on the part of The Auto/Steel Partneany other person named herein that the information is suitable for any general or particulaor free from infringement of any patent or patents. Anyone making use of the information as

all liability arising from such use.

This publication is intended for use by Auto/Steel Partnership memb ers only. For more informadd itional copies of this pub lication, please contac t the Auto/Steel Partnership, 2000 Town Cen

320 Southfield MI 48075 1123 or phone: 248 945 7777 fax: 248 356 8511 web site: www a


4/30

Table of Contents

Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Executive Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.0 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

2.0 Body Measurement Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

2.1 Measurement System Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2 Part Locating System (GD&T) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

3.0 Gage Capability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

3.1 Gage Capability for Checking Fixture Data . . . . . . . . . . . . . . . . . . . . . . . . 3.2 Gage Capability for CMM Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

4.0 Measurement System Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

4.1 Gage Error and Type of Part . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2 Gage Error and Dimensional Characteristics . . . . . . . . . . . . . . . . . . . . . . 4.3 Effect of Dimensioning and Part Locating System (GD&T) on Accuracy . .

4.3.1 Case Study I: Effect of Clamping Sequence . . . . . . . . . . . . . . 4.3.2 Case Study II: Effect of Add itional Clamping Locators . . . . . . .

4.4 Gage Variability and Part-to-Part Variation . . . . . . . . . . . . . . . . . . . . . . . .

5.0 The Effect of the Measurement System on Dimensional Evaluation Processes5.1 Gage Capability and Tolerances . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

5.2 Constrained versus Over constrained Clamping Systems . . . . . . . . . . . . .

6.0 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .


5/30

List of Figures

Figure 1. Measurement Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Figure 2. Body Coordinate System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Figure 3. The 3-2-1 Locating Scheme . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Figure 4. Number of Locator Clamps at Company C versus Company E . . . . . . . . . . . . .

Figure 5. Histogram of Gage Standard Deviation for Checking Fixtures . . . . . . . . . . . . . .

Figure 6. Distribution of % Gage R&R (Goal < 30%) . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Figure 7. Percent of Gage Variation Explained by Repeatability Error . . . . . . . . . . . . . . . .Figure 8. Static versus Dynamic CMM Gag e Repeatab ility Error . . . . . . . . . . . . . . . . . . . .

Figure 9. Distribution of Gage Error for Small Simple and Large Complex Parts . . . . . . . .

Figure 10. Correlation of Gage Error for Right and Left Coordinated Dimensions . . . . . . . .

Figure 11. Histogram of CMM Gage Variation for a One-Piece Body Side Outer Panel . . . .

Figure 12. High Gage Error vs. Datum Scheme . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Figure 13. Gage Error by Part Area . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Figure 14. Dimensional Measurements for an Inner Quarter Panel . . . . . . . . . . . . . . . . . . .

Figure 15. Differences in Mean and Variation for Alternate Clamping Sequenc e . . . . . . . . .

Figure 16. Effect of Clamping Sequence on Dimension #4 . . . . . . . . . . . . . . . . . . . . . . . . .

Figure 17. Body Side Conformance and Clamping Strategies . . . . . . . . . . . . . . . . . . . . . . .

Figure 18. Contribution of Gage Variation to Part-to-Part Variation . . . . . . . . . . . . . . . . . . . .


6/30

List of Tables

Table 1. Gage Variation by Manufacturer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Table 2. CMM vs. Checking Fixture Gage Repeatab ility for One-Piece Body Sides . .

Table 3. Mean and Variation Conformance by Clamping Approach . . . . . . . . . . . . . .

Table 4. Effect of Measurement Instrument on Mean Values: CMM vs. Feeler Gages .

Table 5. Effect of Measurement Instrument on Variation: CMM vs. Feeler Gage Data .

Table 6. Inherent Gage Error and Minimum Tolerance Requirements . . . . . . . . . . . . .


7/30

Preface

This report is one of a series published by theAuto/Steel Partnership Body Systems Analysis

Project Team on stamp ing and assembly variation,body measurement systems and process valida-tion. These reports provide a summary of the proj-ect research and are not intended to be all inclu-

sive of the research effort. Numerous seminarsand workshops have been given to individualautomotive manufacturers throughout the projectto aid in imp lementation and provide d irect techni-

cal support. Proprietary observations and imple-mentation details are omitted from the reports.

This automotive body development report,Automotive Body Measurement SystemCapability, updates ongoing research activitiesby the Body Systems Analysis Project Team and

the Manufacturing Systems staff at The University

of Michigans Office for the Study of AutomotiveTransportation. The purpose of this report is to

quantify the capability of various body measure-ment systems and to examine the impact of themeasurement system on dimensional evaluationprocesses.

A primary goal of this research is to develop new

paradigms that will drive automotive body-in-whitedevelopment and manufacture towards a total

optimized processing system. Previous reportsdescribed fundamental research investigatingsimultaneous development systems for designing,

tooling and assembling bodies, and also flexiblebody assembly. Since the inception of thisresearch program, considerable emphasis has

been focused on dimensional validation of auto-motive body components. A major factor in thedimensional validation process is the role of themeasurement system.

The researchers are indebted to several global

Company, General Motors CorpNUMMI, Opel and Renault. Ea

experiments under production coing hundreds of hours of effort, oft

commitment of numerous productiengineering personnel. Although

to mention each one of these indoffer our sincere appreciation.

The reports represent a culmina

years of effort by the Body SyProject Team. Team membersh

evolved over the course of this pro

J. Aube, General Motors CorporatH. Bell, General Motors CorporatioC. Butche, General Motors Corpor

G. Crisp, DaimlerChrysler CorporaT. Diewald, Auto/Steel PartnershipK. Goff, Jr., Ford Motor Company

T. Gonzales, National Steel CorpoR. Haan, General Motors CorporatS. Johnson, DaimlerChrysler CorpF. Keith, Ford Motor Comp any

T. Mancewicz, General Motors CoJ. Naysmith, Ronart IndustriesJ. Noel, Auto/Steel PartnershipP. Peterson, USX

R. Pierson, General Motors CorpoR. Rekolt, DaimlerChrysler CorporM. Rumel, Auto/Steel PartnershipM. Schmidt, Atlas Tool and Die

The University of Michigan Research Institute conducted

research and wrote the final repor

research team from the ManufacGroup was:

Patrick Hammett, Ph.D.(734-936-1121/[email protected] Baron, Ph.D.(734 764 4704/jaybaron@umich ed


8/30

Executive Summary

The Auto/Steel Partnership (A/SP) is an innova-tive international association that includes

DaimlerChrysler, Ford, General Motors and elevenNorth American sheet steel producers. ThePartnership was formed in 1987 to leverage theresources of the automotive and steel industries to

pursue research projects leading to excellence inthe application of sheet steels in the design andmanufacture of vehicles. The Partnership hasestablished project teams that examine issues

related to steel properties including strength, dentresistance, surface texture and coating weights,as well as manufacturing methods, including

stamping, welding and design improvements.

This automotive body development report up datesongoing research activities by the A/SP Body

Systems Analysis Project Team and the

Manufacturing Systems staff at The University ofMichigans Office for the Study of Automotive

Transportation. The purpose of the study is toquantify the capability of various body measure-ment systems and then to examine the impact ofthe measurement systems on dimensional evalua-

tion processes.

In the automotive industry, the role of sheet metalmeasurement systems is critical and costly

mistakes can result from poor gage designs andmisinterpretation of data. The two most commonsheet metal measurement technologies, hard

gages and coordinate measuring machines, bothtactile (CMM) and optical (OCMM), are usedextensively for die buyoff, process validation and

process control monitoring. The first step prior tousing the measurement system is to verify therepeatability and reproducibility (R&R) of the sys-tem and, to determine accuracy.

Achieving acceptable gage R&R for large, non-

CMMs, acc ounting for about 8R&R variation respectively. Tthumb that gage R&R account

the tolerance is a major fac

tolerances, check point locatfixture design, particularly for

rigid. In order to comply with check points on non-rigid partsmum tolerances of + /- 0.75 mmrigid parts. Because gage R

significant portion of the tolerafixtures, especially hard gages

at detecting process mean control than they are at idenprocess variation.

Since non-rigid panels deflect

sure and from their own weigtors and c lamps are often usederence plane once the panel

checking fixture. The use of mlocators provides both an dilemma. The problem of ovefor measurement is that the c

torts the part and introduces slem when measuring over-consthey can be held as they woassembly process, and there

ment system can help anticipadatum or clamping sequenceorder to shift variation to areasnot be as critical as the inte

mating flanges, for example. Dactual process variation and mracy and focusing attention

assemble is consistent with a

losophy.

The functional build philosophy

advocates that the measuremthe assembly of the part with and holding clamps. Areas measurements are concent


9/30

desired location, and variation transferred to othernon-critical areas of the part, both in the measure-ment fixture and in the assembly fixture. Although

the measurement locations focus on the ability to

assemble parts, over-stressing of panels must beminimized. The ideal functional build fixture

minimizes the amount of over-conssufficient constraints so that paunloading results in consistent a

with minimal inherent stress.


10/30

1.0 Introduction

To evaluate automotive body quality, NorthAmerican manufacturers are incorporating more

data-based decisions to replace subjective opin-ions. Inherent in to this approach, however, is anunderstanding of the quality of the data collected,and hence the effectiveness of the measurement

systems used. This report assesses the strengthsand limitations of automotive body measurementsystems and considers their impact on dimension-al evaluation strategies.

An ideal measurement system produces resultsthat agree exactly with a master standard.

Unfortunately, measurement systems with suchproperties are rare. These systems routinely pro-duce data with measurement biases and variation.Measurement biases are deviations between

measured values and the true values obtained by

using more precise measuring equipment.Measurement variation relates to the inability to

obtain the same value for repeated measurementsof the same part. Automotive manufacturers typi-cally evaluate the impact of measurement systemvariation using gage capability studies, gage

repeatability and reproducibility studies, and otheranalysis methods outlined in the Measurement

Systems Analysis reference manual(1) publishedby the Automotive Industry Action Group (AIAG).

The measurement system plays a critical role inany dimensional evaluation process. In the case of

the automotive body, its role is particularly influen-tial. Body manufacturers measure most partfeatures in absolute space using X, Y, and Z

coordinates rather than as relative distancesbetween points. Absolute space measurementsare more complex, particularly for angledsurfaces. They are also heavily dependent uponthe part locating system, or datum scheme, which

often is difficult for parts lacking rigidity.

system effects limit the abildimensional problems, as somare attributable to the part loc

than the stamping die.

The purpose of this report is to

bility and limitations of the varment systems, including hardand coordinate measuring mporting data are based primari

noted manufacturers

In this report, the various meused in automotive body described first. Section 3 provigage capability for the most wifixture and coordinate meas

Section 4 examines sources compares gage variation andvariation. Section 5 considers

measurement system measurement strategies. Tmeasurement system on the tolerances is examined, alon

over-constrained fixtures to detail components.

This report will show that altho

ment systems typically have serror, they have limitations in ment biases. The lack of rigidirequires manufacturers to vio

locating principles. Althougprinciples by adding secondagage variation, it also crea

biases. In other words, the loca

in measurement fixtures may npositioning in assembly tools. measurement biases is that ma

not simply evaluate a part charits gage readings, but also in rprocesses.


11/30

2.0 Body Measurement Systems

2.1 Measurement System Applications

The most widely used systems to measure auto-motive bodies and their stamped c omponents are

checking fixtures, often called hard gages, andcoordinate measuring machines, the CMMs.CMMs may either be mechanical or optical.Mechanical CMMs are usually stationary, that is,

fixed plates, although portable CMM systems areseeing increased usage. Figure 1 below illustrates

a checking fixture and a stationary coordinatemeasurement machine.

The use of a particular measudepends largely on app lication anobjectives. Typically, manufactur

nate measuring machines for l

parts requiring numerous dimensithe coordinate measuring mach

systems in environmentally contrthe most common, and are consiaccurate and repeatable. Other bflexibility, in terms of adding dime

and that they may be operated uprograms, thereby reducing the

urement personnel to be present.

Portable CMMs are even more

stationary CMMs because addichecks does not require programcan be moved to the process. Thismanufacturers to use these syste

solving during stamping tryout. Soers also use them on the shop flassembly-tooling locators. The pr

with portable CMMs is their limitating the exact location of a paracross a large samp le of parts. Theoperator intensive. Thus, portable

primarily to measure only one or tw

Another type of coordinate meacommon in body manufacturingversion (OCMM). Typically, these

used for on-line measurement reduced cycle time allows themproduction speeds. These on-linereal-time, 100% inspection of bo

sub-assemblies possible. OCMMsmaterial handling problems that reporting large, complex-shaped aspecial CMM inspection room. On

OCMMs, however, is their accurament bias. Manufacturers often pro

Body Side - Check Fixture

Body Side - CMM


12/30

with OCMMs is part locating. Some OCMM usersalign parts mathematically by measuring locatorholes and surfaces. They then reference part char-

acteristics to this datum scheme. Unfortunately,

part measurements based on mathematical align-ment often differ from fixture measurements due to

problems created b y locator hole distortions, partmovement during clamping in fixtures or theeffects of gravity.

Although coordinate measurement systems offer

tremendous flexibility and data collection efficien-cy, they often are not used for process control inpress shops. Generally, OCMMs are consideredtoo expensive and impractical for widespread use

in stamping. CMMs often are considered impracti-cal for smaller stamped parts with few dimensionsbecause of their long processing times. CMM pro-cessing time includes transportation to a special

inspection room, wait time for a measuringmachine to become available, set up time, andmachine cycle time. Long CMM processing times

delay feedback of measurement information whichimpairs process control effectiveness.

Most manufacturers rely on hard checking fixturesto measure stamped p arts for process control. The

principal advantage of checking fixtures is thatmanufacturers can locate them near a press or asub-assembly line, thus providing quick feedbackon process performance. The principal concerns

for manufacturers using checking fixtures are costand measurement capability. Checking fixturesgenerally cost more than CMM holding fixturesbecause manufacturers have to mount checking

rails and data collection bushings at dimensionallocations. In terms of gage capability, checkingfixtures generally are considered less accurateand repeatable than coordinate measurement sys-

tems. This capability generalization will be exam-ined further in the next section.

2.2 Part Locating System

One of the main componentssystem is the part reference

Regardless of the measuremenly all part measurements aredatum scheme describedDimensioning and Tolerancing

These datum schemes providefor all part surfaces and featurdinates. Figure 2 below illustrcoord inate system. This system

tional X, Y, and Z directiona

fore/aft (X), in/out (Y), and up /dThe 0,0,0 point of the car is the

position.

Figure 2. Body Coordin

Holding fixtures used in meas

assembly operations often follscheme to position parts. Uthree locators position a part indirection. Two locators then

Y AXIS RIGHTI/O - R (In/Out) Right

C/L (Centerline)C/C (CrossCar)+ Positive on Right Side

Y AXIS LEFTI/O - L (In/Out) Left

C/L (Centerline)C/C (CrossCar)

- Value entered asNegative on Left Side

0,0,0


13/30

round pins, one fitting a c ircular hole and the othera slot. The pin locates the part in two directions,in/out and fore/aft. The slot then becomes the

other locator for the secondary dirbelow is a schematic representatprinciple using the hole/slot comb

The lack of rigidity for many stamped componentsand assemblies often forces manufacturers to vio-

late the 3-2-1 locating scheme and use additionallocators to position parts in a stable and repeat-

able manner. As a result, the locating scheme forsheet metal is sometimes referred to as n-2-1. The

n denotes the three or more locators needed toposition a part in a primary plane. The number of

additional constraints may vary gmanufacturers. For example, F

shows a similar body side outer two manufacturers. Company C

locators in the in/out direction whas twenty. The effects of differe

schemes on gage error are examquent sections.

Figure 3. The 3-2-1 Locating Scheme

Up / Down

In / Out

Fore /

In / OFore

Clamp

Clamp

Up / Down

Hole

Clamp

Up / Down

Slot

In / Out

Up / Down

Pin

ClampsDetail Fix.


14/30ensions

20%

15%

10%

Median = 0.0395th Percentile = 0.09

3.0 Gage Capability

3.1 Gage Capability for Check Fixture Data

Measurement systems are subject to variation andtherefore, dimensional analysis of a process first,

requires an evaluation of gage capability. Mostmanufacturers evaluate capability using gageR&R studies. gage repeatability refers to the vari-ation in measurements obtained when one opera-

tor uses the same gage for measuring identicalcharacteristics of the same parts. Gage repro-ducibility refers to the variation in the average of

measurements made by different operators usingthe same gage to measure identical characteris-tics of the same parts. The total gage variation,Equation 1, is based on repeatability and repro-

ducibility. To compute the capability of a measur-ing device, manufacturers typically compare therange of gage variation, estimated by 5.15 x gage,to the tolerances, Equation 2.

Equation 1Total Gage Variation: gage= 2

Equation 2

% Gage Capability (Gage R&R

To assess gag e cap ability, the typically uses a 30% rule. Thisrange of gage variation must bthe total tolerance for a part

R&R < 30%. For instance, if thepart characteristic is +/- 0

standard deviation must be l(30% x 1.4 / 5.15 < .08).

Table 1 and Figure 5 belowvariation across several pa

studies. Overall, these studies facturers achieve similar levelNote that although Case Study

percentile value, it exhibitedsigma gage. Since this casesignificantly more parts, it likeestimate of the distribution of g

Table 1. Gage Variation by Manufacturer

Case # Parts / Median 95

Study (# Dimensions) gage

I 4 (34) 0.04

II 61 (428) 0.03

III 12 (309) 0.03

All 77 (771) 0.03


15/30

Figure 6 below shows the distribution of gagecapability for the 700 part dimensions presentedin Figure 5. Over 90% of the dimensions exhibited

a gage R&R less than 30%. In addition, more than

50% of the dimensions had a gage R&R less than

10%. Although the inherent gage ally acceptab le, a small percentagstill have gage error concerns.

section discusses why certain d

larger gage variation.

%o

fDimen

sions

% Gage R&R

10%

0%

< 10% 10-20% 20-30% 30-40% 40%

20%

30%

40%

50%

60%

Median = 0.0395th Percentile = 0.09

Figure 6. Distribution of % Gage R&R (Goal < 30%)

The next step is to determine which of the twocomponents of gage variation, repeatability or

reproducibility, account for the greater proportionof the gage variance. The data in Figure 7 below,

based on Case Study III, suggest that nearly85% of the observed gage error may be attributedto repeatability. The principal cause of this gage

repeatability error relates to the loaof parts in the fixture and not the

measurement probe. Once a partfixture, measurement probes are q

with repeatability less than 0.01 mm unloading between measurement

Dimensions

30%

40%

50%

60%70%

gage

= .03

= .01

2 2= 85%

AVG

AVG

AVG repeatability

reproducibility

repeatability


16/30

% of CMM Explained by Setup (load/unload

%

ofDimension

0%

20%

40%

60%

80%

100%


17/30

The next step is to compare the repeatability of aCMM to that of checking fixtures or hard gages.Unfortunately, data based on identical parts and

holding fixtures are not available to make these

comparisons. Thus, the following analysis repre-sents a general comparison of measurement sys-

tems. Table 2 below summarizes the repeatabilityfor integrated or one-piece body side outer panelsat four manufacturers. These data indicate that forsimilar clamping strategies, gage repeatability

error appears only slightly better for a CMM. Thisresult is not surprising given that gage variation

relates primarily to the load/unload operation and

not the static repeatability of thprobe. These data also suggest over-constrained holding fixtures

a greater influence than the

technology in terms of reducing example, although companies A

similar one-piece body side designed quarter panels, the CMM gaerror at company B is higher thanOne explanation is that com

significantly less clamps in theifixture, 5 versus 11 cross-car clam

Table 2. CMM vs. Checking Fixture Gage Repeatability for One-Piece Body Sides

(Note: Body Side for company C in this table is different than in prior tables)

Measurement # Cross Car Median 95tCompany System Clamps repeatability

A CMM 11 0.04

B CMM 5 0.04

C Check Fixture 10 0.05

G CMM 17 0.01


18/30

4.0 Measurement System Analysis

A fundamental question in evaluating measure-ment systems is whether to separate the analysis

of the gage from the part characteristics. Somemanufacturers maintain that evaluating gagecapability should be independent of the part fea-tures. Here, manufacturers use a subset of part

characteristics to evaluate gage capability.Unfortunately, the distribution of g age error shownpreviously does not support this strategy. In thefollowing sub-sections, several issues are identi-

fied that affect gage error and measurement bias-es.

4.1 Gage Error and Type of P

Figure 9 b elow compares the dvariability of large/complex

parts. Large/complex parts te

distribution of gage variation parts. One reason for this differthe measurement system cla

dimensional characteristics. typically may be c onstrained u1 approach. In this situation, tdoes not appear to significantl

For large/complex parts, howev

tional clamps (n-2-1) can sgage error in certain localized

For example, dimensions in sdue to proximity to locator cllocalized part area, often have than non-stable areas. These i

in the next sub-section.

%o

fD

imensions

0%

10%

20%

30%

40%

50%

60%

70%

0.03 0.06 0.09 >0.09

-90% of Small/Simple

-66% of Large/Complex

Small/Simple Large/Complex

< 0.06

< 0.06gage

gage

gage

Figure 9. Distribution of Gage Error for Small/Simple and Large/Complex Parts

4.2 Gage Error and DimensionalCharacteristics

then low correlation between

dimensions might be expected


19/30

Even within a single part, considerable variabilityin gag e error may exist. Figure 11 below shows thegage error distribution for a relatively uncon-

strained body side of company B. This histogram

suggests that gage error across a pendent of the part characteristured.

LH

0.00

0.00

0.02

0.04

0.06

0.080.10

0.12

0.02 0.04

Correlation, R = 0.75

0.06 0.08 0.10 0.12

gage

RH gage

Figure 10. Correlation of Gage Error for Right and Left Coordinated Dimensions

%

ofDimensions

0%

5%

10%

15%

20%

25%

0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 0.1 >.1

Gage Repeatability ( )repeatability

Median

95th Percentile

= .04

= .13repeatability

repeatability


20/30

To further explore this lack of independence,Figure 12 below illustrates high and low gage vari-ation areas for a body side outer in relation to part

locating clamps. Again, one p redominant theory to

explain why certain dimensions have higher gagevariation is lack of part constraint in certain local

regions and not the measurement technology. Forthis body side, those areas located close toclamps are well constrained and exhibit low sigma

gage measurements ranging frcontrast, areas of the part strained exhibit significantly

measurements, as high as 0.1

straint and the resulting gage certain areas of the part that

strained to also exhibit higher

Figure 12. High Gage Error vs. Datum Scheme

(Clamps designated by )

Sigma Gage~0.10

Sigma Gage~0.02

Sigma Gage~0.10Sigma Gage~0.03

Sigma Gage~0.10 Sigma Gage~0.15

S

S

Sigma Gage

Figure 13 on page 14 illustrates gage error and

localized part rigidity for another body side outerpanel. This figure also suggests that gage error isnot independent of the part characteristic being

ment areas in the quarter pan

Because of this greater gage vrigid areas of large/complex pgreater part variation than th


21/30

Figure 13. Gage Error by Part Area

Sigma Gage~0.050.07

4.3 Effect of Dimensioning and Part LocatingSystem (GD&T) on Accuracy

In addition to repeatability and reproducibility,whether for a CMM or chec k fixture, manufacturersshould also evaluate dimensional measurement

biases. For instance, they should examine whetherthe observed measurement mean biases accu-rately reflect their true means. Manufac turers often

use additional clamps beyond the 3-2-1 locatingscheme to hold a panel in a stable position. Asmentioned previously, this approach can introducemeasurement biases for certain dimensions. This

bias is the deviation of the observed mean fromits true mean.

Traditionally, manufacturers assess the true meanof a part characteristic by using more precisemeasurement equipment. However, given theunique influence of the part locating system on

stamping measurements it is recommended that

teristic, such as a mating flange, ably. Inconsistencies between asand detail measurement fixtures rancies between measurement dat

tioning at time of assembly.

Many of these discrepancies aredatum schemes. Part holding fixtutional clamps beyond 3-2-1 often cment biases by temporarily bend i

clamping. This bending may sheither toward or away from its taspecification. Thus, the observe

dimension can reflect its actual poture effect.

Similarly, the observed part var

include a fixture effect. Non-rigid sponents typically conform to their

Sigma Gage~0.050.07

Sigma Gage~0.050.0

Sigma Gage~0.020.03


22/30

When the locating system of a fixture affects boththe observed mean and variation of non-rigid partdimensions, it becomes an active part of the

measurement system. This contrasts with a pas-

sive measurement system where dimensionalmeans are not dependent on the checking fixture

locating scheme or clamping sequence. Forinstance, if a manufacturer measures the relativedistance between two features, the actual locatingscheme may become less critical if the fixture is

not deforming the part.

Two case studies are presented, showing theeffects of the part locating system on measure-ment biases and variation.

4.3.1 Case Study I: Effect of ClampingSequence

In Case Study I, the effect of clamping sequence

on gage error for a quarter inner panel was con-sidered. This experiment studied the effect of

altering the clamping sequence by changing theorder of the last three clamps (see Figure 14,

below). In the second sequendimension #4 was engaged belocated next to dimensions #

pose of this second sequence

variation in d imension #4 mighit has the smallest assigned t

sions #1, #3, and #4. The sammeasured for each clamping s

Figure 15 on page 16 summarin the mean and variability for

using the two clamping sequenand standard d eviation for dimby altering the clamping seqFigure 16, also on page 16decrease in the standard dev

for this point increased. The however, did result in incredimensions 1 and 3, but these

tolerances than # 4. The mean dimensions did not change forclamps 6 and 7, where the altered.

Clamp

Pt. 6

Pt. 4Pt. 3

Pt. 1

Clamp


23/30

This study shows that by changing the clampingsequences, manufacturers can shift variation toless critical areas without actually changing the

part. It also confirms the widely held belief that

clamping sequence affects d imensional measure-ments. For non-rigid parts, manufacturers can pro-

duce different estimates for dimensional meansand variation depending upon the clampingsequence. The ramifications of these findings aresignificant. Since clamping in assembly tooling

typically occurs simultaneously manually in measurement holdingfacturers must accept some pot

ment biases and variation inconsi

stamping data. They should exbefore reworking or adjusting a p

nominal because of the potential laship between measurement datationing at time of assembly.

Deviationfrom

Nominal(mm)

Check Point

Original Alternate

1.25

1.00

0.75

0.50

0.25

0.001 3 4 6 7 8 S

tandardDeviation

(mm)

Check Point

Original st. dev. Alternate st. d

0.20

0.15

0.10

0.05

0.001 3 4 6 7 8

Figure 15. Differences in Mean and Variation for Alternate Clamping Sequence

0.20

0.40

0.60

0.80

1.00

1.20

1.40

Clamp last Clamp first

DeviationfromN

omina

l(mm)

.20 mm

.45 mm


24/30

4.3.2 Case Study II: Effect of AdditionalClamping Locators

Case Study II compared constrained versus over-

constrained clamping strategies. Figure 17 below

illustrates ten dimensions on a body side innerpanel and the location of two sets of clamps. The

constrained system uses 9 and the over-constrained systexperiment, ten body sides we

both a CMM and a feeler gage

ing systems.

Figure 17. Body Side Conformance and Clamping Strategies

P7

Over-Constrained (17 C/C Clamps

Constrained (9 C/C Clamps)

P6

P5 P4P3

P2

P8

P9

P10

Table 3 on page 18 indicates that the use ofadditional clamps may significantly shift meandimensions and reduce variation. In this study,

three of the ten dimensions shifted more than0.5 mm. Interestingly, these mean shifts were notalways toward nominal. One d imension, P10, shift-

clamping, but to q uestion the assess mean deviations.

This experiment also indicates reductions for several dimensiconstrained clamping system


25/30

Table 4 below compares mean dimensional meas-urements between CMM data and feeler gagedata using both constrained and over-constrained

systems. These data suggest that the CMM had asignificant effect on mean values. Four dimensionsshifted over 0.5 mm between the CMM data and

the feeler gage data. Furthermore, in all caseswhere dimensions shifted, the CMM mean dimen-

sions had greater mean deviations than the feeler

gage data. Table 5 below examinimpac t of the measurement gage oresults of this analysis are mixed.

show significant reductions usingalthough the overall observed pdoes not differ significantly betwee

instruments.

Table 3. Mean and Variation Conformance by Clamping Approach

Average Deviation from Nominal (mm) by Panel Dimension

P1 P2 P3 P4 P5 P6 P7 P8 P9 P

Constrain (9 clamps) -0.54 -0.96 -0.46 0.09 0.10 -0.29 0.70 -0.06 -0.74 0Over-Constrain (17 clamps) -0.20 -0.45 0.15 0.38 0.43 -0.23 0.67 -0.09 -0.55 1

Mean Difference 0.34 0.51 0.61 0.29 0.33 0.06 0.03 0.03 0.19 1

Standard Deviation (mm) by Panel Dimension

P1 P2 P3 P4 P5 P6 P7 P8 P9 P

Constrain (9 clamps) 0.23 0.21 0.19 0.18 0.21 0.16 0.31 0.09 0.15 0Over-Constrain (17 clamps) 0.08 0.03 0.14 0.14 0.25 0.07 0.20 0.17 0.06 0

Statistical Difference?(based F-test, =.05) Dec Dec Dec Dec

Table 4. Effect of Measurement Instrument on Mean Values: CMM vs. Feeler Gages

Average Deviation from Nominal (mm) by Panel Dimension

P1 P2 P3 P4 P5 P6 P7 P8 P9 P

CMM (17 clamps) -0.36 -0.47 -0.76 0.07 -0.08 -0.19 0.61 0.17 -0.49 1

Feeler (17 clamps) -0.04 -0.13 0.18 0.20 0.00 -0.10 0.14 0.04 -0.34 -0

Mean Difference 0.32 0.60 0.58 0.13 0.08 0.09 0.47 0.13 0.15 1


26/30

In practice, manufacturers try to maintain consis-tent locating schemes and clamping sequencesbetween checking fixtures and assembly tooling.

This consistency is needed to obtain measure-

ments that are valid or representative of stampingquality. Maintaining this consistency, however, is

not always feasible. First, many assembly opera-tions use only a subset of the measurement sys-tem locators. Second, when automating assemblyoperations, manufacturers may have to change

the position of datum locators. The lack of consis-tency between locating schemes and clamping

sequences may result in observed measurementsfor stamped parts that are not reflective of theirpositioning in assembly tooling. This has led somemanufacturers to wait until after an assembly eval-uation before altering stamped parts, or employ a

functional build app roach.

Due to the limitations with measuring non-rigid

parts, observed mean deviations may not indicatea problem with a set of dies or a press line.Therefore, manufacturers using a traditional build-to-nominal approach may rework dies unneces-

sarily to correct deviations that result from meas-urement system problems. This research is notsuggesting that all deviations from nominal are aresult of measurement problems, but rather that

approving a stamped part for production is morecomplex than simply comparing individual partmeasurements to design specifications. In manycases, manufacturers must wait until after a part

becomes more rigid in sub-assemblies beforedeciding on whether observed stampingdimensional measurements are reflective of

body quality.

4.4 Gage Variability and Part-to-Part Variation

Most manufacturers conduct gage R&R studies to

verify the capability of their measuring instru-ments. However, they should also consider the

common mathematical relabetween the observed variatio

tion.

Equation 42

observed =

Observed variation is the variaple of parts, a sample stObserved variation may be

measurement system and trueEquation 5 estimates the contrability to the observed process

Equation 5%Gage Contribution =

2g

As the gage error represents aof the observed variation, the cgage is unable to separate pro

that of the gage. The significan

bution is that little value is gment. In other words, if the gequal to short term part-to-part

information is gained by actualal parts over a short run.

Figure 18 on page 20 compare

contribution to the part-to-part part-part, for over 450 part dime

standard deviation is compute

ple of at least 50 panels fromrun. This figure shows that wsmall, or part-part < 0.15 mm, can explain a large portion of

ability. Of the dimensions withdard deviation less than 0.15 a gage contribution over 50%

sions, the usefulness of checkiguishing part-to-part variation error is questionable. This figuwhen part-to-part standard

part-part > 0.30 mm, the gage than 20% This indicates that c


27/30

Understanding the effects of gage error has impli-

cations for determining the number of panels tosample for a tryout or production run. For moststamping dimensions, the short term part-to-part

variation is low, pp < 0.15 mm, and thus, measur-ing large samples of panels from a single run

typically yields minimal value due

separate product and gage variatdoes not suggest, however, tstamped p arts is non-value add ed

measuring large samples over shoin a single run yields minimal value

Figure 18. Contribution of Gage Variation to Part-to-Part Variation

0.00 0.10 0.20 0.30 0.40 0.50 0.60

Part-to-part standard deviation (mm)

100%

80%

60%

40%

20%

0%

Gage not separatingproduct variation fromgage variation

%(

GageVariance/

Part-to-PartVariance

)


28/30

5.0 The Effect of the MeasurementSystem on DimensionalEvaluation Processes

5.1 Gage Capability and Tolerances

One effect of the gage error distribution acrossa part relates to the assigning of dimensionaltolerances. This research suggest that gage may

range from 0.01 to 0.09 mm (median = 0.03)depending on the part characteristic. Table 6below derives minimum tolerance requirements,

given this range of inherent gage variation, inorder to meet a 30% gage error/tolerance ratio.This analysis suggests minimum tolerances of +/-0.3 to +/- 0.75 are needed to meet these gageR&R requirements. Less stable measurement

areas on a part would require the larger minimumtolerances of +/- 0.75 mm.

Table 6. Inherent Gage Error and Minimum ToleranceRequirements

5.2 Constrained versus Over-ConstrainedClamping Systems

A major difference among manufacturers is theiruse of secondary locator clamps for larger non-

rigid parts. Some manufacturers use nearly twiceas many clamps as others for similarly designed

body side panels. This finding sugg ests two clear-ly different strategies. On the one hand, somemanufacturers try to minimize the number of sec-ondary locator clamps to reduce their potential

as long as similar clamping strassembly tools and that part not over-stressing the part

approach is referred to as

because some of the seconadded even though they are

meet gage capability requireondary clamps are used to cment during assembly and added to component part hold

late this movement and madatum schemes. Note that m

these over-constrained systemparts in a free state, prior to einsure that no p art areas are o

One question raised by this

approach is better. The benefconstrained system is typicapart and gage variation. In somtional secondary locators will m

ation, allowing assignment oances. Not surprisingly, of theers in the body side experime

facturers with the tightest tolerconstrained measurement sysdrawback of an over-constraiadditional locators may advers

the part holding fixture. Part may significantly shift due tHistorically, some of these shifnominal, but others may be fu

ing upon the relationship betwtion and the area of the part be

In contrast to the over-constraitain manufacturers seek a m

secondary locators. The princapproach is that manufacturer

tial to over-stress parts during

Determining whether to constrastrain means recognizing cert

Gage Error Minimum Tolerancegage (Gage R&R > 30%)

0.03 +/- 0.3 mm

0.05 +/- 0.45 mm

0.07 +/- 0.6 mm

0.09 +/- 0.75 mm


29/30

before engaging clamps and taking measure-ments. Manufacturers should also recognize thatthe effective use of over-constraining only applies

to large, non-rigid panels such as body sides,

quarter inners, quarter outers, hoods, roofsfenders, floor pans, rear compartment pans and

dash panels. This assertion is important becausemanufacturers should not infer that adopting over-constraining systems will drastically reduce overallvariation as it would likely only impact a relatively

small percentage of body parts that are heavilyinfluenced by clamping strategy. Nevertheless,

these large, non-rigid parts typically are the mostdifficult to approve for production use.

One hypothesis is that over-constraining largenon-rigid parts may provide the best predictor of

metal movement during assembly. The addition ofspot welds deforms non-stable part dimensionsduring assembly. The use of additional secondarylocators could help predict part positioning and

movement during assembly because they consti-tute additional control points. If the principalobjective of stamping measurements is to assess

the potential to build dimensionally correct sub-assemblies, then over-constraining may offer abetter approach. This Project Team intends toexplore more fully the ramifications of using over-

constrained measurement systems in futureresearch.

6.0 Conclusions

In devising a dimensional evaluation strategy forthe automotive body, manufacturers must careful-

ly consider the effects of the measurement system.

This research found that checking fixtures andcoordinate measuring machines are capable ofmeasuring most stamping dimensions with a six

sigma gage spread of 0.24 mm (6 x 0.04). Sincemost stamping tolerances are at least +/- 0.5 mm,

f t ll t R&R

measurement system analysis fromcollection points. Dimensions in

urement areas may yield sigma 0.10 mm. For these high gage v

manufacturers must either add secor assign larger tolerances (+ /- 1 m

dimensions in unstable measuremhigh gag e variation often conformponents during assembly, minimizcontrol them at tight tolerances of

mm.

Another finding of this study is tchecking fixtures exhibit similar variation because the principal error relates to the ability to co

unload parts in fixtures. The staticCMMs or check fixture probes are

static-repeatability is less than 0.01 m

Although body measurement syslow gage variation, they are not ne

rate or representative of part positbly tooling, particularly for larger, This research recommends great

improving the correlation betwemeasurements in holding fixtures

or check fixture, and part positioassembly. Some manufacturers

achieve this by over-constraining parts. This contrasts with the tradof trying to develop datum schegage capability requirements usi

mum number of secondary locator

One concern with measuremmean dimensions at the detail paimpact on dimensional evaluat

Manufacturers using a build-to-nomay unnecessarily rework matin


30/30

AK Steel Corporation

Bethlehem Steel Corporation

DaimlerChrysler Corporation

Dofasco Inc.

Ford Motor Company

General Motors Corporation

Ispat/Inland Inc.

LTV Steel Company

National Steel Corporation

Rouge Steel Company

Stelco Inc.

U.S. Steel Group, a Unit of USX Corporation

WCI Steel, Inc.

Weirton Steel Corporation

Auto/SteelPartnership

This publication was prepared by:

Body Systems Analysis Project Team

The Auto/Steel Partnership Program

2000 Town Center, Suite 320Southfield, Michigan 48075-1123248.356.8511 fax

http://www.a-sp.orgA/SP-9030-4 0100 2M PROGPrinted in U.S.A.

Documents

Automotive Body Measurement