Acta Polytechnica


https://doi.org/10.14311/AP.2022.62.0473
Acta Polytechnica 62(4):473–478, 2022 © 2022 The Author(s). Licensed under a CC-BY 4.0 licence

Published by the Czech Technical University in Prague

EVALUATION OF RESIDUAL STRENGTH OF POLYMERIC YARNS
SUBJECTED TO PREVIOUS IMPACT LOADS

Gabriela Hahna,
Antonio Henrique Monteiro da Fonseca Thomé da Silvab, c,

Felipe Tempel Stumpfd, Carlos Eduardo Marcos Guilhermea, ∗

a Universidade Federal do Rio Grande, Policab – Stress Analysis Laboratory, Av. Itália km. 8, 96203-000 Rio
Grande, Brazil

b Petróleo Brasileiro SA – Petrobras, CENPES R&D Centre, Ilha Cidade Universitária - Cidade Universitária,
21941-915 Rio de Janeiro, Brazil

c Universidade Federal Fluminense – Departamento de Engenharia Mecânica, Rua Passo da Pátria 156 Bl. D.,
24210-240 Niterói, Brazil

d Universidade Federal do Rio Grande do Sul, Departamento de Engenharia Mecânica, Rua Sarmento Leite 425,
90050-170 Porto Alegre, Brazil

∗ corresponding author: carlosguilherme@furg.br

Abstract.
The discovery of oil fields in deep and ultra-deep waters provided an opportunity to evaluate the

use of synthetic ropes, complementarily or alternatively to traditional steel-based mooring lines in
offshore units, mainly because of the former’s lower specific weight. Considering the series of complex
dynamic-mechanical mainly axial loads to which these structures may be subjected, originated from
different sources, such as wind, water current, tide, etc., there may be cases when at least one of
these lines may possibly face an abrupt, shock-like axial load of considerably larger magnitude. The
goal of the present study is to evaluate the residual tensile strength of three different synthetic yarns
(polyester, and two grades of high modulus polyethylene) after exposure to such axial impact loads. It
was observed that, for the tested materials, polyester is the one with the largest impact resistance to the
conditions evaluated herein, mainly because of its comparatively greater energy absorption properties.

Keywords: Offshore mooring, ultra deep-water, impact load.

1. Introduction
Since the end of the World War II, there has been
an increase in the application of synthetic materials,
mainly because of the reduction of their production
costs and their significantly advantageous mechanical
properties [1]. As an example, one can mention the
construction of polymeric ropes, which can be used in
a wide range of sports and industrial applications, such
as climbing, rescue operations, mooring of offshore
structures, shipping operations, etc. [2].

During the 1990s, the offshore oil industry began
to replace the traditional mooring system based on
steel cables and chains by systems consisting mainly
of polyester cables. The main motivation for this shift
was the severe increase in the water depth in which
these structures were now being anchored, requiring
compliant ropes with low specific weight in order to
reduce the overall weight of the floating system [3, 4].
Nowadays, as examples, one can mention the synthetic
fibres typically used for mooring ropes manufacturing:
polyester (PET), high modulus polyethylene (HMPE),
polyamide (PA), liquid crystal polymer (LCP), aramid
and polypropylene (PP).

Apart from the mechanical loads originating from
the movement of the floating unit, such anchoring

systems may be subjected to some degree of envi-
ronmental damage caused, for example, by ultravio-
let incidence and hydrolysis, depending on the fibre
group [5, 6]. Yarn-on-yarn abrasion is another (now a
mechanical) degradation mechanism, even more rel-
evant than the previous ones, that can affect the
material’s mechanical behaviour [7]. For characterisa-
tion purposes considering mooring ropes applications,
static and dynamic stiffness of polymeric multifila-
ments are typically assessed according to ISO 18692 [8]
and ISO 14909 [9, 10].

Polyester yarns have a high mechanical resistance,
good tenacity and abrasion resistance [7, 11, 12].
When exposed to the environment under typical moor-
ing conditions, they do not degrade considerably
and are resistant to hydrolysis and ultraviolet inci-
dence [5, 13]. Polyester is also not biodegradable,
has a negligible creep behaviour at room temperature
and, when exposed to high temperatures, it contracts
instead of expanding [14, 15].

High modulus polyethylene is produced from a gel
spinning process, resulting in a highly crystalline struc-
ture, oriented and extended along the fibre axis, with
many different grades available in the market with
specific properties. In general, HMPE fibres present a

473

https://doi.org/10.14311/AP.2022.62.0473
https://creativecommons.org/licenses/by/4.0/
https://www.cvut.cz/en


G. Hahn, A. H. M. F. T. da Silva, F. T. Stumpf, C. E. M. Guilherme Acta Polytechnica

lower-than-water density allowing its buoyancy in reg-
ular water, which makes it a very interesting choice for
marine applications. As for its mechanical properties,
the fibre has a high tenacity and stiffness as compared
to similar materials. Also, the deformation of HMPE
at rupture is very low, but it shows remarkable creep
behaviour at lower temperatures [16, 17].

There is a lack of previous studies in the literature
regarding the influence of severe and abrupt tensile
loads on the mechanical properties of polymeric mate-
rials applied to mooring ropes. However, it represents
a relevant topic in regards to synthetic fibres used
for mooring and operation lines since during its life-
time, an anchoring or operation line might be exposed
to such loads several times. One important factor
that influences the capacity to support shock loads is
the degree of crystallinity, which is inversely propor-
tional to the resistance to high, instantaneous tensile
forces [18, 19].

Considering the aforementioned, the main goal of
this paper is to assess the residual tensile strength
of polyester and high modulus polyethylene yarns
exposed to a prior shock-like axial load.

2. Materials and methods
2.1. Materials
In the present work, one grade of PET and two dif-
ferent grades of HMPE are evaluated, referred to as
PET, HMPE1 and HMPE2, respectively. The materi-
als have titers of 3300 dtex, 1761 dtex, and 1759 dtex,
respectively.

2.2. Methods
2.2.1. Environmental conditions during tests
All tests were performed with 500 mm long yarn spec-
imens conditioned according to ISO 139:2014, which
determines that the samples must stay for at least two
hours in an environment at 20 ± 2 ◦C and a relative
humidity of 65 ± 4 % prior to any experimental proce-
dure. The tests themselves must also be performed in
such environmental conditions.

2.2.2. Tensile tests in reference unexposed
samples

To determine the YBL reference values, for compari-
son purposes, unexposed fibres were tested according
to ASTM D2256. 30 rupture tests were performed
for 500 mm long yarn samples of PET, HMPE1 and
HMPE2 at 250 mm/min. Prior to tensile testing, sam-
ples were twisted along their axes with 60rounds per
meter. An EMIC DL2000 universal testing machine
with a 1 kN load cell was used.

2.2.3. Impact tests
It is considered that there is a critical velocity, above
which the material shows brittle behaviour. The
British standard BS EN 892:2012 proposes impact
tests in mountaineering ropes, which consists in the
application of an instantaneous tensile force applied

Figure 1. Free-fall diagram. Source [20].

by a free-falling mass (Figure 1) [20]. The input data
to the experiment are the rope length, the free fall
height, the mass of the falling object, and standard
atmospheric conditions, such as temperature and rela-
tive humidity. BS EN 892:2012 brings different proce-
dures depending on the investigated factors, such as
stiffness, cycle number for impact load, transmitted
force in the rope and maximum stretching.

Multifilaments used in the impact test are expected
to present the capacity of dissipation and absorption
of the potential energy. Specimens must be capable
of preserving their original elastic behaviour, granting
structural integrity and not compromising their me-
chanical properties. In the present paper, we apply
such loads to the specimens and investigate eventual
changes in their yarn break load (YBL) in the two
distinct moments: immediately after the impact load,
and 24 hours after the test. The YBL of the tested
material is compared to that of an unexposed sample.

The application of the abrupt tensile forces was per-
formed with increasingly heavier dead weights, typ-
ically from 1 % to 7 % YBL (and higher whenever
applicable) increased by 2 % YBL steps. If samples
did not fail when exposed to the chosen dead weight,
residual strength was evaluated (as described in the
next sections) and a new set of impact tests with
increased dead weight was performed. In all tests,
the dead weight was released from a 250 mm height,
which corresponds to a half of the samples’ length.
A total of 30 untwisted samples were used for each of
the investigated materials.

2.2.4. Residual strength tests
2.2.4.1. Residual strength of partially im-

pacted samples
In order to evaluate the intermediate residual quasi-
static tensile strength of the materials after the ap-
plication of each of the impact loads steps detailed
in Section 2.2.3, those samples that did not fail by

474



vol. 62 no. 4/2022 Evaluation of residual strength of polymeric yarns . . .

Material YBL (N/tex) Specific deformation at break (%)
PET 0.76 ± 0.01 10.90 ± 0.04

HMPE1 3.04 ± 0.15 3.30 ± 0.14
HMPE2 3.08 ± 0.10 3.20 ± 0.09

Table 1. Yarn break load of reference unexposed samples of PET, HMPE1 e HMPE2.

Material
Maximum impact load

promoted by the 1 % YBL
dead weight (N/tex)

YBL
(N/tex) after

impact

Specific
deformation at

break (%)
HMPE1 1.24 ± 0.03 3.25 ± 0.09 3.30 ± 0.12
HMPE2 1.33 ± 0.02 3.23 ± 0.10 3.20 ± 0.12

Table 2. Results of the tensile tests for residual strength evaluation immediately after the impact load of 1 % YBL.

Material
Maximum impact load

promoted by the 3 % YBL
dead weight (N/tex)

YBL (N/tex)
after impact

Specific
deformation at

break (%)
PET 0.31 ± 0.01 0.76 ± 0.01 10.30 ± 0.40

HMPE1 2.14 ± 0.13 3.15 ± 0.17 3.20 ± 0.20
HMPE2 2.22 ± 0.10 3.11 ± 0.11 3.20 ± 0.11

Table 3. Results of the tensile tests for residual strength evaluation immediately after the impact load of 3 % YBL.

rupture were subjected to tensile tests using the pro-
cedure of Section 2.2.2. The time interval between
both experiments was of 1 minute and 30 seconds,
during which the samples were kept under the same
controlled environmental conditions of Section 2.2.1.
The same EMIC DL2000 machine was used.

2.2.4.2. Residual Strength of samples after
resting time

A second round of tensile tests was performed in
order to evaluate the influence of the resting time
of the samples between the impact experiments and
the residual tensile strength measurements. For that,
instead of testing the samples for their residual YBL
immediately after the shock-like events, they were first
left to rest for 24 hours in a controlled environment
(see Section 2.2.1). After this period, the tensile tests
were performed according to the procedure detailed
in Section 2.2.2.

3. Results and discussion
3.1. Yarn break load of reference

(unexposed) samples
Table 1 shows the results for YBL of PET, HMPE1 e
HMPE2, determined according to the experimental
setup detailed in Section 2.2.2 considering 30 samples
each. It can be seen that PET shows a lower YBL
and a higher elongation at break when compared to
both grades of HMPE, as expected.

3.2. Tensile tests immediately after the
impact loads

As detailed in Section 2.2.3, each material was tested
with abrupt tensile loads using an initial dead weight
equivalent to 1 % YBL, which exposed the samples to
a specific impact load. Due to technical difficulties
in measuring the very small force during the tests
for PET (approximately 2.5 N), this material was
excluded from this first experimental batch.

After the tests with 1 % YBL, no visible damage was
observed in any of the HMPE samples. Moreover, no
sample has failed during the impact load. Immediately
after the impact tests, tensile tests were performed
in order to observe eventual changes in the materials’
original YBL. Table 2 shows the results considering
30 samples each.

Although the impact forces applied to the samples
were close to half of the materials’ original YBL, both
HMPE showed an apparent increase in their tensile
strength: HMPE1 increased its YBL by about 7 %
and HMPE2 by about 5 %, when compared to the
materials’ reference YBL (see Table 1). There was
no significant change in the specific deformation at
break. It was observed that the standard deviation
of the HMPE1 samples decreased as compared to the
results in the reference unexposed samples.

Then, additional impact tests were performed, in-
creasing the dead weight to 3 % of the materials’ orig-
inal YBL, now including PET samples. Again, there
was no visible damage in any of the samples after the
application of the sudden axial loads. Table 3 shows

475



G. Hahn, A. H. M. F. T. da Silva, F. T. Stumpf, C. E. M. Guilherme Acta Polytechnica

Material

Maximum impact load
promoted by the 5 %

YBL dead weight
(N/tex)

Number of
samples
impact
tested

Samples
broken
during
impact

YBL
(N/tex)

after
impact

Specific de-
formation
at break

(%)
PET 0.50 ± 0.02 30 0 0.77 ± 0.02 10.50 ± 0.40

HMPE1 2.86 ± 0.12 78 48 Not applicable
HMPE2 2.98 ± 0.10 54 24 3.09 ± 0.12 3.20 ± 0.13

Table 4. Results of the tensile tests for residual strength evaluation immediately after the impact load of 5 % YBL.

Material
Maximum impact

force caused by dead
weight drop (N/tex)

Number of
samples
impact
tested

Samples
broken
during
impact

YBL
(N/tex)

after
impact

Specific de-
formation
at break

(%)
PET 7 % 0.62 ± 0.02 30 10 0.77 ± 0.02 10.50 ± 0.35
PET 9 % 0.69 ± 0.01 47 17 0.77 ± 0.02 10.30 ± 0.41
PET 11 % 1.12 ± 0.02 58 28 Not applicable

HMPE2 6 % 3.09 ± 0.09 52 22 Not applicable

Table 5. Results of the tensile tests for residual strength evaluation immediately after impact load – higher dead
weights.

the results of the tensile tests performed after the
impacts considering 30 samples each.

Here, the axial load for the HMPE samples exceeds
almost 70 % of the materials’ original YBL, while for
polyester, it was close to 50 % of its original YBL.
The results for PET showed no significant difference
when compared to the unexposed reference material
(Table 1). Both HMPE samples showed, after impact,
a higher YBL than that of the reference samples, but
now only 3 % higher for HMPE1 and 1 % for HMPE2.
Again there was no significant change in the specific
deformation at the break. The high standard devia-
tion of HMPE1 may be an indication of permanent
damage caused to the multifilament structure during
the impact tests even thoughthe quasi-static mechani-
cal behaviour was not jeopardized.

Increasing the deadweight to 5 % YBL, more than
50 % of the HMPE1 samples failed by rupture dur-
ing the impact test, which means that the impact
strength of that material was reached. Less than half
of HMPE2 samples showed failure by rupture. Some
of the samples that did not break during the impact
test showed visible, macroscopic damage in their struc-
ture, while all PET samples did not show any visible
structural damage. Table 4 shows the residual tensile
strength test results. Impact sampling was performed
in order to always guarantee around 30 viable samples
for the residual tensile strength tests afterwards.

Results show that, the higher the impact load, the
higher the standard deviation and, in this case, it is
reaching the limits of the impact tolerance for the
HMPE fibres, with a significant amount of fibres al-
ready failing under impact. Again, PET showed a very

small increase in its tensile strength, while the specific
deformation at break remained almost unchanged.

The next set of tests was performed with a dead-
weight of 7 % YBL. As expected, all samples of
HMPE2 have broken during impact, so, for that ma-
terial, the load was decreased to 6 % YBL, in order
to find its impact strength. Following the procedure
of increasing the load by 2 % YBL, PET was further
tested up to 11 % YBL, which was found to be beyond
the material’s impact strength. Table 5 shows these
results.

3.3. Tensile tests 24 hours after the
impact loads

In this section, the effect of the time interval be-
tween the impact test and subsequent tensile tests
on the quasi-static residual strength of the fibres was
observed, aiming to observe any potential microstruc-
tural accommodation. Therefore, because of the sim-
ilar mechanical behaviour of HMPE1 and HMPE2,
only PET and HMPE1 were chosen to undergo this
new set of experiments.

Samples of PET were subjected to an impact load
of 5 % YBL and HMPE1 to 3 % YBL. These values
were chosen because they were found to be the highest
impact loads (among the loads tested in the study)
that do not cause macroscopic damage to the material
(see Section 3.2). After the impact experiment, the
samples were left to rest for 24 hours in a controlled
environment as determined by ISO 139:2014. Table 6
shows the results.

In Table 3, it can be seen that PET has a tensile
strength of 1.12 ± 0.02 N/tex when the tensile test is
performed immediately after the impact test. When

476



vol. 62 no. 4/2022 Evaluation of residual strength of polymeric yarns . . .

Material Number ofsamples

Maximum
impact force

(N/tex)

YBL (N/tex)
after impact

Specific
deformation
at break (%)

PET 5 % YBL 30 0.50 ± 0.02 0.77 ± 0.02 11.20 ± 0.33
HMPE1 3 % YBL 30 2.14 ± 0.13 3.11 ± 0.01 3.20 ± 0.12

Table 6. Results of the tensile tests for residual strength evaluation 24 hours after impact load.

the tensile test was conducted 24 hours after the
impact experiment, the YBL was measured as 1.13 ±
0.02 N/tex. For HMPE1, the equivalent results are
of 3.15 ± 0.17 N/tex (immediately after impact) and
3.11 ± 0.01 N/tex (24 hours after impact). It was
concluded that the materials tested do not restore
their tensile strength after a considerable time interval
of 24 hours after the impact test.

4. Conclusions
The main goal of this article is to evaluate an eventual
loss in tensile strength of polyester and (two grades of)
high modulus polyethylene yarns after an exposure
to abrupt, axial impact loads (as a percentage of
materials’ original YBL). This is made by measuring
the YBL of the unexposed reference material’s YBL,
and comparing it to the YBL of samples previously
exposed to different levels of impact loads.

The obtained results suggest that, among the tested
materials, PET is the one being less affected by a %
YBL impact load, having shown impact strength to
an axial load equivalent to about 9 % of its original
YBL (see Table 5). Both evaluated grades of HMPE,
HMPE1 and HMPE2, presented an impact strength
equal to 5 % YBL and 6 % YBL, respectively (see
Tables 4 and 5). A possible explanation is that PET’s
elongation at break (∼10 %) is significantly larger
than those of HMPE’s (∼3.3 %), which means that
the former is more capable of absorbing strain energy
than the latter.

Because of the absence of similar studies in the liter-
ature, this study is considered to be pioneer in terms
of the assessment of the consequences of axial, abrupt
loads for the posterior tensile strength of polymeric
yarns. The methodology followed here is considered
adequate to be applied when one intends to quan-
titatively compare the impact strength of different
polymeric fibres. It should be noted that due to the
difference between the axial stiffness of the different
tested materials, naturally, the strain rate is expected
not to be the same when this approach, based on
deadweight release, is used. If one intends to apply
exactly the same strain rate to the materials being
compared, more sophisticated experimental apparatus
must be employed. It is also important to note that
the results regarding impact strength can be highly
affected by the temperature during the experiments,
so it is recommended to perform the shock-like exper-
iments at the service temperature of the materials for
more accurate comparisons.

References
[1] Jr. W. D. Callister and D. R. Rethwisch. Materials

Science and Engineering: An Introducion. John Wiley
& Sons, 2007.

[2] A. J. McLaren. Design and performance of ropes for
climbing and sailing. Proceedings of the Institution of
Mechanical Engineers, Part L: Journal of Materials:
Design and Applications 220(1):1–12, 2006.
https://doi.org/10.1243/14644207JMDA75.

[3] V. Sry, Y. Mizutani, G. Endo, et al. Consecutive
impact loading and preloading effect on stiffness of
woven synthetic-fiber rope. Journal of Textile Science
and Technology 3:1–16, 2017.
https://doi.org/10.4236/jtst.2017.31001.

[4] K. H. Lo, H. Xü, L. A. Skogsberg. Polyester rope
mooring design considerations. In Ninth International
Offshore and Polar Engineering Conference. Brest,
France, 1999. ISOPE-I-99-1691999.

[5] J. P. Duarte, C. E. M. Guilherme, A. H. M. F. T.
da Silva, et al. Lifetime prediction of aramid yarns
applied to offshore mooring due to purely hydrolytic
degradation. Polymers and Polymer Composites
27(8):518–524, 2019.
https://doi.org/10.1177/0967391119851386.

[6] M. M. Shokrieh, A. Bayat. Effects of ultraviolet
radiation on mechanical properties of glass/polyester
composites. Journal of Composite Materials
41(20):2443–2455, 2007.
https://doi.org/10.1177/0021998307075441.

[7] S. R. da Silva Soares, V. E. Fortuna, F. E. G.
Chimisso. Yarn-on-yarn abrasion behavior for polyester,
with and without marine finish, used in offshore
mooring ropes. In 9th Youth Symposium on
Experimental Solid Mechanics, pp. 60–64. 2010.

[8] International Organization for Standardization. Fibre
ropes for offshore stationkeeping – Part 2: Polyester
(ISO Standard No. 18692-2), 2019.

[9] International Organization for Standardization. Fibre
ropes for offshore stationkeeping – High modulus
polyethylene (HMPE) (ISO Standard No. 14909), 2012.

[10] F. T. Stumpf, C. E. M. Guilherme, F. E. G.
Chimisso. Preliminary assessment of the change in the
mechanical behavior of synthetic yarns submitted to
consecutive stiffness tests. Acta Polytechnica CTU
Proceedings 3:75–77, 2016.
https://doi.org/10.14311/APP.2016.3.0075.

[11] T. S. Lemmi, M. Barburski, A. Kabziński, K. Frukacz.
Effect of thermal aging on the mechanical properties of
high tenacity polyester yarn. Materials 14(7):1666,
2021. https://doi.org/10.3390/ma14071666.

477

https://doi.org/10.1243/14644207JMDA75
https://doi.org/10.4236/jtst.2017.31001
https://doi.org/10.1177/0967391119851386
https://doi.org/10.1177/0021998307075441
https://doi.org/10.14311/APP.2016.3.0075
https://doi.org/10.3390/ma14071666


G. Hahn, A. H. M. F. T. da Silva, F. T. Stumpf, C. E. M. Guilherme Acta Polytechnica

[12] G. Susich. Abrasion damage of textile fibers. Textile
Research Journal 24(3):210–228, 1954.
https://doi.org/10.1177/004051755402400302.

[13] M. Moezzi, M. Ghane. The effect of UV degradation
on toughness of nylon 66/polyester woven fabrics. The
Journal of The Textile Institute 104(12):1277–1283, 2013.
https://doi.org/10.1080/00405000.2013.796629.

[14] C. Oudet, A. Bunsell. Effects of structure on the
tensile, creep and fatigue properties of polyester fibres.
Journal of Material Science 22:4292–4298, 1987.
https://doi.org/10.1007/BF01132020.

[15] H. Y. Jeon, S. H. Kim, H. K. Yoo. Assessment of long-
term performances of polyester geogrids by accelerated
creep test. Polymer Testing 21(5):489–495, 2002.
https://doi.org/10.1016/S0142-9418(01)00097-6.

[16] P. Davies, Y. Reaud, L. Dussud, P. Woerther.
Mechanical behaviour of HMPE and aramid fibre ropes
for deep sea handling operations. Ocean Engineering

38(17-18):2208–2214, 2011.
https://doi.org/10.1016/j.oceaneng.2011.10.010.

[17] Y. Lian, H. Liu, Y. Zhang, L. Li. An experimental
investigation on fatigue behaviors of HMPE ropes.
Ocean Engineering 139:237–249, 2017.
https://doi.org/10.1016/j.oceaneng.2017.05.007.

[18] E. Hage Jr. Resistencia ao impacto. In S. V.
Canevarolo Jr. (ed.), Técnicas de Caracterização de
Polímeros. Artliber, São Paulo, 2003.

[19] E. L. V. Louzada, C. E. M. Guilherme, F. T. Stumpf.
Evaluation of the fatigue response of polyester yarns
after the application of abrupt tesion loads. Acta
Polytechnica CTU Proceedings 7:76–78, 2017.
https://doi.org/10.14311/APP.2017.7.0076.

[20] I. Emri, A. Nikonov, B. Zupančič, U. Florjančič.
Time-dependent behavior of ropes under impact
loading: A dynamic analysis. Sports Technology
1(4-5):208–219, 2008.
https://doi.org/10.1080/19346182.2008.9648475.

478

https://doi.org/10.1177/004051755402400302
https://doi.org/10.1080/00405000.2013.796629
https://doi.org/10.1007/BF01132020
https://doi.org/10.1016/S0142-9418(01)00097-6
https://doi.org/10.1016/j.oceaneng.2011.10.010
https://doi.org/10.1016/j.oceaneng.2017.05.007
https://doi.org/10.14311/APP.2017.7.0076
https://doi.org/10.1080/19346182.2008.9648475

	Acta Polytechnica 62(4):473–478, 2022
	1 Introduction
	2 Materials and methods
	2.1 Materials
	2.2 Methods
	2.2.1 Environmental conditions during tests
	2.2.2 Tensile tests in reference unexposed samples
	2.2.3 Impact tests
	2.2.4 Residual strength tests


	3 Results and discussion
	3.1 Yarn break load of reference (unexposed) samples
	3.2 Tensile tests immediately after the impact loads
	3.3 Tensile tests 24 hours after the impact loads

	4 Conclusions
	References