Research on dispersion in indenter and reference block for rockwell hardness test


ACTA IMEKO | www.imeko.org December 2020 | Volume 9 | Number 5 | 221 

ACTA IMEKO 
ISSN: 2221-870X 
December 2020, Volume 9, Number 5, 221 - 225 

 
 

RESEARCH ON DISPERSION IN INDENTER AND REFERENCE BLOCK 

FOR ROCKWELL HARDNESS TEST 
 

Takeshi Sawa1 

 
1 Mitutoyo Corporation, 1-20-1 Sakado, Takatsu-ku, Kawasaki-shi, Kanagawa-ken, 213-8533, Japan 

takeshi_sawa@mitutoyo.co.jp  

 

Abstract: 

In this report, the Rockwell indenters and 

hardness reference blocks obtained from different 

three manufacturers were evaluated. And these 

results with the tolerances for the indirect 

verification prescribed in ISO 6508-2 and ASTM 

E18 were compared, it has been suggested that the 

tolerances prescribed in ASTM E18 might be too 

strict in view of actual circumstances. 

Keywords: Rockwell indenter, hardness 

reference block, ISO/IEC 17025, ISO standards, 

ASTM standards 

1. INTRODUCTION 

The importance of ensuring the traceability in 

measurement has been increased in these years, and 

accordingly, the spread of systems certifying tests 

and calibrations in conformity with ISO/IEC 

17025[1] that can publicly give assurance has been 

promoted in a worldwide manner. In addition, the 

Mutual Recognition Agreement (MRA) that 

mutually certifies the conformity assessment results 

of own country as equivalent to another country is 

implemented among many accredited systems, and 

the accredited certificates are valid in any countries. 

It can be said that this system is extremely beneficial 

for globalization of manufacturing and industry. 

The hardness test is no exception, and the number 

of cases where the calibration of hardness reference 

blocks, indenters and hardness testing machines is 

requested to conform to ISO/IEC 17025 is 

increasing. It can be said that the supply source of 

indenters and hardness reference blocks that are 

used for calibration of hardness testing machines 

does not matter if the testing and calibration are 

performed in conformity with accredited systems 

system. However, a hardness value is industrial 

quantity and has no physical reference, therefore, 

the traceability system and uncertainty evaluation 

method for measurement values and calibration 

values are not completely standardized in the global 

aspect under present circumstances. Such present 

circumstances may become a problem for 

calibration of hardness testing machines. We know 

that even differences in hardness values among 

national metrology institute level are a problem for 

calibration of actual hardness testing machines [2]. 

In this report, evaluating the dispersion of Rockwell 

indenters and hardness reference blocks obtained 

from different three manufacturers have been 

conducted for the purpose of investigating the 

realistic circumstances. Then, based on these results, 

the uncertainty of calibration in indirect verification 

of the hardness testing machine was estimated, and 

the consistency with the tolerance value prescribed 

in the standard was investigated. 

2. EXPERIMENT 

Three pieces of Rockwell diamond conical 

indenters, tungsten carbide ball indenters with 

1.5875 mm in diameter and six types of hardness 

reference blocks from three manufacturers are 

obtained. The scales and hardness levels of the six 

hardness reference blocks were 30 HRC, 60 HRC, 

90 HR15N, 30 HRBW, 90 HRBW and 35 HR 

15TW. Three manufacturers are hereinafter referred 

to as Manufacturer A, Manufacturer B, and 

Manufacturer C. Manufacturer A is a famous 

indenter and reference block manufacturer in 

Europe. Manufacturer B and Manufacturer C are US 

manufacturers and a lot of their products are 

widespread in the United States. An accredited 

certificate is issued for all of these indenters and 

reference blocks to prove that calibration of them 

conforming to ISO/IEC 17025 has been conducted. 

SHT-31, a Rockwell hardness standard testing 

machine manufactured by Mitutoyo Corporation, 

was used for evaluation. The specifications and 

accuracy of each part of the testing machine is 

conformed to ISO 6508-3: calibration of the 

hardness reference block [3]. Moreover, the testing 

conditions such as time of loading the testing force 

and duration time were set according to this 

standard.  

 
 

http://www.imeko.org/
mailto:takeshi_sawa@mitutoyo.co.jp


ACTA IMEKO | www.imeko.org December 2020 | Volume 9 | Number 5 | 222 

2.1. Results of evaluating indenters of each 
manufacturer 

Testing on the same reference block with the use 

of obtained indenters and our indenter were 

conducted. Our indenter, i.e. Mitutoyo indenter is 

conformable to ISO 6508-3, and is equivalent to an 

indenter classified in Class A prescribed in ASTM 

E18[4]. The differences between the hardness 

values of the targeted indenters to be evaluated and 

the hardness value of our indenter were evaluated. 

Figure 1 shows the differences of hardness values in 

each scale. The medians were regarded as the 

average of the differences of hardness values in each 

scale. The value PI shows a 95% confidence interval 

of dispersion calculated by multiplying the 

experimental standard deviation of the differences 

of hardness values in each scale by a coverage factor. 

Here, the coverage factor was determined to be 2.31 

on the basis of the Student's t-distribution in 

consideration that the degree of freedom of data in 

each scale is 8. 

 

 
Figure 1: Difference from hardness value with Mitutoyo 

indenter  

 

2.2. Results of evaluating obtained reference 
blocks of each manufacturer 

Testing on the obtained reference blocks with the 

use Mitutoyo indenter were conducted, and the 

measured hardness values were compared with the 

calibration values of the targeted hardness reference 

blocks to be evaluated. Figure 2 shows the 

differences from the calibration values of the 

reference blocks in each scale. The medians were 

regarded as the average of the differences of 

hardness values in each scale. The value PB shows a 

95% confidence interval of dispersion calculated by 

multiplying the experimental standard deviation of 

the differences of hardness values in each scale by a 

coverage factor. Here, the coverage factor was 

determined to be 2.31 on the basis of the Student's 

t-distribution in consideration that the degree of 

freedom of data in each scale is 8. 

 

 
Figure 2: Differences from the calibration values of the 

reference blocks 

3. DISCUSSION 

3.1. Regarding results of evaluating indenters 
Table 1 shows the PI value in Fig. 1 and the 

tolerance of indenter prescribed in ISO 6508-2[5] 

and ASTM E18. For diamond indenter, the 

difference from the reference indenter is prescribed 

in standard as tolerance. There are some PI values 

that are larger than the tolerance. This fact suggests 

that indenters whose tolerance exceeds the 

standards may be on the market.  

-2.0

-1.0

0.0

1.0

2.0

A B C A B C A B C

D
if

fe
r
e
n

c
e
 f

r
o

m
 h

a
r
d

n
e
ss

 v
a

lu
e

w
it

h
 M

it
u

to
y

o
 i

n
d

e
n

te
r
 [

H
R

]

30 HRC 60 HRC 90 HR15N

Indenter manufacturer

PI = 0.6 HRC PI = 0.7 HRC
PI = 1.3 HR15N

Diamond indenter scales

-2.0

-1.0

0.0

1.0

2.0

A B C A B C A B C

D
if

fe
r
e
n

c
e
 f

r
o

m
 h

a
r
d

n
e
ss

 v
a

lu
e

w
it

h
 M

it
u

to
y

o
 i

n
d

e
n

te
r
 [

H
R

] 

30 HRBW 90 HRBW 35 HR30TW

Indenter manufacturer

PI = 1.1 HRBW
PI = 0.4 HRBW PI = 0.4 HR30TW

WC ball indenter scales

-3.0

-2.0

-1.0

0.0

1.0

2.0

3.0

A B C A B C A B CD
if

fe
r
e
n

c
e
 f

r
o

m
 c

a
li

b
r
a

ti
o
n

 v
a

lu
e
 

o
f 

r
e
fe

r
e
n

c
e
 b

lo
c
k

 [
H

R
]

30 HRC 60 HRC 90 HR15N

Reference block manufacturer

PB = 1.0 HRC PB = 0.8 HRC PB = 0.7 HR15N

Diamond indenter scales

-3.0

-2.0

-1.0

0.0

1.0

2.0

3.0

A B C A B C A B CD
if

fe
r
e
n

c
e
 f

r
o

m
 c

a
li

b
r
a

ti
o
n

 v
a

lu
e
 

o
f 

r
e
fe

r
e
n

c
e
 b

lo
c
k

 [
H

R
]

30 HRBW 90 HRBW 35 HR30TW

Reference block manufacturer

PB = 1.9 HRBW PB = 0.7 HRBW PB = 1.0 HR30TW

WC ball indenter scales

http://www.imeko.org/


ACTA IMEKO | www.imeko.org December 2020 | Volume 9 | Number 5 | 223 

For tungsten carbide ball indenter, the difference 

from the calibration value of reference block is in 

standard as tolerance. Except for 30 HRBW, the PI 

value is sufficiently smaller than the tolerance of 

standard. However, this result is a difference from 

the value measured with the same testing machine 

and test conditions, and the uncertainty of the 

calibration value of reference block is considered to 

be larger than measured value. That is, the fact that 

the PI value is smaller is expected result. 

Nevertheless, for 30 HRBW, the PI value exceeds 

the tolerance of ASTM standard. In addition, there 

is a bias of indenters among manufacturers. That is, 

for 30 HRBW, it is suggested that the use of 

indenter from another manufacturer may exceed the 

tolerance of ASTM standard. 

 

Table 1: Evaluation results of indenter and tolerance of 

standard 

 
 

3.2. Estimation of uncertainty for indirect 
verification using results of evaluating 

reference block. 

The disperse i.e. uncertainty of hardness values 

generated in the indirect verification of the testing 

machine was estimated. As factors for the 

uncertainty, we hypothesized the influences of the 

depth measuring device, difference of the indenter 

with respect to the reference indenter, the 

calibration value and uniformity of the hardness 

reference block, and machine hysteresis. It thought 

that the testing force, indentation time, and duration 

time of testing force can be ignored because these 

factors have little influence. The standard 

uncertainty for each factor was estimated based on 

the description in Table 2. If a tolerance is different 

between the ISO standards and ASTM standards, 

stricter tolerance was used for estimation. Table 3 

shows the uncertainty budget for each scale. 

Table 2: Hypothesized uncertainty factors 

 

3.3. Regarding consistency with the tolerance of 
the standard 

Table 4 shows the tolerances in indirect 

verification prescribed in ISO 6508-2, ASTM E18 

and estimated expanded uncertainty. When the 

expanded uncertainty of the hardness value in 

indirect verification is compared with the tolerance 

of the standard, it is considered that the uncertainty 

is within the tolerance range of the ISO standards in 

all the evaluated scales. With regard to the ASTM 

standards, on the other hand, the uncertainty may 

not be within the tolerance range except for 30 

HRBW and 35 HR30TW through the indirect 

verification even with the use of indenters and 

reference blocks that are conformable with the 

standards. The tolerance of 30 HRBW in the ASTM 

standards was revised from ±1.5 HRBW to ±2.5 

HRBW in 2019, and it is thought that review in light 

of present circumstances is conducted. However, 

even for 30 HRBW and 35 HR30TW, the margin 

for the tolerance is small, and the tolerance may not 

be appropriate. 

In this report, the estimated expanded 

uncertainty of the hardness value in indirect 

verification is the result of extracting only main 

uncertainty factors, and is obtained by using the 

smaller tolerance between two types of standards. In 

addition, experimental estimation of indenters and 

hardness reference blocks was conducted only about 

three manufacturers. Thus, the actual uncertainty 

may be even larger. Regardless of the above, when 

the dispersion is compared with the tolerance of the 

ASTM standards, the dispersion is equivalent to or 

Scale
Tolerance

(ISO 6508-2)

Tolerance

(ASTM E18)

Experimental result

(P I value)

30 HRC 0.8 HRC 0.8 HRC 0.6 HRC

60 HRC 0.8 HRC 0.4 HRC 0.7 HRC

90 HR15N 0.8 HR15N 0.5 HR15N 1.3 HR15N

30 HRBW 4.0 HRBW 0.8 HRBW 1.1 HRBW

90 HRBW 2.0 HRBW 0.8 HRBW 0.4 HRBW

35 HR30TW 3.0 HR30TW 0.8 HR30TW 0.4 HR30TW

Uncertainty 

factors 

Hypothesized variation of hardness 

values 

Depth measuring 

device 

Tolerances in verification of depth 

measuring device prescribed in ISO 

6508-2 and ASTM E18 

Diamond 

indenter 

Tolerance of difference in hardness value 

from reference indenter at testing with 

the identical reference block prescribed 

in ASTM E18 

Ball indenter Tolerance of difference in calibration 

value from reference block at testing 

reference block prescribed in ASTM E18 

Uniformity of 

reference block 

Tolerance of difference between the 

maximum and minimum values in 5-

point measurement prescribed in 

standard for reference block, ISO 6508-3 

or ASTM E18 

Calibration value 

of reference 

block 

PB values in Fig. 2 are hypothesized as 

expanded uncertainty. 

Machine 

deformation 

Although the tolerance is ±1.0 HR in ISO 

6508-2 or ASTM E18, it is hypothesized 

to be ±0.3 HR in view of actual 

circumstances. 

http://www.imeko.org/


ACTA IMEKO | www.imeko.org December 2020 | Volume 9 | Number 5 | 224 

exceeds the tolerance. Especially regarding 60 HRC, 

the difference between the tolerance ±0.5 HRC of 

indirect verification prescribed in the ASTM 

standards and the dispersion of the estimated 

hardness values is large. This fact suggests that the 

tolerance of ASTM standards is not realistic. 
 

Table 3: Uncertainty budget for each scale 

 

Table 4: Tolerance of standards and estimated extended 
uncertainty in indirect verification 

 

4. SUMMARY 

The following items including verification and 

evaluation were conducted and described in this 

report. 

• Evaluation of the dispersion of indenters and 

hardness reference blocks which are in the market. 

• Comparison of evaluation result and tolerances 

for indenter. 

• Estimation of uncertainty of hardness value in 

indirect verification of the testing machine based on 

the evaluation result of the reference block. 

• Verification of validity of the tolerance in the 

indirect verification prescribed in the ISO standards 

and ASTM standards. 

 

By conducting the items above, and the 

following were found. 

• For diamond indenter, there is a possibility that 

the indenters exceed the tolerance of standard are in 

market. 

• For tungsten carbide ball indenters, there is a 

possibility that the difference may exceed the 

tolerance of standard for 30 HRBW if the indenter 

manufacturer changes. 

• It was suggested that the estimated dispersion of 

the hardness values generated in indirect 

verification fell under the tolerance range of the ISO 

standards. 

• It was suggested that the estimated dispersion of 

the hardness values generated in indirect 

verification exceed the tolerance range of the 

ASTM standards, and in particular, the tolerance of 

ASTM standard for 60 HRC may be too strict. 

5. CONCLUSION 

A hardness value is not a physical quantity, but 

an industrial quantity obtained by a prescribed 

testing method. The fact is that testing machines and 

testing conditions vary within the prescribed 

tolerance range and relatively large dispersion of 

hardness values is observed even in a national 

standard level. If the fact that this is a destructive 

30 HRC

Factor Distribution Divisor

Depth measuring 0.5 HRC Rectanglar √3 0.289 HRC

Indenter 0.4 HRC Rectanglar √3 0.231 HRC

Uniformity of block 0.35 HRC Rectanglar √3×√5 0.090 HRC

Uncertainty of block 1.0 HRC Normal 2 0.500 HRC

Machine hysteresis 0.3 HRC Rectanglar √3 0.173 HRC

0.652 HRC

1.3 HRC

60 HRC

Factor Distribution Divisor

Depth measuring 0.5 HRC Rectanglar √3 0.289 HRC

Indenter 0.4 HRC Rectanglar √3 0.231 HRC

Uniformity of block 0.2 HRC Rectanglar √3×√5 0.052 HRC

Uncertainty of block 0.8 HRC Normal 2 0.400 HRC

Machine hysteresis 0.3 HRC Rectanglar √3 0.173 HRC

0.574 HRC

1.1 HRC

90 HR15N

Factor Distribution Divisor

Depth measuring 0.5 HR15N Rectanglar √3 0.289 HR15N

Indenter 0.5 HR15N Rectanglar √3 0.289 HR15N

Uniformity of block 0.25 HR15N Rectanglar √3×√5 0.065 HR15N

Uncertainty of block 0.7 HR15N Normal 2 0.350 HR15N

Machine hysteresis 0.3 HR15N Rectanglar √3 0.173 HR15N

0.569 HR15N

1.1 HR15N

30 HRBW

Factor Distribution Divisor

Depth measuring 0.5 HRBW Rectanglar √3 0.289 HRBW

Indenter 0.8 HRBW Rectanglar √3 0.462 HRBW

Uniformity of block 0.5 HRBW Rectanglar √3×√5 0.129 HRBW

Uncertainty of block 1.9 HRBW Normal 2 0.950 HRBW

Machine hysteresis 0.3 HRBW Rectanglar √3 0.173 HRBW

1.116 HRBW

2.2 HRBW

90 HRBW

Factor Distribution Divisor

Depth measuring 0.5 HRBW Rectanglar √3 0.289 HRBW

Indenter 0.8 HRBW Rectanglar √3 0.462 HRBW

Uniformity of block 0.35 HRBW Rectanglar √3×√5 0.090 HRBW

Uncertainty of block 0.7 HRBW Normal 2 0.350 HRBW

Machine hysteresis 0.3 HRBW Rectanglar √3 0.173 HRBW

0.676 HRBW

1.4 HRBW

35 HR30TW

Factor Distribution Divisor

Depth measuring 0.5 HR30TWRectanglar √3 0.289 HR30TW

Indenter 0.8 HR30TWRectanglar √3 0.462 HR30TW

Uniformity of block 0.5 HR30TWRectanglar √3×√5 0.129 HR30TW

Uncertainty of block 1.0 HRBW Normal 2 0.500 HR30TW

Machine hysteresis 0.3 HR30TWRectanglar √3 0.173 HR30TW

0.770 HR30TW

1.5 HR30TWExpanded uncertainty (k  = 2)

Combined standard uncertainty

Expanded uncertainty (k  = 2)

Combined standard uncertainty

Expanded uncertainty (k  = 2)

Combined standard uncertainty

Expanded uncertainty (k  = 2)

Combined standard uncertainty

Expanded uncertainty (k  = 2)

Combined standard uncertainty

Expanded uncertainty (k  = 2)

Combined standard uncertainty

Range (±)

Range (±)

Range (±)

Range (±)

Range (±)

Range (±)

Standard uncertainty

Standard uncertainty

Standard uncertainty

Standard uncertainty

Standard uncertainty

Standard uncertainty

Scale
Tolerance

(ISO 6508-2)

Tolerance

(ASTM E18)

Analysis

(Expanded uncertainty)

30 HRC 1.5 HRC 1.0 HRC 1.3 HRC

60 HRC 1.5 HRC 0.5 HRC 1.1 HRC

90 HR15N 2.0 HR15N 0.7 HR15N 1.1 HR15N

30 HRBW 4.0 HRBW 2.5 HRBW 2.2 HRBW

90 HRBW 2.0 HRBW 1.0 HRBW 1.4 HRBW

35 HR30TW 3.0 HR30TW 1.5 HR30TW 1.5 HR30TW

http://www.imeko.org/


ACTA IMEKO | www.imeko.org December 2020 | Volume 9 | Number 5 | 225 

test in which absolute reference does not exist is 

considered, it can be said that such dispersion is 

generated as a matter of course. On the other hand, 

under such circumstances, accredited systems and 

traceability are utilized. This type of issue may be a 

big problem that cannot be solved by company 

alone for manufacturers of hardness testing 

machines in each country. We hope that verification 

of consistency with public standards becomes 

emphasis for the sake of the industrial world. 

 

6. REFERENCES 

[1] ISO/IEC 17025:2017 General requirements for the 

competence of testing and calibration laboratories. 

[2] T. Sawa, "Current state of Rockwell hardness in the 

world", Proceedings of IMEKO World Congress, 

Belfast, UK, 2018. 

[3] ISO 6508-3:2015 Metallic materials Rockwell 

hardness test: Calibration of reference blocks. 

[4] ASTM E18-20 Standard Test Methods for 

Rockwell Hardness of Metallic Materials. 

[5] ISO 6508-2:2015 Metallic materials Rockwell 

hardness test: Verification and calibration of testing 

machines. 

 
 

http://www.imeko.org/