30 2 2014 (66) METALURGIA

THE APPLICATION OF ANALYSIS OF VARIANCE (ANOVA) TO DIFFERENT

EXPERIMENTAL RESULTS OF C45 MEDIUM - CARBON STEEL

Ioan MILOSAN

Transilvania University of Brasov, Faculty of Materials Science and Engineering, Brasov, Romania

Received on 10/01/2014, accepted on 17/02/2014

==============================================================================================
Key words: analysis of variance (ANOVA), tensile strength, properties C45 medium - carbon steel.

Abstract: Analysis of variance (ANOVA) is the most efficient method available for the analysis of
experimental data. Analysis of variance is a method of considerable
complexity and subtlety, with many different variations, each of which applies
in a particular experimental context. The paper presents an exemple of
calculig for the one-way repeted measures ANOVA applied for the results of
the tensile strength properties in the case of C45 medium - carbon steel. It is
shown that by applying a subzero treatment, it has a great influence on the
tensile strength properties in the case of of C45 medium - carbon steel. A one
way repeated measures ANOVA is used for calculation in Excel.
Prof. Ph. D. Eng
Ioan MILOSAN
variations, each of which applies in a
1. INTRODUCTION particular experimental context.

Analysis of variance (ANOVA) is the 2. THE PURPOSE OF ANALYSIS OF

method used to compare continuous
measurements to determine if the
measurements are sampled from the same or
different distributions. It is an analytical tool
used to determine the significance of factors
on measurements by looking at the
relationship between a quantitative "response
variable" and a proposed explanatory "factor."
This method is similar to the process of
comparing the statistical difference between
two samples, in that it invokes the concept of
hypothesis testing. Instead of comparing two
samples, however, a variable is correlated
with one or more explanatory factors,
typically using the F-statistic. From this F-
statistic, the P-value can be calculated to see
if the difference is significant [1].
Analysis of variance (ANOVA) is the
most efficient method available for the
analysis of experimental data. Analysis of
variance is a method of considerable
complexity and subtlety, with many different
VARIANCE
A One-Way Analysis of Variance is a
way to test the equality of three or more
means at one time by using variances [2].
Assumptions
- The populations from which the samples
were obtained must be normally or
approximately normally distributed [3].
- The samples must be independent.
- The variances of the populations must be
equal.
Hypotheses
The null hypothesis will be that all
population means are equal (H0: μ1 = μ2 = μ3
…= μn) and the alternative hypothesis is that at
least one mean is different (H1: at least one of
the means is different).
In the following, lower case letters apply
to the individual samples and capital letters

Correspondence to: Ioan MILOSAN , milosan@unitbv.ro, Transilvania University of Brasov, Faculty of Materials Science and Engineering,
Brasov, Romania
(66) METALURGIA
apply to the entire set collectively. That is,
”n” is one of many sample sizes, but ”N” is
the total sample size.
Grand Mean ( ) so there are k-1 degrees of freedom.
The grand mean of a set of samples is the
total of all the data values divided by the total
sample size.
This requires that you have all of the
sample data available to you, which is usually
the case, but not always. It turns out that all
that is necessary to find perform a one-way
analysis of variance are the number of
samples, the sample means, the sample
variances, and the sample sizes.
Another way to find the grand mean is to
find the weighted average of the sample
means. The weight applied is the sample size.
Total Variation
The total variation (not variance) is
comprised the sum of the squares of the
differences of each mean with the grand
mean.
SS(T) = )2
There is the between group variation and
the within group variation. The whole idea
behind the analysis of variance is to compare
the ratio of between group variance to within
group variance. If the variance caused by the
interaction between the samples is much
larger when compared to the variance that
appears within each group, then it is because
the means aren't the same.
Between Group Variation
The variation due to the interaction
between the samples is denoted SS(B) for
Sum of Squares Between groups. If the
sample means are close to each other (and
therefore the Grand Mean) this will be small.
Correspondence to: Ioan MILOSAN , milosan@unitbv.ro, Transilvania University of Brasov, Faculty of Materials Science and Engineering,
Brasov, Romania
2 2014 31
SS(B) = )2 (4)
There are ”k” samples involved with one
data value for each sample (the sample mean),
The variance due to the interaction
between the samples is denoted MS(B) for
Mean Square Between groups. This is the
between group variation divided by its
(1)
degrees of freedom.
It is also denoted by
Within Group Variation
The variation due to differences within
individual samples, denoted SS(W) for Sum
of Squares Within groups. Each sample is
considered independently, no interaction
between samples is involved.
2
SS(W) = (5)
The degrees of freedom is equal to the
(2) sum of the individual degrees of freedom for
each sample. Since each sample has degrees
of freedom equal to one less than their sample
sizes, and there are k samples, the total
degrees of freedom is k less than the total
sample size: df = N - k.
The variance due to the differences
within individual samples is denoted MS(W)
for Mean Square Within groups. This is the
(3) within group variation divided by its degrees
of freedom. It is also denoted by . It is the
weighted average of the variances (weighted
with the degrees of freedom).
F test statistic
Recall that a F variable is the ratio of two
independent chi-square variables divided by
their respective degrees of freedom. Also
recall that the F test statistic is the ratio of two
sample variances, well, it turns out that's
exactly what we have here. The F test statistic
is found by dividing the between group
variance by the within group variance.
F= -- (6)
 32
The degrees of freedom for the numerator
are the degrees of freedom for the between
group (k-1) and the degrees of freedom for the
denominator are the degrees of freedom for
the within group (N-k).
2. EXPERIMENTAL PROCEDURE AND
RESULTS
The studied C45 medium - carbon steel,
has the following chemical composition (% in
weight): 0.47% C; 0.69 % Mn; 0.2% Si;
0.021%S; 0.021%P.
Heat treatment to improve is composed of
a group of operations as follows: quenching
followed by high tempering.
All specimens for testing the resilience
have undergone tempering operation. These
were heated in an electric baking temperature
of 8500C, while keeping time to equalize the
temperature in samples being 20 minutes.
Cooling energy has been made in water;
the water temperature is at room temperature,
i.e. 200C. After quenching operation, structure
which is obtained is martensite hardening, a
non equilibrium structure. The quenched
samples were performed such tests Rockwell
hardness, resulting in values: 55 ÷ 58 HRC.
After quenching, samples were subjected
to high-tempering operation.
Tempering operation, unlike the
annealing and quenching operations, is not a
unique stand-alone operation; it applies only
after quenching operations and aims to
facilitate specific changes only in hardened
areas.
The changes are intended to bring the
equilibrium structure obtained by quenching
(quenching martensite), close to equilibrium
structures with distinct properties.
The samples for testing the resilience of
the high-tempering operation of 6000C, took
15 minutes during the maintenance to
equalize temperature. Following high-
tempering operation, a tempering soorbite
structure was obtained. The high-tempering
samples were performed such tests Rockwell
hardness, resulting in values: 36 HRC for all
specimens.
Correspondence to: Ioan MILOSAN , milosan@unitbv.ro, Transilvania University of Brasov, Faculty of Materials Science and Engineering,
Brasov, Romania
2 2014 (66) METALURGIA
After heat treatment for improvement, has
passed the test at low temperature impact
bend. This test was performed inside the
ROMAN S.A. Company.
Before testing, all samples were measured
to see if they fall within prescribed limits
contained in standards (SR EN 10045-1:1993)
and then he passed the test at low temperature
impact bend.
Impact bend test at low temperatures was
conducted according to standard SR EN
10045-1:1993 [4].
The test was performed according to
standard SR EN 1400-1998, with the
following specifications [4]:
1. The samples are cooled in cooling
facilities consist of a thermally insulated
container, where specimens are placed on
media such as to ensure their contact on all
sides with the cooling medium are immersed.
2. As a refrigerant (until -750 C) was used
ethanol and carbon dioxide snow (CO 2
solid), which has a temperature of
vaporization at 194.6 K, a critical temperature
to 1.6 10 kg/m 3 and the latent heat of 13700
J/kg.
3. Duration of maintaining the
temperature of the samples specific cooling
facility must be at least 5 minutes to cool in
refrigerant. The samples were kept in the
container for 10 minutes.
4. To measure temperature using a
thermometer must fall at least 0.5 accuracy
classes and have a maximum value of a
division 10C.
5. For handling and removal of samples
from the cooling plant, has been used tongs
cooled to the temperature of samples.
6. Breaking the samples was performed
within 5 seconds of their removal from the
installation of cooling.
7. The samples were focused on to a
pendulum hammer-centering device, for the
hammer-notch pendulum strikes the opposite
side.
8. The test was performed for six different
temperature levels, for each temperature level
being tested a total of three samples.
(66) METALURGIA 2 2014 33

After the bending test in cold shock We - the potential energy consumed to
standard (SR EN 10045-1:1993), were break the specimens, [J];
obtained different values of potential energy S0 – specimen section [cm2],
consumed “We “.
After determining the amount of potential SO = ac b (8)
energy consumed to break samples, the
impact strength values were determined using where:
the formula [5, 6, 7]: ac - is located at the notch surface to the
opposite side;
We b is the specimen side.
KCU 2  (7) The impact strength values are presented
S0
in table 1.
where:

Table 1. The impact strength values obtained from bending test at low temperature shock
Nr. KCU2 [J/cm2]
crt. 0 0C – 15 0C – 30 0C – 45 0C – 60 0C – 75 0C
1 92.5 88.5 85.5 57.5 36 8.5
2 94 93.5 82 68.5 30.5 9
3 92.5 91.5 89 60 34 9

3. CALCULATION OF ONE-WAY you have measured something a few times [8,

ANOVA IN EXCEL
A one way repeated measures ANOVA is
used when you have a single group on which
9, 10].
The impact strength values obtained from
bending test at low temperature shock, ”Data
Analysis” for calculation in Excel are
presented in table 2.

Table 2. Data Analysis

Parallel Observation Samples, k = 6
n=3 0° C -15° C -30° C -45° C -60° C -75° C
Obs 1 92,5 88,5 85,5 57,5 36 8,5
Obs 2 94 93 82 60,5 33,5 9
Obs 3 92,5 91,5 89 60 34 9,5

Steps in Excel: (4) The ANOVA output will appear on a

(1) From Data Analysis it was selected new worksheet.
ANOVA;
(2) It was selected all the results; The Summary Table of the ANOVA-Single
(3) It was selected the level of significance Factor is presented in table 3.
(The default is 5% or 0.05.);

Correspondence to: Ioan MILOSAN , milosan@unitbv.ro, Transilvania University of Brasov, Faculty of Materials Science and Engineering,
Brasov, Romania
 34 2 2014 (66) METALURGIA

Table 3. Summary Table of the ANOVA-Single Factor

Anova: Single Factor
SUMMARY
Groups Count Sum Average Variance
Column 1 3 279 93 0,75
Column 2 3 273 91 5,25
Column 3 3 256,5 85,5 12,25
Column 4 3 178 59,33333 2,583333
Column 5 3 103,5 34,5 1,75
Column 6 3 27 9 0,25

The Source of Variation of the ANOVA-Single Factor is presented in table 4.

Table 4. Source of Variation of the ANOVA-Single Factor

ANOVA
Source of P-
Variation SS df MS F value F crit
Between
Groups 17779,78 5 3555,956 934,4117 4,13E-15 3,105875
Within Groups 45,66667 12 3,805556

Total 17825,44 17

4. CONCLUSION the critical value (F crit), i.e. 934,4117>

(a) After cooling, maintaining samples at low
temperatures and shock bending test at these
temperatures, it has been a decrease in
toughness with decreasing values of
temperature observation from studying the
table shows the recorded data were obtained
from bending.
(b) It is assumed that this decrease in
mechanical properties, default values of
impact strength, with temperature is due to
residual austenite Arez transformation into
martensite, which has a strong influence on
the mechanical properties of alloys.
Martensite properties are dependent on the
percentage carbon, thus high carbon content,
reaching hardness of 66 HRC.
(c) From table 4 it was compares the two
values of the F test statistic: calculate value
(F) and critical value ( F crit). Note that in this
case the calculated value (F) is greater than
3,105875
(d) In short, an ANOVA test will test us that
at least one mean is different, but an
additional test must be conducted to
determine which mean(s) is/are different.
REFERENCES
[1] R. A. Armstrong, F. Eperjesi, B.
Gilmartin, The application of analysis of
variance (ANOVA) to different experimental
designs in optometry, In: Ophthalmic and
Physiological Optics, 22 (2002), 248–256.
[2] S. M . Măruşteri, Noţiuni fundamentale
de biostatistică, Ed. University Press, ISBN
973-7665-11-2, Târgu-Mureş, 2006.
[3] D. Taloi, Optimizarea proceselor
metalurgice- Aplicaţii în metalurgie, E.D.P.
Bucureşti, 1997.
[4] 4.*** Lista Standarde SR En -Lista
standardelor europene adoptate ca standarde
române, Available

Correspondence to: Ioan MILOSAN , milosan@unitbv.ro, Transilvania University of Brasov, Faculty of Materials Science and Engineering,
Brasov, Romania
(66) METALURGIA 2 2014 35

[5] from: www.scribd.com/doc/Lista- [10] J. Budde, Variance analysis and linear

Standarde-SR-En contracts in agencies with distorted
[6] I. Tapalagă, Industrial cryogenics performance measures, In: Management
Criogenia industrială, E.T. Bucharest, Accounting Research (September, 2009),
Romania, 1986. 166-176.
[7] A. Bensely, A. Prabhakaran, D. Mohan [11] K. T. Wing, Using enhanced cost models
Lal, G. Nagarajan, Enhancing the wear in variance analysis for better control and
resistance of decision making, In: Management Accounting
[8] case carburized steel by cryogenic Quarterly (Winter, 2005), 27-35.
treatment, Cryogenics, 45, (2005), 747-754. [12] M. Yahya-Zadeh, A new framework for
[9] D. Catana, Study regarding cryogenic capacity costing and inventory variance
temperatures resilience of the bearings steels, analysis, In: Journal of Applied Management
Metalurgia, vol. XVIII, 4, (2013), 80-82. Accounting Research (Summer, 2011), 61-81.

DATA BASES
 According to paragraph 2.6 of license agreement between EBSCO Publishing Inc. (U.S.A.) and FUNDAŢIA METALURGIA ROMÂNĂ
(ROMÂNIA) the journals METALURGIA and METALURGIA INTERNATIONAL are included in EP Products, starting from July 1st,
2006.
 These journals are included on EBSCO’s site: www.ebscohost.com , chapter „Computers and Applied Sciences Complete”, pozitions 1026
and 1027.
 Starting from January 1st, 2007 the journal METALURGIA INTERNATIONAL is also in the SCOPUS database, belonging to
ELSEVIER BIBLIOGRAPHIC DATABASES – Amsterdam (Netherlands).
 Starting from january 1-ST ,2010,the journal METALURGIA 0461-9579 and METALURGIA INTERNATIONAL 1582-2214 are included
in PRO-QUEST ABSTRACTS IN TECHNOLOGY AND ENGINEERING, CSA TECHNOLOGY RESEARCH DATABASE AND CSA /
ASCE CIVIL ENGINEERING ABSTRACTS. For further details,please access: http://www.csa.com/ids70/serials_source_list.php?db=civil-set-c
Starting from january 1-ST ,2011, the journal METALURGIA INTERNATIONAL 1582-2214 is included in American Chemical Society.
For further details,please access: http://www.cas.org/

 The “METALURGIA, ISSN 0461-9579 &
METALURGIA INTERNATIONAL, ISSN: 1582-
2214” magazines receives manuscripts of papers including
basic scientific research and industrial research in the
following fields: metallurgy, materials science and
engineering and different relating processes.
 Original papers not previously publishing in any other
journal, or not sent for publishing before, are accepted.
 After publication, the copyright is transferred to the
publishing house.
 Every manuscript will be referred, their reports form
the basis of the Editor’s decision.
 Any paper from this journal can’t be reproduced,
diffused or used in any other way without the written
agreement of the editor .
 Manuscripts and unpublished documents (photos, flyers, cards) submitted to
redaction, on the initiative of the authors, according to law will be not returned.
 In accordance with PENAL CODE the legal responsibility for the content of the
article belongs to author.
 Also, in case of press agents and cited person, the legal responsibility belongs to
them.
 The copyright of parts from this publication is allowed with the written agreement
of the editor.
Copyright © 2014
METALURGIA, ISSN 0461-9579 & METALURGIA INTERNATIONAL,
ISSN: 1582-2214
METALURGIA INTERNATIONAL
83, Calea Griviţei, sector 1, Postal code 010705, postal office 12, Bucharest,
Romania
Tel.: +(40)-0372926401; +(40)-0724537051
+(40)-0722696187; +(40)-0724296800
+(40)-0722311272; +(40)-0735547316

E-mail: redactia@metalurgia.ro; See also web: www.metalurgia.ro

REDACTION

Correspondence to: Ioan MILOSAN , milosan@unitbv.ro, Transilvania University of Brasov, Faculty of Materials Science and Engineering,
Brasov, Romania
Copyright of Metalurgia is the property of Fundatia Metalurgia Romana and its content may
not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's
express written permission. However, users may print, download, or email articles for
individual use.