1.Fundamentals of stripe coatingFromJPCL,January 2014
More items forQuality Control
Paint failures on bolted connection that had not been stripe
coatedPhotos courtesy of Corrosion Control Consultants and Labs,
Inc.
Consider the following scenario, which points out one of the
worst disappointments in the painting of steel structures.The owner
carefully plans a project to include a well-written specification,
careful material evaluation and selection, a qualified contractor,
and thorough inspection of the work. The project is done on time,
within budget, and with no claims for extra work. Two years later,
visual inspection of the project reveals that 99% of the painting
work shows no signs of failure. Yet, essentially every edge, bolt,
and weld is rusting.What happened?The project specification did not
require striping or stripe coating of all edges and welds during
the painting work. Is this the problem? Maybemaybe not.What is
Striping or Stripe Coating?A stripe coat is a coat of paint applied
only to edges or to welds on steel structures before or after a
full coat is applied to the entire surface. The stripe coat is
intended to give those areas sufficient film build to resist
corrosion.1Therefore, striping, as it is sometimes called, is the
process of painting the edges of a surface or welds to give them
extra protection. Striping is done before priming or before the
application of a full coat of paint.1(In this article, the terms
stripe coating and striping are used interchangeably.)SSPC-PA 1,
Shop, Field, and Maintenance Painting of Steel, includes the
following advice about stripe coating.2 If stripe coating is
specified for a project, then all corners, crevices, rivets, bolts,
welds, and sharp edges should receive a stripe coating with the
priming paint before the steel receives a full coat of primer. The
stripe coat should extend at least 1 in. (2 cm) from the edge. To
prevent removal of the stripe coat by later application of the
primer, the stripe coat should be allowed to set to touch before
the full coat of primer is applied. (However, it should not be
permitted to dry long enough to allow rusting of the unprimed
steel.) Alternatively, the stripe coat may be applied after a
complete coat of primer, especially if a long drying period for the
stripe coat would allow the uncoated steel to deteriorate. Tinting
of the stripe coat is advisable to promote contrast. Stripe coating
is most effective on edges that are rounded by grinding.The
specification notes that stripe coating is advantageous in
preventing coating breakdown on edges, etc., in very corrosive
surroundings, but it is an expensive operation and may only be
justified when it is believed that the cost will be compensated for
by extra service life of the coating system.Is Stripe Coating
Necessary?Stripe coating of edges, bolts, and welds is often
specified because liquid paints tend to flow away from these parts.
This is a result of surface tension in the paint film and shrinkage
of the paint film during curing.If this occurs, the paint film at
or near the edges will be thinner than elsewhere on the painted
surface, and the result can be early corrosion failure in these
areas. This can become a critical issue when the paint is failing
on the nuts, bolts, rivets, and welds, because these are the items
holding the structural pieces together.
Edge failure on stiffener that was not stripe coated
Edge failure evident on yellow angle
The benefits of striping are two-fold. First, it tends to fill
in small voids, laps, and irregularities in the substrate (such as
porosity in welds). Second, if allowed to cure to the point of
tackiness, striping tends to retard the next full coat of paint
material from flowing away from edges.High-solids paints are less
prone to thinning at edges than low-solids paints because they
generally have faster setting time, higher viscosity, and lower
surface tension.At one time, most structural steel painting work
was done with low-solids, relatively slow-curing, oil-based (alkyd)
materials. The fact that the industry has moved toward the use of
faster setting, higher solids coating materials, which exhibit less
tendency to flow away from edges after application, does not mean
that stripe coating is not necessary.The corrosiveness of the
environment will often determine whether stripe coatings are
needed. Stripe coating is often considered most cost-effective in
highly corrosive environments such as the insides of tanks and
marine or chemical exposures. In moderately corrosive environments
such as those frequently wet by fresh water, coating choice and
good control of the application without stripe coating may be
adequate to protect the structure cost-effectively. In mild
environments such as those with low humidity or indoors, striping
is not necessary.Stripe Coating TechniquesSince the original ATB on
stripe coating was published in 2001, SSPC has issued SSPC-PA Guide
11, Guide to Protection of Edges, Crevices, and Irregular Surfaces.
Published in 2008, the guide discusses the reasons for employing
extra corrosion protection measures on edges, corners, crevices,
bolt heads, welds, and other irregular steel surfaces, as well as
various protection options such as edge grinding, chamfering, and
application of stripe coats. Some details, including the advantages
and limitations of specific methods of obtaining additional coating
thickness by stripe coating, are described to assist the
specification writer in assuring that the project specification
will address adequate corrosion protection.While Guide 11 should be
consulted for projects that may include stripe coating application,
each specifier and paint applicator must interpret the necessity,
means, and methods for stripe coating for each individual
project.Therefore, the following information is provided for the
reader, based on the authors experiences and interaction with
various paint manufacturers, specifiers, and applicators. This
information is not meant to be comprehensive; for more specific
questions on stripe coating application, you should refer to Guide
11.When Should Stripe Coating be Specified?Stripe coating should be
specified when the history of the structure indicates that edge
failure of the paint system has been a problem. Consideration also
should be given to specifying stripe coating in a severely
corrosive environment, or if the paint manufacturer recommends
stripe coating.Is Stripe Coating an Additional Coat of Paint?Owners
and contractors have disagreed about whether stripe coating is an
additional coat of paint. That depends on what the specification
says. The need to cover a blast cleaned surface is paramount in
corrosive environments. Therefore, the logical course is to apply
the stripe coating after the primer. In this case, the stripe
coating is clearly an extra step. On the other hand, in moderate
environments or if there are not a lot of edges, it may be possible
to apply the stripe coat just prior to the full primer. Then the
contractor may have workers applying the stripe coat in front of
workers applying the primer, and both of them using paint from the
same cans. This process would not necessarily be considered an
extra step.Which Generic Paints Warrant Consideration of Stripe
Coating?For the most part, low-solids/low-viscosity paints (such as
alkyds) tend to benefit from stripe coating. In general,
fast-setting paints (such as inorganic zincs) and
high-solids/high-viscosity paints (such as epoxy mastics) do not
draw away from edges. However, striping does apply additional
coating thickness to edges that might not have received enough
paint originally.Which Coating Layers Warrant Stripe
Coating?Keeping in mind that the primary benefit of stripe coating
is compensation for possible reduced coating thickness at sharp
edges and irregularities in the substrate, it is reasonable to
conclude that only the primer should be striped. After application
of the primer, substrate irregularities are covered.Applying stripe
coats to all layers of paint can cause more harm than good. Too
much paint increases stresses in a coating film, thereby causing
cracking or peeling. The tendency of liquid paint to pull away from
edges is reduced once a layer of primer has been applied. It is
quite common to measure 750 micrometers (30 mils) of paint or more
on a surface near edges where a three-coat system of 300450
micrometers (1218 mils) was specified with stripe coating of all
three layers.Should Stripe Coating Be Applied Before or After the
Full Coat of Primer?If a high degree of surface cleanliness is
specified, such as SSPC-SP 10/NACE No. 2, Near-White Blast Cleaning
(the equivalent of Sa 2 in ISO 8501-1), the applicator has only a
short period of time, depending upon atmospheric conditions, to
prime the steel substrate before flash rusting occurs. To preclude
flash rusting, the entire substrate probably should be primed first
and the stripe coating applied later. The stripe coat should then
be tinted so that it is obvious where the stripe coat was applied
and if any areas were missed.Is Thinning Required for the Striping
Material?If stripe coating with a particular paint material is
specified, the application data sheet should be consulted for
thinning instructions for the application method selected. For
instance, if the stripe coat is to be applied by brush, the
thinning instructions for brush application should be followed. No
extra thinning should be done. Too much solvent in the paint,
especially when the stripe coat is applied before the primer, will
require more time for the stripe coating to become tacky. Solvent
entrapment, bubbling, or pinholing can occur.Should a Thickness Be
Specified for a Stripe Coat?Since irregular surfaces are one of the
places stripe coating is used, it may be difficult or impossible to
get an accurate dry film thickness reading. Nevertheless, it is
important to remember that if total dry film thickness is exceeded
by applying both a stripe coat and a full coat, then film defects
may result. To achieve a stripe coat that is not excessively thick,
the specifier may require that the paint be applied to produce a
visual color change on the affected areas and not specify a
particular wet or dry film thickness. It should be noted that only
a portion of the paint applied directly to an edge flows away, so
only a small amount of additional paint is needed to bring the
coating on an edge to the same thickness as on flat surfaces.What
Application Methods Should Be Used for Stripe Coating?The specifier
and applicator must first examine the required qualities of the
stripe coating to determine the optimum method of application. In
general, the required qualities of stripe coating are filling voids
and other irregularities in the affected substrate areas; providing
a tacky surface for subsequently applied full coats of paint to
adhere to; and not exceeding the optimum dry film thickness for the
stripe coat in combination with the full coat.Two application
methods meet the requirements of these three qualities: brushing
and spraying with conventional or air-assisted airless equipment.
The specifier should permit all of these application methods for
stripe coating, depending on specific job conditions. For instance,
brushing can be used for stripe coating of small, complex shapes
(such as lattice members and bolted connections), whereas
conventional spraying is appropriate for the edges of large
structural shapes.Application methods that can deposit relatively
high volumes of paint (e.g., rolling with a heavy nap roller or
airless spraying) should be avoided to prevent excessive dry film
thickness and possible film defects. (This assumes that the stripe
coating or full layer of primer is being applied while the
underlying material is still tacky. More latitude in application
methods can be allowed if a full layer of primer is applied and
allowed to cure until its dry-to-recoat time. Then, the stripe
coating can be thought of as an additional coat of paint being
applied to the primer.)Edge Retentive CoatingsYou have probably
also heard of edge retention coatings, which claim to have an edge
coating thickness similar to that of a nearby flat plate. The
question youre asking is, can I use one of these coatings, or do I
still need to carry out stripe coating?The answer is simplestripe
coating still needs to be carried out, as it serves more than one
purpose. In addition to increasing the film thickness at the edge
of plates or beams, stripe coating carried out by brush is better
at wetting the surface and forcing the paint into cracks and
crevices, over weld beads and bold heads, and other areas which are
subject to premature failure.ConclusionStriping or stripe coating
is used to extend the life of certain paint systems in corrosive
environments. It compensates for liquid coatings that flow away
from edges of steel structures, thus reducing the dry film
thickness. For stripe coating to be beneficial and cost-effective,
the specifier must consider the configuration of the structure to
be painted and the type of paint system to be applied. Stripe
coating should be limited to one coat of paint to avoid overly
thick coating systems. Proper stripe coating application is needed
to avoid defects in the paint film that can cause other problems
besides early rusting, for which the stripe coating was
applied.Editors Note: The original ATB on stripe coating was
written by Jon R. Cavallo, P.E., of Corrosion Control Consultants
and Labs, Inc. (Eliot, ME) for the May 2001JPCL. It was slightly
updated for this issue byJPCLTechnical Editor Brian Goldie.
References1. SSPC Protective Coatings Glossary (Pittsburgh, PA,
USA: SSPC: The Society for Protective Coatings, 2011), p. 201.2.
SSPC-PA 1, Shop, Field, and Maintenance Painting of Steel
(Pittsburgh, PA, USA: SSPC: The Society for Protective Coatings,
April 2000), p. 13.The Journal Of Protective Coatings &
Linings2014 Technology Publishing Company
Comment from Michael Deaton, (2/26/2014, 7:03 AM)
After supervising the 1 million square foot Innerbelt project in
Cleveland last year and dealing with the very intense inspection by
Mr. Dave Nolan, owner of Quality Control Services, stripe coating
was an essential part of the coatings application. There is over a
half a million bolts on this project and the finish coat is white,
therefore the stripe coat must provide a paint tight seal. Painters
utilized 4" cigar or weeny rollers to apply 1st the organic zinc,
then macropoxy 646 and finally the acrylic polyurethane to all
bolts, edges, welds, etc. The finish coat only required striping
where the airless gun could not access, but both the primer coat
and the intermediate required full striping. This striping was very
time consuming and should be factored into any bid.
Comment from Tom Selby, (2/26/2014, 12:49 PM)
It makes more sense to get all blasted surfaces covered with the
first coat of paint so that there is no compromising of the quality
of the initial blast. After that coat is dry a contrasting color
can be used to stripe coat with a brush or weenie roller.
Comment from Billy Russell, (2/26/2014, 4:35 PM)
3.Testing adhesion of multi-coat system
When should the adhesion of an applied coating or lining
multi-coat system be tested?From Karen Fischer Amstar of
WNYAdhesion testing should be performed for one of two basic
reasons: if the specification calls for it as a qualifying test for
acceptance of the coating system, or if there is a failure or
suspected failure in the coating system (material and/or methods)
that cannot be evaluated (or fully evaluated) by non-destructive
methods.One must always keep in mind that an adhesion test is a
destructive test, so the resulting test area now becomes a repair
that could, in itself, fail. This is especially important to keep
in mind in the case of linings or any system that will be in
immersion service, mechanical service, or a chemical/harsh
environmental service. Because it may be necessary to perform
adhesion testing in multiple areas (depending on the size of the
suspected areas), there will be multiple repairs. Destructive
testing should always be the last method employed, not the first
method, when evaluating a coating for suspected or obvious
failure.From James Albertoni CA Department of Water ResourcesSome
good instances where a multi-coat system should be tested for
adhesion include if the re-coat window is missed, if the topcoat is
not specifically recommended by the manufacturer to be compatible
with the basecoat, if the basecoat and topcoat are from two
different manufacturers, or if it is suspected one of the coats was
mixed slightly off ratio. Most importantly, the system should
always be tested for adhesion if the spec calls for it.From Daniel
Liu APCFirst, it will be up the specifier to decide if an adhesion
test is required, and, if so, the specification should include not
just the testing requirement but also the acceptance criteria and
tester type. From my experience in the field of tank coating, this
test is normally not required in the specification because it is a
destructive method. Any repair area creates a weak point in the
lining, so the more repair areas you have, the more weak spots you
have.However, it is quite necessary to make a proper adhesion test
recommended by the paint maker when application has or may have
deviated from the specification, such as by exceeding the recoat
interval, not maintaining the proper level of humidity, or using
the wrong mixing ratio for plural-component coatings. Adhesion is
quite important for tank coatings that are to be immersed in
liquid. But passing the adhesion test does not mean the whole
coating system is conclusively qualified for service. The test is
only a reference.From Tom Swan M-TESTIts important to note that if
an adhesion test is called for in the specification, the document
should specify failure criteria as well as the pull tester to be.
All pull testers pull at different rates, and, when I discuss pull
tests with most people, they have no idea what the pass/fail
criterion is or what adhesion tester to use.If you want to use
adhesion testing for pass/fail testing, the specifications should
specify the minimum pull required and what test instrument to use.
Also note that if you stop the test before the coating fails, there
is no guarantee that the pull fixture will not take off the coating
when you try to remove the fixture, nor does passing the test
guarantee that the adhesion or coating integrity was not affected
by the pull.From Manpreet Singh Spiecapag Australia PTY LTDIf the
clients specification calls for adhesion testing, the paint system
should be simulated on a test specimen of the same material class,
100 mm2and 6 mm thick. ISO 4624 describes the method of performing
the adhesion test. Acceptance criteria, unless specified by the end
user, shall be a minimum of 7 MPA at 23 C, and, at 65 C, no more
than 40% decrease from pull-off at 23 C.From Atanas Cholakov
ACTAdhesion should be tested after the complete cure of the coating
system. Information on curing can be acquired from the paint
suppliers technical representative. In the product data sheet,
curing is highlighted in a table in accordance with different
ambient temperatures and other conditions.From Trevor Neale TF
Warren GroupCritical service specifications typically call for
adhesion testing, so I assume this question relates to field
painting where weather and other delays are often unavoidable and
formal adhesion testing is not part if the job/project
specification. If there is any suspicion that adhesion may be
compromised, then the appropriate form of adhesion testing is
recommended to ensure the complete system integrity. This is simply
a CYA procedure to avoid possible conflicts, or worse, premature
failures.From Bryant Chandler Greenman-Pedersen, Inc.Adhesion
testing on coatings must be done after the proper cure time at the
correct temperature. This enables the coating to develop the full
physical properties. If the coating is tested prematurely, the
results often will not meet the specified minimum requirement. The
test may or may not be destructive, depending on the thickness of
the coating/substrate, and whether or not the test is continued
until coating disbondment.As called for in ASTM D 7234 (adhesion
testing of coatings on concrete), scoring around the dolly down to
the substrate will require a repair even if the test does not go to
failure and stops at the minimum test value; a thick coating system
(>3040 mils) on a metallic substrate may require scoring if
called for in the specification. If the test can be stopped at the
minimum value specified without causing coating failure, than the
dolly can be removed, often times by striking the dolly with a
sharp blow from the side or carefully inserting a sharp 5:1 tool
(putty knife) at the glue line and shearing off the dolly.
Repairing the top surface may be required but is much better than
having to repair the total coating system.When testing thermal
spray coatings, always perform the adhesion tests before
application of the seal coat. Tests performed after seal coat
application will result in test values that are two to three times
the value of virgin thermal sprayed coating.
The Journal Of Protective Coatings & Linings2013 Technology
Publishing Company
4. The case ofbubbles, and pinholes, and blisters, oh
my!FromJPCL,September 2013
JamesD.MachenPCS, KTA-Tator, Inc.James D. Machen is a Senior
Coatings Consultant with KTA-Tator, Inc., a coatings consulting
engineering firm and distributor of inspection instruments, where
he has been employed for over 20 years. Machen is an SSPC-certified
Protective Coatings Specialist, a NACE-certified Coatings Inspector
Level 3 (Peer Review), and a Level II Inspector in accordance with
ASTM D4537. He performs coating failure analyses, coating system
recommendations, specification preparation, and major project
management for a variety of clients in the transportation, water
and waste, power generation, chemical processing, and marine
industries. He holds a B.A. from the Pennsylvania State
University.RichardBurgessKTA-Tator, Inc., Series EditorThis months
Case from the F-Files describes the problem of bubbles, pinholes,
and blisters in a polyurethane finish coat applied to new
structural steel members at a coal-fired power generation plant.
Many of the pinholes and bubbles were so small that they were
difficult to detect with the unaided eye. Many of the largest
blisters on the webs of structural members were very flat and
shallow and also difficult to detect by eye. These conditions
became more difficult to see overtime as thin layers of dirt from
normal plant operating processes formed on the surface of the
polyurethane finish coat. This case file illustrates that
interacting variables, rather than a single cause, can combine to
cause a failure.BackgroundThe specification required that the
structural steel be blast cleaned in the shop in accordance with
SSPC-SP 6/NACE No. 3, Commercial Blast Cleaning. Following blast
cleaning, a two-coat system, consisting of a moisture cured
urethane (MCU) zinc-rich primer and an aliphatic polyurethane
finish, was shop-applied. The MCU primer was specified to be
applied at a dry film thickness (DFT) of 2.5 to 3.5 mils, and the
polyurethane finish was to be applied at a DFT of 4.0 to 5.0 mils.
The total two-coat DFT was to be 6.5 to 8.5 mils.
Fig. 1: Sections of newly-coated steel members at a coal-fired
power plant displayed blistering and other signs of coating
failure. Photos courtesy of James D. Machen, KTA-Tator, Inc.
Field touchup work was specified to be SSPC-SP 2, Hand Tool
Cleaning, and/or SSPC-SP 3, Power Tool Cleaning, followed by the
application of a coat of surface-tolerant epoxy mastic (4.0 to 6.0
mils DFT) and a finish coat of polyurethane (4.0 to 5.0 mils
DFT).The steel was delivered to the project site for sequenced
erection. In mid-summer, near the completion of the project,
blistering and peeling were observed. At that time, the shop
contractor mobilized a field team to make repairs. Repairs were
reported to have been performed using low-pressure water cleaning
(4,0005,000 psi), in conjunction with hand and power tool cleaning,
to identify and remove defective areas, which were then touchup
repaired.In the spring of the next year, additional coating defects
were discovered and field touchup was again performed. However, the
same problems reportedly continued to appear. As a result of the
continuing problems, an independent investigation of the coating
problem was undertaken.Field Investigation
Fig. 2: Close-up of typical concentrations of small, fine
blisters in the polyureathane finish coat
The tests and inspections performed during the field
investigation were those typically associated with failure
investigations, and included the following. A visual assessment was
performed to determine the degree and distribution of coating
defects (in this instance bubbles, pinholes, blisters, and
peeling). Total coating thickness was measured using a Type 2
electronic film thickness gage operated according to ASTM D7091,
Nondestructive Measurement of Thickness of Nonmagnetic Coatings on
a Ferrous Base. The number of coatings present and the thickness of
each were determined using a destructive coating thickness gage as
described in ASTM D4138, Standard Test Methods for Measurement of
Dry Film Thickness of Protective Coating Systems by Destructive
Means. An integral portable microscope (50X) was used to observe a
cross-section of the applied coating. The number of coating layers
and thickness of each were measured. Further, evidence of intercoat
contamination, voids, underlying rust, mill scale, and pinholes was
recorded. Adhesion testing was conducted using Method A (X-Cut) of
ASTM D3359, Measuring Adhesion by Tape Test. Method A involves
cutting an X through the coating to the substrate using a razor
knife. Pressure sensitive tape is placed over the X-cut, then
rapidly removed. The amount of coating detached by the tape is
rated in accordance with the ASTM rating scale. Ratings of 4A and
5A are considered to represent good adhesion, 2A to 3A represent
fair adhesion, while 0A and 1A represent poor adhesion. The coating
system was removed in small areas, and the substrate was examined
for under-film corrosion or mill scale. Active under-film corrosion
may be associated with the coating failure and may also contribute
to a shortened life of the system.Coating samples at both failing
and non-failing areas were removed for laboratory analysis, and
digital images of the typical field coating conditions were
obtained.Visual ObservationsThe structural steel consisted
primarily of vertical and horizontal I-beam members. Both intact
and fractured (peeling) blisters were observed. Blisters were
observed on virtually all members inspected. Some of the blisters
appeared to be fractured as a result of someone physically scraping
the areas, while others appeared to have cracked and fractured on
their own. Blistering ranged in size from concentrations of very
fine blisters (approximately 1/64 to 1/128 of an inch in diameter)
up to single blisters with diameters of approximately 2 to 3
inches. Both irregularly shaped and circular blisters were
observed. The fine concentrations of blisters were located
primarily on beam flanges and in the corner areas where the webs
and flanges meet. Larger shallow blisters were generally located on
the webs of the I-beams. The fine blisters and larger shallow
blisters on the webs were more difficult to see, oftentimes
becoming visible only when viewed at the proper angle with sunlight
hitting the surface after the film of surface dirt and grime was
removed.
Fig. 3: Blisters formed in the polyurethane finish coat on a
flange
Upon scoring around the perimeter of the larger blisters or
areas of concentrated fine blisters with a razor knife, the full
blister area could be removed. Upon removal, a portion of the
zinc-rich primer remained on the steel surface, and a portion
remained attached to the backside of the removed blister (cohesive
break within the zinc primer). Both faces of the split primer films
contained a visible white powder-like residue.Areas that had been
repaired by field touch-up were visible across the structure.
Blisters were still visible in some touch-up areas. It was not
apparent if the blisters had reoccurred in the touch-up area or if
some of the blisters were not completely removed and touch-up
material was applied over them.Coating ThicknessThe results of the
total system thickness measurements from various locations on the
structural steel are summarized below. Minimum DFT (mils): 6.3
Maximum DFT (mils): 15.7 DFT Average (mils): 13.2Destructive film
thickness measurements most often identified two distinct layers of
paint on the steel. In some instances where touch-up repairs had
been made, additional coats were apparent, and three to five
individual layers were evident. When two coats were present, the
first coat was a metallic gray/green and ranged from 4 to 10 mils;
the second coat was dark green and ranged from 3 to 7 mils.
Fig. 4: Blisters in the polyurethane finish coat on a lateral
brace web
AdhesionAdhesion of the coating system in and immediately around
blistered areas was rated poor (0A to 1A rating); however, adhesion
of the coating system in blister-free areas was rated fair to good
(3A to 4A rating). The adhesion test process consistently forced a
break within or at the surface of the zinc-rich primer
layer.Substrate ExaminationThe substrate was examined at
destructive film thickness measurement areas and sample acquisition
areas. Because a thin layer of zinc-rich primer remained adherent
to the steel surface, a thorough inspection of the underlying
substrate was not possible. However, under 50X power illuminated
magnification of the destructive coating thickness gage, a
roughened bright metal substrate was sometimes visible. This
evidence suggests that the steel surface had been abrasive blast
cleaned.Laboratory InvestigationThe laboratory investigation
consisted of visual and microscopic examination, infrared
spectroscopy and scanning electron microscopy-energy dispersive
x-ray spectroscopy (SEM-EDS). The test methods and results are
described below.
Fig. 5: Formation of whitish-colored zinc salts on the surface
of the zinc-rich primer, beneath areas where the blistered finish
coat was removed
Microscopic ExaminationMicroscopic examination of the samples
was conducted using a digital microscope with magnification to
200X. The samples had between two and five coating layers. Coating
layer thickness measurements, obtained by laboratory microscopic
methods, are inTable 1.
TABLE 1Coating Layer Thickness MeasurementsSample #Coating
Layers and Thickness (mils)
Sample 1(Fine Blisters)Two LayersGreenTopMetallic
GrayBottom3.06.93.87.3
Sample 2(Fine Blisters)Two LayersGreenTopMetallic
GrayBottom2.24.42.33.6
Sample 3(Fine Blisters)Two LayersGreenTopMetallic
GrayBottom3.86.05.27.2
Sample 4(Large Blisters)Two LayersGreenTopMetallic
GrayBottom4.98.45.07.9
Sample 5(Non-Failing)Two LayersGreenTopMetallic
GrayBottom6.98.52.63.9
Sample 6(Non-Failing Repair Area)Five LayersGreenTopLight
GreenGreenGreenMetallic
GrayBottom2.04.02.55.54.06.03.05.05.29.9
Sample 7(Non-Failing Repair Area)Four
LayersGreenTopGrayGreenMetallic
GrayBottom4.05.53.54.01.83.53.95.2
Sample 8(Single Blister)Three LayersGreenTopGrayMetallic
GrayBottom2.95.83.16.83.98.0
Infrared SpectroscopyInfrared spectroscopic analysis revealed
the following. The spectrum obtained of the green top-coat was
consistent with a urethane. Water (moisture) and crystalline silica
were also indicated. The spectrum obtained of the gray primer was
most consistent with a zinc urethane. No distinct characteristic
bands are associated with zinc coatings although the baseline noise
appearance was consistent with a zinc coating (confirmed by
elemental analysis).SEM-EDSSEM-EDS analysis revealed that the white
powdery substance on the gray surface of the primer was primarily
zinc. Other elements detected included magnesium, aluminum, and
silicon.ConclusionsThe field investigation and laboratory analysis
identified multiple variables that contributed to the blistering
coating problems on the structural steel.
Fig. 6: Close-up of zinc salt formation on the zinc-rich primer
surface, beneath the removed blistered finish coat
The zinc-rich primer used on the project was a MCU material.
MCUs react with moisture (atmospheric humidity or other moisture
source) to cure. During the curing reaction with moisture, carbon
dioxide gas (CO2) is liberated as a reaction product. The CO2gas
escapes from the coating film in a process commonly referred to as
out-gassing. When a lot of moisture is available, MCUs cure at an
accelerated rate, and CO2formation and out-gassing increase. When
an additional paint layer is applied while the MCU is still
out-gassing, the release of CO2from the MCU can be inhibited. The
gas must now pass out of the MCU and through the newly applied
layer. Depending on the state of drying and curing of the newly
applied layer, some CO2gas may escape, and some may become trapped
in the new film. The CO2that escapes produces pinholes or craters
when the topcoat has begun to gel, while CO2that is trapped creates
sufficient pressure to form bubbles through the cross-section and
at the surface of the new film.Laboratory microscopic examination
of the paint samples consistently revealed that pinholes and
bubbles were present in the green topcoat layer applied over the
MCU primer. This evidence indicates that the MCU zinc-rich primer
was top-coated with the polyurethane before the primer had
sufficiently cured.Infrared spectroscopic analysis of the green
polyurethane finish coat identified bound moisture within the film.
In order for moisture to become bound within this layer, the
moisture would have had to have been present on the MCU zinc-rich
primer layer over which the polyurethane finish was applied. This
evidence indicates that the surface of the MCU zinc-rich primer
where defects occurred (i.e., bubbling, pinholes) was damp when the
polyurethane was applied.Field thickness measurements and
laboratory microscopic measurements revealed that the MCU zinc-rich
primer was often applied above the specified DFT range of 2.5 to
3.5 mils. Destructive thickness measurements and laboratory
microscopic measurements indicated DFTs of up to 7 mils and 9.9
mils respectively. Excessive primer thickness prolongs the dry and
cure time of the primer; as a result, the CO2out-gassing is also
prolonged, serving to increase the likelihood of pinholes and
bubbling.The polyurethane finish coat was also applied above the
specified DFT range of 4.0 to 5.0 mils, with measurements up to 8.7
mils in some instances. These thicker films could slow the escape
of the CO2or trap it, possibly contributing to increased bubble and
pinhole formation.The white powdery residue on the backside of the
detached blister area and on the substrate was identified as zinc
oxide in the laboratory. Zinc oxide (white rust) is produced as the
zinc dust in the primer oxidizes. This finding indicates that the
MCU zinc-rich primer layer was performing as designed: providing
galvanic/sacrificial corrosion protection to the carbon steel
substrate. Moisture (rain, condensing moisture) was gaining access
to the MCU zinc-rich primer through the voids (i.e., pinholes,
fractured bubbles) in the polyurethane finish coat. The moisture
served as the electrolyte, allowing the MCU zinc-rich primer to
oxidize. Moisture condensing on the steel was likely contaminated
with sulfides from the coal-fired power generating station.
Water-soluble salts such as sulfides, in combination with moisture,
increased the corrosivity of the exposure
environment.RecommendationsThe defective areas (i.e., bubbles,
pinholes) were identified and removed by high-pressure water
cleaning. Industry experience has shown that water pressures in
excess of 4,000 psi are usually effective for revealing and
removing defective coatings. However, because each individual
project is unique, some experimentation is needed to arrive at the
optimal cleaning pressure. It was ultimately determined that the
best removal method involved the use of a zero-degree, rotating tip
on the pressure washer gun, with careful observation to maintain
the equipment manufacturers gun-to-work-piece distance and dwell
times. In areas where pressure washing was not entirely effective,
supplemental mechanical cleaning with power tools (i.e., power
sanding) was used. Once the defective coating was completely
removed, any coating that remained was probed with a dull putty
knife as described in SSPC-SP 2 and SSPC-SP 3, Hand Tool and Power
Tool Cleaning, respectively. Remaining coating that passed the dull
putty knife test criteria was considered tightly adherent for
touchup repairs. The periphery of touchup areas was feather-edged
to provide a smooth transition from the repair area to surrounding
intact coatings.Once surface preparation was accomplished, touchup
proceeded using the field touchup system, consisting of a coat of
epoxy mastic followed by a matching green polyurethane finish
coat.The Journal Of Protective Coatings & Linings2013
Technology Publishing Company
5. Measuring Dry Film Coating Thickness According to SSPC-PA
2FromJPCL,April 2013
William D. CorbettPCS, KTA-Tator, Inc.
WilliamD.CorbettPCS, KTA-Tator, Inc.Bill Corbett is the Vice
President and Professional Services Group Manager for KTATator,
Inc., where he has been employed for 33 years. He chairs SSPC
committees C.3.2 on Dry Film Thickness and C.6 (Education). He is
an SSPC-approved instructor for four SSPC courses, and he holds
SSPC certifications as a Protective Coatings Specialist, Protective
Coatings Inspector (Level 3), and Bridge Coatings Inspector (Level
2). He is also a NACE Level 3-certified Coatings Inspector. He was
the co-recipient of the SSPC 1992 Outstanding Publication Award,
co-recipient of the 2001JPCLEditors Award, recipient of SSPCs 2006
Coatings Education Award, and recipient of SSPCs 2011 John D. Keane
Award of Merit.Coating thickness shall be measured in accordance
with SSPC: The Society for Protective Coatings Paint Application
Standard No. 2 (SSPC-PA 2) is a simple enough statement, yet this
common specification requirement is often misinterpreted or
regarded as a document that simply states how to measure the dry
film thickness (DFT) of coatings, something we already profess to
know how to do. Yet the requirements of SSPC-PA 2 regarding gage
calibration, verification of gage accuracy and adjustment
procedures, the number of measurements to obtain, and the tolerance
of the measurements are complex and should be fully understood by
the specification writer before invoking PA 2 in a contract.
iStock
On more than one occasion, I have heard the question, When did
SSPC-PA 2 and dry film thickness measurement become so complicated?
In fact, when you take a close look, measuring DFT isnt that
complex. We have allowed it to become more technologically complex
while making the data easier to analyze. We can gather hundreds of
gage readings in a relatively short time; batch the measurements;
print the data or upload it to a computer for graphing; report the
highest, the lowest, the mean, and standard deviation of the
collected data; incorporate digital images of the structure or
coated area; and even program the gage to produce an audible signal
if a spot measurement is outside of the tolerance range. I am no
doubt leaving out other bells and whistles, but my point is that
while we are able to do a lot with the readings obtained, measuring
DFT involves four or five basic steps. Step 1: Instrument
Calibration Step 2: Verification of Gage Accuracy on Certified
Coated Standards or Certified Shims Step 3: Base Metal Reading
Acquisition or Gage Adjustment (using certified or measured shims)
Step 4: Measurement of Coating Thickness Step 5: Correction for
Base Metal Reading (if acquired).After a brief review of the
history of SSPC-PA 2, this article will describe each of the five
steps, based on the 2012 edition of SSPC-PA 2. Special attention
will be given in the article to how PA 2 addresses the required
number of coating thickness measurements; the acceptability of gage
readings, spot measurements, and area measurements; nonconforming
thickness; measuring DFT on coated edges; and measuring DFT on pipe
exteriors.Some HistorySSPC-PA 2 was originally published as a
temporary standard 40 years ago in 1973 (73T) as Measurement of Dry
Coating Thickness with Magnetic Gages. The standard referenced
gages like the one shown inFig. 1, which are now all but obsolete.
The standard has been updated on multiple occasions. Until 2012,
the most recent technical changes were published in May 2004, with
a minor editorial revision in 2009 to one of the appendices
(regarding measurements on test panels). The SSPC Committee on Dry
Film Thickness Measurement began revising and updating the 2004
version in 2007. The revisions took five years to complete. The
latest edition of the standard (Procedure for Determining
Conformance to Dry Coating Thickness Requirements) is dated May
2012 and was made available to the industry in July 2012.
Fig. 1: One type of magnetic gage referenced in original SSPC-PA
2 for measuring dft Figures courtesy of the author except where
otherwise indicated
In nearly the same timeframe, the 2005 version of ASTM D7091,
Standard Practice for Nondestructive Measurement of Dry Film
Thickness of Nonmagnetic Coatings Applied to Ferrous Metals and
Nonmagnetic, Nonconductive Coatings Applied to NonFerrous Metals
was being revised and updated. It, too, was published in 2012. The
most current version of the ASTM standard focuses on proper gage
use, while SSPC-PA 2 focuses primarily on the frequency of
measurements and the acceptability of the acquired measurements.
References to the frequency of measurements were removed from the
ASTM standard. The two documents are designed to be used together.
It is important to note that both documents address the measurement
of the DFT of coatings on ferrous and non-ferrous metal substrates.
Before 2012, SSPC-PA 2 addressed measurement of coatings only on
steel, a ferrous metal. (The sidebar onp. 32in this article
summarizes the key changes made to PA 2 in 2012.)Summary of Changes
to SSPC-PA 2: 2004 Version and 2012 Version20042012
Measurement of Dry Coating Thickness with Magnetic
GagesProcedure for Determining Conformance to Dry Coating Thickness
Requirements
Addressed measurement of coatings on steel onlyAddresses
measurement of coatings on ferrous and non-ferrous metal
surfaces
No referenced standards sectionASTM D7091 and SSPC Guide 11
included by reference
Definitions section included Calibration; Verification;
Adjustment; Coating Thickness Standard; Shim (foil); Dry Film
Thickness Reference Standard; Accuracy; Structure.Definitions
section includes Gage Reading, Spot Measurement and Area
Measurement only. All definitions related to gage calibration,
accuracy and adjustment are incorporated by reference in ASTM
D7091. Spot measurement definition was expanded.
No. of Area Measurements based on the size of the structureNo.
of Area Measurements based on the size of the area of coated
surface
Isolation of nonconforming areas required measurement of each
100 square foot area painted during the work shift.The magnitude of
nonconforming thickness assessed by obtaining spot measurements in
eight equally spaced directions radiating outward from the
nonconforming area
Recommended specifying minimum & maximum thickness range; if
no range was specified, thickness value was considered minimum
(with no maximum)If a single value is specified and the coating
manufacturer does not recommend a range, the minimum and maximum
thickness range is established at 20% of the stated value
Contained a minimum gage accuracy requirement to qualify for
useNo minimum gage accuracy requirement to qualify for use
Conformance to Specified Thickness:Gage Readings:
UnrestrictedSpot Measurements: 20% of specified rangeArea
Measurements: Per SpecificationConformance to Specified
Thickness:Five different Coating Thickness Restriction Levels
established. If no Restriction Level is specified, default is based
on 2004 conformance requirement.
Notes section contained principles of gage operation and various
substrate/surface conditions that may affect measurements;
overcoating; and correcting for low/high thickness.Notes section
includes Overcoating and Correcting for Low/High Thickness only
ASTM D7091 describes principles of gage operation and various
substrate/surface conditions that may affect measurements.
Contained 6 appendicesContains 8 appendices. Six appendices from
2004 version included. Added two: Method for Measuring DFT of
Coating on Edges Method for Measuring DFT of Coated Steel Pipe
Exterior
Gage TypesSSPC-PA 2 addresses two types of DFT gages, both of
which are supplied by a variety of manufacturers. Magnetic pull-off
gages are categorized as Type 1 (Fig. 2).
Fig. 2: Example of magnetic DFT gage categorized as Type 1 in
SSPC-PA 2
These gages were designed in the 1950s. While their use has
declined, they are still readily available and used by some. For
these gages, a permanent magnet is brought into direct contact with
the coated surface. The force necessary to pull the magnet from the
surface is measured and converted to coating thickness, which is
displayed on a scale on the gage. The operating principle is
simple. Less force is required to remove the magnet from a thick
coating, while more force is required to remove the magnet from a
thinner one. The scale is not linear, as will be discussed
below.Electronic gages are categorized as Type 2 (Fig. 3). These
gages use electronic circuitry to convert a reference signal into
coating thickness and are more popular than Type 1 gages. They are
typically regarded to be faster, more accurate, and easier to
use.
Fig. 3: Example of electronic DFT gage categorized as Type 2 in
SSPC-PA 2
Gage Calibration, Accuracy Verification, and AdjustmentTo help
assure the reliability of the coating thickness measurements, ASTM
D 7091 describes three operational steps that must be performed
before taking the measurements. These steps are (1) gage
calibration, (2) verification of gage accuracy and (3) gage
adjustment. The steps are incorporated by reference in SSPCPA 2 and
are completed before obtaining coating thickness measurements to
determine conformance to a specified coating thickness range. The
steps to verify the accuracy of the gage are based on the principle
thatyou check the gage by measuring a known thickness before you
use the same gage to measure an unknown thickness.Verification of
gage accuracy is typically performed using certified coated
thickness standards (for Type 1 or Type 2 gages) or certified shims
(Type 2 gages). Adjustment of Type 2 gages to compensate for
substrate characteristics (described later) is typically performed
using certified shims. Measured shims (individually labeled with a
stated thickness value) commonly supplied with Type 2 gages can
also be used for gage adjustment.Dry film thickness gages are
calibrated by the equipment manufacturer, its authorized agent, or
an accredited calibration laboratory (under controlled conditions).
A test certificate or other documentation showing traceability to a
national metrology institution is required. While there is no
standard time interval for re-calibration, an interval can be
established based on experience, the work environment, and/or the
internal equipment calibration procedures of the company using the
gage. A one-year calibration interval is a typical starting point
suggested by gage manufacturers.Verifying Gage AccuracyTo guard
against measuring with an inaccurate gage, SSPC-PA 2 requires that
gage accuracy be verified (at a minimum) at the beginning and end
of each work shift according to the procedures described in ASTM D
7091. If a large number of measurements are being obtained, the
user may opt to verify gage accuracy during measurement acquisition
(for example, hourly). If the gage is dropped or suspected of
giving erroneous readings during the work shift, its accuracy
should be rechecked.Verifying the Accuracy of Type 1 GagesThe
accuracy of Type 1 (magnetic pull-off) gages is verified by placing
the gage probe onto a certified coated thickness standard
(Figs.4and5). A one-point or two-point accuracy verification
procedure can be performed; typically, the two-point verification
provides greater accuracy. If a one-point verification procedure is
adopted, the coated standard should be selected based on the
intended range of use. For example, if the intended use is between
4 and 6 mils, then a five-mil coated standard is appropriate. Using
the same example, if a two-point verification procedure is adopted,
then a two-mil and an eight-mil set of coated standards (slightly
below and above the intended range of use) is appropriate.
Fig. 4: (top) and 5 (bottom) Verifying the accuracy of Type 1
gages using certified coated thickness standards
The final step in the process is to obtain a set of base metal
readings (BMRs) to compensate for substrate characteristics
including (but not limited to) substrate metallurgy, geometry,
thickness/thinness, and roughness (Fig. 6). These readings
represent the effect of the substrate conditions on the coating
thickness measurement device. SSPC-PA 2 states that a minimum of 10
(arbitrarily spaced) locations should be measured (one reading per
location) and then averaged. This average BMR is then deducted from
subsequent coating thickness measurements to remove any effect of
the base metal surface and its conditions.
Fig. 6: Obtaining base metal readings with Type 1 gage
Because Type 1 gages cannot be adjusted, some gage operators
believed that a correction value could be applied to the coating
thickness readings to compensate for the inaccuracy of the gage.
For example, if a gage reading was 5.7 mils on a five-mil coated
standard, a 0.7-mil correction value could be deducted (by the gage
operator) from subsequent coating thickness measurements. However,
because Type 1 gages are non-linear, one cannot assume a linear
(mil-for-mil) correction value across the full range of the gage.
While the gage may be out of tolerance by 0.7 mils at 5 mils, it
may be out of tolerance by more or less than 0.7 mils at a
different thickness. Accordingly, SSPC-PA 2 states that the
practice of using a linear correction value is not
appropriate.However, Note 6 in the standard states,A correction
curve can be prepared by plotting the actual gage readings against
the stated values on the (coated) test blocks (standards).
Subsequent coating thickness measurements can be corrected by
plotting the measurements along the correction curve. The
correction curve may or may not cover the full range of the gage,
but should cover the intended range of use. The Base Metal Readings
(BMR) described in 6.1 may also need to be plotted on the
correction curve.This requirement makes Type 1 gages very difficult
to use. While some gage operators may simply subtract a fixed
amount (for example, 0.5 mils) from any reading, such a practice is
not in compliance with SSPC-PA 2.Verifying the Accuracy of Type 2
GagesThe accuracy of Type 2 (electronic) gages can be verified by
placing the gage probe onto a certified coated thickness standard
(described for Type 1 gages) or certified shims (Figs.7and8). The
certified shim should be placed onto a smooth, uncoated metal
surface to remove any effect of the surface roughness during this
process. A one-point or two-point accuracy verification procedure
can be performed (as described earlier for Type 1 gages).
Fig. 7: Verifying accuracy of Type 2 gage on a certified coated
standard
Fig. 8: Verifying accuracy of Type 2 gage using a certified
shim
Adjusting Type 2 GagesThe final step in the process is to adjust
the gage on the surface to which the coating will be applied.
Adjustment is accomplished by placing a certified or measured shim
(or shims) onto the prepared, uncoated metal surface and adjusting
the gage (when feasible) to compensate for substrate
characteristics including (but not limited to) substrate
metallurgy, geometry, thickness/thinness, and roughness (Fig. 9).
The gage reading is adjusted to match the thickness of the shim,
which effectively removes any influence from the underlying
surface.
Fig. 9: Adjusting Type 2 gage using a measured shim on the
surface to which the coating will be applied
This step sounds reasonably straightforward but poses several
hidden challenges. First, once the surface is coated (for example,
with a primer), an uncoated surface may no longer be available for
subsequent gage adjustments, so the user may want to have a similar
uncoated surface prepared and reserved for future gage adjustments
on a given project. Naturally, this surface must be representative
of the metallurgy, geometry, thickness/thinness, and roughness of
the actual surface, which can be a challenging requirement.Second,
some Type 2 gages cannot be adjusted. In such cases, the user will
need to obtain BMRs from the prepared, uncoated substrate
(described earlier for Type 1 gages). While many Type 2
(electronic) gages have a zero-set function, the gages should never
be adjusted to zero unless the surface is smooth.Required Number of
Coating Thickness MeasurementsThe section of SSPC-PA 2, Required
Number of Measurements for Conformance to a Thickness
Specification, causes many users confusion, which can result in
either under- or over-inspection. Arguably the most critical
section in the document, Section 8, describes how many areas to
check, the size of the areas, the number of measurements to obtain
in each area, and the steps to take if spot or area measurements do
not conform to the specification.SSPC-PA 2 contains three
definitions that are critical to understanding this next area of
discussion. Gage Reading: A single instrument reading. Spot
Measurement: The average of at least three gage readings made
within a 4-cm (1.5-inch) diameter circle. Acquisition of more than
three gage readings within a spot is permitted. Any unusually high
or low gage readings that are not repeated consistently are
discarded. The average of the acceptable gage readings is the spot
measurement. Area Measurement: The average of five spot
measurements obtained over each 10 m2(100 ft2) of coated surface,
or increment (portion) thereof.An area is defined as approximately
100 square feet. Within each area, five randomly spaced spots are
selected. Each spot consists of a 1.5-inch diameter circle. A
minimum of three gage readings is obtained in each spot,
culminating in a minimum of 15 gage readings within an area.
Unusually high or low gage readings that cannot be repeated
consistently are discarded. The average of the three acceptable
gage readings is the spot measurement; the average of five spot
measurements is the area measurement.Figure 10, from Appendix 1 in
SSPC-PA 2, depicts an approximate 100-square-foot area containing
gage readings and spot measurements.
Fig. 10: Approximate 100-square-foot area containing gage
readings and spot measurements, as depicted in Appendix 1 of
SSPC-PA 2. Courtesy of SSPC
The number of areas that must be measured for coating thickness
varies, depending on the size of the coated area. There are three
categories of coated area: less than 300 square feet; 300 to 1,000
square feet; and greater than 1,000 square feet. For areas
containing less than 300 square feet of coated surface, every
100-square-foot area must be measured for coating thickness. For
areas of coating 300 to 1,000 square feet, three random areas are
selected and measured. For areas of coating exceeding 1,000 square
feet, three random areas are selected from the first 1,000 square
feet, along with one additional area for each additional 1,000
square feet.Because areas of coating often exceed 1,000 square
feet, our example will be based on this third tier (>1,000
square feet). Lets assume that the total coated area (perhaps the
area coated during a work shift, although SSPC-PA 2 does not equate
coated area with work shift) is 12,500 square feet. A total of 15
areas must be measured (three in the first 1,000 square feet and
one additional area in each of the 12 remaining 1,000-square-foot
areas or portions thereof). This culminates in a total of 75 spot
measurements (15 x 5) and a minimum of 225 gage readings (15 x 5 x
3). If spot measurement variances result in area measurements that
do not meet the specification, then additional spot measurements
are acquired (radiating outward in eight directions from the
nonconforming area) to determine the magnitude of the
non-conforming thickness. This process is described later in this
article.Acceptability of Gage Readings, Spot Measurements, and Area
MeasurementsWhile individual gage readings that are unusually high
or low (and cannot be repeated consistently) can be discarded,
there are limitations on the thickness values representing the spot
measurements (the average of three gage readings). A minimum
thickness and a maximum thickness are normally specified for each
layer of coating. However, if a single thickness value is specified
and the coating manufacturer does not provide a recommended range
of thickness, then the minimum thickness and maximum thickness for
each coating layer are established by SSPC-PA 2 at 20% of the
stated value. For example, if the specification requires 3 mils DFT
and the coating manufacturer does not provide any additional
information regarding a recommended thickness range, then, by
default, the specified range is established as 2.43.6 mils. Because
the coating may not perform at the lower thickness, it is important
for the specifier to indicate an acceptable range for each coating
layer. To assist the specifier, the 2012 edition of SSPC-PA 2
incorporates a Restriction Level Table (Fig. 11). The Table enables
the specifier to select from five different restriction levels
related to spot and area measurements.
Fig. 11: Coating Thickness Restriction Levels (as shown inTable
1of SSPC-PA 2, Section 9)Courtesy of SSPC
Level 1 is the most restrictive and does not allow for any
deviation of spot or area measurements from the specified minimum
and maximum thickness, while Level 5 is the least restrictive.
Depending on the coating type and the prevailing service
environment, the specifier can select the DFT restriction level for
a given project. The specifier may also invoke a maximum thickness
threshold for Level 5 Spot or Area Measurements for a generic
product type and/or service environment that will not tolerate an
unlimited thickness. If no Restriction Level is specified, then the
default is Level 3, which is based on the 2004 version of SSPC-PA 2
(what many users of the standard have become accustomed to).For the
purpose of final acceptance of the total DFT, the cumulative
thickness of all coating layers in each area must be no less than
the cumulative minimum specified thickness and no greater than the
cumulative maximum specified thickness.For example, assume that the
specification requires a four- to six-mil application of primer.
The actual minimum and maximum spot and area thickness requirements
are shown inFig. 12for each of the five restriction levels.
Fig. 12: Coating Thickness Restriction Levels Based on a
Four-to-Six-Mil RequirementDerived using the 2012 edition of
SSPC-PA 2, Table 1, Coating Thickness Restriction Levels
Determining the Magnitude of Nonconforming ThicknessAnother
change in the 2012 version of the standard is the procedure for
identifying nonconforming areas (Fig. 13). In the 2004 edition, if
spot or area measurements were out of conformance, each
100-square-foot area coated during the work shift had to be
measured, and nonconforming areas had to be demarcated. On a larger
structure with multiple applicators, the measurement and
documentation process could be extensive, so the approach was
changed in the 2012 revision. If a nonconforming area is
identified, spot measurements are made at five-foot intervals in
eight equally spaced directions radiating outward from the
nonconforming area, as shown inFig. 13.
Fig. 13: Depiction of procedure for identifying nonconforming
areas, as described in the 2012 edition of SSPC-PA 2.Courtesy of
SSPC
If there is no place to measure in a given direction, then no
measurement in that direction is necessary. Spot measurements are
obtained in each direction (up to the maximum surface area coated
during the work shift) until two consecutive conforming spot
measurements are acquired in that direction, or until no additional
measurements can be made. Acceptable spot measurements are defined
by the minimum and maximum values in the contract documents. No
allowance is made for variant spot measurements (for example, 20%),
which is consistent with the practice followed when determining the
area DFT.On complex structures or in other cases where making spot
measurements at five-foot intervals is not practical, spot
measurements are taken on repeating structural units or elements of
structural units. This method is used when the largest dimension of
the unit is less than 10 feet. Spot measurements are obtained on
repeating structural units or elements of structural units until
two consecutive units in each direction are conforming or until
there are no more units to test.Non-compliant areas are demarcated
using removable chalk (or another specified marking material) and
documented. All of the area within five feet of any non-compliant
spot measurement is considered non-compliant. For a given
measurement direction or unit measurement, any compliant area or
unit preceding a non-compliant area or unit is designated as
suspect, and, as such, is subject to re-inspection after corrective
measures are taken.Appendices to the StandardThere are eight
appendices in the 2012 version of SSPC-PA 2. Two of the eight
appendices were added in 2012 (the remaining were in the 2004
edition) and are highlighted below. The appendices to SSPC-PA 2 are
not mandatory but may be invoked by contract documents.Appendix 6:
Method for Measuring the Dry Film Thickness of Coatings on EdgesFor
decades, the industry was cautioned about taking coating thickness
measurements within one inch of an edge, let alone on an edge.
However, several Type 2 (electronic) gage manufacturers offer a
variety of probe configurations, some of which are less affected by
proximity to edges and are designed to better measure the thickness
of coatings on edges (Fig. 14). Obviously, the gage operator should
consult the gage manufacturers instructions before measuring
coating thickness on edges.
Fig. 14: One of a variety of Type 2 gage probe configurations
designed to better measure DFT of coatings on edges
Before measuring coating thickness on edges, the user should
verify the gage and probe for accuracy by placing a thin, flexible
shim (certified or measured) onto the prepared, uncoated edge.
Adjustments to the gage may or may not be required. This procedure
also verifies that the probe configuration will accommodate the
edge configuration before acquiring coating thickness data.Once
verification of accuracy and adjustments are made, a minimum of
three gage readings are taken within 1.5 linear inches of coated
edge. The average of the gage readings is considered a spot
measurement. The number of spot measurements along the edge will
vary, depending on the total length of the coated edge.Appendix 7:
Method for Measuring Dry Film Thickness on Coated Steel Pipe
ExteriorAppendix 7 was added to accommodate pipe coaters that need
to determine coating thickness conformance on non-flat (or
non-plate) areas, including smaller pipe sections on a cart or rack
and longer pipe spools.Pipe sections loaded onto a cart or rack can
be considered a complete unit (Fig. 15). The total number of spot
and area measurements is based on the total square footage of pipe
on the cart or rack. The square footage is calculated as shown onp.
35.
Fig. 15: (top and bottom): Appendix 7 of the 2012 edition of
SSPC-PA 2 describes a method for measuring DFT on non-flat steel,
such as pipe sections that can be loaded on racks or carts. Photos
courtesy of Turner Industries Group, L.L.C.
Some carts may have several small pipe sections, and the total
coated surface may exceed 100 square feet. In this case, a Pipe DFT
Frequency Factor shown below may be invoked. Pipe DFT Frequency
Factor 2 = (length of each pipe x circumference) x number of pipe
sections on cart or rack = (number of spot measurements) x 2 Pipe
DFT Frequency Factor 3 = (length of each pipe x circumference) x
number of pipe sections on cart or rack = (number of spot
measurements) x 3 Pipe DFT Frequency Factor 4 = (length of each
pipe x circumference) x number of pipe sections on cart or rack =
(number of spot measurements) x 4 Pipe DFT Frequency Factor 5 =
(length of each pipe x circumference) x number of pipe sections on
cart or rack = (number of spot measurements) x 5 Pipe DFT Frequency
Factor 6 = (length of each pipe x circumference) x number of pipe
sections on cart or rack = (number of spot measurements) x 6Based
on the example above, if Pipe DFT Frequency Factor 4 was invoked,
20 spot measurements would be taken (5 spots x Frequency Factor
4)Pipe spools that are not loaded onto a rack or cart are typically
measured individually (Fig. 16). The number and locations of spot
measurements are based on Appendix 7s Table A7 (Fig. 17). Three
sets of four circumferential spot measurements should be obtained
on pipe spools less than 10 feet in length.
Fig. 16: DFT of pipe spools not loaded on cart or rack are
typically measured individually.
Fig. 17: Number and Locations of Spot MeasurementsPipe Spools
(Table A7 from 2012 edition of SSPC-PA 2, Appendix 7)Courtesy of
SSPC
ConclusionSSPC-PA 2 has undergone significant changes in an
attempt to make it more complete; more in concert with ASTM D7091;
easier to use in the shop and field; and more flexible in providing
the specifier with options for coating thickness restrictions based
on the type of structure, the coatings to be applied, and the
service environment. SSPC-PA 2 and ASTM D7091 are both undergoing
additional technical and editorial changes to bring them into even
greater alignment with one another.Get the Latest Standards on Dry
Film Thickness of CoatingsThe 2012 edition of SSPC-PA 2, Procedure
for Determining Conformance to Dry Coating Thickness Requirements,
is available from the SSPC: The Society for Protective Coatings
throughsspc.org, under the Standards tab at the top of the home
page.The 2012 edition of ASTM D7091, Standard Practice for
Nondestructive Measurement of Dry Film Thickness of Nonmagnetic
Coatings Applied to Ferrous Metals and Nonmagnetic, Nonconductive
Coatings Applied to NonFerrous Metals, is available from ASTM
International throughastm.orgunder the Standards at the top of the
navigation bar on the site.
Change is never easy. Communicating the new requirements of this
standard to the industry is challenging but essential. One conduit
is through training and education. For example, SSPC offers a short
course, Using SSPC-PA 2 Effectively, that was recently updated to
reflect changes made to the standard. Free webinars are available
through SSPC/JPCLfor those who cannot participate in instructor-led
training. Updates to SSPC and other industry-provided inspector
training and certification courses (and the associated instructor
education) will be critical to fully understanding and effectively
communicating the requirements of this highly regarded industry
standard.The Journal Of Protective Coatings & Linings2013
Technology Publishing Comp
6.Testing the cure of IOZFromJPCL,April 2013Is the MEK rub test
a conclusive test to check the cure of inorganic zinc coatings?From
Rob Francis ConsultantThe solvent rub test for cure of ethyl
silicate inorganic zinc (IOZ) coatings, as described in ASTM D4752,
is the accepted test for checking if such a coating is cured. This
is especially critical before overcoating, but even a single coat
system can dry without curing if applied under low humidity
conditions. Other tests that are used include scraping with the
edge of a coin (the Quarter test in the USA) or simply a
fingernail. If significant zinc powder is produced with either
test, the coating is considered uncured. However, these tests are
considered more subjective than the solvent rub test.An uncured
coating will result in considerable zinc removal with a few double
rubs, while a cured coating will be little affected with the 50
double rubs required. It should be noted that even a fully cured
coating will show some zinc discoloration on the white rag. While
as with any test, it may be possible to get ambiguous results, this
test, in most situations, will be definitive, and certainly
superior to any other simple field test. If there is any doubt
regarding the cure, leave the coating (if the humidity is high
enough) or water mist it and retest.From Simon Hope BIS Salamis
(M&I)The MEK rub test is only subjective as proof of cure for
IOZ coatings. The test is only valid for the actual area tested and
can be applied to the whole item only by extrapolation of the
result.Confidence in the result can vary wildly, depending on time,
humidity, and temperature, because the curing mechanism of IOZ is
totally dependent on the integration of water into the silicate
precursor to create the matrix to support the metallic zinc. Hence,
the best advice is that once touch dry, fresh water washing
enhances the cure mechanism. High humidity and water washing will
give confidence, and MEK rub then gives verification.From Gary Hall
ConsultantI am answering the question about the MEK test as it
relates to testing a coating in the field.The ASTM test method that
pertains to measuring cure of inorganic zinc coatings/primers by
solvent resistance is ASTM D4752, Standard Practice for Measuring
MEK Resistance of Ethyl Silicate (Inorganic) Zinc-Rich Primers by
Solvent Rub. This method has been shown to correlate well with the
results obtained with an analytical chemical test called diffuse
reflectance infrared spectroscopy, by which the degree of cure of
inorganic zinc (IOZ) primers can be accurately determined. It
should be noted that the results of D4752 might not indicate when
full cure has been achieved because the coating may become
resistant to MEK before full cure occurs.A hardness test may also
be used if the coating manufacturer can provide the appropriate
hardness data. One such test is ASTM D2240, which uses a Shore
durometer. The Shore durometer test gives the amount of indentation
made by a specific needle. Because the ASTM D2240 test will deform
the coating surface under the needle, this type of test is best
performed on a companion piece of substrate coated and cured in the
same manner as the coating on the substrate on the actual
project.Problem Solving Forum questions and answers are published
inJPCLand its sister daily electronic publication,PaintSquare News.
Upcoming questions inJPCLinclude the following. Whose
responsibility is safety on a bridge coating site? If an inorganic
zinc (IOZ) coating has not fully cured because of low humidity, can
water be sprayed onto the IOZ-coated surface to accelerate the
cure? What action should be taken if an inorganic zinc coating
fails the MEK (methyl ethyl ketone) rub test? What causes amine
blush in epoxy topcoats? Do water treatment processes to stop the
transfer of invasive marine species in ballast water affect the
performance of ballast tank coatings? How soon does metallizing
need to be sealed after it is applied to concrete on
bridges?Responses toJPCLquestions can be submitted
[email protected]. Readers may also propose
questions.Readers can also respond to PSF questions posted
onPaintSquare Newsand can propose questions onPaintSquare
Newsatpaintsquare.com/psf.
7. Basics of corrosion of steel for applicatorsFromJPCL,January
2013JoePikasA good coating job requires the right steps to be
performed to achieve the protection needed. It is important to know
why things are done, as well as how the various steps are
performed. A primary reason for using protective coatings is
corrosion protection.For the purposes of this series, corrosion of
steel is defined as the destruction of steel by an electrochemical
process that is characterized or recognized by the formation of
rust or pits. To understand how protective coatings protect a steel
surface, the nature of corrosion must be understood why it occurs
and how it can be prevented.Steel is manufactured by taking the
mined ore and adding a large amount of energy to it in the blast
furnace. This produces an unstable metal. Nature does not like all
that energy stored in the steel. So upon exposure to the
atmosphere, especially moisture and oxygen, this energy is
released, and the iron returns to its natural stable stateiron ore.
Rust, therefore, is nothing more than a pure form of iron ore
(oxides). Protection of steel from corrosion involves methods to
retard this natural release of energy (Fig. 1).
Fig. 1: One approach to slowing the natural corrosion of steel
and appearance of rust is the use of protective coatings.Courtesy
ofJPCL
To understand how coatings protect steel, we must understand the
four conditions required for corrosion to occur. Unless all four of
these conditions are present, corrosion will not occur. These four
conditions are: a positive pole (a cathode), a negative pole (an
anode), an electrical conductor, and an electrolyte.The terms anode
and cathode have technical definitions in electrochemistry, but for
our purposes, we will use them to refer to areas on a substrate or
materials of different electrical potentials. The electrical
conductor is a means of conducting electricity, similar to the
copper wiring in your house. An electrolyte is a liquid solution
(usually water) that also can conduct electricity.To help
illustrate these terms and how corrosion happens, lets look at the
dry cell (battery). A battery represents a beneficial use of
corrosion, though the process is the same as corrosion that occurs
with steel.A battery has two terminals. Typically, one is connected
to a carbon rod running down the center of the battery, while the
other is connected to the outer casing, which is made of zinc.
These are the two dissimilar materials of different electrical
potential, which serve as cathode and anode. If you have ever taken
a battery apart, you would have seen that there is a pasty material
between the casing and the carbon rod. This substance is the
electrolyte.If you wanted to use the battery, you would connect the
wires to something such as a flashlight. The wires are the
electrical conductor. Once the wires are connected, the flashlight
will keep on glowing until the battery has become corroded. In the
battery, it is actually the zinc casing that is consumed and
corroded.How does this example of a battery relate to corrosion of
steel? You would see that steel is not a smooth, uniform material
if you looked at it under high magnification (Fig. 2). It actually
consists of very small grains or grain boundaries. This means that
steel has spots on it with slightly different electrical
potentials.
Fig. 2: Peaks and valleys of profiled clean steelCourtesy of
KTA-Tator
Adding stresses to the steel also creates areas of different
electrical potential. This normally occurs by such processes as
differential heating of the steel during treatment, bending or
cutting the steel, or even hitting it with a hammer. Any of these
processes adds small amounts of energy to the steel. So by its very
nature, steel contains spots of different electrical potential, or
anodes and cathodes.What about the electrical conductor? This was
the wire in the example of the battery.Does steel conduct
electrical current? It certainly does, so wires are not needed. As
you can see from this explanation, steel contains anodes and
cathodes, and is an electrical conductor. It already contains three
of the four conditions necessary for corrosion.The fourth condition
needed is the electrolyte or liquid that can conduct electricity.
Where does it come from? Normally, atmospheric moisture that
condenses on the surface serves as an electrolyte. It can be in the
form of rain, dew, or simply humidity in the air. Some structures
either are used to hold water or are used in water. They are
constantly exposed to an electrolyte.Our atmosphere is laden with
moisture at all times, even in desert areas, although to a esser
degree. Most steel surfaces are exposed to dew at night and water
vapor during the daytime hours in addition to the normal rainfall.
In highly industrial areas such as Houston, Los Angeles, and New
York, airborne chemical contaminants contain substances called
ions.The point to be made is that steel (like most other
manufactured metals) contains three of the four conditions needed
for corrosion. The most common way to slow down corrosion is to
isolate the steel from the electrolyte. Therefore, the major
function of the coating is to keep moisture off the steel. Some
coatings, such as zinc primers, perform other functions, which will
be discussed in subsequent Bulletins.There are a few common forms
of corrosion that a painter will see regularly: general corrosion,
galvanic corrosion, pitting corrosion, and crevice corrosion.
General corrosion takes place fairly evenly over the metal.
Usually, it begins as spots or freckles and becomes progressively
worse.Galvanic corrosion occurs when dissimilar metals are in
contact. The more active metal (the anode) corrodes to protect the
less active metal (the cathode). For example, if a brass valve were
connected to a steel pipe, the steel would corrode to protect the
brass. The steel at the fitting would be consumed rather quickly,
first appearing as a thinning of the metal and ultimately resulting
in penetration. Another example of galvanic corrosion is mill scale
on steel. Steel is more active than mill scale, so when corrosion
conditions are present, the steel will corrode to protect the mill
scale.Pitting corrosion occurs when the corrosion forces are
concentrated in a small area. Metal loss is into the steel rather
than over the surface. The rust pits that form have serious
consequences because the pits represent metal section loss. This
can result in perforation if the structure is a tank or a vessel,
and loss of structural integrity no matter what the structure
is.Crevice corrosion (Fig. 3) is another common form seen on
structures, and occurs when there is a small space between
structural elements, be they metal-to-metal or metal-to-non-metal.
Examples of places where crevice corrosion can occur are
back-to-back angles, where steel plates overlap, around rivets and
bolts, near tack welds, and any other place where a small opening
is present. What happens is that moisture gets into the crack and
completes the corrosion circuit. The moisture gets trapped in the
crevice and accelerates the corrosion compared to the surrounding
area. The corrosion reactions are greatest at the bottom of the
crevice, so metal loss is concentrated in that area.
Fig. 3: Severe corrosion is seen at a weld seam.Courtesy of
KTA-Tator
There are many other forms of corrosion that a painter may see,
including microbiologically influenced corrosion (deep isolated
pitting as shown inFig. 4), cavitation corrosion, or erosion
corrosion, to name a few. In most of these cases, the corrosion
reaction is accelerated by another factor beyond the general
corrosion reaction explained above.
Fig. 4: Microbiologically influenced corrosion on piece of steel
pipeCourtesy of the author
To stop corrosion, all that is needed is to eliminate one of the
factors that produce the reaction. It is impossible to eliminate
the environment, and it is cost-prohibitive to make steels that
corrode at a slow rate. Therefore, coatings are often used to
prevent corrosion by eliminating contact between the environment
(electrolyte) and the steel substrate. Coatings are therefore a
barrier material.A coating may also be applied to enhance
appearance. However, to protect against corrosion for a period of
time, it is necessary for coatings to possess features that make
them effective barriers.By isolating steel from the electrolyte, a
good protective coating can prevent corrosion for extended periods
of time. The better the application, the longer the coating will
serve its useful purpose.On the other hand, a poor coating job may
lead to the expense of premature failure, which requires reblasting
and recoating. On large projects, such premature failure can cost
hundreds of thousands of dollars or more. Good coating work can
also save steel structures from unnecessary and costly
deterioration. It is estimated that the cost of corrosion in the
U.S. each year runs in the billions of dollars. Your work as an
applicator can help significantly to reduce these losses.Upcoming
Applicator Training BulletinsThe following are among the upcoming
Applicator Training Bulletins.Basics of Corrosion and Coatings How
Coatings Protect Steel Basics of Concrete Deterioration for
Applicators How Coatings Protect ConcreteSurface Preparation Why
Surface Preparation Is Important Introduction to Surface
Preparation of Concrete Mechanical Methods of Preparing Concrete
Chemical Cleaning of Concrete Power Tool Cleaning for Steel Using
Paint Strippers on Steel and Concrete Setting Up Air Abrasive
Blasting Systems Techniques of Air Abrasive Blasting Containing
Dust and Debris during Air Abrasive Blasting Setting Up and
Operating Wet Blasting Equipment Using High Pressure
WaterjettingApplication Product and Application Data Sheets Mixing
and Thinning Paint Basic Training in Brush and Roller Application
The Basics of Conventional Air Spraying Using High Volume, Low
Pressure Spray Equipment Using Airless Spray Equipment Introduction
to Plural Component Spraying Applying Two-Pack Epoxies and
Polyurethanes Applying Zinc-Rich Coatings Applying Water-Borne
Coatings Applying High Solids Coatings Applying Polyureas Applying
Floor Coatings and Toppings Special Concerns about Applying
Coatings in the ShopQuality Control The Effects of Weather on
Cleaning and Coating Work Conforming with Job Requirements Records
of Work and Working Conditions Why Good Housekeeping Is Important
Assessing Surface Cleanliness and Profile Assuring Quality during
Abrasive Blasting Assessing Quality of Wet Methods of Surface
Preparation Computing Film Thickness and Coverage Measuring Dry
Film Thickness Measuring Adhesion of Coatings Recognizing and
Correcting Paint Application Deficiencies Quality Control in
Coating ConcreteSafety and Health Safety Considerations for
Abrasive Blasting Anticipating Job Hazards Respiratory Protection:
Hazards and Equipment Fit Testing Procedures for Respirators Job
Hazards during Climbing and High Work An Introduction to Confined
Spaces Identifying and Controlling Job Hazards When Working around
High Voltage Using Lighting Safely Protection against Worker
Exposures during Hazardous Paint Removal Safety for Applicators
Working near Process Equipment Safety with Solvents and Paint
Strippers Where To Get Help in Safety and HealthEditors Note:This
article marks the return of JPCLs Applicator Training Bulletin, a
series first published from 1988 to 1992. The series was intended
to help industrial contractor firms train blasters and painters.
The original series was developed and written by the Coatings
Society of the Houston Area in collaboration with SSPC, and Lloyd
Smith edited it. The series was subsequently collected in one
volume. From 1992 to 1993, a separate series on safety was
developed and written under the direction of KTA-Tator. In 1997,
the series was updated and expanded. Beginning this month, the
series will again cover the basics of corrosion, surface
preparation, application, quality control, and safety. The series
will be updated and expanded where necessary. Because the basic
theory of corrosion, how coatings protect steel, and the importance
of surface preparation have changed little since 1988, some of the
original articles will also appear with minor revisions, including
this first one on what applicators need to know about corrosion.
Written by Joe Pikas, with Transco Corporation at the time, the
article was first published in the July 1988 JPCL, then updated and
re-published in the April 1997JPCL.
8. Can in-process QC prevent premature coating
failures?FromJPCL,January 2013
William D. Corbett, PCS, KTA-Tator, Inc.
PCS, KTA-Tator, Inc.Bill Corbett is the Professional Services
Business Unit Manager for KTA-Tator, Inc., where he has been
employed for 31 years. He is an SSPC-approved instructor for three
SSPC courses, and he holds SSPC certifications as a Protective
Coatings Specialist, Protective Coatings Inspector, and Bridge
Coatings Inspector. He is also a NACE Level 3-certified Coatings
Inspector. He was the co-recipient of the SSPC 1992 Outstanding
Publication Award, co-recipient of the 2001JPCLEditors Award,
recipient of SSPCs 2006 Coatings Education Award, and recipient of
SSPCs 2011 John D. Keane Award of Merit.RichardBurgessSeries
Editor, KTA-Tator, Inc.For decades we have heard that the incidence
of premature coating failure would decline by explicitly requiring
the contractor to control the quality of workmanship (via contract
document language) using properly trained (and equipped) quality
control personnel. In this Case from the F-Files, well take a brief
look at five case history failures and assess whether quality
control inspection of the work as it proceeded could have prevented
the failure from occurring, or whether it would have happened
despite the efforts of knowledgeable quality control
personnel.Defining Quality ControlThe ISO definition states that
quality control is the operational techniques and activities that
are used to fulfill requirements for quality.1 This definition
could imply that any activity, whether serving the improvement,
control, management, or assurance of quality could be a quality
control activity. Quality control is a process for maintaining
standards and not for creating them. Standards are maintained
through a process of selection, measurement, and correction of
work, so only the products or services that emerge from the process
meet the standards. In simple terms, quality control prevents
undesirable changes in the quality of the product or service
supplied. The simplest form of quality control is illustrated
inFig. 1. Quality control can be applied to particular products; to
processes that create the products; or to the output of the whole
organization, by measuring its overall quality performance.
Fig. 1: The generic control processFigure and photos courtesy of
KTA-Tator
Quality control is often regarded as a post-event activity, that
is, a means of detecting whether quality has been achieved and
taking action to correct any deficiencies. However, one can control
results by installing sensors (e.g., inspection check points)
before, during, or after the results are created. It all depends on
where you install the sensor, what you measure, and the
consequences of failure.1The Joint Certification Standard for Shop
Application of Complex Protective Coating Systems (AISC SPE/SSPC-QP
3 420-10) defines quality control as the inspection of work.
Inspection includes but is not limited to confirming that
procedures are met; workers are properly qualified; equipment is
appropriate and in acceptable working order; and the proper
materials are used and are in compliance with inspection
criteria.Lets take a look at a few case studies to see whether
implementation of a quality control program using trained, properly
equipped inspectors makes a difference.Case Study No. 1:Mirror,
MirrorBackground:A contract was awarded to remove and replace the
existing coating system on a large riveted structure. The
specification required abrasive blast cleaning to achieve a
Near-White blast (SSPC-SP 10/NACE 2), followed by two coats of a
polyamide epoxy (standard gray) and one coat of polyurethane
topcoat. Six months after the contract was completed, corrosion was
observed (Fig. 2).
Fig. 2: Corrosion products on the back sides of the rivets and
edges after six months service
Cause:Corrosion products remained on the back side of the rivets
that were not subjected to direct impact by the abrasive stream
during blast cleaning. The coating was also applied from one
direction, causing thin areas of coating on the back side of the
rivets and the adjacent flat areas of the steel plate. Inadequate
attention was given to the coating along the edges.Avoidance
Through Quality Control Inspection?The QC inspector should have
carefully examined the difficult access areas after surface
preparation and application of each coating layer