Standards for Metallography - George Vander Voort · 2013. 2. 4. · ISO Metallography Standards...

Standards for MetallographyStandards for Metallography

George F. Vander VoortDirector, Research & Technology

Buehler Ltd.Lake Bluff, Illinois USA

Why Do We Need Standards?

• Standards are prepared by experts on the subject and they represent a summary of the current state of knowledge on the

best practices to use.

• If everyone uses the same, best method then results between laboratories will have better reproducibility (i.e., agreement

between laboratories)

• National and international standards are continuously revised and kept up-to-date with changes in technology

Types of StandardsTypes of Standards• Company standards – relate best to the local needs, but the

methods may be poorly developed and defined

• National standards – developed by industry/academic experts in the particular field but the practices and quality of the

standards vary from country to country

• International standards – these are written, and balloted in a number of countries with leading technology and should

represent the best ideas on a given test method (ISO)

• ISO standards tend to be short in length and contain only the basic information required

• ASTM test method standards give more background details and have precision and bias data based on interlaboratory test

programs (“round robins”)

ISO Metallography StandardsISO Metallography Standards• 21 standards on hardness testing (8 on HV, 8 on HK and 6 on HRC) – too many! HV test is broken into standards for loads <

200gf and loads >200 gf

• 3 Grain Size Methods (one on Fe, one on Cu)

• 2 on Graphite Characterization (flake vs nodular)

• 7 on Coating Thickness Measurement

• 1 on the Microstructure of Hard Metals

• 1 on the Sulfur Print Test

• 1 on Carbide Segregation in Bearing Steels

• 1 on Manual Point Counting

• 2 on Nonmetallic Inclusion Rating (macro vs micro)

How are ISO Standards Created?How are ISO Standards Created?

• Committees set up with representatives from member and observer countries (later rarely come to meetings)

• Each committee chaired by a country, the secretariat –usually a national standards writing organization

• Representatives are often bureaucrats who bring specific technical experts to the meeting depending upon the topics

• Committees meet irregularly, when there is business

• All standards must be reviewed every 5 years

• If a standard must be revised, it is done usually by one person (and anyone they wish to utilize) from one country

• Revisions are only balloted within the committee

Creating an ASTM StandardCreating an ASTM Standard

• Each of the more than 130 ASTM committees can create standards (of various types) on their subject of interest.

• Committee E-4 on Metallography writes test methods standards, so the committee need not be “balanced” between

people representing producers and purchasers.

• A standard is created when a need is shown to exist and a task group can be formed with enough people with the

needed expertise to write a draft

• The draft is balloted in the task group until all agree that it is acceptable to go to Subcommittee ballot

• After it passes subcommittee ballot, it must pass a committee ballot, and then a society ballot.

ASTM Standards ASTM Standards –– Upkeep ProcessUpkeep Process

• Every ASTM standard must be reviewed every 5 years

• A task group is assigned to review the standard

• It decides if the standard is acceptable as written, if technology has changed and it must be modified, or that the

standard is of no value and can be made obsolete

• For any of the above actions, a ballot is required. If it is to be revised, a task group is given the job to make the

necessary changes. The revised draft must go through ballots within the task group, the subcommittee, the committee and

the society

• If it is to be withdrawn or re-approved as is, this decision must be balloted, but it is a simple ballot usually

ASTM Metallography StandardsASTM Metallography Standards

• Terminology (E7)

• Specimen Preparation (E3, E340, E407, E768, E1558, E1920, E2015)

• Macrostructural Evaluation (E381, E1180)

• Light Microscopy (E883, E1951)

• Quantitative Metallography (E45, E112, E562, E930, E1077, E1122, E1181, E1245, E1268, E1382,

E2109)

• XRD, SEM, TEM (E81, E82, E766, E963, E975, E986, E2142)

• Microindentation Hardness Testing (E384)

EE--4 Standards for Quantitative Metallography4 Standards for Quantitative Metallography

• Inclusion rating

• Grain Size

• Volume Fraction

• Characterization of Second Phases

• Case Depth/Decarburization

• Degree of Banding

EE--4 Standards for Quantitative Metallography4 Standards for Quantitative Metallography

Determine for the method defined its

Precision and Bias

(accuracy usually cannot be determined)

Using Interlaboratory Test Programs,

Commonly called “round robins”

Quantitative Metallography

Numerical measurements of microstructural features

1. Surface Gradients

Standard metrology methods

2. Matrix Microstructures

Stereological MeasurementsMetrology Measurements

STEREOLOGY

Extrapolation of measurements made on a two-dimensional sectioning plane to

determine the three-dimensional characteristics of the microstructure

Measurements may be 0-, 1-, or 2-dimensional (i.e., points, lines, areas)

STEREOLOGY

Matrix Microstructural Measurements

• Planar (flat) Surface Images

• Non-planar (curved) Surface Images

• Projected Images

Planar Surface Images

Flat, polished and etched surfaces require no additional corrections and are the simplest to employ. Surface

relief in preparation must be minimized and etching depth must be

minimal.

Non-Planar Surface Images

SEM images of fractures depict the rough surface as being flat. However, the surfaces are not flat but exhibit hills and valleys that

vary with the fracture mode and mechanism. The measurements must be

corrected by determining the surface roughness by, for example, using vertical sections. Otherwise, all measurements are

biased.

Projected Images

Images created using transmitted light or electrons (as in TEM thin foils) sense the

structure within a volume of material. Hence, measurements reflect data in

volume and the results must be corrected knowing the thickness or depth of the image plane. If not, biased data will be

obtained.

Statistical Analysis

• Mean (average)

• Standard Deviation

• 95% Confidence Interval

• % Relative Accuracy

• Tests to evaluate the significance of differences between mean values

Statistical Analysis

Statistical precision of the data is mainly a function of the number of measurements made. This is why

image analysis can produce significantly better data than

manual procedures.

Statistical Analysis

Mean (Average)

X = ———∑ Xi

N

Xi are the individual values andN is the number of measurements

Statistical Analysis

Standard Deviation – distribution of the individual values around the mean

S = [ —————— ]∑ ( xi – x)2

N - 1

1/2

Statistical Analysis

95% Confidence Interval

95% CI = ———ts(N)1/2

t is the Students’ t value for a 95% CI and for N-1 degrees of freedom

Statistical Analysis

% Relative Accuracy

%RA = ———— x 10095% CI

X

10% RA is a good target, especially for manual measurements, but is difficult to achieve when the volume fraction is <2%, even with image

analysis equipment.

Accuracy vs. Precision

To determine the accuracy of a measurement, we must know the true value

by some reference method. For microstructural measurements, we never know the true value by any independent referee method. Therefore, we can only

assess the precision of our measurements in terms of the scatter around the mean value.

Specimen Preparation

Image analyzers require correctly prepared specimens - better quality

than for manual measurements.

You cannot measure what cannot be seen!

Stereological SymbolsP = PointL = Line

A = Area (planar)S = Surface (curved)

V = VolumeN = Number

Symbols can be combined, e.g., VV, NA, LA, SV

The following slide shows a synthetic microstructure consisting of 30 spherical

particles with three different diameters to illustrate certain measurements

Volume Fraction – A Measure of the Concentration of a Second

Phase Constituent

VV = ————∑ Vα

VT

But, there is no simple way to directly measure the volume per unit volume of a constituent!

Volume fraction can be assessed from the area fraction, linear

fraction or point fraction, that is -

VV = AA = LL = PP

For manual measurements, PP is the easiest method and most efficient (i.e., best

precision for a given amount of work)

Areal Analysis – Area Fraction

Earliest measurement procedure, used with minerals. Can only be done

manually on structures that are coarse and consist of simple geometric shapes. The method is very precise for a given

field, but too time consuming to measure a large number of fields.

AA = ———A∑ Aα

AT

111

2

3 32

2 4

4

55

67

63

84

7

89

5 10 9

6 141110 12 13

Calculate the Area Fraction, AA

Calculate the area of each spherical particle (circular in cross section) based on a diameter

measurement and a count of the number of each size particle. We will assume that the

image is at 500X magnification. The diameters of the three circular particles

are:12.6, 21.6 and 34 µm. The areas of the circular particles are: 124.69, 366.44 and

907.92 µm2. The test area measures 512 by 380 µm or 194560 µm2.

Calculate the Area Fraction, AA

AA = —————————————————[(6x907.92) + (10x366.44) + (14x124.69)]

(512 x 380)

AA = 0.056 = 5.6%

Point Fraction – Point CountingASTM E 562

Superimpose a grid composed of points over the microstructure. In practice, points are hard to see, so we use crosses or intersecting vertical and horizontal test lines. The intersection is the “point”. The point must be in the constituent to

be a “hit”. If it is a tangent “hit”, count it as one-half. Calculate PP by:

PP = ——∑ Pα

PT

Point Counting GridsThe optimum number of “points” in a point

counting grid is a function of the volume fraction to be measured and is determined from

the equation, P = 3/VV, where the volume fraction is a fraction, not a percentage. So, as VV decreases from 0.5 (50%) to 0.01 (1%), P

varies from 6 to 300.

The following 100-point grid is convenient to use as each “hit” is 1%.

Use of the Point-Counting GridThe grid consists of ten horizontal and vertical lines, yielding 100 points (the

intersection points of the lines). This is superimposed over the microstructure and the image is scanned, usually from upper

left to the lower right, while noting the number of points that are inside the

constituent of interest, and those on phase boundaries (weighed as one-half a “hit”).

This is repeated for N fields. Then, the point fraction is calculated and is an estimate of

the volume fraction.

1

2

3

54

7

Calculation of PP

In this example there were 7“hits” where the “points” were inside the constituent of interest, and no tangent hits. So, the point

fraction is calculated as:

PP = ——— = 0.07 = 7%7100

To obtain good data, more fields must be evaluated.

Image Analysis vs. Manual

For manual work, to obtain the best precision, point count more fields as the

field-to-field variability has a greater influence on precision than the precision in measuring a single field. The adage is “do

more, less well” – that is, put less effort into measuring each field and do more fields.

For IA work, all of the pixels in the field are used. Thus, the precision per field is higher, but the time per field is very small. Hence, even if N is the same, the %RA is better.

Intersections Per Unit Length, PL

PL is a measure of the number of point intersections with phase or grain

boundaries per unit length of test line. It is calculated from:

PL = ————∑ Pα

LT

Pα is the number of intersections and LT is the true line length (line length/magnification)

Intersections/Unit Length, PL

To illustrate this calculation, let us superimpose a series of horizontal test

lines, such as used in the lineal analysis, over the synthetic microstructure.

Intersections/Unit Length, PL

The point intersections are indicated in the next slide.

12

34

56

78

9

1011

12

1314 15

1617

1820

2223

2425

2627

28

2930

31

3233

34

35

19

21

Intersections/Unit Length, PL

In the example, Pα is 35. If each of the 10 lines is 256 mm long, and the magnification is 500X, then,

LT is 5.12 mm, and

PL = ——— = 6.84 mm-1355.12

Interceptions Per Unit Length, NL

NL is a measure of the number of interceptions with phase or grain

particles per unit length of test line. It is calculated from:

NL = ————∑ Nα

LT

Nα is the number of interceptions and LT is the true line length (line length/magnification)

Interceptions/Unit Length, NL

To illustrate this calculation, let us superimpose a series of horizontal test

lines, such as used in the lineal analysis, over the synthetic microstructure.

Interceptions/Unit Length, NL

The particle interceptions are indicated in the next slide.

12 3

4 5 6

7.57

8.5

9.5

10

10.5 11.5

12.5

13.5

14.515.5

17.5

16.5

Interceptions/Unit Length, NL

In the example, Nα is 17.5. If each of the 10 lines is 256 mm long, and the magnification is 500X,

then, LT is 5.12 mm, and

NL = ——— = 3.42 mm-117.55.12

Number Per Unit Area, NA

The number of particles per unit area, NA, is a measure of the quantity of

particles, that is the number density. NA is related to the number per unit

volume, NV, which can only be determined by serial sectioning. It is

determined by:

NA = ————∑ Nα

AT

Number Per Unit Area, NA

To illustrate the determination of NA, let us count the number of particles in our

synthetic microstructure and then divide by the test area. The synthetic

microstructure with 30 particles is shown in the next slide.

Number Per Unit Area, NA

The test area measures 256 x 190 mm and the magnification is 500X. Therefore, NA

is given by 30 particles divided by the true test area:

NA = ————————— = 154.2 mm-2

(256/500)x(190/500)30

Average Particle Area, A

The average particle size, as measured by the area, can be determined from a ration

of the field measurements, AA and NA, without use of individual particle area

measurements from:

A = ——AA

NA

Average Particle Area, A

The area fraction was determined previously by areal analysis, lineal

analysis and point counting, that is, AA, LL and PP. Of these, the AA value is the

most precise. NA was also determined. So, the average cross-sectional area of the

particles is:

A = ——— = 0.0003632 = 363.2 µm20.056154.2

Average Particle Area, A

Using image analysis, we can measure the area of each particle, add all the areas, and

divide by the number of particles. As we have particles with a perfect circular cross-section, we can measure the diameter and calculate the area of each particle. Then,

sum the areas and divide by the number of particles. The average area is:

A = —————∑ Aαi

Nα

Average Particle Area, A

A = ——— = 361.9 µm210857.630

Comparison of Average Areas

Average area based on the area fraction divided by the number per unit area = 363.2 µm2

Average area based on actual measurements was 361.9 µm2

363.2 ≅ 361.9

Mean Center-to-Center Spacing, σ

The mean spacing between particle centers, in all directions, is given simply

by the reciprocal of NL:

σ = ⎯⎯1

NL

Mean Center-to-Center Spacing, σ

In the case of our synthetic microstructure, NL was determined as

3.42 interceptions per mm. So, σ is:

σ = ⎯ = 0.2924 mm = 292.4 µm13.42

Mean Edge-to-Edge Spacing, λ

The mean edge-to-edge spacing between particles, known as the mean free path, is a

good structure-sensitive parameter. λ is calculated from:

λ = ⎯⎯⎯⎯1 - AA

NL

Mean Edge-to-Edge Spacing, λ

In the case of our synthetic microstructure, AA and NL were determined. So, λ is:

λ = —⎯⎯⎯ = 0.276 mm = 276µm1 – 0.056 3.42

The MFP is a very structure-sensitive parameter used in ASTM E 1245 to characterize second-phase particles

Degree of Orientation, Ω

Ω = ——————(PL)⊥ - (PL)||

(PL)⊥ + 0.571(PL)||

Usually expressed as a percentage (%)

Used in ASTM E 1268 to assess the degree of banding or orientation in structures viewed on a longitudinal plane

Interlamellar Spacing

σr = ——1NL

σt = ——σr

2

Where σr is the mean random spacing and σt is the mean true spacing.

Measure NL with random test lines, such as with a circle, rather than directed test line (perpendicular to lamellae)

Grain Size Measurement

Types of Grain Sizes

• Non-twinned(ferrite, BCC metals, Al)

• Twinned FCC Metals (austenite, Cu, Ni)

• Prior-Austenite

(Parent Phase in Q&T Steels)

Grain Size Measurements

• Number of Grains/inch2 at 100X: G• Number of Grains/mm2 at 1X: NA

• Average Grain Area, µm2 : A• Average Grain Diameter, µm: d

• Mean Lineal Intercept Length, µm: l

Grain Size Measurement MethodsComparison Chart Ratings

Shepherd Fracture Grain Size RatingsJeffries Planimetric Grain Size

Heyn/Hilliard/Abrams Intercept Grain Size Snyder-Graff Intercept Grain Size

2D to 3D Grain Size distribution Methods

Definition of ASTM Grain Size

n = 2 G-1

n = number of grains/in2 at 100X

G = ASTM Grain Size Number

ASTM Grain Size, GG n G n

1 1 6 32

2 2 7 64

3 4 8 128

4 8 9 256

5 16 10 512

ASTM Standards for Grain Size

ASTM E 112: For equiaxed, single-phase grain structures

ASTM E 930: For grain structures with an occasional very large grain

ASTM E 1181: For characterizing duplex grain structures

ASTM E 1382: For image analysis measurements of grain size, any type

Comparison Chart RatingsLook at a properly etched microstructure,

using the same magnification as the chart, and pick out the chart picture closest in size to the

test specimen. If the grain structure is very fine, raise the magnification, pick out the closest chart picture and correct for the difference in magnification according to:

G = Chart G + QQ = 6.64Log10(M/Mb)

where M is the magnification used and Mb is the chart magnification

Jeffries Planimetric Method

n1 = number of grains completely inside the test circle

n2 = number of grains intercepting the circle

NA = f[ n1 + (n2/2)]

f = Jeffries multiplier

f = magnification2/circle area

Jeffries Planimetric Method

Average Grain Area = A = ——

G = (-3.322LogA) – 2.955

1NA

n1 = 68 and n2 = 41

Jeffries Planimetric Method - Example

For the preceding micrograph,

n1 = 68 and n2 = 41

And

f = —— = ——— = 0.497M2

A1002

20106.2

Jeffries Planimetric Method - Example

NA = f[n1 + (n2/2)]

NA = (0.497)[68 + (41/2)]

NA = 44.02 mm-2

Jeffries Planimetric Method - Example

A = —— = 0.0227 mm21NA

d = (A)1/2

G = 2.5

Jeffries Planimetric Method - Example

This is an austenitic Mn steel, solution annealed and aged to precipitate a pearlitic phase on the grain boundaries (at 100X).

There are 43 grains within the circle (n1) and there are 25 grains intersecting the circle (n2). The test circle’s area is 0.5 mm2 at 1X.

Jeffries Planimetric Method - Example

NA = f[n1 + (n2/2)]

f = [(1002)/5000]

NA = 2[43 + (25/2)] = 111 mm-2

G = [3.22Log10(111)] – 2.954 = 3.8

(Of course, more than one field should be measured to get good statistical results)

Heyn/Hilliard/Abrams Intercept Method

N = number of grains intercepted

P = number of grain boundary intersections

NL = ——NLT

PL = ——PLT

where LT is the true test line length

Heyn/Hilliard/Abrams Intercept Method

Apply a test line over the microstructure and count the number of grains intercepted

or the number of grain boundary intersections (easier for a single-phase grain

structure). After you count N or P, divide that number by the true line length to get

NL or PL.

Intercept Counts (N)

1/2

1 1

1

1

1 1/2

The test line intercepted 5 whole grains and the line ends fell in two grains. These are weighted as ½ an interception. So the

total is 6 intercepts (N=6).

Intersection Counts (P)

1 1 1 1 1 1

The test line has intersected 6 grain boundaries. The ends within the grains are not important in intercept counting.

So, P=6 for the intercept count.

Heyn/Hilliard/Abrams Intercept Method

Mean Lineal Intercept, l = — = ——1NL

11PL

G = [6.644Log10(NL or PL)] – 3.288

G = [-6.644Log10(l)] – 3.288

Note: Units are in mm-1 (for NL and PL) or mm (for l)

Heyn/Hilliard/Abrams Intercept Method

If the grain structure is not equiaxed, but shows some distortion of the grain shape, use straight test lines at various angles, or simply

horizontal and vertical with respect to the deformation axis of the specimen.

Alternatively, you can use test circles, such as the ASTM three-circle grid (three concentric

circles with a line length of 500 mm). This test pattern averages the anisotropy.

Heyn/Hilliard/Abrams Intercept Method

Example of three concentric test circles for point counting.

To illustrate intercept counting, note that there are 41, 25 and 20 grains intercepted (N) by the three concentric circles.

Intercept Counting Example

LT = 11.4 mm

N = 41 + 25 + 20 = 86

NL = —— = 7.54 mm-18611.4

l = —— = 0.133 mm17.54

G = [-6.644Log10(0.133)] – 3.288 = 2.5

Intercept Grain Size Example – Single Phase

This is a 100X micrograph of 304 stainless steel etched electrolytically with 60% HNO3 (0.6 V dc, 120 s, Pt cathode) to suppress etching of the twin

boundaries. The three circles have a total circumference of 500 mm. A count of the grain boundary intersections yielded 75 (P=75).

Intercept Grain Size Example – Single Phase

PL = ——— = 15 mm-175500/100

l = —— = 0.067 mm115

G = [-6.644Log10(0.067)] – 3.288 = 4.5

Intercept Method for Two-Constituents

Nα = Number of α grains intercepted

LT = Test line length/Magnification

VVα = Volume fraction of the α phase

lα = ———VVα(LT)

Nα

Intercept Method for Two-Constituents

This 500X micrograph of Ti-6242 was alpha/beta forged and alpha/beta annealed, then etched with Kroll’s reagent. The circumference of the three circles is 500 mm. Point counting revealed an alpha phase volume fraction of 0.485 (48.5%). 76 alpha

grains were intercepted by the three circles.

Intercept Method for Two-Constituents

lα = ———————— = 0.006382 mm(0.485)(500/500)76

G = [-6.644Log10(0.006382)] – 3.288 = 11.3

Particle Size Measurement

Six ways to measure particles on a polished cross section.

Particle Size Measurement

Volumetric Diameter, dV

Diameter of a sphere with the same volume as the particle

V = — dV3π

6

Particle Size Measurement

Feret’s Diameter, dF

Mean distance between pairs of parallel tangents to the projected

outline of the particle

Particle Size Measurement

Projected Area Diameter, dA

Diameter of a circle with the same area as the projected area of the

particle.

A = — dA2π

4

Particle Size Measurement

Perimeter Diameter, dP

Diameter of a circle with perimeter length the same as the projected outline

of the particle

P = π dP

Shape Descriptors

SPHERICAL

Global shaped

(Circular on a cross section through the particle)

Shape Descriptors

ACICULAR

Needle-like in three dimensions

Shape Descriptors

FLAKY

Irregular plate-like shape

Shape Descriptors

DENDRITIC

Branched,

Tree-like in three dimensions

Shape Descriptors

LENTICULAR

Lens-like shape

Shape Descriptors

FIBROUS

Regular or irregular

Tread-like shape

Shape Descriptors

ANGULAR

Sharp edged or

Roughly polyhedral shaped

Shape Descriptors

GRANULAR

Approximately equidimensional

Irregularly shaped

Shape Descriptors

IRREGULAR

Lacking any symmetry

Shape Factors

Elongation Ratio or

Anisotropy Ratio

AR = ———LengthWidth

Shape Factors

Sphericity (Roundness)

S = ———4πAP2

S = 1 for a circular feature; S < 1 for other shapes

Sometimes this equation is reversed

Perimeter – Sensitive to Magnification

Try to use shape factors that do not require a perimeter measurement, especially when the particles are

small (<30 pixels/particle)

Perimeter-Free Shape Factor

Measure the maximum Feret’s diameter, dFmax, and calculate the area of the circle

with that diameter.

AF = π(———)dFmax

22

SFPF = —————AF

Ameasured

Degree of Nodularity of Graphite

% Nod. = ——————————100(∑Ai with SF ≥ 0.6)

∑Ai

Ai is the area of a graphite particle

Variations in Graphite Shape

Nodular Iron

Compacted Graphite Cast Iron

Flake Gray Cast Iron

Magnification bars are all 100 µm long

Shape Factors - Example

Histogram of sphericity shape factors for two flake gray iron specimens, a compacted graphite specimen and a nodular graphite specimen.

ASTM E 1245

A Stereological Procedure to Characterize Discrete Second-Phase Particles

Uses field and feature-specific measurements.

While the measurements employ stereological parameters, they may be made on only one plane,

for example, the longitudinal. If the three-dimensional values are desired, then additional test

planes must be assessed.

ASTM E 1245

Measure or calculate:

Area Fraction, usually in %

Number per mm2, NA

Average Length in µm

Average Area in µm2

Mean Free Path in µm

ASTM E 1245

Area fractions of inclusions on three parallel planes, same specimens, of Alloy 718.

ASTM E 1245

Number per sq. mm of inclusions on three parallel planes, same specimens, for Alloy 718.

ASTM E 1245

Average area of inclusions on three parallel planes, same specimens, for Alloy 718

ASTM E 1245

Mean free path of the inclusions on three parallel planes, same specimens, for Alloy 718.

ASTM E 1245

Sulfide area fractions on 12 specimens taken along an as-cast bar of 303 stainless steel, at the mid-radius location.

ASTM E 1245

Sulfide number per sq. mm taken at 12 locations along an as-cast bar of 303 stainless steel at the mid-radius location.

ASTM E 1245

Average area of sulfides taken at 12 locations along an as-cast bar of 303 stainless steel, at the mid-radius location.

ASTM E 1245

Mean free path for sulfides at 12 locations along an as-cast bar of 303 stainless steel, at the mid-radius location.

ASTM E 1245

Distribution of area fractions of the sulfides for 106 bars of wrought 303 stainless steel, at the mid-radius location, longitudinal plane.

ASTM E 1245

Distribution of number per sq. mm of sulfides for 106 bars of wrought 303 stainless steel, at the mid-radius location, longitudinal plane.

ASTM E 1245

Distribution of average area of sulfides for 106 bars of wrought 303 stainless steel, at the mid-radius location, longitudinal plane.

ASTM E 1245

Plot of sulfide length vs. area measurements for each of 106 wrought bar specimens in wrought 303 stainless steel.

Point Counting Inclusions

Point counting of inclusions is tedious and imprecise. This work used 100 fields measured with a 100-point grid, but the 95% confidence limits are

poor, typical for volume fractions below 2%.

Lineal Analysis of Inclusions

A Hurlbut counter was used (one hour per specimen) to measure the lineal fraction of inclusions. Again, the precision of the measurements is poor.

Image Analysis Inclusion Measurement

Image analysis measurement of the inclusions using 1080 fields (grouped in 12 sets of 90) gave better precision in less time than the manual measurements.

Image Analysis Inclusion Measurement

Inclusions measured by image analysis using different magnifications (field sizes) shows the influence of the number of fields on the mean value.

Image Analysis Inclusion Measurement

The relative accuracy of the inclusion volume fractions improved with increasing number of fields measured and is poorest for the highest

magnification (small field size increases field-to-field variability).

Image Analysis Inclusion Measurement

The relative accuracy of the inclusion area fractions improved (decreased) as the area measured increased.

Future of StandardizationFuture of Standardization• More standards are being created each year due to technology

changes, e.g., due to new materials being developed

• More automation is being introduced into test methods and these changes must be rolled into all standards affected by the

technology changes

• Microstructural test methods are moving to quantification by stereological principles, using image analyzers, rather than using

qualitative chart methods

• More statistical analysis procedures are being introduced intotest methods that yield numerical data

• Precision and bias information, based on interlaboratory test programs are being added to standards that produce numerical

data