Investigation of different liquid Solutions for Dye Penetration Tests used in

Standard IEC 62217 for Design and Routine Testing

by B. Hill, V. Sklenicka, F. Schmuck on behalf of

CIGRE WG B2.21

G. BESZTERCEY, E. BROCARD, C. PONS, R. HILL, D. KOTEK, R.E. MACEY, M. ANGEL PÉREZ

LOUZAO, A. PIGINI, G. PIROVANO, R. WESLEY SALLES GARCIA, J. SEIFERT, F. SCHMUCK
(Convener), M.R.SHARIATI, V. SKLENICKA, V. VOSLOO, G. WATT, T. HAYASHI, X. LIANG, B. H.
GAN, T. GILLESPIE, J. HWANG, A. KHALILPOUR, B. STAUB, R. MATSUOKA, K. NAITO, I.
GUTMAN, A. PHILLIPS, R. GARCIA FERNANDEZ

1 Introduction
The internal cores of composite insulators are usually made of glass-fibre reinforced plastic
(FRP). This can be in the form of rods for suspension and post type insulators or tubes for
apparatus insulators. Rods are typically manufactured by a continuous pultrusion process,
with usually only unidirectional glassfibre reinforcement in the axial direction. The filament
winding process is commonly used to manufacture tubes. The winding angle of the fibres is
varied according to the application requirements, such as optimising the tube for bending
strength, pressure resistance or both stress factors. With the first issue of an IEC product
standard for composite insulators in the year 1992 [1], two tests were introduced as design
tests to verify the quality of the rod material and later on in 1998 for hollow core insulators
as well [2]. One of them, the water diffusion test is a combined test, where after 100h of
boiling time, a defined voltage is applied. The test criteria are that the current through the
pre-stressed rod specimen must not exceed the value of 1 mA and no puncture or flashover
shall occur. The water diffusion test mainly identifies any weakness of the resin system in
terms of hydrolysis as well as any interface problems between fibres and resin, such as
inappropriate sizing use for example. The other test is a dye penetration test, which uses a
defined solvent enriched with colour pigments (dye). The intention of this test is to check the
specimens for glass fibre capillaries. Due to the ease of this 15 minute test, this test is also
frequently used as a quality test during FRP rod pultrusion operations.

1/9
2 Need for Investigation
IEC standards are subject to continued maintenance cycles to keep the standard content up
to date with the product evolution. In light of this, two new standards were introduced: IEC
62217 [3] in 2005 and the 2nd Edition of IEC 61109 [4] in 2008. The dye used in the previous
Edition was Fuchsine, which has been identified as a known carcinogen . For this reason, it
was generally replaced by Methin dye. The following parts of the standards show the
difference:

IEC 61109 1st Edition

1% alcohol solution of Fuchsine (1 g Fuchsine in 100 g Ethanol)

IEC 61109 2nd Edition/IEC 62217 1st Edition

1% Methyl or Ethyl alcohol solution of a red/violet Methin dye (Astrazon or Basonil).

Keeping in mind that many different grades of Astrazon exist, the test conditions could vary
significantly and provide different results. For this reason, CIGRE WG B2.21 was assigned to
the investigation of this situation, in order to provide recommendations to IEC WG 11 which
is in charge of maintaining the standards mentioned above. An example of the simple test
set-up is shown in Fig. 1.

Fig. 1: Test principle

2/9
3 Initial Investigations
The first systematic investigations were performed on 63mm rod specimens taken randomly
from production [5]. The results of the tests done according the specified standard but with
different dyes, are summarized as follows:

TIME TO FAILURE
DYE CHEMICAL FAMILY PASS/FAIL
(min)

Pararosaniline
Basic Fuchsine > 60 PASS
Chloride
ASTRAZON Red Violet
Methin 20 PASS
3RN-01
ASTRAZON BR Red BN
Methin > 60 PASS
200%
ASTRAZON BR Red 4G
Cyanine Dye > 15 PASS
200%
ASTRAZON Red BBL
Azo Dye > 60 PASS
200%
ASTRAZON FBL 200% Azo Dye 7 FAIL
ASTRAZON FBL 200% Azo Dyestuff
10 FAIL
LZ Preparation
ASTRAZON Red GTL Azo Dyestuff
> 15 PASS
200% Preparation

Melacryl Fuchsine Triarylmethane > 15 PASS

This study concluded that

• different grades of Astrazon used will give different test results

• a correlation to the previously used Fuchsine test solution with more than ten years
experience is not shown
• the toxicity of some Astrazon grades must be highlighted and excluded for use in the
standards.

4 Further Work conducted in B2.21

The work within the working group had the following focus:

• Can the dye solution be more clearly defined?

• Can the length of the test specimens be increased to the same as for the water
diffusion test (e. g. all specimens for both tests equal to 30 mm)?

3/9
Capillary effects, especially for glass capillaries are well described in literature and a widely
accepted formula is

where:
• γ is the liquid-air surface tension (energy/area)
• θ is the contact angle θ
• ρ is the density of liquid (mass/volume)
• g is acceleration due to gravity (length/time2)
• r is radius of tube (length).

A comparison of the difference in surface tension between water and Ethanol at 25°C, shows
values of 72 dyn/cm and 23 dyn/cm respectively. The value of Ethanol would increase if it is
diluted with water. The Ethanol`s lower surface tension would cause a lower capillary effect,
however the other material factors like density and contact angle on the capillary wall
contribute positively to the effect (for example, the density of Ethanol is 0.8 g/cm3 versus
water at 1.0 g/cm3).

A simple test was carried out to show the influence of the different liquids (Fig. 2).

Fig. 2: Effect of surface tension of different liquids on a silicone

rubber surface

For the test, a hydrophobic surface (silicone rubber) was chosen to show the effect of the
surface tension of different liquids. Water forms the expected drop on the surface (left).
Water enriched with 1% by mass Astrazon still exhibits a droplet shape, however with a
slightly reduced contact angle (middle). A test solution as specified, consisting of 1% by mass
Astrazon in an Ethanol solution wets the surface completely (right). Not shown is that pure
Ethanol shows less surface spreading than the test solution on the right. The drop volume
was identical for all cases.

Furthermore, the situation in a FRP material was considered to be more complex, hence
experimental work with real rod specimens produced for insulators was added to the
investigation. The tests were performed on a Round Robin basis to achieve results with a

4/9
good statistically confirmation. At all times, the results were benchmarked with the previous
test solution consisting of Fuchsine and Ethanol.

4. 1 Comparison to Basonil
Basonil was one of the alternative types of dye. It was investigated whether the choice of the
dye families Basonil or Astrazon has an influence on the test. A very simple test was
performed on one identical specimen.

Fig. 3: Test with Basonil and Ethanol Fig. 4: Test with Astrazon and Ethanol

The specimen was placed in a Basonil/Ethanol test solution as specified. Within the specified
test time of 15 minutes, red marks were easy to identify, which meant a non-conformance
to the standards (Fig. 3). The same specimen was then thoroughly cleaned and placed in a
Astrazon/Ethanol solution as specified by the standard (Fig. 4). The difference in comparison
to Fig. 3 is significant, as only one wick was indentified with the question of whether the
capillary was still filled by Basonil and Ethanol. More systematic investigations were
performed in this respect, as well as tests using an increase in specimen height.

A rod diameter of 51 mm was selected for this investigation. Five specimens were placed in
each of the test solutions. The number of wicks was counted at certain time intervals and
the average value calculated. The results of the tests with specimens of different heights are
shown in Fig. 5 and Fig. 6.

5/9
The specimens with the standardized height of 10 mm failed the test after placement in all
the test solutions (Fig. 5). It is obvious that the test solutions containing Basonil are more
severe than the liquid with Fuchsine and Astrazon.

Wicking as a Function of Time (51 mm Rod, 10 mm Height)

70

60
Astrazon/Ethanol
50 Astrazon/Methanol
Fuchsin/Ethanol
Number of Wicks

40 Fuchsin/Methanol
Basonyl Red 560 Red/Ethanol
Basonyl Red 560 Red/Methanol
30

20

10

0
0 2 4 6 8 10 12 14 16 18 20
Time in Minutes

Fig. 5: Wicking test with 10 mm rod specimens in different test solutions

In contrast to this result, the 30 mm long specimens placed in the Fuchsine solutions and the
Astrazon/Methanol solution passed the test (Fig. 6). Using the testing with Fuchsine as a
reference, an increase in specimen height would reduce the severity of the test. Rod quality
differentiation would only achieve an equivalent result, when Basonil is used. This tendency
was also found for other rod diameters investigated.

Wicking as a Function of Time (Rod 51 mm, 30 mm Height)

70 Astrazon Red/Ethanol
Astrazon Red/Methanol
Fuchsin/Ethanol
60 Fuchsin/Methanol
Basonyl Red 560 Red/Ethanol
50 Basonyl Red 560 Red/Methanol
Number of Wicks

40

30

20

10

0
0 10 20 30 40 50 60 70
Time (minutes)

Fig. 6: Wicking test with 30 mm rod specimens in different test solutions

This test philosophy was continued in the Round Robin testing, with rod specimens from
different suppliers.

6/9
4. 2 Summary of the Round Robin Testing
The following tables show the results achieved with different rods and different
combinations of solvent and dye. In contrast to the tests of chapter 4.1, new specimens
were used for each test, which explains the scattering of some results. The abbreviations
reflect the following combinations:

• A/E=Astrazon+Ethanol,
• B/E=Basonil+Ethanol,
• F/E=Fuchsine+Ethanol,
• A/M=Astrazon+Methanol,
• B/M=Basonil+Methanol,
• F/M=Fuchsine+Methanol.

For all tests, a 1% dye solution was prepared as specified by the standard.

Rod Specimen A
Test Specimen Specimen Time for first Penetration
Duration Diameter Height Number of penetrated Specimens/Total Specimen Number
(min) (mm) (mm) Combination of Dye and Solvent
A/E B/E F/E A/M B/M F/M
105 40
19 10 > 120 > 120 > 120 > 120
(3/5) (4/5)
100 60 15
51 10 > 120 > 120 > 120
(5/5) (5/5) (5/5)
30
120 75 10 > 120 > 120 > 120 > 120 > 120
(1/5)
19 30 > 120 > 120 > 120 > 120 > 120 > 120
40 60
51 30 > 120 > 120 > 120 > 120
(5/5) (2/5)
76 30 > 120 > 120 > 120 > 120 > 120 > 120
19 30 > 60 > 60
60 51 30 > 60 > 60
76 30 > 60 > 60

Rod Specimen B
Test Specimen Specimen Time for first Penetration
Duration Diameter Height Number of penetrated Specimens/Total Specimen Number
(min) (mm) (mm) Combination of Dye and Solvent
A/E B/E F/E A/M B/M F/M
16 10 > 120 > 120 > 120 > 120 > 120 > 120
51 10 > 120 > 120 > 120 > 120 > 120 > 120
76 10 > 120 > 120 > 120 > 120 > 120 > 120
120
16 30 > 120 > 120 > 120 > 120 > 120 > 120
51 30 > 120 > 120 > 120 > 120 > 120 > 120
76 30 > 120 > 120 > 120 > 120 > 120 > 120
16 30 > 60 > 60
60 45 30 > 60 > 60
76 30 > 60 > 60

7/9
Rod Specimen C
Test Specimen Specimen Time for first Penetration
Duration Diameter Height Number of penetrated Specimens/Total Specimen Number
(min) (mm) (mm) Combination of Dye and Solvent
A/E B/E F/E A/M B/M F/M
17 10 > 60 > 60
60 44 10 > 60 > 60
76 10 > 60 > 60

Rod Specimen D
Test Specimen Specimen Time for first Penetration
Duration Diameter Height Number of penetrated Specimens/Total Specimen Number
(min) (mm) (mm) Combination of Dye and Solvent
A/E B/E F/E A/M B/M F/M
17
31.5 10
(10/10)
20
38 10 > 60
(10/10)
60
10 5
44 10
(2/10) (10/10)
17 2
48 10
(6/10) (2/10)

Rod Specimen E
Test Specimen Specimen Time for first Penetration
Duration Diameter Height Number of penetrated Specimens/Total Specimen Number
(min) (mm) (mm) Combination of Dye and Solvent
A/E B/E F/E A/M B/M F/M
20.5 10 > 120 > 120
120
30 10 > 120 > 120

Rod Specimen F1
Test Specimen Specimen Time for first Penetration
Duration Diameter Height Number of penetrated Specimens/Total Specimen Number
(min) (mm) (mm) Combination of Dye and Solvent
A/E B/E F/E A/M B/M F/M
1
120 88 10 > 120
(2/10)

Rod Specimen F2
Test Specimen Specimen Time for first Penetration
Duration Diameter Height Number of penetrated Specimens/Total Specimen Number
(min) (mm) (mm) Combination of Dye and Solvent
A/E B/E F/E A/M B/M F/M
60 88 10 > 60 > 60

8/9
Considering the tables, the individual results are sometimes contradicting. For example, F1
would pass F/E-testing and fail A/E-testing. In contrast to this, example D passed both, but
with F/E had a penetration time of between 20 and 60 minutes for all specimens, which
constitutes a pass as well, as the time was > 15 minutes. This can be attributed to the fact
that the specimens were cut from different rods or batches.

Comparing examples A and B, where the full matrix was tested, a good reference correlation
was found for A/E and F/E. For this reason, the combination of Astrazon BR 200 diluted in
Ethanol was identified as providing the most equivalent results to the previously used
Fuchsine and Ethanol test solution. This is valid, when the height of the specimens remains
at 10 mm as tested in the past.

4. 3 Severity of Testing with Dyes and Alcohol

Under service conditions, the level of moisture ingress is important for composite insulators,
hence the presence of water is the stress factor. A simple test was performed on the
identical specimen shown in Fig. 3 and 4 in chapter 4. 1. The same specimens were placed in
a test solution made up of water and Astrazon. No penetration was observed in all cases.
This is in line with the change of surface tension shown in Fig. 2. These results indicate that
the test with dyes and alcohol is more severe in comparison to the expected stress by water.
It also confirms the historical experience since 1992 that rod qualities, which passed the
Fuchsine test with Ethanol, perform reliably in service.

5 Recommendation for IEC TC 36 WG 12

The dye penetration test is a procedure, which can be easily utilized as a design test or/and
routine test during FRP rod pultrusion manufacture. Using the experience with Fuchsine and
Ethanol as reference, a test solution made of Astrazon BR 200 and Ethanol as well as
Methanol showed equivalent results. Due to the easier availability of Methanol, the test
solution for standard IEC 62217 is recommended to be Astrazon BR 200 and Methanol to
tighten the corresponding test procedures for the relevant product standards. The height of
the test specimens is to be kept at 10 mm.

6 References
[1] IEC 61109, Ed. 1, 1992: Composite insulators for a.c. overhead lines with a nominal
voltage greater than 1 000 V - Definitions, test methods and acceptance criteria.
[2] IEC 61462 TR-2, 1998: Composite insulators – Hollow insulators for use in outdoor
and indoor electrical– Definitions, test methods, acceptance criteria and design
recommendations.
[3] IEC 62217 Ed. 1, 2005: Polymeric insulators for indoor and outdoor use with a
nominal voltage > 1 000 V - General definitions, test methods and acceptance criteria.
[4] IEC 61109, Ed. 2, 2008: Composite suspension and tension insulators for a.c.
overhead lines with a nominal voltage greater than 1 000 V - Definitions, test
methods and acceptance criteria.
[5] McQuarrie, T. S.: Composite insulators for overhead lines - tests, evaluation of
Fuchsine and red/violet Methin as dye penetration fluids for core rods, presented at
meetings of NEMA and ANSI in 2005

9/9