1
Designation: E 45 – 05 e

Standard Test Methods for

Determining the Inclusion Content of Steel 1
This standard is issued under the fixed designation E 45; the number immediately following the designation indicates the year of original
adoption
adopt ion or, in the case of revisio
revision,
n, the year of last revision.
revision. A numb
number
er in parentheses indicates
indicates the year of last reapproval.
reapproval. A supers
superscript
cript
epsilon (e) indicates an editorial change since the last revision or reapproval.

This standard has been approved for use by agencies of the Department of Defense.

1
e NOTE—Table 1 was editorially corrected in February 2007.

1. Sco
Scope
pe 2. Referenced Documents
1.1 Th
1.1 Thes
esee te
test
st memeth
thod
odss co
cove
verr a nunumb
mber
er of rereco
cogn
gniz
ized
ed 2.1   ASTM Standards:   2
procedures for determining the nonmetallic inclusion content D 96 Test Method for Water
Water and Sediment
Sediment in Crude Oil by
of wroug
wrought ht steel
steel.. Macr
Macroscop
oscopic
ic meth
methods
ods incl
include
ude macr
macroetc
oetch,
h, Centrifuge Method (Field Procedure)
fracture, step-down, and magnetic particle tests. Microscopic E 3   Guide for Preparation of Metallographic Specimens
methods include five generally accepted systems of examina- E 7  Terminology Relating to Metallography
tion. In these microscopic methods, inclusions are assigned to E 381   Method
Method of Macr
Macroetc
oetch
h Testin
esting
g Steel Bars, Bille
Billets,
ts,
a cat
catego
egory
ry bas
based
ed on simsimila
ilarit
rities
ies in mormorpho
pholog
logy
y, and not Blooms, and Forgings
necessari
neces sarily
ly on their chemi
chemical
cal ident
identity
ity.. Meta
Metallog
llographi
raphicc tech
tech-- E 709   Guide for Magnetic Particle Examination
niques that allow simple differentiation between morphologi- E 768  Practice for Preparing and Evaluating Specimens for
cally similar inclusions are briefly discussed. While the meth- Automatic Inclusion Assessment of Steel
ods are primarily intended for rating inclusions, constituents E 1245  Practice for Determining Inclusion or Second-Phase
such as carbides, nitrides, carbonitrides, borides, and interme- Constituent Content of Metals by Automatic Image Analy-
tallic
tallic phases
phases may be rat rated
ed usi
using
ng som
somee of the mic micros
roscop
copicic sis
methods. In some cases, alloys other than steels may be rated E 1444   Practice for Magnetic Particle Examination
using
usi ng one or mor moree of the
these
se met
method
hods;s; the methods
methods will be E 1951   Guide
Guide for Cal Calibr
ibrati
ating
ng Ret
Reticl
icles
es and Lig
Light
ht Mi
Micro
cro--
described in terms of their use on steels. scope Magnifications
1.2 Thi
Thiss pra
practi
ctice
ce cov
covers
ers pro
proced
cedure
uress to per
perfor
form
m JK-
JK-typ
typee Standards:3
2.2   SAE Standards:
inclusion ratings using automatic image analysis in accordance J422,   Recommended
Recommended Pract Practice
ice for Dete
Determin
rmination
ation of Inclu
Inclu--
with microscopic methods A and D. sions in Steel
1.3
1.
-
3 DeDepe
pend
ndin
ingg on ththee ty
type
pe of ststee
eell an
andd ththee pr
prop
oper
erti
ties
es 2.3   Aerospace Material Specifications:3
Material Specifications:
-

required, either a macroscopic or a microscopic method for

` 
  ,
  ,
` 
` 
AMS 230 2300,
0,   Premium
Premium Airc
Aircraft
raft-Qual
-Quality
ity Steel Clea
Cleanline
nliness:
ss:
` 

determining the inclusion content, or combinations of the two Magnetic Particle Inspection Procedure
  ,
  ,
  ,
  ,
` 
` 
` 

methods, may be found most satisfactory.

` 
-
` 
-
` 
AMS 2301,   Aircraft
Aircraft Qual
Quality
ity Steel Clean
Cleanlines
liness:
s: Magne
Magnetic
tic
  ,

1.4 These test methods

methods deal only with recommended
recommended test Particle Inspection Procedure
  ,
` 
  ,
  ,
` 
  ,

methods and nothing in them should be construed as defining AMS 2303,   Aircraft Quality Steel Cleanliness: Martensitic
` 
  ,
  ,
` 
-
-
-

or establishing limits of acceptability for any grade of steel. Corrosion-

Corro sion-Resis
Resistant
tant Steel
Steelss Magn
Magnetic
etic Part
Particle
icle Insp
Inspectio
ectionn
1.5 The values
values stated in SI units are to be regarded
regarded as the Procedure
standard.
stand ard. Value
Valuess in paren
parenthes
theses
es are conversions
conversions and are ap- AMS
AM S 230
2304,
4,   Special
Special Airc
Aircraf
raft-Qu
t-Quali
ality
ty Stee
Steell Cle
Cleanli
anlines
ness:
s:
proximate. Magnetic Particle Inspection Procedure
1.6   This sta
standa
ndardrd does not purport
purport to add
addre ress
ss all of the Standards:4
2.4   ISO Standards:
safe
safety
ty co
conc
ncererns
ns,, if an
anyy, as
asso
soci
ciat
ated
ed wi
with
th it
itss us
use.
e. It is th
thee ISO 3763,   Wrought Steels—Macroscopic Methods for As-
responsibility of the user of this standard to establish appro- sessing the Content of Nonmetallic Inclusions
 priate safety and health practices and determine the applica-
bility of regulatory limitations prior to use.
2
For referenced ASTM standards, visit the ASTM website, www.astm.org, or
contact ASTM Customer Service at service@
[email protected] rg. For   Annual Book of ASTM 
astm.org.
Standards volume
Standards  volume information, refer to the standard’s Document Summary page on
1
These test methods are under the jurisdiction of ASTM Committee E04 on the ASTM website
website..
3
Metallography and is the direct responsibility of Subcommittee E04.09 on Inclu- Available from Society of Automotive Engineers (SAE), 400 Commonwealth
sions. Dr., Warrendale, PA 15096-0001.
4
Current edition approved Nov. 1, 2005. Published December 2005. Originally Available from American National Standards Institute (ANSI), 25 W. 43rd St.,

approved in 1942. Last previous edition approved in 2002 as E 45 – 97 (2002). 4th Floor, New York, NY 10036.

Copyright © ASTM International, 100 Barr Harbor Drive, PO Box C700, West Conshohocken, PA 19428-2959, United States.

Copyright ASTM International

1
Provided by IHS under license with ASTM
No reproduction or networking permitted without license from I HS Not for Resale
 

1
E 45 – 05 e

ISO 4967,  Steel—Determination of Content of Nonmetallic place inclusions into one of several composition-related cat-
Inclusions—Micrographic Methods Using Standard Dia- egories (sulfides, oxides, and silicates—the last as a type of 
grams oxide). Paragraph 12.2.1
Paragraph  12.2.1 describes
 describes a metallographic technique
2.5   ASTM Adjuncts: to facil
facilitat
itatee inclu
inclusion
sion disc
discrimi
riminati
nation.
on. Only those incl inclusion
usionss
Inclusions in Steel Plates I-r and II 5 present at the test surface can be detected.
Four Photomicrographs of Low Carbon Steel6 4.3 The macr macroscop
oscopicic test methods evaluate
evaluate larlarger
ger surf
surface
ace
areas than microscopic test methods and because examination
3. Terminology
is visual or at low magnifications, these methods are best suited
3.1   Definitions: for detecting larger inclusions. Macroscopic methods are not
3.1.1
3.1.1 For defidefinit
nition
ionss of ter
terms
ms use
usedd in thi
thiss pra
practi
ctice,
ce, see suitable for detecting inclusions smaller than about 0.40 mm
Terminology   E 7.7. (1 ⁄ 64
64   in.)
in.) in le leng
ngth
th an
andd ththee memeththod
odss do no nott di
disc
scri
rimi
minanate
te
3.1.2 Ter
Terminology
minology   E 7   includes
includes the term   inclusion count ; inclusions by type.
since
since somsomee met
method
hodss of the
thesese tes
testt me
metho
thods
ds inv
involv
olvee len
length
gth
4.4 The microscopic
microscopic test methods
methods are employed
employed to charac-
measurements or conversions to numerical representations of 
terize inclusions that form as a result of deoxidation or due to
leng
length
thss or cocoun
unts
ts,, or boboth
th,, th
thee te
term
rm   inclusion
inclusion rati ng   is
rating
limited solubility in solid steel (indigenous inclusions). These
preferred.
inclusions are characterized by morphological type, that is, by
3.2   Definitions of Terms Specific to This Standard:
size, shape, concentration, and distribution, but not specifically
3.2.1   aspect
aspect rati
ratioo—the length-to-wi
length-to-widthdth ratio of a micrmicro-
o-
by composition. The microscopic methods are not intended for
structural feature.
3.2.2   discontinuous stringer —three
— three or more Type B or C assess
ass essing
ing the con conten
tentt of exo
exogengenous
ous inc
inclus
lusion
ionss (th
(those
ose frofromm
inclusions aligned in a plane parallel to the hot working axis entrapped slag or refractories).
and offset by no more than 15 µm, with a separation of less than 4.5 Becau
Because se the inclu
inclusion
sion population
population within
within a given lot of 
40 µm (0.0016 in.) between any two nearest neighbor inclu- steel varies with position, the lot must be statistically sampled
sions. in order to assess its inclusion content. The degree of sampling
must be adequate for the lot size and its specific characteristics.
3.2.3
,and   cate-type
inclusion
silicate-
sili type inclu —for
inclusion s, definitions
typessions, see Termi of sulfide-,
erminolog
nologyy E 7. alumina-
Globular
Globu lar Materi
Mat erials
als wit
withh ver
veryy low inc inclus
lusion
ion con
conten
tents
ts may be mor moree
oxide, in some methods refers to isolated, relatively nonde- accurately rated by automatic image analysis, which permits
formed inclusions with an aspect ratio not in excess of 2:1. In more precise microscopic ratings.
other methods, oxides are divided into deformable and nonde- 4.6 Resul
Results ts of macr
macroscop
oscopic ic and microscopic
microscopic test meth methodsods
formable types. may be used to qualify material for shipment, but these test
3.2.4   JK inclusion rating—a method of measuring nonme- methods do not provide guidelines for acceptance or rejection
tallic inclusions based on the Swedish Jernkontoret procedures;
-
-
` 
  ,
purposes.
purp oses. Quali
Qualificati
fication
on crit
criteria
eria for asses
assessing
sing the data devel
devel--
  ,

Methods A and D of these test methods are the principal JK

` 
` 
` 
  ,
  ,
oped
op ed by th thesesee memeth
thod
odss ca
cann be fo foun
undd in ASTM pr prod
oducuctt
  ,

rating methods, and Method E also uses the JK rating charts.

  ,
` 
` 
` 
` 
-
standards or may be described by purchaser-producer agree-
3.2.5   stringer —an
—an individual inclusion that is highly elon-
` 
-
` 
  ,
  ,
` 
ments.
ment s. By agreagreement
ementss betwe
between en produ
producercer and purch
purchaser
aser,, this
  ,

gated in the deformation direction or three or more Type B or

  ,
` 
  ,
` 
  ,
practice may be modified to count only certain inclusion types
  ,

C inclusions aligned in a plane parallel to the hot working axis

` 
-
-
- and thi thickn
ckness
esses,
es, or onlonlyy tho
thosese inc
inclus
lusion
ionss abo
above
ve a cercertai
tainn
and offset by no more than 15 µm, with a separation of less than severity level, or both. Also, by agreement, qualitative prac-
40 µm (0.0016 in.) between any two nearest neighbor inclu- tices may be used where only the highest severity ratings for
sions. each inclusion type and thickness are defined or the number of 
3.2.6   threshold setting—isolation of a range of gray level fields containing these highest severity ratings are tabulated.
values exhibited by one constituent in the microscope field. 4.7 TheThesese tes
testt me
metho
thods
ds are intende
intended d for use on wro wrough
ughtt
3.2.7   worst-field rating—a rating in which the specimen is metallic structures. While a minimum level of deformation is
rated for each type of inclusion by assigning the value for the not specified, the test methods are not suitable for use on cast
highest
highe st sever
severity
ity rating observed of that inclusion
inclusion type any- structures or on lightly worked structures.
where on the specimen surface.
4.8 Guideli
Guidelinesnes are pro provid
videded to ratratee inc
inclus
lusion
ionss in ste
steels
els
4. Signi
Significanc
ficancee and Use treate
tre atedd wit
with h rar
raree ear
earth
th add
additi
itions
ons or cal calciu
cium-b
m-bear
earing
ing com
com--
pounds. When such steels are evaluated, the test report should
4.1 The
These
se tes
testt met
method
hodss cov
cover
er fou
fourr mac
macros
roscop
copicic and five
describe the nature of the inclusions rated according to each
microscopic
microscop ic test methods (manual and imag imagee analy
analysis)
sis) for
inclusion category (A, B, C, D).
describin
descr ibing
g the incl
inclusion
usion content of stee
steell and proc
procedure
eduress for
expressing test results. 4.9 In addition
addition to the Practice
Practice E 45 JK rati ratings,
ngs, basic (such
(such
4.2 Incl
Inclusion
usionss are characterize
characterizedd by size
size,, shape
shape,, conce
concentra
ntra-- as used in Practice  E 1245) 1245) stereological measurements (for
tion, and dist
distribu
ribution
tion rather than chemical composition.
composition. Al- exampl
exa mple, e, the vol volume
ume frafracti
ction
on of sul sulfide
fidess and oxi oxides
des,, the
though compositions are not identified, Microscopic methods number of sulfides or oxides per square millimeter, the spacing
between incl
between inclusion
usions,
s, and so fort
forth)
h) may be sepa
separatel
rately
y dete
deter-
r-
mined and added to the test report, if desired for additional
5
Available from ASTM Headquarters. Order ADJE004502. inform
informati
ation.
on. Thi
Thiss pra
practi
ctice,
ce, how
howeve
everr, doe
doess not add
addres
resss the
6
Available from ASTM Headquarters. Order ADJE004501. measurement of such parameters.

Copyright ASTM International

2
Provided by IHS under license with ASTM
No reproduction or networking permitted without license from I HS Not for Resale
 

1
E 45 – 05 e

MACROSCOPIC
MACROSCOPIC METHODS which are the main concern in most cases, are not uniformly
distributed and the spaces between them are relatively large, so
5. Macroscopical Test
Test Methods Overview
thatt the cha
tha chance
ncess of rev
reveal
ealing
ing the
them
m are bet
better
ter when lar
larger
ger
5.1   Summary: specimens are examined.
5.1.1   Macro-etch
Macro-etch Test —The
— The mamacrcro-
o-et
etch
ch te
test
st is us
used
ed to 5.2.2
5.2. 2 Specimen
Specimenss for macmacros
roscop
copic
ic exa
examin
minati
ation
on may be
indicate inclusion content and distribution, usually in the cross quickly prepared
prepared by machining
machining and grinding.
grinding. A highly polished
polished
section or transverse to the direction of rolling or forging. In surface is not necessary. The macroscopic methods are suffi-
some instances, longitudinal sections are also examined. Tests ciently sensitive to reveal the larger inclusions.
are prepared by cutting and machining a section through the 5.3   Disadvantages:
desired area and etching with a suitable reagent. A solution of  5.3.1
5.3.1 These tes testt me
metho
thods
ds do not dis distin
tingui
guish
sh amo
amongng the
one part hydrochloric acid and one part water at a temperature different inclusion shapes.
of 71 to 82°C (160 to 180°F) is widely used. As the name of  5.3.
5.3.2
2 They ar aree no
nott su
suit
itab
able
le fo
forr th
thee de
dete
tect
ctio
ion
n of smsmal
alll
this test implies, the etched surface is examined visually or at globular inclusions or of chains of very fine elongated inclu-
low magmagnifi
nificat
cation
ion for inc inclus
lusion
ions. s. Det
Detail
ailss of thithiss tes
testt are sions.
included in Method E
Method  E 381. 381.  The nature of questionable indica- 5.3.3 The magnetic particle
particle method can lead to incor incorrect
rect
tions should be verified by microscopic examination or other interpre
interpretati
tation
on of micr
microstr
ostructur
uctural
al feat
features
ures such as strea
streaks
ks of 
means of inspection. retained
retained auste
austenite,
nite, micr
microsegr
osegregat
egation,
ion, or carbi
carbides
des in cert
certain
ain
5.1.1.1
5.1 .1.1 Sulfide
Sulfidess are rev reveal
ealed
ed as pit pitss whe
when n the sta standa
ndard
rd alloys; this is particularly likely if high magnetization currents
etchant described in 5.1.1
in  5.1.1 is is used. are employed.
5.1.1.2 Only large oxides oxides are reverevealed
aled by this test method.
method.
5.1.2   Fracture Test —The —The fracture test is used to determine MICROSCOPIC
MICROSCOPIC METHODS
thee pr
th pres
esen
ence
ce an
and d loloca
cati
tion
on of in incl
clus
usio
ions
ns as sh show
own n on th thee
fracture of hardened slices approximately 9 to 13 mm (  ⁄ 8  to   1 ⁄ 2 3 6. Microscopic Test
Test Methods Overview
in.) thick. This test is used mostly for steels where it is possible 6.1 Micr
Microscop
oscopicic methods are used to characterize
characterize the size,
to obtain a hardness of approximately 60 HRC and a fracture distr
distribu
ibutio
tion,
n, num
numberber,, and typ
typee of inc
inclus
lusion
ionss on a pol polish
ished
ed
grai
gr ain
n sisize
ze of 7 or fin finer
er.. Tesestt sp
spec
ecim
imenenss shshou
ould
ld nonott ha
have
ve specimen surface. This may be done by examining the speci-
excess
exc essive
ive ext
extern
ernalal ind
indent
entati
ations
ons or not notche
chess tha
thatt gui
guidede the men
me n wi
with
th a liligh
ghtt mi
micrcros
osco
cope
pe an
andd rerepo
portrtin
ing
g th
thee ty
type
pess of 
fractu
fra cture.
re. It is des
desira
irable
ble tha
thatt fra
fractu
cturere be in the lon longit
gitudi
udinal
nal inclusions encountered, accompanied by a few representative
direct
dir ection
ion app
approx
roxima
imateltely
y acr
across
oss the cen center
ter of the sli slice.
ce. The photomicrographs. This method, however, does not lend itself 
fractured surfaces are examined visually and at magnifications to a unif
uniform
orm repor
reporting
ting style. There
Therefore,
fore, standard reference
reference
up to approximately ten diameters, and the length and distri- charts
char ts depic
depicting
ting a seri
series
es of typic
typical
al incl
inclusio
usion n config
configurat
urations
ions
bution
but ion of incinclus
lusion
ionss is not
noted.
ed. Hea
Heatt tin
tintin
ting,
g, or blubluein
eing,
g, wil
willl (size, type, and number) were created for direct comparison
increase visibility of oxide stringers. ISO 3763 provides a chart with
wit h the mic
micros
roscop
copic
ic fiel
field
d of vie
view
w. A metmethodhod usi
using
ng ima
image
ge
method
met hod for frafractu
cturere sur
surfac
facee inc
inclus
lusion
ion ratrating
ings.s. In som
somee in- analysis to make these comparisons has also been developed.
stances, indications as small as 0.40 mm (1 ⁄ 64 64  in.) in length are 6.2 Variou
ariouss refe
reference
rence charts of this nature have been de-
recorded. vised such as the JK chart7 and the SAE chart found in SAE
5.1.3   Step-Down
Step-Down Meth Method  od —The
— The ste step-d
p-down
own tes testt met
methodhod is Recomm
Rec ommend
ended
ed Pra
Practi
ctice
ce J42
J4222 of the SAE Han Handbo
dbook.ok. The
used
use d to detdeterm
ermine
ine the pre presen
sence
ce of inc inclus
lusion
ionss on ma machichined
ned microscopic methods in Test Methods E 45 use refined com-
surfaces of rolled or forged steel. The test sample is machined parison charts based on these charts. Method A (Worst Fields),
to spe
specifi
cified
ed dia
diamet
metersers bel
belowow the sur surfac
facee and sur survey
veyed ed for Method
Met hod D (Lo(Loww Inc
Inclus
lusion
ion Content)
Content) and Met Method
hod E (SA (SAMM
inclusions
inclu sions under good illu illumina
mination
tion with the unaided eye or Rating)) use charts
Rating charts bas
based
ed on the JK cha chartrt whi
whilele Met
Method
hod C
with low magni
magnificati
fication.on. In some instances,
instances, test samp samplesles are (Oxides and Silicates) uses the SAE chart. ISO Standard 4967
machined to smaller diameters for further examination after the also uses the JK chart.
original diameters are inspected. This test is essentially used to 6.3 No chart can represent all of the various types types and forms
determine the presence of inclusions 3 mm (1 ⁄ 8  in.) in length of inclusions. The use of any chart is thus limited to determin-
and longer. ing the content of the most common types of inclusions, and it
5.1.4   Magnet
Magnetic ic Parti
Particle
cle Meth
Method od —The
—The magn magnetic
etic part
particle
icle must
mu st be ke
kept
pt in mimind
nd th
that
at su
such
ch a dedete
term
rmin
inatatio
ionn is no
nott a
method is a variation of the step-down method for ferromag- complete metallographic study of inclusions.
netic materials in which the test sample is machined, magne- 6.4 An altaltern
ernate
ate to com
compar
pariso
ison
n (ch
(chart
art)) met
method
hodss suc
suchh as
8
tized, and magnetic powder is applied. Discontinuities as small Methods A, C and D may be found in Method B. Method B
as 0.40 mm (1 ⁄ 64 64  in.) in length create magnetic leakage fields (Lengt
(Le ngth)
h) is use
usedd to det
determ
ermine
ine inc
inclus
lusion
ion conconten
tentt bas
based
ed on
that attract the magnetic powder, thereby outlining the inclu- length
length.. Onl
Onlyy inc
inclus
lusion
ionss 0.1
0.127
27 mm (0.(0.005
005 in.in.)) or lon
longer
ger are
sion.
sio n. See Pra Practi
cticece   E 1444   and Gui Guide de   E 709   on magn magnetic
etic recorded regardless of their type. From this method one may
particle examinations for more details of the procedure. Refer
to Aerospace Materials Specifications AMS 2300, AMS 2301,
7

AMS 2303, and AMS 2304. The JK chart derives its name from its sponsors Jernkontoret, the Swedish
Ironmasters Association.
5.2   Advantages: 8
Notee tha
Not thatt whi
while
le the
these
se meth
methods
ods are call
called
ed com
compar
pariso
ison
n char
chartt met
method
hods,
s, the
5.2.1 These test methods facilitate
facilitate the examination
examination of speci- procedure used may also consist of length measurements or counts of inclusions, or
mens with large surface areas. The larger inclusions in steel, both.
--`,,```,,,,````-`-`,,`,,`,`,,`---

Copyright ASTM International

3
Provided by IHS under license with ASTM
No reproduction or networking permitted without license from I HS Not for Resale
 

1
E 45 – 05 e

obtain data such as length of the longest inclusion and average sections. Exercise care, however, to crop specimens of suffi-
inclusion length. In addition, photomicrographs may also be cientt lengt
cien length
h from the bill
billets
ets for for
forging;
ging; otherwise,
otherwise, there is
taken to characterize the  background inclusions  that were not dangerr of the shear-dragg
dange shear-dragged ed ends being incorporated
incorporated in the
long enough to measure. specimens. Such distorted material will give a false result in
6.5 The advantages
advantages of the microscopic
microscopic methods
methods are: the inclusion determination. To avoid this, it is helpful to saw
6.5.1 Incl
Inclusion
usionss can be characteriz
characterized ed as to their size, type, thee en
th endsds of ththee bi
bill
llet
et le
leng
ngth
th fo
forr fo
forg
rgin
ing
g an
and
d to tatake
ke th
thee
and number. specimen from the middle of the forged length.
6.5.2 Extre
Extremely
mely small inclusions
inclusions can be revealed.
revealed. 6.11 Several of the methods described
described in these test methods
6.6 A dis disadv
advant
antage
age of the mic micros
roscop
copicic me
metho
thods
ds is thathatt requir
req uiree tha
thatt a spe
specifi
cificc are
areaa of the prepared
prepared surface
surface of the
individual rating fields are very small (0.50 mm 2). This limits specimen is surve
specimen surveyed,
yed, and all the sign
significan
ificantt incl
inclusion
usionss ob-
the practical size of the specimen, as it would simply take a served be recorded and expressed in the results. The reported
prohibitive number of fields to characterize a large specimen. resu
result
lt fo
forr ea
each
ch sp
spec
ecim
imen
en ex
exam
aminined
ed is
is,, th
ther
eref
efor
ore,
e, a momorere
The result obtained by a microscopic characterization of the accurate representation of the inclusion content than a photo-
inclus
inclusion
ionss in a lar
large
ge sec
sectio
tion
n is gov
govern
erned
ed by cha chance
nce if loc
local
al micrograp
micr ograph h or diagr
diagram.
am. A disad
disadvanta
vantage
ge of the Worst
Worst Fiel
Fieldd
variations in the inclusion distribution are substantial. The end approach
appr oach is that no such distributi
distribution
on of inclusion
inclusion rati
ratings
ngs is
use of the product determines the importance of the micro- obtained.
scopic
sco pic res
result
ults.
s. Exp
Experi
erienc
encee in int
interp
erpret
reting
ing the
these
se res
result
ultss is 6.12 To make comparisons
comparisons possible between different
different heats
necessary in order not to exaggerate the importance of small and different parts of heats, the results shall be expressed in
inclusions in some applications. such a manner that an average for the inclusion content of the
differ
different
ent spe
specim
cimens
ens in the hea
heatt can be obtobtain
ained.
ed. Whe
When n the
6.7 In determining
determining the inclusion
inclusion conte
content,nt, it is important
important to
lengths of the inclusions are measured, the simplest number is
realiz
realizee tha
that,t, wha
whatev
tever
er met
method
hod is use used,
d, the res result
ult act
actual
ually
ly
thatt for the agg
tha aggreg
regate
ate length
length of all the inc inclus
lusion
ionss per area
applies only to the areas of the specimens that were examined.
examined; however, it may be desirable not merely to add the
For pra
practi
ctical
cal rea
reason
sons,
s, suc
suchh spe
specim
cimens
ens are rel relati
ativel
vely
y sma
smallll
lengths
leng ths but also to weig
weight
ht the inclusions
inclusions accor
according
ding to thei
theirr
compared with the total amount of steel represented by them.
For the inclusion determination to have any value, adequate individual lengths.
and the total numberThe length of the
of inclusions maylargest inclusion
also be found
expressed.
sampling is just as necessary as a proper method of testing.
6.8 Steel often differs
differs in incl
inclusio
usion n cont
content
ent not only from
7. Samp
Sampling
ling
heat to heat, but also from ingot to ingot in the same heat and
even in different portions of the same ingot. It is essential that 7.1 To obta
obtain
in a reas
reasonabl
onablee esti
estimate
mate of inclu
inclusion
sion variations
variations
the unit lot of steel, the inclusion content of which is to be within a lot, at least six locations, chosen to be as representa-
determ
determine
ined,
d, sha
shall
ll not be lar larger
ger tha
than n one hea heat.t. Suf
Sufffici
icient
ent tive of the lot as possible, should be examined. In this context,
samples should be selected to represent the lot adequately. The a lot shall be defined as a unit of material processed at one time
exact sampling procedure should be incorporated in the indi- and subjected to similar processing variables. In no case should
vidual product requirements or specifications. For semifinished more than one heat be in the same lot. For example if a lot
products, the specimens should be selected after the material consists of one heat, sampling locations might be in the product
has been sufficiently cropped and suitable discards made. If the obtained from the top and bottom of the first, middle, and last
locations of the different ingots and portions of ingots in the usable
usa ble ingots
ingots in the pouring
pouring seq
sequen
uence.
ce. For str
strand
and cas
castt or
heat
heat cacann
nnot
ot be id iden
enti
tifie
fiedd in th thee lo
lott be
beining
g tetest
sted
ed,, ra
randndom
om bottom
bot tom pou
pourr pro
proces
cessin
sing,
g, a sim
simila
ilarr sam
sampli
pling
ng pla
plann per hea
heatt
    -

sampling should involve a greater number of test specimens for should be invoked.     -
    -
        `
  ,
  ,
        `
  ,
        `
  ,

an equivalent weight of steel. A value for the inclusion content 7.2 For cases
cases in whi
whichch a defi
definit
nitee loc
locati
ation
on wit
within
hin a hea
heat,
t,
  ,
        `
  ,
  ,

of an isolated piece of stee

steel,
l, even if accur
accuratel
ately
y deter
determine
mined,
d, ingot, or other unit lot is unknown, statistical random sampling         `
    -
        `
    -
        `
        `

should not be expected to represent the inclusion content of the with a greater number of specimens should be employed.         `
        `
  ,
  ,
  ,
  ,
        `

whole heat. 7.3 Ratings obtained will will vary with the amount
amount of reduction         `
        `
  ,
  ,
        `
    -

6.9 The size and shape of the wrought steelsteel product tested
tested of the product. For mate
material
rialss accep
acceptanc
tancee or for comparison
comparison     -

has a marked influence on the size and shape of the inclusions. among heats, care must be taken to sample at the correct stage
During reduction from the cast shape by rolling or forging, the of processing.
inclusions are elongated and broken up according to the degree
of reduction of the steel cross section. In reporting results of  8. Test Specimen Geometry
inclusio
inclusion
n dete
determi
rminati
nations,
ons, the
therefo
refore,
re, the siz
size,
e, shap
shape,
e, and 8.1 The minimum
minimum polished surface
surface area of a spec
specimen
imen for
method of manufacture of the steel from which the specimens the mi
micro
crosco
scopic
pic det
determ
ermina
inati
tion
on of inc
inclus
lusion
ion con
conten
tentt is 160
were cut must be stated. In comparing the inclusion content of  mm2 (0.25 in.2). It is recommended that a significantly large
different steels, they must all be rolled or forged as nearly as area should be obtained so that the measurements may be made
possible to the same size and shape, and from cast sections of  within the defined area away from the edges of the sample. The
about the same size. Specimens cut lengthwise or parallel to polished surface must be parallel to the longitudinal axis of the
the direction of rolling or forging shall be used. product. In addition, for flat-rolled products, the section shall
6.10 It may be convenient,
convenient, in order to obtai
obtain
n more readily also
also be per
perpen
pendic
dicula
ularr to the rol
rollin
ling
g pla
plane;
ne; for rou
rounds
nds and
comparable results, to forge coupons from larger billets. These tubular shapes, the section shall be in the radial direction. In all
forged sections may then be sampled in the same way as rolled cases, the polished surface shall be parallel to the hot-working

Copyright ASTM International

4
Provided by IHS under license with ASTM
No reproduction or networking permitted without license from I HS Not for Resale
 

1
E 45 – 05 e

NOTE  1—This method is also applicable to round sections.

NOTE   2—a   denotes surface removal.
FIG. 1 Quarter Section Specimen from Square Section for
Magnetic Particle Test, Machine Only

NOTE  1—Inch-pound equivalents: 12 mm = 0.47 in.; 19 mm = 0.75 in.

FIG. 3 Specimen from 1 1 ⁄ 2-in. (38.1 mm) Round Section for
Microscopic Test

NOTE  1—Method also applicable to square sections.

NOTE   2—a  denotes distance equal to surface removal.
FIG. 2 Quarter Section Specimen from Round Section for
Magnetic Particle Test, Forging and Machining

axis. Studies have demonstrated that inclusion length measure-

ments are significantly affected if the plane of polish is angled
more than 6° from the longitudinal hot-working direction. 9
8.1.1 Secti
Sections
ons less than 0.71 mm in thic
thickness
kness shall
shall not be
analyzed using Test Methods E 45.
8.2   Thick Section (Product
(Product Secti
Section
on Size Greater
Greater than 9.5
mm (0.
(0.375
375 in.
in.)) Thi
Thick,
ck, Suc
Suchh as For
Forgin
gings,
gs, Bil
Billet
let,, Bar
Bar,, Sla
Slab,
b, FIG. 4 Specimen from Large Bar or Billet for Microscopic Test
Plate, and Pipe):
8.2.1
8.2 .1 For wid
widee pro
produc
ducts,
ts, the one
one-qu
-quart
arter
er poi
point
nt alo
along
ng the
product
produ ct width is comm
commonly
only used to provi
provide
de repr
represent
esentativ
ativee specimenss poli
specimen polished
shed and exam
examined
ined.. Other areas,
areas, such as the
center and the surface, may be examined as well, provided the
material.
8.2.2 For round sections,
sections, the manner of cutt cutting
ing a spec
specimen
imen sampling procedure used is stated in the results. A billet or bar
from a 38 mm (1.5-in.) diameter section is shown in  Fig. 3.
-
-
`  3. A about 50 to 100 mm (2 to 4 in.) round or square is the preferred
  ,
  ,
` 

disk at least 12 mm (0.474 in.) thick is cut from the product.

` 
` 
  ,
size from which specimens should be taken; however, larger or
  ,

smalle
sma llerr siz
sizes
es may be use
used,
d, pro
provid
videded the pro
produc
ductt siz
sizes
es are
  ,
  ,
` 

The quarter-section indicated in Fig.

` 
` 
` 
-
in  Fig. 3 is
3  is cut from the disk and
` 
-

the shaded area is polished. Thus the specimen extends at least

` 
  ,
  ,
reported with the results.
` 
  ,
  ,

12 mm along the length of the product from the outside to the

` 
  ,
` 
8.3  Thin Sections (Product Section Sizes 9.5 mm (0.375 in.)
  ,

Thick or Less; Strip, Sheet, Rod, Wire, and Tubing)—Full cross

  ,
` 
-

center. -
-

8.2.3 For large sections

sections,, each spec
specimen
imen shall
shall be take
takenn from section longitudinal specimens shall be cut in accordance with
the mid-radius location, as shown by the shaded area in Fig. in  Fig. 4.
4. the following plan:
The specimen
specimen face to be pol polish
ished
ed ext
extend
endss at lea
least
st 12 mm 8.3.1 For 0.95 to 9.5-mm (0.0375 to 0.375
0.375 in.) cross section
parallel to the longitudinal axis of the billet and at least 19 mm thicknesses inclusively, a sufficient number of pieces from the
(0.75 in.) in the longitudinal radial plane, with the polished same sampling
sampling point are mount
mounteded to provi
provide
de appro
approxima
ximately
tely
2 2
face midway between the center and the outside of the billet. 160 mm (0.25 in. ) of polished specimen surface. (Example:
Such
Suc h mid
midway
way sam
sampli
pling
ng is use
usedd to dec
decrearease
se the number
number of  For a sheet 1.27 mm (0.050 in.) thick, select seven or eight
longitudinal pieces uniformly across the sheet width to provide
one specimen).
9
8.3.2 For cross section thickness
thicknesses
es less than 0.95 mm, ten
Allmand,, T. R., and Col
Allmand Coleman
eman,, D. S., “The Ef
Effect
fect of Sect
Section
ioning
ing Errors on
Microscopic
Microsco pic Determin
Determinations
ations of Non-Me
Non-Metallic
tallic Inclusions in Steels,”
Steels,”   Metals
Metals and  longit
longitudi
udinal
nal pie
pieces
ces fro
from
m eac
each
h sam
sampli
pling
ng loc
locati
ation
on sha
shall
ll be
 Materials,, Vol 7, 1973, pp. 280–283.
 Materials mounted to provide a suitable specimen surface for polishing.

Copyright ASTM International

5
Provided by IHS under license with ASTM
No reproduction or networking permitted without license from I HS Not for Resale
 

1
E 45 – 05 e

(Dependent
(Depe ndent on mate
material
rial thickness
thickness and piece lengt
length,
h, the pol- pixel dimensions in both horizontal and vertical orientations.
ished specimen area may be less than 160 mm 2. Because of  Check the instruction manual to determine how corrections are
practical
practical difficul
difficulties
ties in mount
mounting
ing a grou
group
p of more than ten made for those systems that do not use square pixels.
pieces,
piece s, the reduc
reduced
ed speci
specimen
men area will be consi
considere
deredd suf
sufffi- 10.2.1 Follo
Follow
w the manu
manufact
facturer’
urer’ss reco
recommen
mmendati
dations
ons in ad-
cient.)
cient.) Note that when using the comparison
comparison proc
procedure
eduress of   justing the microscope light source and setting the correct level
Methods A, C, D and E, the thickness of the test specimen of illumination for the television video camera. For systems
cross
cross sec
sectio
tion
n sho
should
uld not be les
lesss tha
than
n the defi
defined
ned min
minimu
imum m with 256 gray levels, the illu
illumina
mination
tion is gener
generally
ally adjusted
adjusted
dimension of a single field of view. Therefore, the minimum until the as-polished matrix surface is at level 254 and black is
thickness required is 0.71 mm for Methods A, D, and E, and at zero.
0.79 mm for Method C. Thinner sections should be rated by 10.2.2 For modern image analyzers analyzers with 256 gray levels, levels,
other means. with the illumination set as described in   10.2.1, 10.2.1,   it is usually
possible to determine the reflectance histogram of individual
9. Prep
Preparati
aration
on of Speci
Specimens
mens inclusions as an aid in establishing proper threshold settings to
9.1 Meth
Methods
ods of speci
specimen
men preparatio
preparationn must be such that a discrimi
disc riminate
nate between oxides and sulfi sulfides.
des. Oxide
Oxidess are darke
darkerr
polished, microscopically flat section is achieved in order that and usually exhibit gray levels below about 130 on the gray
the sizes and shapes of inclusions are accurately shown. To scalee while the ligh
scal lighter
ter sulfides generally
generally exhi
exhibit
bit values be-
obtain satisfactory and consistent inclusion ratings, the speci- tween about 130 and 195. These numbers are not absolute and
men must have a pol polish
ished
ed sur
surfac
facee fre
freee of art
artifa
ifacts
cts such as will vary somew
somewhat hat for dif
differe
ferentnt stee
steels
ls and dif
differe
ferentnt imag
imagee
pitting, foreign material (for example, polishing media), and analyzers
anal yzers.. Afte
Afterr sett
setting
ing the thres
threshold
hold limi
limits
ts to discr
discrimin
iminate
ate
scratches. When polishing the specimen it is very important oxid
ox ides
es an
and d su
sulfi
lfidedes,
s, us
usee th
thee   flicker
flicker metho
method d   of switc
switching
hing
that the inclusions not be pitted, dragged, or obscured. Speci- back-a
bac k-and-
nd-for
forth
th bet
betwee
ween n the livlivee inc
inclus
lusion
ion ima
image ge and the     -
    -
    -
        `

mens must be examined in the as-polished condition, free from detected (discriminated) image, over a number of test fields, to
  ,
  ,
        `
  ,
        `
  ,

the effects of any prior etching (if used). ensure that the settings are correct, that is, detection of sulfides
  ,
        `
  ,
  ,
        `
    -
        `

9.2 Metallographic specimen

specimen preparation must be carefully or oxides by type and size is correct.     -
        `
        `
        `
        `
  ,
  ,
  ,
  ,
        `

con
contr
troll
olled
ed
andtoimage
produc
pro ducee acc
accept
eptabl
ablee qua
qualit
lity
y sur
surfac
faces
es for bot
bothh
        `

manual analysis. Guidelines and recommendations 11. Classification of Inclusions and Calculation
Calculation of          `
  ,
  ,
        `
    -

Severities
    -

are giv
given
en in PraPracti
ctice
ce   E 3,
3 ,   Test
Test Meth
Methods
ods E 45, and Prac
Practice
tice
E 768.
768. 11.1 In these microscopic methods,
methods, inclusions are classified
classified
9.3 Incl
Inclusion
usion retention
retention is gener
generally
ally easier to accom
accomplis
plish
h in into four categories (called Type) based on their morphology
hardened
harde ned steel speci
specimens
mens than in the annealed condition.
condition. If  and two sub subcat
catego
egorie
riess bas
based
ed on thetheir
ir wid
width
th or dia
diamet
meter
er..
inclusion retention is inadequate in annealed specimens, they Categories A-Sulfide Type, B-Alumina Type, C-Silicate Type
should be subjected to a standard heat treatment cycle using a and D-Globular Oxide Type define their shape while categories
relatively low tempering temperature. After heat treatment, the Heavy and Thin describe their thickness. Although the catego-
specimen must be descaled and the longitudinal plane must be ries contain
contain chemi
chemical
cal names that imply knowledge
knowledge of theitheirr
reground
regro und below any decar decarburiz
burization
ation.. This recom
recommend
mendation
ation chemical content, the ratings are based strictly on morphology.
only applies to heat-treatable steel grades. The chemical names associated with the various Types were
9.4 Mount
Mounting ing of speci
specimens
mens is not required
required if unmounted
unmounted derived from historical data collected on inclusions found in
specimens can be properly polished. these shapes or morphologies. The four categories, or Types,
are partitioned into Severity Levels based on the number or
10. Calibration and Standardization length of the particles present in a 0.50 mm2 field of view.
10.1 Recom
Recommend
mended ed calibration
calibration guidelines
guidelines can be found in These Severity
Severity Level
Levelss and inclusion
inclusion Types are depic
depicted
ted in
Guide  E 1951.
Guide E 1951. Plate I-r and their numerical equivalents are found in  Tables 1
10.2
10.2 For image analysis
analysis,, a sta
stage
ge mic
microm
rometeeterr and a rul
ruler
er,, and 2.
2.
both
bot h cal
calibr
ibrate
ated
d aga
agains
instt dev
device
icess tra
tracea
ceable
ble to a rec recogn
ognize
izedd 11.1.1
11 .1.1 Type A and C inclinclusion
usionss are very similar
similar in size and
national standards laboratory, such as the National Institute for shape. Therefore, discrimination between these Types is aided
Standards and Technology (NIST), are used to determine the by meta
metallogr
llographic
aphic techniques.
techniques. Type A-Sulfide are light gray
magn
ma gnifi
ifica
cati
tion
on of th thee sy
syst
stem
em anandd ca
cali
libr
brat
atee th
thee sy
syst
stem
em in while Type C-Silicate are black when viewed under brightfield
accordance with the manufacturer’s recommended procedure. illumination. Discrimination between these types may also be
For exa
exampl
mple,e, the rul
ruler
er is sup
superi
erimpo
mposed
sed ove
overr the mag
magnifinified
ed aided by viewing the questionable inclusions under darkfield or
image of the stage micrometer on the monitor. The apparent cross-pol
cros s-polariz
arized
ed illu
illumina
mination
tion where prope
properly
rly polis
polished
hed sulfi
sulfide
de
(magnified) distance between two known points on the stage inclusions are dark and silicate inclusions appear luminescent.
micrometer is measured with the ruler. The magnified distance 11.2 The B-type stringers consist
consist of a number (at least three)
is di
divi
vide
dedd by th thee tr
true
ue didist
stan
ance
ce to dedete
term
rmininee th
thee sc
scre
reen
en of round or angular oxide particles with aspect ratios less than
magnification. The pixel dimensions can be determined from 2 tha
thatt are ali
aligne
gned
d nea
nearly
rly parallel
parallel to the defdeform
ormati
ation
on axi
axis.
s.
thee nu
th numb
mberer of pipixe
xels
ls fo
forr a kn
know
own n ho
hori
rizo
zontntal
al or ve vert
rtic
ical
al Particles within 615 µm of the centerline of a B-type stringer
dimension on the monitor. Divide the known length of a scale are considered to be part of that stringer. The Type C-Silicate
or mask by the number of pixels representing that length on the stringers consist of one or more highly elongated oxides with
monitor to determine the pixel size for each possible screen smooth
smo oth sur
surfac
faces
es ali
aligne
gned
d par
parall
allel
el to the def
deform
ormati
ation
on axi
axis.
s.
magnification. Not all systems use square pixels. Determine the Aspect ratios are generally high, $ 2. The maximum permitted

Copyright ASTM International

6
Provided by IHS under license with ASTM
No reproduction or networking permitted without license from I HS Not for Resale
 

1
E 45 – 05 e

TABLE 1 Minimum Values

Values for Severity Level Numbers parameters for classification into the Thin or Heavy category
(Methods A, D, and E) A,B  are listed in   Table 2. 2.   An inclusion whose width varies from
(mm (in.) at 1003, or count) Thin to Heavy along its length shall be placed in the category
Severity A B C DC  that best represents its whole. That is to say, if more of its
0.5 3.7(0.15) 1.7(0.07) 1.8(0.07) 1 length falls into the Heavy range, classify it as Heavy. See 11.8
See  11.8
1.0 12.7(0.50) 7.7(0.30) 7.6(0.30) 4 for instructions on reporting inclusions that exceed the limits of 
1.5 26.1(1.03) 18.4(0.72) 17.6(0.69) 9
2.0 43.6(1.72) 34.3(1.35) 32.0(1.26) 16
Table 1 
1   or
or Table
 Table 2.
2.
2.5 64.9(2.56) 55.5(2.19) 51.0(2.01) 25 11.4
11 .4 Inc
Incluslusion
ionss thi
thinne
nnerr tha
than
n the 2 µm min
minim
imum
um lis
listed
ted in
3.0 89.8(3.54) 82.2(3.24) 74.6(2.94) 36 Table 2 are
2  are not rated. That is, their lengths or numbers are not
3.5 118.1(4.65) 114.7(4.52) 102.9(4.05) 49
4.0 149.8(5.90) 153.0(6.02) 135.9(5.35) 64 included in the determination of Severity.
4.5 189.8(7.47) 197.3(7.77) 173.7(6.84) 81 11.5
11 .5 After classificat
classification
ion by type and thick
thickness,
ness, the sever
severity
ity
5.0 223.0(8.78) 247.6(9.75) 216.3(8.52) 100 levels are determined for the inclusions within 0.50 mm2 test
(µm (in.) at 13, or count) areas based upon the total Type A sulfide lengths per field, the
Severity A B C DC  total Type B or C stringer lengths per field, and the number of 
0.5 37.0(.002) 17.2(.0007) 17.8(.0007) 1 isolat
isolated
ed D-t
D-type
ype inc
inclus
lusion
ionss per fiel
field.
d. The
Thesese val
values
ues can be
1.0 127.0(.005) 76.8(.003) 75.6(.003) 4 reported according to the length or number in each 0.50-mm 2
1.5 261.0(.010) 184.2(.007) 176.0(.007) 9 field or as the length per unit area or number per unit area
2.0 436.1(.017) 342.7(.014) 320.5(.013) 16
2.5 649.0(.026) 554.7(.022) 510.3(.020) 25 (mm2), but the me measu
asurem
rement
entss mus
mustt be mad
madee on concontig
tiguou
uouss
3.0 898.0(.035) 822.2(.032) 746.1(.029) 36 0.50 mm2 test areas. Severities
Severities are calc
calculate
ulated
d base
basedd on the
3.5 1181.0(.047) 1147.0(.045) 1029.0(.041) 49
limits given in Table
in  Table 1.
1. Note that these values are the minimum
4.0 1498.0(.059) 1530.0(.060) 1359.0(.054) 64
4.5 1898.0(.075) 1973.0(.078) 1737.0(.068) 81 length or number for each class. In general, severity values
5.0 2230.0(.088) 2476.0(.098) 2163.0(.085) 100 (calculated as described below) are rounded downward to the
(mm/mm2(in./in.2), or count/mm2) neares
nea restt who
whole
le or hal
halff uni
unit.
t. For ste
steels
els wit
with
h par
partic
ticula
ularly
rly low
Severity A B C DC  inclusion contents, severity values may be rounded down to the
-
-
` 
  ,

0.5 0.074(1.88) 0.034(.864) 0.036(.914) 2 nearest quarter

ser.. orHow
tenth value, per agreement
e of the between producer
  ,
` 

and pur
purcha
chaser Howeveever,
r, bec
becaus
ause way D inc inclus
lusion
ion
` 
` 
  ,
  ,
  ,
  ,
1.0 0.254(6.45) 0.154(3.91) 0.152(3.86) 8
1.5 0.522(3.64) 0.368(9.35) 0.352(8.94) 18 counts are defined (for 1 inclusion, the severity is 0.5 and for
` 
` 
` 
` 
-
` 
-
2.0 0.872(22.15) 0.686(17.32) 0.640(16.26) 32
` 
  ,
  ,
` 
  ,
2.5 1.298(32.97) 1.110(28.19) 1.020(25.91) 50
0 inclusions, the severity is 0), there can be no subdivisions
  ,

between 0 and 0.5 severities.

` 
  ,
` 
  ,
  ,
` 
3.0 1.796(45.59) 1.644(41.76) 1.492(37.90) 72
3.5 2.362(59.99) 2.294(58.27) 2.058(52.27) 98
-
-
-
11.6
11.6 Calcu
Calculati
lationon of the severity
severity number for TypeType A, B, and
4.0 2.996(76.10) 3.060(77.72)† 2.718(
2.718(69.
69.04)
04) 128
4.5 3.796(96.42) 3.946(100.2) 3.474(88.24) 162
C inclusions is based on a log-log plot of the data in  Table 1 on 1  on
5.0 4.460(113.3) 4.952(125.8) 4.326(109.9) 200 Minimum Values for Inclusion Rating Numbers (Methods A
A
Note that length values in this table have been changed to be compatible with and D). Such plots10 reveal a linear relationship between the
automated rating methods. The significant length changes occurred at minimum severity numbers and the minimum total sulfide length (Type
rating levels of  1 ⁄ 2  where manual methods are least accurate. Inclusion counts for
Type D inclusions have also been revised. In this case, the changes are greatest
A) and the minimum total stringer length (Types B & C) per
for high counts, which are above the levels of material acceptance standards. 0.50-mm2 field for each severity level as shown in Figs. in  Figs. 10-12.
10-12.
B 
VanderVoort,
Vande rVoort, G. F., and Wilson, R. K., “Nonmetallic Inclusions and ASTM A least-sq
least-squar
uaree fit to the data in   Table Table 1   has bebeen
en usused
ed to
Committee E04,” Standardization
E04,”  Standardization News , Vol 19, May 1991, pp 28–37.
C 
Maximum aspect ratio for Type D inclusions is < 2. †
produce
prod uce the rela
relations
tionships
hips in   Table
Table 6, which can be used to
†
Editorial corrections made February 2007. calculate the severity of Type A, B, and C inclusions, either
thin or thick. The antilog is determined and rounded downward
TABLE
TABLE 2 Inclusion Width and Diameter Parameters to the nearest half-severity value.
(Methods A and D) A 11.7
11.7 Calcu
Calculati
lationon of the severity
severity numbe
numbers
rs for D-typ
D-typee oxide
oxidess
Thin Series Heavy Series is done in the same manner as for Types Types A, B, and C inclusions
Inclusion
Type Width, min, Width, max, Width, min, Width, max, except that the criterion is the number of oxides rather than
µm µm µm µm their length. Fig.
length.  Fig. 13 shows
13  shows a log-log plot of the data in Table
in  Table
A 2 (.00008) 4 (.00016) >4 (.00016) 12 (.0005) 1.
B 2 (.00008) 9 (.00035) >9 (.00035) 15 (.0006) 11.8 The fields shown in Plate I-r representrepresent the total lengths
lengths
C 2 (.00008) 5 (.0002) >5 (.0002) 12 (.0005)
D 2 (.00008) 8 (.0003) >8 (.0003) 13 (.0005)
of the A inclusio
inclusions, ns, the tot
total
al str
string
inger
er len
length
gthss of B and C
A
inclusions, the number of D inclusions, and their respective
Any inclusion with maximum dimensions greater than the maximum for the
Heavy Ser
Heavy Series
ies mus
mustt be rep
report
orted
ed as   oversized  
oversized    accompan
accompanied
ied with its actu
actual
al limiting widths or diameters. If any inclusions are present that
dimensions. are longer than the fields shown in Plate I-r, their lengths shall
be recorded separately. If their widths or diameters are greater
than the limiting values shown in Plate I-r and  Table 2, 2,  they
separation between particles in a stringer is 40 µm. Any oxides shall be recorded separately. Note that an oversize A, B, or C
that have aspect ratios < 2, and are not part of a B- or C-type
stringer,, are rated as D-ty
stringer D-types.
pes. No other shape restriction
restriction is
applicable. 10
Vand
anderer Voor
oort,
t, G. F., and Gol
Golden
den,, J. F., “Au
“Autom
tomatin
ating
g the JK Inclusio
Inclusion
n
11.3 After the inclusions are categorized
categorized by Type,
Type, they must Analysis,”  Microstructural Science,
Analysis,” Microstructural Science, Vol 10, Elsevier Science Publishing
Publishing Co., Inc.,
be cat
catego
egoriz
rized
ed by thi
thickn
ckness
ess or dia
diamet
meter
er.. Inc
Inclus
lusion
ion wid
width
th NY, 1982, pp. 277–290.

Copyright ASTM International

7
Provided by IHS under license with ASTM
No reproduction or networking permitted without license from I HS Not for Resale
 

1
E 45 – 05 e

    -
    -
    -
        `
  ,
  ,
        `
  ,
        `
  ,
  ,
        `
  ,
  ,
        `
    -
        `
    -
        `
        `
        `
        `
  ,
  ,
  ,
  ,
        `
        `
        `
  ,
  ,
        `
    -
    -

NOTE
and   1—The
Heavy seriessquare
square mask
of Types A,will yield
B, C, anda D
field area
is on theofleft.
0.50Several
mm onoversized
the specimen
Typesurface.A graphic
D are depicted onrepresenta
repr
theesentation
right tion of the maxim
maximum
for convenience. um thickness of the Thin
FIG. 5 Suggested Reticle or Overlay Grid For Methods A, D, and E

NOTE   1—Syst
1—Systematic
ematically
ally scan the entire masked
masked area. Methods A, B, C, and E permi
permitt adjustment of the field locations in order to maximi
maximize
ze a severi
severity
ty
level number or facilitate a measurement. For Method D, the fields must remain contiguous and only features within the field are compared to Plate I-r.
NOTE  2—Method D will require a larger (10 3 17 mm) test area to facilitate placement contiguous, 0.71 mm square fields to total 160 mm2
placement of enough contiguous,
of polished surface area.
FIG. 6 Typical Scan Pattern for Microscopic Methods

inclusion or inclusion stringer still contributes to the determi- appear at the bottom of Plate I-r. A Type D globular oxide may
nation of a field’s Severity Level Number. Therefore, if an A, not exceed an aspect ratio of 2:1.
B, or C inclusion is oversized either in length or thickness that 11.9 Oxides located at the tips of Type
Type A-Sulfide inclusions
inclusions
portion that is within the field boundaries shall be included in are rated at Type D- Globular Oxides unless they are close
the appro
appropria
priate
te Thin or Heavy severity
severity level measurement
measurement.. enough
enou gh to
toge
geth
ther
er to memeet
et th
thee re
requ
quir
irem
emen
ents
ts of a Typ ypee
Likewise, if an oversize D inclusion is encountered in a field, B-Alumina.
it is also included in the count that determines the D heavy 11.10
11 .10 The indigenous
indigenous inclu
inclusions
sions in stee
steels
ls deoxi
deoxidize
dizedd with
rating.
rating. For refe
referenc
rence,
e, illu
illustra
stration
tionss of lar
large,
ge, globu
globular
lar oxide
oxidess rare earth elem
elements
ents or calci
calcium-c
um-contai
ontaining
ning mate
material
rialss are also

Copyright ASTM International

8
Provided by IHS under license with ASTM
No reproduction or networking permitted without license from I HS Not for Resale
 

1
E 45 – 05 e

12. Metho
Method
d A (Worst
(Worst Fields)11
Fields)
12.1   Manual
Manual Introduct ion—Th
Introduction —Thisis tes
testt met
method
hod req
requir
uires
es a
survey of a 160 mm2 (0.25 in.2) polished surface area of the
specimen at 1003. The field size shall equal an area equivalent
to 0.50 mm2 (0.000779 in.2) on the specimen surface as defined
by a square with 0.71 mm (0.02791 in.) long sides (see  Fig. 5). 5).
Each 0.50 mm2 field is compared to the square fields depicted
in Plate I-r in a search for the worst field, that is, the highest
severity rating, of each inclusion Type A, B, C, and D for both
the   Thin   and   Heavy  series. The severity level of these worst
fields shall be reported for every specimen examined.
12.2   Manual Procedure:
12.2
12 .2.1
.1 Eith
Eitherer of twtwo
o te tech
chni
niqu
ques
es ma
may y be emempl
ploy
oyed
ed to
2
achieve a 0.50 mm field of view. One method is to project the
1003   micros
microscop
copee ima
image
ge ont
onto
o a vie
viewin
wing g scr
screen
een tha
thatt has a
NOTE   1—One
1—One uni
unitt eq
equa
uals
ls 0.
0.12
127
7 mm (0
(0.0
.005
05 in.
in.)) on the spe
speci
cime
men
n
square mask with 71.0 mm (2.79 in.) sides drawn on it. Another
surface. option is to use a reticle made for the microscope, which will
FIG. 7 Suggested Reticle or Overlay Grid for Method B superimpose the required square mask directly onto the field of 
view (see Fig.
(see  Fig. 5).
5).
12.2
12 .2.2
.2 To
To be begi
gin,
n, ou
outl
tlin
inee th
thee re
requ
quir
ired
ed te
test
st ar
area
ea on th thee
classified by morphology and thickness with the added require- specimen surface using either an indelible marker or a carbide-
ment that compositional information be given in the report. For tipped scribe. Place the specimen on the microscope stage and
example, rare earth or calcium-modified sulfides with an aspect start the examination with a field in one of the corners of the
ratio   $   2 are rated as A-types by their total length per field marked test area. Compare this field to the images on Plate I-r.
according to the limits of  Table
  Table 1 and
1  and the width limits of  Table Table Classify the inclusion content of the field based on the rules
2. However, for aspect ratios < 2 and if they are not part of a listed in Section   11
11 concerning
 concerning Type, and thickness of inclu-
stringer, they are rated as D-types by their number per field sions present. Record the severity level in whole numbers from
according to the number limits of  Table  Table 1 and
1  and the width limits 0 to 3.0 for each inclusion type (A, B, C, and D) that most
of    Tabl
of  Tablee 2. In boboth
th ca
case
ses,
s, a ge
gene nera
rall de
desc
scri
ript
ptio
ion
n of th thei
eirr resembles the field under observation. (See Table
(See  Table 1 if
1  if required
    -

compositi
compo sition
on must be provi
provided
ded to avoiavoid d conf
confusion
usion.. Becau
Becausese to report severity levels > 3.0). Do this for both the Thin and     -
    -
        `
  ,
  ,
        `

they
the y are sulfides
sulfides wit
withh a D-t
D-type
ype morpholo
morphology gy,, the
theyy ma
mayy be Heavy series. It is important to note here that if a field of    ,
        `
  ,
  ,
        `
  ,

inclus
inc lusion
ionss fal
falls
ls bet
betwee
weenn two sev
severi
erity
ty lev
levels
els,, its valvalue
ue is
  ,
        `

referred to as   DS .
    -
        `
    -
        `
        `

rounded down to the lower severity level. For example, when         `
        `
  ,

11.11
11. 11 Compl
Complexex inclu
inclusions
sions,, such as oxysuoxysulfides
lfides or duple
duplex x   ,
  ,
  ,

using Plate I-r, a field that contains fewer inclusions, or less         `
        `
        `

inclus
inc lusion
ions,
s, are als
also
o rat
rated
ed acc
accord
ordinging to the their
ir mor
morpho
pholog
logy:
y:   ,
  ,
        `

inclusion length than Severity Level Number 1, is counted as a

    -
    -

whether they are stringered or elongated (for aspect ratios $ 2)

0.
or isolated (not part of a stringer and aspect ratio < 2); and then
12.2.3 Move the microscope
microscope stage to reveal an adjacent
adjacent field
by thickness. Isolated, globular particles are rated as D-types
and repea
repeatt the comp
compariso
arison
n proce
procedure
dure.. Conti
Continue
nue this process
by their average thickness. Complex Ds  may be predominantly
until the required polished surface area of the specimen has
(>50 % by area) sulfides or oxides and should be identified as
such. For example, if the oxide area is greater in a globular been scanned.requires
This method A typical scan configuration
adjustment is shown instage
of the microscope  Fig. to
6.
6.
oxysulfide, it could be called a D OS  type. Stringered complex
maximize an inclusion severity level. That is, the field of view
particles are rated by the aspect ratio of the individual particles;
is adj
adjust
usted
ed usi
using
ng the mimicro
crosco
scope
pe sta
stage
ge con
contro
trols,
ls, suc
such
h tha
thatt
if < 2, they are B-types, if  $ 2 they are A- or C-types (separat
(separatee
inclusions are moved inside the square mask in order to locate
by gray level). For those complex inclusions with aspect ratios
the  worst field . In practice, the rater is actually scanning the
$   2, they are classified as A-types if more than 50 % of the
specimen and stopping only when a potential   worst field   of 
area is sulfide and C-types if more than 50 % of the area is
each type and thickness is in view.
oxide.
oxi de. Rep
Report
ort the com
compos
positi
ition,
on, in gen
genera
erall ter
terms,
ms, to avo
avoid
id
12.3   Manual Expression of Results :
confusion, and state the nature of the inclusions, for example,
12.3.1 The averages
averages of the worst fields for each inclusion
inclusion
“globular calcium aluminates encapsulated with a thin film of 
typee in all the spe
typ specim
cimens
ens of the lot shashall
ll be cal
calcul
culate
ated
d in
calcium-manganese sulfide,” or “irregular aluminates partially
accordance with the Severity Level Numbers given at the sides
or fully embedded in manganese sulfide stringers.”
of Pl
Plat
atee I-
I-rr or   Table
Table 1.   An exa
exampl
mplee sho
showin
wing g the ave
averag
rages
es
11.12 If producer-purchaser
producer-purchaser agreements limitlimit the analysis to
obtained for six specimens examined is given in  Table 4. 4.
only certain inclusion types, thickness categories, or severity
12.3.2 Oversize inclusions with with widths or diameters greater
greater
limits, the scheme in Section   11 11 can
 can be modified to analyze,
than the limiting values
values shown on Plate I-r (and Table
(and  Table 2)2) shall
measure, and store only the data of interest. It may also contain be recorded separately.
procedures to perform basic (see Practice E
Practice  E 1245)
1245) stereological
measurements to supplement the JK analyses. Such measure-
ments are not covered by this practice. 11
This method is similar to the Jernkontoret Method, Uppsala, Sweden (1936).

Copyright ASTM International

9
Provided by IHS under license with ASTM
No reproduction or networking permitted without license from I HS Not for Resale
 

1
E 45 – 05 e

NOTE  1—d = disconnected, vd = very disconnected, and g = grouped.

NOTE  2—0.51 mm = 0.02 in.
    -

FIG. 8 Designation of Length and Weight of Inclusions (4 Units Long)     -

    -
        `
  ,
  ,
        `
  ,
        `
  ,
  ,
        `
  ,
  ,
        `
    -
        `
    -
        `
        `
        `
        `
  ,
  ,
  ,
  ,
        `
        `
        `
  ,
  ,
        `
    -
    -

NOTE   1—One
1—One uni
unitt eq
equa
uals
ls 0.
0.12
127
7 mm (0 (0.0
.005
05 in.
in.)) on the spe
speci
cime
men
n
surface. Dimensions equal actual distance on the specimen surface and
2

will yield a field area of 0.83 mm .

FIG. 9 Suggested Reticle or Overlay Grid for Method C

12.3.3 If desired,
desired, the predominant
predominant chemical
chemical type of incl
inclu-
u-
sions may be determined and recorded as sulfide, silicate, or
oxid
ox ide.
e. If th
thee ch
char
arts
ts ar
aree us
used
ed to ra
rate
te ca
carb
rbid
ides
es or ni
nitr
trid
ides
es,,
chemical composition information may also be determined and
reported. FIG. 10 Relationship between Severity Rating and the Minimum
12.4   Image Analysis Introduct ion—The inclusions
Introduction inclusions on the Total Sulfide Length for Plate I-r and Table
and  Table 1
surface
surface of a prope
properly
rly prepa
prepared
red as-po
as-polish
lished
ed meta
metallogr
llographic
aphic
specimen are viewed with a high-quality, metallurgical micro-
scop
scope.
e. Th
Thee br
brig
ight
ht-fi
-fieleld
d im
imag
agee is pi
pick
cked
ed up by a su suit
itab
able
le down on th
down thee st
stag
agee pl
plat
atee an
and
d ho
hold
ld in pl
plac
acee wi
with
th th
thee st
stag
agee
television camera and transferred to the image analyzer screen. clamps. With an upright-type microscope, place the sample on
Image analysis software is then used to evaluate the inclusion a slide and level the surface using clay or plasticene and a
content of the material based
based on microscopic
microscopic Methods A and D. hand-l
han d-leve
evelin
lingg pre
press.
ss. Cer
Certai
tain
n upr
uprigh
ightt mic
micros
roscop
copes
es can be
12.5   Image Analysis Procedure: equipped with an autoleveling stage for mounted specimens. If 
12.5.1 Plac
Placee the spec
specimen
imen on the microscope
microscope stage so that the sample must be leveled using clay, the tissue paper placed
the specimen surface is perpendicular to the optical axis. With between the specimen surface and the leveling press ram may
an inverted-type microscope, simply place the specimen face- adhere to the surface and present artifacts for measurement. In

Copyright ASTM International

10
Provided by IHS under license with ASTM
No reproduction or networking permitted without license from I HS Not for Resale
 

1
E 45 – 05 e

FIG. 11 Relationship between Severity Rating and the Minimum FIG. 12 Relationship between Severity Rating and the Minimum
Total B-Type Stringer Length for Plate I-r and  Table 1 Total C-Type Stringer Length for Plate I-r and  Table 1

TABLE 3 Worst-Field Inclusion

Inclusion Ratings (Method A)
some cases, adherent tissue
tissue can be blown off the specspecimen
imen Severity LevelsA
surface. An alternative procedure to avoid this problem is to Type A Type B Type C Type D
Specimen
place an aluminum or stainless steel ring form, which has been Thin Heavy Thin Heavy Thin Heavy Thin Heavy
flattened slightly in a vise to an oval shape, between the sample 1 2 1 2 1 1 0 2 1
and the ram. If the specimen was mounted, the ring form will 2 3 1 2 1 0 1 2 2
rest
rest on
only
ly on ththee su
surf
rfac
acee of th
thee mo
moun
unti
ting
ng ma
mate
teri
rial
al.. If th
thee 3 2 1 2 1 0 0 2 2
4 2 1 2 1 1 0 2 1
specimen is unmounted but with a surface area substantially 5 2 1 2 1 0 1 2 1
2
greater than the 160-mm area required for the measurement, 6 3 1 2 1 0 0 2 1
the ring form can rest on the outer edges of the specimen for Average 2.3 1.0 2.0 1.0 0.3 0.3 2.0 1.3
flattening and thus avoid contact with the measurement area. A
See 12.3.1
See  12.3.1..
Align
Ali gn the specimen
specimen on the stage so tha thatt the inclusio
inclusions
ns are Oversized inclusions—Type D, width; 17 micrometers.
aligned parallel to the  x -direction
-direction of the stage movement, that
is, horizontal on the monitor screen. Alternatively, if program-
ming
min g is fac
facili
ilitat
tated,
ed, ali
align
gn the incinclus
lusio
ions
ns par
paral
allel
lel to the analyzer is capable of such a procedure. 12 If the system cannot
 y-dir
-directi
ection
on of the stage movem
movement,
ent, that is, the longi
longitudin
tudinal
al work in this manner, that is, if the inclusions in each field must
direction is vertical on the monitor screen. be dis
discri
crimin
minate
ated
d by typ
type,
e, mea
measur
sured,
ed, and a sevseveri
erity
ty lev
level
el
12.5.2 Check the microscope
microscope light source
source for corre
correct
ct align- assigned on a field-by-field basis, then the magnification must
ment and adjust the illumination to the level required by the be chosen so that the field area is as close to 0.50 mm 2 as
television video camera.
12.5.3 The inclusions
inclusions can be examined and discr discrimin
iminated
ated 12
Forget,
For get, C., “Im
“Impro
proved
ved Meth
Method
od for E1E1122
122 Ima
Image
ge Ana
Analys
lysis
is Non
Nonmeta
metallic
llic
by type using magnifications other than 1003  and field areas
Ratings,”  MiCon 90: Advances in Video Technology for Microstructural
Inclusion Ratings,” MiCon
other than 0.50 mm 2 as long as the severity measurements are 1094,, American Society for Testing
Control,, ASTM STP 1094
Control Testing and Materials, Philadelphia,
2

based upon the required

required 0.50-
0.50-mm
mm fiel
field
d are
area,
a, if the image
image 1991, pp. 135–150.

--`,,```,,,,````-`-`,,`,,`,`,,`---

Copyright ASTM International

11
Provided by IHS under license with ASTM
No reproduction or networking permitted without license from I HS Not for Resale
 

1
E 45 – 05 e

TABLE
TABLE 4 Example of Inclusion Rating (Method D)
Number of Fields in Each Specimen Average of Six
Severity Level Number Specimen Number Specimens
1 2 3 4 5 6 Thin Heavy
Type A
0.5 Thin 65 60 50 65 37 56 55.5
Heavy 9 8 12 6 16 8 9.8
1.0 Thin 19 15 31 8 12 10 15.8
Heavy 4 3 4 1 2 1 2.5
1.5 Thin 1 3 2 0 1 0 1.2
Heavy 0 0 0 0 0 0 0
2.0 Thin 1 0 0 0 0 0 0.2
Heavy 0 0 0 0 0 0 0
2.5 Thin 0 0 0 0 0 0 0
Heavy 0 0 0 0 0 0 0
Type B
0.5 Thin 13 8 7 6 11 10 9.2
Heavy 0 0 0 1 1 0 0.3
1.0 Thin 13 14 10 6 12 12 11.2
Heavy 0 0 0 0 2 1 0.5
1.5 Thin 1 6 6 3 3 2 3.5
Heavy 0 0 0 0 0 0 0
2.0 Thin 0 2 1 0 1 1 0.8
Heavy 0 0 0 0 0 0 0
2.5 Thin 0 1 0 0 1 0 0.3
Heavy 0 1 0 0 0 0 0.2
Type C
0.5 Thin 0 0 0 0 1 0 0.2
Heavy 0 0 0 0 0 0 0
1.0 Thin 0 0 0 0 0 0 0
Heavy 0 0 0 0 0 0 0
1.5 Thin 0 0 0 0 0 0 0
Heavy 0 0 0 0 0 0 0
2.0 Thin 0 0 0 0 0 0 0
Heavy 0 0 0 0 0 0 0
2.5 Thin 0 0 0 0 0 0 0
Heavy 0 0 0 0 0 1 0.2
Type D
0.5 Thin 35 33 28 32 47 29 34.0
Heavy 9 4 5 6 9 9 7.0
1.0 Thin 13 10 20 9 12 41 17.5
Heavy 0 2 2 1 2 4 1.8
1.5 Thin 0 0 4 0 0 6 1.7
Heavy 0 0 0 0 0 0 0
2.0 Thin 0 0 0 0 0 0 0
Heavy 0 0 0 0 0 0 0
2.5 Thin 0 0 0 0 0 0 0
Heavy 0 0 0 0 0 0 0
Max D Size 0.0305 mm 0.0254 mm 0.0254 mm
(0.0012 in.) (0.001 in.) (0.001 in.)

Oversized inclusions—Type D, width; 17 micrometers.

TABLE 5 SAM Rating

Rating (Method E) results. The magnification chosen should produce a maximum
B-Type rating A,B 
D-Type Rating A,C 
calibration factor of 1.3 microns at 100 3.
No. of No. of No. of 12.5.4
12.5.4 Sel
Select
ect the gra
gray-l
y-leve
evell thr
thresh
eshold
old set
settin
tings
gs to per
permit
mit
“B” “B” “D”
Observed Observed Observed   Units independent detection of sulfides and oxides (see 10.2.2
(see  10.2.2)).
Thin Heavy Heavy
Fields Fields Fields
12.5.5 When detec
detecting
ting sulfides,
sulfides, a fals
falsee image (called
(called the
not recorded 0.5 not recorded 0.5 5 0.5 (1)
not recorded 1.0 2 1.0 2 1.0 (2)
halo effect ) may be detected around the periphery of oxides in
3 1.5 1 1.5 1 1.5 (3) the same field. This problem can be corrected by the use of an
1 2.0 0 2.0 0 2.0 (4) auto-delineation feature or by application of appropriate algo-
0 2.5 0 2.5 0 2.5 (5)
rithms
rithms to the binary image. Choice of the most satisfactor
satisfactory
y
A
Total area observed = 1.5 in.2
B 
approach depends upon the image analysis system used.
SAM rating = (3 3 1.5
1.5)) + (1 3 2) + (2 3 1) + (1 3 1.5
1.5)) = 10 4 1.5 =
1.5  = 7.
C 
SAM rating = (5 3 1) + (2 3 2) + (1 3 3) = 124 1.5
1.5 =
 = 8. 12.5.6
12.5.6 Set the sta
stage
ge con
contr
trols
ols to mov
movee the specime
specimen n in a
square or rectangular pattern with contiguous field alignment
so that a minimum of 160 mm2 is examined and evaluated.
possi
possible.
ble. A devia tion of less than 60.05 mm2 from the required
deviation Other
Oth er mea
measur
sureme
ement
nt are
areas
as ma
mayy be use
used
d bas
based
ed on pro
produc
ducer
er--
0.50-mm2 areareaa wil
willl not sig
signifi
nifican
cantly
tly imp
impair
air mea
measur
sureme
ement
nt purchaser agreements.
--`,,```,,,,````-`-`,,`,,`,`,,`---

Copyright ASTM International

12
Provided by IHS under license with ASTM
No reproduction or networking permitted without license from I HS Not for Resale
 

1
E 45 – 05 e

TABLE 6 Regression Equations for Severity Rating Calculations

TABLE Calculations
(Based on the Four Alternate Ways of Expressing A, B, or C
Lengths or Two Ways to Express D Counts in  Table 2)
2)
1. Lengt
Length
h in in. at 1003 or count per field
A Log(Se
Log(Sevv.) = [0.560
[0.5 6052
522L
2Log
og(A
(A)])] + 0.
0.16
1688
8870
70
B Log(
Log(Se
Sevv.) = [0.4
[0 .462
6263
631L
1Log
og(B
(B)])] + 0.
0.24
2410
1092
92
C Log(
Log(Se
Sevv.) = [0.4
[0 .480
8073
736L
6Log
og(C
(C)])] + 0.
0.25
2521
2106
06
D Log
og(S
(Se
e v.
v.) = [ 0.
0. 5Lo
5Log
g(D
(D)]
)] - 0 .3
.3 01
01 02
02
2. Leng
Length
th in mm at 1003 or count per field
A Log(Se
Log(Sevv.) = [0.561
[0.5 6173
739L
9Log
og(A
(A)]
)] − 0.
0.62
6200
003
3
B Log(
Log(Se
Sevv.) = [0.4
[0 .463
6333
336L
6Log
og(B
(B)]
)] − 0.
0.41
4101
017
7
C Log(
Log(Se
Sevv.) = [0.4
[0 .479
7973
731L
1Log
og(C
(C)]
)] − 0.
0.42
4213
132
2
D Log
og(S
(Se
e v.
v.) = [ 0.
0. 5Lo
5Log
g(D
(D)]
)] − 0.
0.30
3010
1022
3. Lengt
Length
h in µm at 13 or count per field
A Log(Se
Log(Sevv.) = [0.561
[0.5 6173
739L
9Log
og(A
(A)]
)] − 1.
1.18
1817
177
7
B Log(
Log(Se
Sevv.) = [0.4
[0 .463
6333
336L
6Log
og(B
(B)]
)] − 0.
0.87
8735
35
C Log(
Log(Se
Sevv.) = [0.4
[0 .479
7973
731L
1Log
og(C
(C)]
)] − 0.
0.90
9010
105
5
D Log
og(S
(Se
e v.
v.) = [ 0.
0. 5Lo
5Log
g(D
(D)]
)] − 0.
0.30
3010
1022
4. Length per unit area (mm/mm2) or count per unit area (no./mm 2)
Length
A Log(Se
Log(Sevv.) = [0.561
[0.5 6173
739L
9Log
og(A
(A)]
)] − 0.
0.33
3343
434
4
B Log(
Log(Se
Sevv.) = [0.4
[0 .463
6333
336L
6Log
og(B
(B)]
)] − 0.
0.37
3770
7021
21     -
    -
    -
        `

C Log(
Log(Se
Sevv.) = [0.4
[0 .479
7973
731L
1Log
og(C
(C)]
)] − 0.
0.39
3937
3723
23   ,
  ,
        `
  ,
        `

D Log
og(S
(Se
e v.
v.) = [ 0.
0. 5Lo
5Log
g(D
(D)]
)] − 0.
0.45
4515
1544   ,
  ,
        `
  ,
  ,
        `
    -
        `
    -
        `
        `

NOTE   1—Choo
1—Choosese the equatio
equations
ns to calcul
calculate
ate the inclusion severity (both         `
        `
  ,
  ,
  ,

thin and heavy series) based on the nature of the measurement used; all
  ,
        `
        `
        `
  ,
  ,

approaches give the same severity values.         `

    -
    -

NOTE   2—Round
2—Round of offf the sev
severi
erity
ty num
number
ber down
downwar
wardd to the nea
neares
restt
half-severity level (or, if desired, to the nearest one-quarter or one-tenth
value). For D-type inclusions, because we have only whole integer counts,
and 0.5 is the severity for one inclusion in a field (a field has an area of 
0.5 mm2), there cannot be a D severity of 0.25 or any one-tenth value
below 0.5, except for 0 if there are no ratable Ds present. FIG. 13 Relationship between Severity Rating and the Minimum
NOTE   3—To determine the severity value using the above equations, Number of Globular D-Type Inclusions for Plate I-r and Table
and  Table 1
take the Log (base 10) of the measured value, multiply by the indicated
value, subtract or add the indicated value, then take the antilog and round
downward as described above. 12.5.10 The use of randomly
randomly selected, contiguously
contiguously aligned
fields may not produce true worst field (Method A) ratings.
Vali
alid
d wor
worstst fiel
fieldd rat
rating
ingss req
requir
uiree adv
advanc
anced
ed ima
imagege ana
analys
lysis
is
12.5.7 Use a previ
previously
ously written
written computer
computer program to sepa- technology, for example, use of a 0.50-mm2 mask that can be
rate the inclusion images by type and thickness, then calculate moved
mov ed any
anywhe
wherere wit
within
hin the 160 -mm2 te
160-mm test
st ar
area
ea ususin
ing
g an
severi
sev eritie
tiess by len
length
gth or num
number
ber based on the rules listed
listed in algorithm that controls the mask movement by maximizing the
Section 12
Section  12.. The program should also store results, control stage severity values.
movements (if an automated stage is used), and generate the 12.5.11 For quantitative inclusion descriptions,
descriptions, blank fields
test report. (that is, those that contain no visible inclusions of a particular
12.5.8 If the width of an A inclusion,
inclusion, or a B or C stringer,
stringer, type and width) may be differentiated from non-ratable fields
varies and becomes less than 2 µm over part of its length, (that
(that is, fiel
fields
ds wit
with h inc
inclus
lusion
ionss   #   2 µm in wi widt
dth,
h, or wiwith
th
detect as much of it as possible and calculate the severity based inclusion lengths or stringer lengths below the minimum limit
on the detected length. For specimens from wrought products for 0.5 severity).
with
wit h hig
highh deg
degree
reess of red
reduct
uction
ion,, whe
where
re the ma
major
jority
ity of the 12.5.12 The program should incorporateincorporate procedures to deal deal
inclus
inc lusion
ionss are < 2 µm thi thick,
ck, bas
based
ed on pro
produc
ducer
er-pu
-purch
rchase
aserr with
wit h fiel
fieldsds tha
thatt con
contai
tainn art
artifa
ifacts
cts,, eit
either
her fro
fromm pol
polish
ishing
ing or
agreement, the minimum thickness of the thin series can be set cleaning, or from dust settling on the specimen, and so forth.
at a lower value, such as 0.5 µm, or the lower limit can be The alignmen
alignmentt of Type A, B, and C inc inclus
lusion
ionss in wro
wrough
ughtt
dropped. Detection of these thinner inclusions will require use specimens generally will not deviate by more than 620° from
of a higher magnification with a resultant field size less than the longitudinal direction. Depending on the system and the
0.50 mm2; hence, field data must be combined, as described in natu
na ture
re of th thee ar
arti
tifa
fact
ct,, it ma
may y be po poss
ssib
ible
le to de deve
velo
lop
p an
12.5.3,, to obtain valid ratings.
12.5.3 algorithm that will recognize such artifacts and remove them
12.5.9 An array is estaestablis
blished
hed in the computer
computer memory to from
from th thee bi
bina
nary
ry imimag
age.e. By re rest
stri
rict
ctin
ingg th
thee or
orie
ient
ntat
atio
ion
n of 
tabulate the number of fields that were rated according to the detected
dete cted features within this limit, cert certain
ain arti
artifact
factss (for ex-
Thin and Heavy limits of the four inclusion types for eleven ample, deep scratches not removed during polishing) can be
possible severities from 0 to 5 in half-level increments. After recognized and deleted from the binary image, if their orien-
each
each fie
field
ld is rarate
ted
d an
and
d th
thee se
seve
veri
riti
ties
es ar
aree co
comp
mpututed
ed,, th
thee tation is greater than this limit. If this cannot be done, the field
appropriate array locations are incremented to store the results. should be rejectable, that is, no test results from the field should

Copyright ASTM International

13
Provided by IHS under license with ASTM
No reproduction or networking permitted without license from I HS Not for Resale
 No reproduction or networking permitted without license from I HS Not for Resale

1
E 45 – 05 e

be stored. In such a case, another field should be analyzed to 13.2.1 This method utilizes
utilizes a pattern of parallel
parallel lines whose
replace the rejected field, if this is possible. If a rejected field spacing is such that the distance between lines is equivalent to
cannot
can not be rep
replac
laced
ed in the sam
samee run
run,, it may be pos
possib
sible
le to 0.127 mm (0.005 in.) on the specimen surface when viewed at
evaluate and rate the additional fields required in a subsequent 1003. ThThis
is di
dist
stan
ance
ce sh
shal
alll be re
refe
ferr
rred
ed to as on
onee ununit
it.. Th
Thee
run
run (d(do
o no
nott ra
rate
te fie
field
ldss al
alre
read
adyy ra
rate
ted)
d).. Go
Good
od pr
prep
epar
arat
atio
ion
n pattern may be drawn on (or taped to) a viewing screen, in
practices will minimize the need to reject fields with artifacts. which case the physical distance between lines would be 12.7
In no case should the test results for a measurement area less mm (0.5 in.) since the specimen is magnified 100 times. An
than 160 mm2 be mathmathemat
ematical
ically
ly extra
extrapolat
polated
ed or conv
converted
erted alternate technique would be to have a reticle made that will
(for example, because of rejected fields) in an effort to produce superimp
supe rimpose
ose the requi
required
red pattern directly
directly onto the imag
imagee as
2
data for a 160-mm area. seen through the eyepieces of the microscope.  Fig. 7 shows
7  shows a
12.6   Image Anlysis Expression of Results : recommended measurement grid for use with Method B. Note
12.6.1 The number of fields of each inclusion type
-
-
` 
  ,
type (A, B, C, that the parallel lines are contained in a mask to aid in the
  ,

and D) and thickness (thin and thick) are reported for each
` 
` 
` 
  ,
  ,
  ,
  ,
indexing of fields.
severity from 0 to 5 in whole or half-severity level increments.
` 
` 
` 
` 
-
` 

For steels with very low inclusion contents, severities may be

-
` 
  ,
13.2.2
13.2 .2 To
To be
begi
gin,
n, ou
outl
tlin
inee th
thee re
requ
quir
ired
ed te
test
st ar
area
ea on ththee
  ,

specimen surface using either an indelible marker or a carbide-

` 
  ,

computed
compu ted to one-
one-quart
quarter
er or one-t
one-tenth
enth severity level incr
incre-
e-
  ,
` 
  ,
` 
  ,
  ,
` 

ments. Note that for D-type inclusions, because one inclusion

-
-
-
tipped scribe. Place the specimen on the microscope and start
per field is a severity of 0.5, by definition, there can be no the examination with a field in one of the corners of the marked
D-severity levels between 0 and 0.5. test area. Measure and record all inclusions in this field that are
12.6.2 If desired, based on producer-purchaser
producer-purchaser agreements,
agreements, one uni
unitt lon
long
g or lon
longer
ger.. Inc
Inclus
lusion
ionss sep
separa
arated
ted by a dis
distan
tance
ce
modifications of the reported data may be made, for example, greater than one unit shall be classified as two inclusions and
reports for only certain inclusion types, thicknesses, or severity not be considered as one stringer. The length of an inclusion
values
values.. Oth
Other
er mod
modific
ificati
ations
ons may incinclud
ludee onl
only
y wor
worst-
st-fiel
field
d shall be rounded down to the next whole unit and only whole
severi
sev erity
ty rat
rating
ingss or the num
number
ber of fiel
fields
ds at the wor
worst-
st-fiel
field
d units will be recorded. For example, if an inclusion measures
1

severity ratings. 2  ⁄ 2   units, it shall be recorded as a “2.” If an inclusion lies

12.6.3 If desired by producer-purcha
producer-purchaser ser agreement, an index partially outside of the field, that is, part of its length lies in
may be calculated to describe the inclusion content. what will become Field Number 2, move the field slightly in
12.6.4 To produce average results
results for more than one spec speci-
i- order that its entire length may be measured.
men per lot, the average number of fields for each severity 13.2.3 Move the microscope stage stage to view an adjacent field.
field.
rating,
rating, inclu
inclusion
sion type, and thickness
thickness may be calc calculate
ulated
d (see Repeat the measurement procedure. Take care that any inclu-
Table 4)4). sion measured in the previous field is not remeasured. Continue
12.6.5 Data for inclusions
inclusions or stri
stringers
ngers that are oversized
oversized in thiss pro
thi proces
cesss unt
until
il the req
requir
uired
ed pol
polish
ished
ed sur
surfac
facee are
areaa of the
either length or width, or both, should be reported separately. specimen
spec imen has been scanned. A typical scan config configurati
uration
on is
Reportt the inclusion
Repor inclusion type, measmeasured
ured width, and lengtlengthh (for shown in Fig.
in  Fig. 6.
6.
Types A, B, and C).
13.3   Expression of Results :
12.6
12 .6.6
.6 Fields
Fields wiwith
th ze
zero
ro se
seve
veri
rity
ty le
leve
vels
ls ma
mayy be fu furt
rthe
herr
clas
classi
sifie
fied,
d, if de
desi
sire
red,
d, as eieith
ther
er bl
blan
ankk (n
(noo in
incl
clus
usio
ions
ns of a 13.3.1 The det
13.3.1 determ
ermina
inatio
tion
n for eac
each
h spe
specim
cimen
en sha
shall
ll be di-
particular type and width category are present) or non-ratable vided into two parts, as follows:
(inclusions are present but their length is below the 0.5 severity 13.3.1
13. 3.1.1
.1 The len length
gth of the lonlonges
gestt inc
inclus
lusion
ion shall be re-
limit or their width is < 2 µm), or their diameter is < 3 µm. corded first. It shall be supplemented to describe the inclusion
12.6.7
12. 6.7 Inf
Inform
ormati
ation
on per
pertai
tainin
ning
g to the comcompos
positi
ition
on of the widt
widthh by a susupe
persrscr
crip
iptt T fo
forr th
thin
in or H fo forr he
heav
avy
y. A ththin
in
inclusions (Types A to D) may be provided if desired. For rare inclusion is defined as being 10 µm (0.0004 in.) or less in width
earth- or calcium-treated steels, or other steels with nontradi- over more than 50 % of its entire length. Likewise, a heavy
tional deoxidation approaches, the chemical composition of the inclusion must have a thickness of 30 µm (0.012 in.) or more
inclusions, in general terms, must be reported with each rating. over the majority of its length. Inclusions greater than 10 µm
Microanal
Micr oanalytic
ytical
al tech
technique
niquess may be requi
required
red to obtai
obtain
n such but less than 30 µm wide shall not be represented by a T or H
information if the operator is not able to identify the inclusions superscript. Supercripts   d   (disconnected),   vd   (very disco
discon-
n-
by light optical examination. nected),
nected), and   g   (group
(grouped)
ed) may alsalso
o be use
usedd to des
descri
cribe
be the
12.6.8 Supplementary stereological
stereological data determined
determined during degree of connectivity or clustering as illustrated in  Fig. 8. 8.
analysis may be included in the test report as desired. Stan-
13.3.1.2 The average length
13.3.1.2 length of all inclusions
inclusions one unit and
dardization of such test data is not governed by this practice
longer in length, but excluding the longest inclusion, shall be
(see Practice E
Practice  E 1245)
1245).
reported as a single number, followed by a superscript denoting
13. Meth
Method
od B (Length) the number of inclusions averaged.
13.1   Introduction—This test method requires a survey of a 13.3.2 When required,
required, a seri
series
es of compa
compariso
rison
n photo
photomicr
micro-
o-
2
160 mm polished surface area of the specimen at 1003. Any graphs at 1003, which illustrates
illustrates all othe
otherr nonme
nonmetall
tallic
ic par
par--
incl
inclus
usio
ion n wh
whos
osee le
leng
ngth
th is 0.
0.12
127
7 mm or lo
long
nger
er is to be ticles
ticles pres
present,
ent, may be used to char character
acterize
ize the back
backgroun
ground
d
measured and individually tallied. appearance of the specimen. If used, these shall be labeled A,
13.2   Procedure: B, ... etc
etc.,
., in ord
order
er of inc
increa
reasin
sing
g inc
inclus
lusion
ion pop
popula
ulatio
tion.
n. The

Copyright ASTM International

14
Provided by IHS under license with ASTM
 No reproduction or networking permitted without license from I HS Not for Resale

1
E 45 – 05 e

specific photomicrographs used shall be mutually agreed upon shown in   Fig. 6. 6. It is permissible, and will be necessary at
between the interested parties.13 times,
tim es, to adjadjust
ust the mic
micros
roscop
copee sta
stage
ge suc
suchh tha
thatt the ententire
ire
13.3.3 The following
following is an expr
expressio
ession
n of resu
results
lts for a sing
single
le stringer may be viewed within the mask. The rater’s objective
d 3
specim
spe cimen
en by thi
thiss met
methodhod:: 6 -2 -A. Thi
Thiss ind
indica
icates
tes tha
thatt the is to fin
find d th
thee lo
long
ngesestt ox
oxid
idee an
andd si
sili
lica
cate
te st
stri
ring
nger
erss in th thee
longest inclusion observed was six units long, that three other specimen. Therefore, in practice, the rater is actually scanning
inclusions were observed whose average length was two units, the spe
specim
cimen en and stostoppi
pping
ng onl
only
y whe
when n a pot
potent
ential
ial   longest 
and that the background inclusions were similar in appearance stringer  is
  is in view.
to the A figure from a background photomicrographic series. 14.3   Expression of Results :
13.3.4
13.3.4 The res
result
ultss for all spe
specim
cimens
ens from a lot shall
shall be 14.3.1
14.3.1 The ma maxim
ximumum lenlength
gth of eac
each h typ
typee of inc
incluslusion
ion,,
tabulated
tabulated.. If req
requir
uired,
ed, the pre
predom
domina
inant
nt typ
typee of inc
inclus
lusion
ionss usually a series of individual particles in a stringer, is generally
(sulfides, silicates, or oxides) shall be recorded. used to evaluate a specimen. The silicate photomicrographs are
used for deformable-oxide inclusions, and the oxide photomi-
14. Meth
Method
od C (Oxi
(Oxides Silicates)14
des and Silicates) crographs
crog raphs for all non-d
non-defor
eformabl
mablee oxide
oxide,, or hard
hard-type
-type,, incl
inclu-
u-
14.1   Introduction—This method requires a survey of a 160 sions. For example, a specimen may be classified 0-5 (oxide)
mm2 polished surface area of the specimen at 100 3. Each field and S-4 (silicate) to indicate that the longest non-deformable
on ththee sp
spec
ecim
imen
en sh
shal
alll be ex exam
amin
ined
ed fo
forr th
thee pr
pres
esen
ence
ce of  oxidee incl
oxid inclusio
usionn seen was comp comparabl
arablee to Oxide Photomicro-
Photomicro-
non-deformable alumina oxide and deformable silicate string- graph 5, and the longest deformable-oxide inclusion seen was
ers and rated by comparison to Plate II. The longest stringer of  comparable to Silicate Photomicrograph 4.
each
eac h inc
inclus
lusion
ion type (“O
(“O”” for alumina
alumina oxi
oxides
des and “S” for 14.3.2 Modifications, such as suffix suffix numerals,
numerals, may be used
silicates) shall be reported, per the designations of Plate II, for to indicate the number of long inclusions noted or the exact
every specimen examined. Note that sulfides are not rated by length of a particular inclusion when it is over the maximum
this method. length indicated by the photomicrographs.
14.2   Procedure:
14.2.1
14. 2.1 Thi
Thiss met
method
hod utiutiliz
lizes
es a rec
rectan
tangul
gular
ar ma
mask
sk tha
thatt wil
willl 15. Metho
Method
d D (Low Inclusion
Inclusion Content)
Content)
2 2
presen
pres entt a fiefield
ld ar
area
ea of 0.0.83
83 mm (0.00
(0.001289
1289 in. ) on the 15.1  Manual Introduction—This test method is intended for
specimen surface. The rectangular mask shall have sides equal
application to steel with low inclusion contents, as the severity
to 0.79 3 1.05 mm (0.03125 3 0.04125 in.) on the specimen
levels shall be reported in   1 ⁄ 2 increments. It requires a survey of 
surface (see Fig.
(see  Fig. 9).
9).
a 160 mm2 polished surface area of the specimen at 1003.
14.2.2 Eith
Either
er of two techn
techniques
iques may be used to mask off a
Every square 0.50 mm2 (0.000779 in.2) field on the polished
field of the required size. One method is to project the 100 3
- surface is examined for inclusion Types A, B, C, and D and
image from the microscope to a viewing screen equipped with
-
` 
  ,
  ,
` 
` 
` 
compar
com pared
ed wit
withh the squ
square
are fiel
fields
ds dep
depict
icted
ed on PlaPlate
te I-r
I-r.. The
a rectangular mask having sides 79.0   3  105.0 mm. Another
  ,
  ,

result of this   every field   comparison is recorded and tallied.

  ,
  ,
` 
` 

option is to have a reticle made for the microscope, which will

` 
` 
-

15.2   Manual Procedure:

` 
-
` 

superimpo
super imposese the requi
required
red rect
rectangul
angular
ar mask directly onto the
  ,
  ,
` 
  ,
  ,
` 
  , 15.2.1
15. 2.1 A field shall
shall be defined
defined as a squ square
are with
with 0.7
0.711 mm
field of view.
` 
  ,
  ,

(0.02791 in.) long sides. See  Fig. 5. 5.  This will result in a field
` 
-
-

14.2
14 .2.3
.3 To
To bebegi
gin,
n, ou
outl
tlin
inee th
thee re
requ
quir
ired
ed te
test
st ar
area
ea on th
thee
-

area of 0.50 mm2 on the specimen. Either of two techniques

specimen surface using either an indelible marker or a carbide-
may be employed to achieve the square field. One method is to
tipped scribe. Place the specimen on the microscope and start
project the 1003 microscope image onto a viewing screen that
the examination with a field in one of the corners of the marked
has a square mask (with 71.0 mm sides) drawn on it. Another
tes
testt are
area.
parallel a.toThe
the longer
longe r sid
rolling side
e of theCompare
direction. rectangular
rectangu lar mask
this shall
sha
field with ll the
be option is to have a reticle made for the microscope, which will
superimpose the required square mask directly onto the field of 
images on Plate II and record the number of the frame that
view (see Fig.
(see  Fig. 5).
5).
most resembles the oxide or silicate stringers, or both, present.
It is important to note that if an inclusion’s size falls between 15.2
15 .2.2
.2 To
To be begi
gin,
n, ou
outltlin
inee th
thee re
requ
quir
ired
ed tetest
st ar
area
ea on th thee
two of the numbered frames on Plate II, it shall be rounded specimen surface using either an indelible marker or a carbide-
down to the lower whole number. Also, stringered inclusions tipped scribe. Place the specimen on the microscope and start
shall
sha ll be cla
classi
ssified
fied as two disdisti
tinct
nct inc
inclus
lusion
ionss whe
when n the
they
y are thee ex
th exam
amin
inat
atio
ion
n wi
with
th a fie field
ld in ononee of ththee co
corn
rner
erss of th thee
separa
sep arated
ted by at lealeast
st 40 µm (0.(0.001
0016 6 in.
in.)) on the specime
specimen n specim
spe cimen.en. Com
Comparparee thi
thiss fiel
fieldd wit
withh the imimage
agess on Pla
Plate
te I-r
I-r..
surface or offset by more than 15 µm. Classify the inclusion content of the field based on the rules
14.2.4 Move the microscope
microscope stage to reveal an adjacent
adjacent field listed in Section 12 concerning Type, and thickness of inclu-
and repeat the comparison procedure with Plate II. Continue sions
sio ns pre
presen
sent.
t. Rec
Record
ord the Sev Severi
erity
ty Lev
Levelel Num
Numberber for eaceachh
thiss pro
thi proces
cesss unt
until
il the req
requir
uired
ed pol
polish
ished
ed sur
surfac
facee are
areaa of the inclusion type (A, B, C, and D) that most resembles the field
specimen
speci men has been scanned. A typical scan config configurat
uration
ion is under observation. Do this for both the Thin and Heavy series.
It is important to note that if a field of inclusions falls between
two severity levels it is rounded down to the nearest severity
13
A series of four photomicrographs of low carbon steel, previously printed as
level. Therefore, a field that contains fewer inclusions, or less
inclusion length, than severity level   1 ⁄ 2  is recorded as a 0.
partt of Pra
par Practi
ctice
ce E 45
45,, may be ob
obtai
tained
ned fro
from
m AST
ASTM M HeaHeadq
dquar
uarter
ters.
s. Ord
Order
er
ADJE004501. 15.2.3 Move the microscope
microscope stage to reveal an adjacent
adjacent field
14
This method is similar to SAE Recommended Practice J422. and repeat the comparison procedure with Plate I-r. The fields

Copyright ASTM International

15
Provided by IHS under license with ASTM
 No reproduction or networking permitted without license from I HS Not for Resale

1
E 45 – 05 e

shall be contiguous and only inclusions or portions of inclu- rest on

rest only
ly on th thee su
surf
rfac
acee of th
thee mo
moun
unti
ting
ng ma
mate
teri
rial
al.. If th
thee
sions that fall within the square mask shall be considered. It is specimen is unmounted but with a surface area substantially
not acceptable practice to move an inclusion into the square greater than the 160-mm2 area required for the measurement,
field simply to prevent its intersection with the sides of the the ring form can rest on the outer edges of the specimen for
mask. Continue this process until the required polished surface flattening and thus avoid contact with the measurement area.
area of the specimen
specimen has been rated rated.. A typi
typical
cal scan config
configura-
ura- Align
Ali gn the specimen
specimen on the stage so tha thatt the inclusio
inclusionsns are
tion is shown in Fig.
in  Fig. 6.6. aligned parallel to the  x -direction
-direction of the stage movement, that
15.2.4
15. 2.4 In concontra
trastst with MetMethod
hod A, thi thiss is an   every
every field  is, horizontal on the monitor screen. Alternatively, if program-
rating
rating met
method
hod.. The arb arbitr
itrary
ary fiel
fieldd bou
boundandarieriess cre
create
atedd by ming
mi ng is fac
facil
ilita
itated
ted,, ali
align
gn the incinclus
lusion
ionss par
parall
allel
el to the
stepwise movement through the sample should not be altered  y-dire
-directio
ction
n of the stage move
movement
ment,, that is, the longitudinal
longitudinal
or adjusted. Record the severity level shown on the side of  direction is vertical on the monitor screen.
Plate I-r selected for each inclusion type (A, B, C, or D) that 15.5.2 Check the microscope
microscope light source for corr
correct
ect align-
appears most like the field under observation for both the thin ment and adjust the illumination to the level required by the
and hea
heavyvy ser
series
ies.. Rep
Report
ort eaceach
h fiel
fieldd con
contaitainin
ningg inc
inclus
lusion
ionss television video camera.
equivalent to or greater than the 0.5 severity level. See Table See  Table 1 15.5.3 The incl
inclusio
usions
ns can be exami
examined
ned and discr
discrimin
iminated
ated
for values of Severity Level Numbers > 3.0. by type using magnifications other than 100 3  and field areas
15.3   Manual Expression of Results : other than 0.50 mm2 as long as the severity measurements are
15.3.1 The number of fields of each inclusion type type (A, B, C, based
base d upon the requ ired 0.50-mm2 fiel
required field
d are
area,
a, if the image
image
15
and D of Plate I-r) found for both the thin and heavy series analyzer is capable of such a procedure. If the system cannot
shall be recorded for each specimen in terms of the Severity work in this manner, that is, if the inclusions in each field must
Level Numbers 0.5 to 3.0. be dis
discri
crimin
minate
atedd by typ
type,
e, mea
measur
sured,
ed, and a sevseveri
erity
ty lev
level
el
15.3.2
15. 3.2 If any fiel field d or inc
inclus
lusion
ion is fou
found
nd thathatt exc
exceed
eedss the assigned on a field-by-field basis, then the magnification must
limits of severity level 3.0 (displayed on Plate I-r and listed in be chosen so that the field area is as close to 0.50 mm 2 as
Table 1), it shalshalll be reco
recorded
rded separately
separately.. Likew
Likewise ise oversize possible.
poss ible. A devi
deviation of less than 60.05 mm2 from the required
ation
2

inclusions with widths or diameters greater than the limiting 0.50-mm are areaa wil
willl not sig
signifi
nifican
cantly
tly imp
impair
air mea
measursureme
ementnt
values
val ues shown on Pla Plate
te I-r (and   Table
Table 2)   shall
shall be recorrecorded
ded results. The magnification chosen should produce a maximum
separately. calibration factor of 1.3 microns at 100 3.
15.5.4
15. 5.4 Sel
Select
ect the gragray-l
y-leve
evell thr
thresh
eshold
old set
settin
tings
gs to perpermit
mit
15.3.3 To averaverage
age the resuresults
lts of more than one speci specimen,
men,
independent detection of sulfides and oxides (see 10.2.2(see  10.2.2)).
the average of the number of fields found for each inclusion
15.5.5 When detec detecting
ting sulfides,
sulfides, a fals
falsee image (called
(called the
rating
rating numbe
numberr and type in the vari various
ous specimens
specimens exam examined
ined
halo effect) may be detected around the periphery of oxides in
within a lot may be calculated as illustrated in  Table 4. 4.
the same field. This problem can be corrected by the use of an
15.3.4 If desired,
desired, the predominant
predominant chemical
chemical type of incl inclu-
u-
auto-delineation feature or by application of appropriate algo-
sions may be determined (using, for example, energy disper-
rithms
rith ms to the binary image. Choice of the most satisfactor satisfactory y
sive x-ray spectroscopy on a scanning electron microscope).
approach depends upon the image analysis system used.
15.4   Image Analysis Introduct ion—The inclusions
Introduction inclusions on the 15.5.6
15. 5.6 Set the sta stage
ge con
contr
trols
ols to mov
movee the specime
specimen n in a
surface
surfa ce of a propeproperlyrly prepa
preparedred as-po
as-polish
lished
ed meta
metallogr
llographic
aphic square or rectangular pattern with contiguous field alignment
specimen are viewed with a high-quality, metallurgical micro- so that a minimum area of 160 mm2 is examined and evaluated.
scop
scope.e. Th
Thee brbrig
ight
ht-fi
-fiel
eld
d imimagagee is pi
pick
cked
ed up by a su suit
itab
able
le Other
Oth er mea
measur
sureme
ement nt are
areas
as ma
may y be use
usedd bas
based
ed on pro produc
ducer
er--
television camera and transferred to the image analyzer screen. purchaser agreements.     -
    -
    -
        `

Image analysis software is then used to evaluate the inclusion

  ,
  ,

15.5.7 Use a previ

previously
ously written
written computer
computer program to sepa sepa--         `
  ,
        `
  ,

content of the material. rate the inclusion images by type and thickness, then calculate
  ,
        `
  ,
  ,
        `
    -
        `

15.5   Image Analysis Procedure: severities based on length or number based on the rules listed
    -
        `
        `
        `
        `
  ,
  ,

15.5.1 Plac
Placee the specspecimen
imen on the microscope
microscope stage so that in Section   12
12..  The program should also store results, control
  ,
  ,
        `
        `
        `
  ,

the specimen surface is perpendicular to the optical axis. With

  ,

stage movements (if an automated stage is used), and generate

        `
    -
    -

an inverted-type microscope, simply place the specimen face- the test report.
down
do wn on th thee st
stag
agee plplat
atee anandd ho
hold
ld in pl plac
acee wi withth th
thee st
stag
agee 15.5.8 If the width of an A inclusion,
inclusion, or a B or C stringer,
stringer,
clamps. With an upright-type microscope, place the sample on varies and becomes less than 2 µm over part of its length,
a slide and level the surface using clay or plasticene and a detect as much of it as possible and calculate the severity based
hand-l
han d-leve
evelin
lingg pre
press.
ss. Cer
Certai
tainn upr
uprigh
ightt mic
microsroscope
copess can be on the detected length. For specimens from wrought products
equipped with an autoleveling stage for mounted specimens. If  with
wit h hig
high
h deg
degreereess of red
reduct
uction
ion,, whe
where
re the majmajori
ority
ty of the
the sample must be leveled using clay, the tissue paper placed inclus
inc lusion
ionss are < 2 µm thi thick,
ck, bas
based
ed on proproduc
ducer
er-pu
-purch
rchase
aserr
between the specimen surface and the leveling press ram may agreement, the minimum thickness of the thin series can be set
adhere to the surface and present artifacts for measurement. In
some cases, adherent
adherent tisstissue
ue can be blown off the spec specimen
imen
surface. An alternative procedure to avoid this problem is to 15
Forget,
For get, C., “Im
“Impro
proved
ved Meth
Method
od for E1E1122
122 Ima
Image
ge Ana
Analys
lysis
is Non
Nonmeta
metallic
llic
place an aluminum or stainless steel ring form, which has been
Ratings,”  MiCon 90: Advances in Video Technology for Microstructural
Inclusion Ratings,” MiCon
flattened slightly in a vise to an oval shape, between the sample Control,, ASTM STP 1094
Control 1094,, American Society for Testing
Testing and Materials, Philadelphia,
and the ram. If the specimen was mounted, the ring form will 1991, pp. 135–150.

Copyright ASTM International

16
Provided by IHS under license with ASTM
 No reproduction or networking permitted without license from I HS Not for Resale

1
E 45 – 05 e

at a lower value, such as 0.5 µm, or the lower limit can be 15.6.2 If desired, based on producer-purchaser
producer-purchaser agreements,
dropped. Detection of these thinner inclusions will require use modifications of the reported data may be made, for example,
of a higher magnification with a resultant field size less than reports for only certain inclusion types, thicknesses, or severity
0.50 mm2; hence, field data must be combined, as described in values
values.. Oth
Other
er mod
modific
ificati
ations
ons may inc includ
ludee onl
only
y wor
worst-
st-fiel
field
d
15.5.3,, to obtain valid ratings.
15.5.3 severi
sev erity
ty rat
rating
ingss or the num
number
ber of fiel
fields
ds at the wor
worst-
st-fiel
field
d
15.5.9 An array is esta
establis
blished
hed in the computer
computer memory to severity ratings.
tabulate the number of fields that were rated according to the 15.6.3 If desired by producer-purchas
producer-purchaser er agreement, an index
Thin and Heavy limits of the four inclusion types for eleven may be calculated to describe the inclusion content.
possible severities from 0 to 5 in half-level increments. After 15.6.4 To prod
produce
uce average results
results for more than one speci
speci--
each
ea ch fie
field
ld is ra
rate
ted
d anandd th
thee se
seve
veri
riti
ties
es ar
aree co
comp
mput
uted
ed,, th
thee men per lot, the average number of fields for each severity
appropriate array locations are incremented to store the results. rating, inclusion type, and thickness may be calculated (see
Table 4).
4).
15.5.10 The use of randomly
randomly selected, contiguously
contiguously aligned
15.6.5 Data for inclusions
inclusions or stri
stringers
ngers that are oversized
oversized in
fields may not produce true worst field (Method A) ratings.
either length or width, or both, should be reported separately.
Vali
alid
d wor
worst
st fiel
fieldd rat
rating
ingss req
requir
uiree adv
advanc
anced
ed im
image
age ana
analys
lysis
is
2
Reportt the inclusion
Repor inclusion type
type,, meas
measured
ured width, and lenglength
th (for
technology, for example, use of a 0.50-mm mask that can be
Types A, B, and C).
moved
mov ed any
anywhe
wherere wit
within
hin the 160 -mm2 te
160-mm test
st ar
area
ea us
usin
ingg an
15.6.6
15.6.6 Field
Fieldss wit
withh zer
zero
o sev
severi
erity
ty lev
levels
els may be fur furthe
therr
algorithm that controls the mask movement by maximizing the clas
classi
sifie
fied,
d, if de
desi
sire
red,
d, as ei eith
ther
er bl
blan
ank
k (n(no
o in
incl
clus
usio
ions
ns of a
severity values. particular type and width category are present) or non-ratable
15.5.11 For quantitative inclusion
inclusion descriptions, blank fields (inclusions are present but their length is below the 0.5 severity
(that is, those that contain no visible inclusions of a particular limit or their width is < 2 µm), or their diameter is < 3 µm.
type and width) may be differentiated from non-ratable fields 15.6.7 7 Info
Informat
rmation
ion pertaining
pertaining to the composition
composition of the
(that
(that is, fiel
fields
ds wit
withh inc
inclus
lusion
ionss   #   2 µm in wi
widt
dth,
h, or with
with inclusions (Types A to D) may be provided if desired. For rare
inclusion lengths or stringer lengths below the minimum limit earth- or calcium-treated steels, or other steels with nontradi-
for 0.5 severity). tional deoxidation approaches, the chemical composition of the
15.5.12 The program should incorporate
incorporate procedures to deal deal inclusions, in general terms, must be reported with each rating.
with
wit h fiel
fieldsds tha
thatt con
contai
tain
n art
artifa
ifacts
cts,, eit
either
her fro
fromm pol
polish
ishing
ing or Microana
Micr oanalytic
lytical
al tech
technique
niquess may be requi
required
red to obta
obtain
in such
cleaning, or from dust settling on the specimen, and so forth. information if the operator is not able to identify the inclusions
The ali
alignm
gnmentent of Type A, B, and C inc inclus
lusion
ionss in wro
wrough
ughtt by light optical examination.
specimens generally will not deviate by more than 620° from 15.6.8 Supplementary stereological
stereological data determined
determined during
the longitudinal direction. Depending on the system and the analysis may be included in the test report as desired. Stan-
natu
na ture
re of th thee ar
arti
tifa
fact
ct,, it ma
may y be po poss
ssib
ible
le to dedeve
velo
lop
p an dardization of such test data is not governed by this practice
algorithm that will recognize such artifacts and remove them (see Practice E
Practice  E 1245).
1245).
from
from th thee bi
bina
nary
ry imimag
age.
e. By re rest
stri
rict
ctin
ingg th
thee or
orie
ient
ntat
atio
ion
n of  16. Metho
Method
d E (SAM Rating)
detected
detec ted feat
features
ures within this limit, certa certain
in arti
artifact
factss (for ex-
16.1   Introduction—T
—Thihiss te
test
st me
meth
thod
od is usused
ed to rarate
te th
thee
ample, deep scratches not removed during polishing) can be
inclusion content of steels in a manner that reflects the severity
recognized and deleted from the binary image, if their orien-
and fre
freque
quency
ncy of occ
occurr
urrenc
encee of the larlarger
ger B- and D-T D-Type
ype
tation is greater than this limit. If this cannot be done, the field 2
inclusions. It will result in a survey of a 160 mm polished
should be rejectable, that is, no test results from the field should
surface of the specimen at 100 3.
be stored. In such a case, another field should be analyzed to 16.2   Procedure:
replace the rejected field, if this is possible. If a rejected field
16.2.1
16. 2.1 A field shall
shall be defined
defined as a squ square
are with
with 0.7
0.71
1 mm
cannot
can not be rep
replac
laced
ed in the sam
samee run
run,, it may be pospossib
sible
le to
(0.02791 in.) long sides. See  Fig. 5. 5.  This will result in a field
evaluate and rate the additional fields required in a subsequent
area of 0.50 mm2 on the specimen. Either of two techniques
run
run (d(do
o nonott ra
rate
te fie
field
ldss al
alre
read
adyy ra
rate
ted)
d).. Go
Good
od prprep
epar
arat
atio
ion
n
may be employed to achieve the square field. One method is to
practices will minimize the need to reject fields with artifacts.
project the 1003 microscope image onto a viewing screen that
In no case should the test results for a measurement area less
has a square mask (with 71.0 mm sides) drawn on it. Another
than 160 mm2 be math mathemat
ematical
ically
ly extra
extrapolat
polated
ed or conv
converted
erted
option is to have a reticle made for the microscope, which will
(for example, because of rejected fields) in an effort to produce
superimpose the required square mask directly onto the field of 
data for a 160-mm2 area.
view.
15.6   Image Anlysis Expression of Results : 16.2
16 .2.2
.2 To
To be
begi
gin,
n, ou
outl
tlin
inee th
thee re
requ
quir
ired
ed te
test
st ar
area
ea on th thee
15.6.1 The number of fields of each inclusion type type (A, B, C, specimen surface using either an indelible marker or a carbide-
and D) and thickness (thin and thick) are reported for each tipped scribe. Place the specimen on the microscope and start
severity from 0 to 5 in whole or half-severity level increments. the examination with a field in one of the corners of the marked
For steels with very low inclusion contents, severities may be test area. Compare this field with the images on Plate I-r. Rate
computed to one-
computed one-quart
quarter
er or one-t
one-tenth
enth severity level incr
incre-
e- only the B and D type inclusions using the following criteria.
ments. Note that for D-type inclusions, because one inclusion 16.2.3 A rating of B-type inclusions
inclusions is obtained by compar-
per field is a severity of 0.5, by definition, there can be no ing each field of the specimen with the fields in Plate I-r  (
 (Table
Table
D-severity levels between 0 and 0.5. 1   may als
also
o be use
used).
d). Rec
Record
ord all B-T
B-Thin
hin fields observed
observed at

--`,,```,,,,````-`-`,,`,,`,`,,`---

Copyright ASTM International

17
Provided by IHS under license with ASTM
 No reproduction or networking permitted without license from I HS Not for Resale

1
E 45 – 05 e

severity levels of 1.5 or higher and all B-Heavy fields observed 16.3.4 All oversized B- and D-Type inclusions
inclusions are reported
at each severity level of 1.0 or higher. See Table See  Table 2   for width along with their actual lengths or widths, or both.
and diameter parameters. Classify a field with size of inclu-
sions intermediate between configurations in Plate I-r or  Table 17. Test Report
1   as the low
lowerer inc
inclus
lusion
ion rat
rating
ing.. An incinclus
lusion
ion whowhosese wid
width
th 17.1 Pertinent information
information regarding the origin and identity
identity
varies from Thin to Heavy along its length shall be placed in of the test specimen should be reported along with the data
the category that best represents its whole. requirements covered in the “Expression of Results” section of 
16.2.4 Class
Classify
ify broken B-ty B-types
pes as two distinct inclusions
inclusions each test method.
when they are separated by at least 40 µm (0.0016 in.) or offset 17.2 Repor
Report,
t, also
also,, the following
following infor
informati
mation:
on:
by more than 15 µm on the specimen surface. If two or more 17.2.1 Date of test,test,
B-types appear in one microscope field, their summed length 17.2.2 Rater
Rater’s
’s name,
determines the inclusion rating number. 17.2.3 Plant location,
location,
16.2
16.2.5
.5 When
When an A- A-ty
type
pe susulfi
lfide
de hahass foform
rmed
ed a co compmple
lex
x 17.2.4 Heat number;
number; and
inclusion with either a B- or D-type oxide, the inclusion shall 17.2.5 Speci
Specimen
men identificati
identification
on code and any other uniq
unique
ue
be rated as a B- or D-type provided its oxide volume is the data (such as a lot number) that can provide traceability within
predominant (>50 % by area) chemical type. the seller’s organization.
16.2.6 A rating
rating of D-type inclusions
inclusions is obtai
obtained
ned by recor
record-
d-
ing all D-Heavy fields with a rating of 0.5 or higher. See Table See  Table 18. Prec
Precisio
ision
n and Bias
2  for width and diameter parameters. Fields of 0.5 severity are
18.1 Studi
Studies
es of JK ratiratings
ngs made by difdifferen
ferentt labo
laborator
ratories
ies
counted as one unit; fields of 1.0 severity as two units; fields of 
have sho
have shown
wn tha
thatt the
there
re is an inh
inhere
erent
nt pro
proble
blem
m in incinclus
lusion
ion
1.5 severity as three units; and so on. The minimum inclusion
identification, chiefly in discrimination between Type A (sul-
numbers for D-type are printed on Plate I-r and listed in Table in  Table
fides) and C (silicate) deformable oxide inclusions. Hence, the
1.
accura
accuracy
cy of JK ratrating
ingss can be sev
severe
erely
ly infl
influen
uenced
ced by suc
suchh
16.2.7 Move the microscope
microscope stage to reveal an adjacent adjacent field
and reprepeat
eat the com compar
pariso
ison n pro
proced
cedure
ure wit
withh Pla
Platete I-r
I-r.. Thi
Thiss pro
proble
blems
ms.. by
influenced Thetotal
accur
accuracy
acy of contents.
inclusion Method
Met hod A, C, and
As the D rat
rating
inclusion ingss is
content
method requires adjustment of the microscope stage in order to increases, the accuracy of such ratings decreases.
maximize
maxi mize the inclu
inclusion
sion Severity
Severity Level Number.Number. That is, the 18.2 For steels that are rated
rated to 0.5 Severity Level Numbers
Numbers
field of view is adjusted using the microscope stage controls
on Plate I-r, worst field ratings are generally accurate within
such that inclusions are moved inside the square mask in order
61 severity number and may be within   60.5 severity at low
to dedete
term
rmininee ththee ma
maxiximu
mum m seseve
veri
rity
ty of ra rate
teab
able
le B- an and
d
inclusion content. In general, the accuracies of rating of Type
D-Types. Continue this process, being careful not to rate any
B and D incl
inclusion
usionss are better than for Type
Type A and C inclusions.
inclusions.
inclusion more than once, until the required polished surface
Also, the accuracy of the thin series is generally better than for
area of the specimen
specimen has been rated rated.. A typi
typical
cal scan config
configura-
ura-
the Heavy series, regardless of the inclusion type.
tion is shown in Fig.
in  Fig. 6.
6.
18.3 For steels that must be rated to whole Severity Level Level
16.2.8 If any inclusions
inclusions are present that are longer longer than the Numbers using Plate I-r, the accuracies are generally poorer,
fields
fiel ds sho
shownwn in Pla Plate
te I-r
I-r,, the
their
ir len
length
gthss sha
shall
ll be recrecord
orded
ed approaching 62 at the highest severity levels. The same trends
separatel
separ atelyy. If thei
theirr width
widthss or diam
diameter
eterss are greater than the apply here regarding A and C versus B and D Types and Thin
limiting values shown in Plate I-r and  Table 2, 2,   they shall be versus Heavy. Greater inaccuracies will occur if inclusions are
recorded separately. Note that an oversize B or D inclusion still misidenti
misi dentified.
fied. The accur
accuracy
acy of inclu
inclusion
sion field count
countss usin
usingg
contributess to the determinati
contribute determinationon of a field’
field’ss Sever
Severity
ity Level Method D is not as good as for the worst field ratings. A good,     -

Number
Num ber.. The
Theref
refore
ore,, if a B inc
inclus
lusion
ion is ove
oversi
rsized
zed eit
either
her in
    -
    -

accurate Method D rating requires considerable effort.

        `
  ,
  ,
        `
  ,

leng
length
th or ththic
ickn
knes
ess,
s, th
that
at po
port
rtio
ionn th
that
at is wi
with
thin
in th
thee fie
field
ld 18.4
18.4 The acc
accura
uracy
cy of Met
Method
hod C ratrating
ingss is sig
signifi
nifican
cantly
tly
        `
  ,
  ,
        `
  ,
  ,

boundaries shall be included in the appropriate Thin or Heavy

        `
    -

influenced by misidentification of S Type (deformable oxide)

        `
    -
        `
        `
        `

severity level measurement. Likewise, if an oversize D inclu- inclusion

inclusions.
s. When such prob
problems
lems are not encou
encounter
ntered,
ed, stee
steels
ls
        `
  ,
  ,
  ,
  ,
        `

sion is encountered in a field, it also is included in the count with low inclusion contents will agree within   61 unit, while
        `
        `
  ,
  ,
        `
    -

that determines the D heavy rating.

    -

steels with high inclusion contents will agree within 62 units

16.3   Expression of Results: of severity. Method C, Plate II, is only used to rate oxides,
16.3.1 Results are expressed in terms
terms of two rating numbers never sulfides.
reflecting B-type and D-heavy type inclusion contents. 18.5 The precision
precision of ratings
ratings made by the use of Plate I-r
16.3.2 The number of B-type fields fields recorded
recorded at each sever-
sever- generally agrees with the chart severity increments used but
ity level times the severity level is summed (see  Table 5) 5) and may in certain cases be slightly higher. For very low inclusion
normalized by dividing by the total rated area, in square inches, content steels, automatic image analysis methods (as covered
of all samples. The nearest whole number is recorded as the by Practices E 1122 and  E 1245)
1245) are preferable where ratings
rating. below the minimum rating (1 ⁄ 2) are possible. Note that micro-
16.3.3 The number
number of D unit
unitss is summed (see Table
(see Table 5)
5) and scopic Methods A and D stipulate minimum sizes for  ratable
normalized by dividing by the total rated area, in square inches, inclusions; thus a field or a specimen may contain inclusions
of all samples. The nearest whole number is recorded as the that are identifiable but not ratable because they are below the
rating. minimum size for a non-zero rating.

Copyright ASTM International

18
Provided by IHS under license with ASTM
 No reproduction or networking permitted without license from I HS Not for Resale

1
E 45 – 05 e

18.6
18. 6 For the ima
image
ge ana
analys
lysis
is pro
proced
cedure
ures,
s, whe
whenn the sam
samee 18.6.2 Interlaborator
Interlaboratory
y test variability has
has not been evaluated
specimen is reanalyzed immediately, starting over at the same but may be expected to be greater. This variability will be at a
location and re-measuring the same fields, reproducibility is minimum
minimum if each labor
laboratory
atory cont
controls
rols speci
specimen
men prepa
preparati
ration
on
extremely good. Worst-field ratings are usually identical, but according to the guidelines in Practice E
Practice  E 768.
768.
may occasionally show a variation of one-half severity limit for 18.6.3 Use of a man
18.6.3 manual
ually
ly ope
operat
rated
ed sta
stage,
ge, rat
rather
her than an
one of the eight possible ratings (A to D, thin and thick). The
-
-
` 
  ,
  ,
` 
` 
automated stage, may introduce bias into the field selection.
number of fields at each severity level for each inclusion type
` 
  ,
  ,
  ,
  ,
` 

and thickness generally varies by less than 5 %.

` 
` 
` 
-
` 
-
19. Keyw
Keywords
ords
18.6.1 If a rate
rated
d specimen
specimen is re-p
re-polish
olished
ed and rated again on
` 
  ,
  ,
` 
  ,
  ,
` 
  ,

a par
parall
allel
el pla
plane
ne by the sam
samee lab
labora
orator
tory
y, the res
result
ultss wil
willl be 19.1 alumina;
alumina; autom
automatic
atic imag
imagee analy
analysis;
sis; compl
complex
ex inclu
inclu--
` 
  ,
  ,
` 
-

reasonably reproducible. Worst-field ratings will usually vary sions; fracture test; globular inclusions; inclusions; inclusion
-
-

by no mor
moree tha
than
n one
one-ha
-half
lf sev
severi
erity
ty lev
level
el for several
several of the rating; inclusion stringers; JK inclusion rating; light micros-
inclusion types and thickness categories but larger variations copy; macroetch test; magnetic-particle method; oxide; SAM
are occasionally encountered due to the inherent variability of  rating; silicate; steel; step-down method; stringer; sulfide
the inclusion content.

ASTM International takes no position respecting the validity of any patent rights asserted in connection with any item mentioned 
in this standard. Users of this standard are expressly advised that determination of the validity of any such patent rights, and the risk 
of infringement of such rights, are entirely their own responsibility.

This standard is subject to revision at any time by the responsible technical committee and must be reviewed every five years and 
if not revised, either reapproved or withdrawn. Your comments
comments are invited either for revision of this standard or for additional standards 
and should be addressed to ASTM International Headquarters. Your comments will receive careful consideration at a meeting of the 
responsible technical committee, which you may attend. If you feel that your comments have not received a fair hearing you should 
make your views known to the ASTM Committee on Standards, at the address shown below.

This standard is copyrighted by ASTM International, 100 Barr Harbor Drive, PO Box C700, West Conshohocken, PA 19428-2959,
United States. Individual reprints (single or multiple copies) of this standard may be obtained by contacting ASTM at the above 
addres
address
s or at 610
610-832
-832-95
-9585
85 (pho
(phone)
ne),, 610
610-832
-832-95
-9555
55 (fa
(fax),
x), or serv
service
ice@as
@astm.
tm.org
org (e-m
(e-mail
ail);
); or thro
through
ugh the AST
ASTM
M web
websit
site 
e 
(www.astm.org).

Copyright ASTM International

19
Provided by IHS under license with ASTM