![Equivalent sun years in Arizona and Florida [12]](https://www.researchgate.net/profile/Jenny-Zhu-9/publication/326561943/figure/tbl1/AS:1093148218523648@1637638253092/Equivalent-sun-years-in-Arizona-and-Florida-12_Q320.jpg)
Content uploaded by Jenny Zhu
Author content
All content in this area was uploaded by Jenny Zhu on Nov 16, 2021
Content may be subject to copyright.
Effects of Natural Sunlight on
Fiberglass Reinforced Polymers for Crossarms
Jiayi Jenny Zhu and Mike S Schoenoff
Department of Engineering
GEOTEK, LLC
Stewartville, MN-55076, USA
jzhu@geotekinc.com
Fiberglass crossarms have been successfully used in the electrical
utility industry to improve grid reliability over the past 25 years.
Because they are engineered with fiberglass reinforced polymer
(FRP) materials, they are well suited to meet the demands of
continuous outdoor use while in service, provided they are
properly engineered and protected from the effects of outdoor
weathering, including ultraviolet (UV) exposure from natural
sunlight. Most fiberglass crossarm manufactures purport the use
of synthetic veils and UV inhibitors formulated into the resin
system as features to mitigate the effects of outdoor weathering.
Some fiberglass crossarm manufactures include the use of an
outer UV coating to provide further protection from the effects of
outdoor weathering. This paper explores the effects of natural
sunlight based on accelerated outdoor weathering test per ASTM
G90 on test coupons cut from commercially available fiberglass
crossarms which were either with or without a UV coating. The
paper will present side-by-side testing results including images of
samples with or without exposure to sunlight under ASTM G90,
total color change, and analysis from a material performance
perspective as well as an application perspective, and will address
factors which can affect the service life of fiberglass crossarms..
Fiberglass Crossarms, Fiberglass Reinforced Polymer
Composites, Accelerated Weathering Test
I. INTRODUCTION
Fiber reinforced polymers (FRP) are advanced composite
materials which have been used as fiberglass crossarms to
replace traditional wood crossarms in the utility industry for
several decades[1]. It is well known that fiberglass crossarms
are more uniform in strength, lighter in weight, and last longer
than wood crossarms. In 2011, an extensive test program was
developed and completed by Powertech, a research subsidiary
of BC Hydro, for the comparison of accelerated aging,
electrical, mechanical, and structural properties of crossarms
made by wood, steel, FRP, and other specialty composite
materials[2]. The test results showed that fiberglass crossarms
held structural loads well, resisted lightning strikes, and had
long term durability. However, outdoor weathering stability
of fiberglass crossarms has not been well studied. It was
reported that a series of FRP utility poles installed in Hawaii
in the early 1960s were removed from service after 50 years
of service because of fiber blooming concerns from ultraviolet
(UV) light exposure, not because of structural reasons[3]. Since
all synthetic and natural polymers absorb solar UV radiation
and undergo photolytic, photo-oxidative, and thermos-
oxidative reactions that result in polymer degradations[4 – 7], it
is important to understand and monitor the degree of changes
due to UV-induced polymer degradations and how they affect
the longevity and service life of fiberglass crossarms.
Currently, fiberglass crossarms are formed by the
pultrusion process through which continuous glass fiber
rovings are combined with glass fiber mats and synthetic
surfacing veils, impregnated with thermoset resin, and pulled
through a heated die to cure into the specified crossarm
structures. The pultrusion process is an effective and
economical manufacturing method for producing high
performance FRP composite crossarms. The use of synthetic
surfacing veils in the composite construction ensures a
smooth, resin-rich surface which helps to control fiber
blooming, minimizes accumulations of contaminates, and
reduces the risk of electrical tracking. Since the resin-rich
outer surface is composed of the same thermoset resin used
for binding glass fibers in the center core of fiberglass
crossarms, it bears similar weathering stability as the center
core thermoset resin. To improve surface protection of
fiberglass crossarms, a coating with specially designed
surface properties can be applied on the outer layer of a
crossarm. Since coatings can be applied after the pultrusion
process and have minimal impact on mechanical properties of
fiberglass crossarms, the chemistry for the coating materials
can be very different than the thermoset resins used for the
center core of fiberglass crossarms and therefore can be
designed or selected to offer superior surface protection
including excellent weathering stability for the fiberglass
crossarms.
To study the effects of natural sunlight on fiberglass
crossarms, we selected five commercially available fiberglass
crossarms for accelerated outdoor weathering test per ASTM
G90[8]. They were produced by five different manufacturers
and were either with or without a UV protective coating.
Upon completion of the twelve months of accelerated outdoor
weathering exposure, we examined the exposed areas for
surface smoothness and fiber blooming based on digital
images and water drop techniques. We also conducted
quantitative analysis on the test specimens which are related
to surface changes and service lives for fiberglass crossarms
101
2018 IEEE Rural Electric Power Conference
2153-3636/18/$31.00 ©2018 IEEE
DOI 10.1109/REPC.2018.00023
by measuring time-to-track based on arc tracking test per
ASTM D2303[9 ] and total color changes per CIELAB color
scales.
II. EXPERIMENTATIONS
A. Test Specimens
Test specimens for the accelerated outdoor weathering
evaluations were flat panels in the size of 50 mm x 140 mm.
They were cut from five commercially available fiberglass
crossarms which were purchased in the open marketplace in
2014 in new or like-new condition (i.e. never placed into
service) and were produced by five different manufacturers,
designated as A to E. According to the manufacturers’
product descriptions at the time of acquisition, three of them
(i.e., A, B, and C) had a UV protective coating on the outer
surface of the crossarms and two of them (i.e., D and E) had
no coating on the outer surface of the crossarms”.
B. Accelerated Outdoor Weathering Test
Accelerated outdoor weathering test was completed by Q-
Lab Arizona (Buckeye, AZ) per ASTM G90 for a test period
of twelve months using a Q-TRAC® Natural Sunlight
Concentrator[10] which automatically tracks the sun with a
solar tracking device and concentrates sunlight onto the test
specimens with a set of mirrors. Total accumulated UV
radiation on test specimens were calculated based on the
exposed solar energy recorded by the solar tracking device
and the reflectance efficiency of the mirrors used in the Q-
Trac tester.
Four replicas were cut from each of the fiberglass
crossarms and were tested with the outer surface of the
crossarms facing the sunlight. To prevent overheating on the
surface of test specimens during daytime and to simulate the
subtropical weather on the test specimens with the Q-Trac
tester, water was sprayed onto the test specimens periodically
by following Cycle 1 defined by ASTM G90 – i.e., 8 min.
every hour during daytime and 8 min. every 3 hours between
9:00 pm and 6:00 am.
C. Color Measurements
Color changes on the test specimens before and after
twelve months of accelerated outdoor weathering test were
measured by Q-Lab Arizona (Buckeye, AZ) using an X-Rite
Color i5 spectrophotometer with D65 illuminant. CIELAB
(L*, a*, b*) color scale was used for the color measurements
and the total color change ('E*) was calculated by equation
(1):
'E*= [(L* - Lo*)2 + (a* - ao*)2 + (b* - bo*)2]1/2 (1)
Where L*, a*, and b* were measured on the specimens
after twelve month of the accelerated outdoor weathering test;
Lo*, ao*, and bo* were measured on the specimens without
exposing to the accelerated outdoor weathering test.
D. Surface Analysis
Digital images on surfaces of the test specimens were
taken by using an iPhone 6S. Microscope images on surfaces
of a specimen with coating and a specimen without coating
were taken by using an AmScope microscope attached with a
MU1000 microscope digital camera (www.amscope.com).
Surface smoothness on the test specimens were analyzed
by water drop technique. All test specimens were rinsed under
tap water and pat-dried with clean paper tower prior to the
analysis. A colored water solution was prepared by adding
one drop of food die (Red Food Color, McCormick) into 50
mL of water. Water drops were placed on the test surface by
using a 1 mL disposable pipette and freshly prepared colored
water solution. Digital images were taken at intervals of 0 and
5 min. after water drops were placed on the surface of each
test specimen.
E. Arc Tracking Test
The two selected specimens for microscope images (i.e.,
the one with coating and the one without coating after
exposing to twelve months of accelerated outdoor weathering
test) along with the two controls without exposing to the
accelerated weathering test were submitted for arc tracking
test which was conducted by ELTEK Labs (St. Charles, MO)
according to the time-to-track test method described in section
10 per ASTM D2303. The recorded value of time-to-track is
the time when tracking reaches a distance of 25 mm above the
bottom electrode under a constant voltage of 2.5 kV and a
constant contaminant feed of 0.1% ammonium chloride with
0.02% X-100 surfactant.
Arc tracking test on additional specimens were also
evaluated but not included in the discussions due to an artifact
identified on the test specimens which was related to the
manufacturing process and was not resulted from outdoor
weathering effects[11].
III. RESULTS AND DISCUSIONS
A. Accelerated Outdoor Weathering Exposure
When a FRP crossarm is placed into service in the field, it
is exposed to natural sunlight. UV radiation from sunlight can
be absorbed by the polymers in fiberglass crossarms, resulting
in slow yet progressive damages to the polymers due to UV-
induced polymer degradations. The rate and severity of the
damage is affected by the amount of UV radiation absorbed
by polymers which is related to the time and location that
polymers are exposed to the outdoor weathering. Fig. 1 shows
plots of accumulated monthly UV radiation (295 – 385 nm)
and direct solar radiation (295 – 3000 nm) recorded by Q-Labs
located in Arizona and Florida in 2014, 2015, and 2016,
respectively[12]. It appears that despite the daily and seasonal
variations in UV and solar radiations, the accumulated
monthly radiations from one year to another remained
relatively constant at each location. Therefore, every twelve-
month or annual accumulated UV or solar radiation at each
location can be used for the estimation and computation for
accelerated weathering evaluations.
For the test specimens after twelve months of accelerated
outdoor weathering test which was completed per ASTM G90
in Nov. 2015 by Q-Lab in Arizona, the calculated total UV
and solar radiations were 1,363 MJ/m2 and 48,483 MJ/m2,
102
respectively. Based on the average annual data collected by
Q-Lab through years of actual sunlight measurements[10], as
listed in Table 1, the equivalent years of outdoor weathering
exposure on the test specimens in Arizona (arid) or in Florida
(subtropical) can be estimated according to the plots shown in
Fig. 2 – i.e., (i) The calculated total UV radiation of 1,363
MJ/m2 is approximately equal to ~ 4 years in Arizona or ~ 5
years in Florida. (ii) The calculated total solar radiation of
48,483 MJ/m2 is approximately equal to ~ 6 years in Arizona
or ~ 7 years in Florida.
B. Surface Changes after Accelerated Weathering Test
Digital images of the test specimens after twelve months
of accelerated outdoor weathering test were shown in Fig.3.
The top and bottom sections of each specimen were not
exposed to any sunlight while the middle section of each
specimen were exposed to the twelve months of concentrated
sunlight. It appeared that the exposed surface areas on the
three test specimens of A, B, and C having a UV-protective
coating remained smooth while the exposed areas on the two
test specimens of D and E having no coating became rough.
Microscope images on the surfaces of specimens A and D
were shown in Fig. 4. It was clear that both the exposed and
non-exposed surface areas on specimen A remained smooth
while the exposed surface area on specimen D showed
shinning exposed glass fibers, indicating some of the
polymers covering the glass fibers on the exposed surface area
of specimen D were lost due to UV-induced polymer
degradations, otherwise known as fiber blooming. Since fiber
blooming on fiberglass crossarms can lead to poor aesthetics,
difficulty in handling by linemen due to fiber glass slivers, and
potentially electrical failure, the early onset of fiber blooming
can result in shortened service life for fiberglass crossarms[3].
Figure 1. Plots of monthly UV radiation (295 –
385 nm) and
direct solar radiation (295
–
3000 nm) in 2014, 2015, and 2016 vs the
12 calendar month in a year. Data were accumulated based on daily
recorded values by Q
-
Labs located in Arizona and Florida,
respectively
[12]
.
Table 1 Equivalent sun years in Arizona and Florida [12]
Arizona Florida Arizona Florida
1 333 280 8,004 6, 588
2 666 560 16,008 13, 176
3 999 840 24,012 19, 764
4 1,332 1, 120 32,016 26, 352
5 1,665 1, 400 40,020 32, 940
Equivalent
Sun Years
Total UV Radiation, MJ/m2
Total Solar Radiation, MJ/m2
Figure 2. Plots of average total UV radiation and total solarv
radiation vs equivalent sun years based on data recorded by Q
-
Lab, as
listed in Table
1[12]. Calculated total UV radiation and total solar
radiation on the tested specimens after twelve months of accelerated
outdoor weathering exposure were also marked on the plots.
Figure 3. Digital image of the tested specimens: top and bottom
sections on the
tested surface of each specimen were not exposed to any
sunlight while the middle section were exposed to the twelve months of
concentrated sunlight.
Figure 4. Microscope images on top views of specimens A and D, top
and bottom sections on each test surface were exposed and non
-
exposed to
the 12 months of concentrated sunlight, respectively
.
103
The change in surface smoothness on test specimens can
also be evaluated by colored water drops placed on the test
surfaces, as shown in Fig. 5. It was apparent that the three
water drops on the upper non-exposed areas of each test
specimen had well-defined circular edges which remained
unchanged in Fig. 5a and 5b when images were taken
immediately and 5 min. after water drops were placed on the
test surfaces, respectively. The water drops on the exposed
areas of the test specimen A, B, and C looked similar to those
placed on the non-exposed areas, indicating no significant
changes to the exposed surface areas on these three test
specimens of which the surface was covered by a UV
protective coating. While water drops on the exposed areas of
the test specimens D and E appeared to be deformed in Fig.
5a, indicating the surface underneath the water drops was
rough. The exposed area of specimen D in Fig. 5a also
showed stained surface areas around water drops due to the
diffusion of colored water into the substrate, indicating the
presence of micro-voids which were resulted from the loss of
UV degraded polymers. Fig. 5b further showed that the water
drops on the exposed area of specimen D were diffused
completely into the substrate within five minutes and that the
stained area around water drops on exposed area of specimen
E were clearly visible due to a slower diffusion rate of colored
water into the substrate comparing to that of specimen D.
For fiberglass crossarms placed into service in the field, a
rough surface with exposed glass fibers and micro-voids can
trap contaminates and cause arc tracking or flashover,
resulting in service failure for fiberglass crossarms. To
compare the arc tracking resistance between specimen A and
D which were non-exposed or exposed to the twelve months
of concentrated sunlight, Table 2 listed measured values of
time-to-track per ASTM D2303 under a constant voltage of
2.5 kV and a constant contaminant feed of 0.1% ammonium
chloride with 0.02% X-100 surfactant. It showed that
measured values of time-to-track on the smooth surfaces of
the non-exposed specimens A and D were about the same –
i.e., 12 and 11 min., respectively. After exposing to twelve
months of concentrated sunlight, measured values of time-to-
track for specimens A and D dropped by 33% and 73%.,
respectively. These results suggest that arc tracking
progressed more rapidly on the rough surface with exposed
glass fibers and micro-voids of the exposed specimen D
comparing to the relatively smooth surface of exposed
specimen A. Since arc tracking leaves a carbonized path on
the non-conductive surface of fiberglass crossarms, it can turn
it into conductive and cause flashover, resulting in electrical
failure for the fiberglass crossarms. Therefore, the rough
surface resulted from UV-induced polymer degradations on
fiberglass crossarms without coating may lead to shorter
service lives than those with a UV protective coating.
Another common sign of surface degradation resulted
from UV-induced chemical reactions is faded color. Table 3
listed average values of measured CIELAB colors (L*, a*, b*)
for the five sets of test specimens before and after 12 months
of concentrated sunlight exposure. Average values and
standard deviations of calculated total color changes ('E*) per
equation (1) were also listed in Table 3. It showed that the
average values of 'E* for the test specimens of A, B, and D
were between 1 ~ 3 while the average values of 'E* for the
test specimens D and E were about 8, indicating 'E* for the
test specimens without coating were at least 2X higher than
those with a UV-protective coating. Apparently, the surface
of fiberglass crossarms without any coating was significantly
less color stable with more UV-induced chemical changes
than those with a UV-protective coating.
IV. CONCLUSIONS
UV radiation from natural sunlight can cause slow, yet
progressive damages to FRP polymers used in fiberglass
crossarms due to UV-induced polymer degradations. Despite
daily and seasonal changes in weather, research and scientific
data revealed that annual accumulation of UV or solar
radiation at a geological location is relatively constant. Our
twelve-month accelerated outdoor weathering test per ASTM
G90 in Arizona is estimated to be approximately equal to
about 4 ~ 6 years in Arizona under arid and dry weather or
about 5 ~ 7 years in Florida under subtropical and humid
weather.
Table 3 Measured CIELAB color values (L*, a*, b*) and
calculated total color changes ('E*)
L
o
*a
o
*b
o
*L* a* b* ''E* Std.
AYes 71.29 -1. 31 - 0.06 72. 21 -1.49 -0.67 1.11 0.05
BYes 72.45 -3. 75 - 0.07 73. 49 -3.75 -0.81 1.28 0.12
CYes 62.27 -2. 06 - 1.45 63. 75 -1.41 -4.27 3.25 0.05
DNo 73. 88 0. 21 -2. 17 78.03 - 1.57 4.17 7.79 0. 32
ENo 71. 80 -0. 49 9.81 79.98 -1.03 8.65 8.28 0.77
Specime n
ID
Color Measurements
Total Color Change
Coating
Table 2 Measured values of time-to-track per ASTM D2303 on
specimens A and D
Non-Exposed Exposed Change, %
AYes 12 8 -33%
DNo 11 3 -73%
Time-to-Track, min
Specime n ID
Coating
Figure 5. Digital image of test surfaces with colored water
drops: 5a) Images were taken immediately after water drops were
placed on test surfaces; 5b) Images
were taken five minutes after
water drops were left on test surfaces. The upper and lower sections
on each test surface were non
-exposed and exposed to the twelve
months of outdoor weathering test, respectively.
104
Surface analysis based on digital and microscope images
as well as water drop evaluations on the test specimens
indicated that those with a UV protective coating maintained
a smooth surface after the twelve-month accelerated outdoor
weathering test while those without coating showed rough
surface with fiber blooming and vacant micro-voids on areas
exposed to concentrated sunlight. Results of arc tracking test
per ASTM D2303 and measured values of total color changes
per CIELAB color scale also indicated that those with a UV-
protective coating retained better arc tracking resistance and
color stability after exposing to the twelve-month accelerated
outdoor weathering test than those without coating.
Since fiberglass crossarms are placed into service under
direct sunlight, it is apparent that those with a UV-protective
coating will have better protections against UV-induced
polymer degradations from daily exposure to natural sunlight
and therefore will have a longer service life than those without
any coating.
DISCLAIMER
1. GEOTEK performed mechanical strength testing in connection with
the tests underlying this publication, but the strength tests were
inconclusive and are not referenced herein.
2. Other environmental factors aside from UV radiation that may affect
crossarm degradation and coating reliability, such as pollution, freeze-
thaw cycling, wind-blown particulates, salt fog, and/or bird feces, were
not considered as part of the testing underlying this publication. As Q-
Lab document LL-9031 states, while the accelerated UV testing will
result in the "same amount of UV deposited over five years of Florida
(subtropical) testing" ... "[t]his is not meant to imply that the
degradation that occurs over one year of Q-TRAC natural sunlight
concentrator testing will necessarily be the same as five years of
Florida testing." Q-Lab document LL-9031 further states: "As with all
accelerated testing, the amount of acceleration depends on many
variables such as material composition, mode of degradation,
temperature response and moisture."
3. ASTM G90 states that the "relative durability of materials in natural or
field exposure can be very different depending on the location of the
exposure because of differences in UV radiation, time of wetness,
temperature, pollutants, and other factors. Therefore, even if results
from a specific accelerated test condition are found to be useful for
comparing the relative durability of materials exposed in a particular
exterior location, it cannot be assumed that they will be useful for
determining relative durability for a different location." The
accelerated testing underlying this publication was performed in
Arizona.
4. Q-Lab document LU-8030 states that "Joules do not reflect variations
in degradation caused by differences in exposure to moisture,
temperature, or wavelength spectrum of the light source.
Characterization and control of these other parameters is often more
important than Joules of radiant dosage." While the accelerated testing
described herein was not timed in Joules, one may note that Q-Lab
document LU-8030 states: "Exposures of equal Joules do not
necessarily produce equivalent degradation."
REFERENCES
[1] Headley, M. “A Market with Staying Power”, Composites
Manufacturing, March/April, 2016, pp. 14 – 19
[2] Powertech Laboratories, “Testing and Performance Evaluation of BC
Hydro Cross Arm Alternatives – Transmission Innovations Cross
Arms”, BC Hydro, Vancouver, BC, Canada, 2011.
[3] Utility Products, “Use of Composites Increases in the Utility and
Telecommunication Industries”, Vol 16, issue-2, Feb 1, 2012,
http://www.utilityproducts.com/articles/print/volume-16/issue-
2/readers-choice-awards/line-construction-maintenance/use-of-
composites-increases-in-the-utility-and-telecommunication-
industries.html
[4] Andrady, A.L., et al “Effects of Increased Solar Ultraviolet Radiation
on Materials”, J. Photochemistry and Photobiology B: Biology 46,
1998, pp. 96 - 103
[5] Rabek, J.F. Polymer Photodegradation, London: Chapman and Hall,
1995
[6] Scott, G. Polymer Degradation and Stabilization, London: Elsevier
Applied Science, 1990
[7] Wypych, J. Weathering Handbook, Toronto, Canada: Chemtec
Publishing, 1990
[8] ASTM G90 “Standard Practice for Performing Accelerated Outdoor
Weathering of Nonmetallic Materials Using Concentrated Natural
Sunlight”, ASTM International, West Conshohocken, PA, 2010, DOI:
10.1520/G0090-10
[9] ASTM D2303 “Standard Test Methods for Liquid-Contaminant,
Inclined-Plane Tracking and Erosion of Insulating Materials”, ASTM
International, West Conshohocken, PA, 2013, DOI: 10.1520/D2303-
13, www.astm.org
[10] Technical Bulletin LL-9031.3 © 2011, “Q-TRAC Natural Sunlight
Concentrator”, Q-Lab Corporation, www/q-lab.com, accessed on Feb.
23, 2016. http://www.q-lab.com/documents/public/002697f8-4405-
4421-9055-b86f7430e6a2.pdf?ReturnUrl=/resources/technical-
bulletins.aspx
[11] Arc track testing for both the non-exposed and exposed specimens E
showed the same value of time-to-track at 1 minute which was
significantly lower than what we would expected. Upon examination
of the test specimens E at the completion of the arc track testing, we
found that the tracking took a path along a seam on the test surface that
was present in both the non-exposed and the exposed test specimens.
A similar seam was not observed on the test surface of specimens A or
D. We believe the seam present in the test specimens E was an artifact
of the manufacturing process and not the result of outdoor weathering
effects. Based on this finding, the results of arc tracking test for
specimens E were not included in the discussions of outdoor
weathering effects for this paper.
[12] Recorded monthly weather summary data by Q-Labs at Arizona site
and Florida site, respectively.
https://www.myweathertest.com/WeatherData.aspx.
105
