Let’s Talk About In-Process and Post-Application Adhesion Testing

KTA’s Certified Coating Inspector Forum  Volume 2, Issue No. 9 – September 2023 

William Corbett, COO 

AMPP Senior Certified Coating Inspector & Certified Protective Coating Specialist 

KTA’s Certified Coating Inspector Forum is designed to provide professional development/continuing education on standards, inspection practices, new instruments, and other topics to help keep certified AMPP and FROSIO coating inspectors current. It represents the views of the author and KTA-Tator, Inc. It may or may not represent the views of AMPP: The Association for Materials Protection & Performance, even though SSPC, NACE, and AMPP standards are frequently referenced in the content. 


I thought I’d begin this issue of the Certified Coating Inspector Forum on adhesion testing by discussing various definitions of adhesion. When I Googled, “which is the best definition of adhesion in science,” I got: “Adhesion is the ability of one thing to stick firmly to another.” Simple enough. ASTM D3359, Standard Test Methods for Rating Adhesion by Tape Test defines it as “The molecular attraction and mechanical bonds between a coating and its substrate,” which is bit more technical. And although the testing we do is called “adhesion,” we are also assessing another characteristic of the coating, which is cohesion. A Google definition of cohesion is “The sticking together of particles of the same substance.” But I wanted something related to measuring adhesion (and cohesion) simply stated, so I came up with: A determination of how well a coating is attached to a substrate, how well coating layers are attached to one another, and the inner strength of any individual coating layer. So now that we have a basis for what we are measuring, let’s dive into the testing itself.  

While some time is dedicated to learning about the basics of adhesion testing in AMPP CIP 2 training, I feel it is a topic worthy of more in-depth discussion, especially as it relates to in-process adhesion testing of thermal spray coatings, since they are becoming more widely specified. This issue of the Certified Coating Inspector Forum focuses on the following topics related to adhesion testing: 

  1. Why Measure Coating Adhesion Properties? 
  1. Selecting an Adhesion Testing Method 
  1. Measuring Adhesion by the Tape Test 
  1. Evaluating Adhesion by the Knife Test 
  1. Measuring Pull-off Adhesion of Coatings on Steel 
  1. Measuring Pull-off Adhesion of Coatings on Concrete 
  1. Reporting Tensile (pull-off) Adhesion Test Data 
  1. In-process Testing of Thermal Spray Coatings (TSC) 

Why Measure Coating Adhesion Properties? 

Given that adhesion testing of a coating or lining is typically a destructive test, why would we do it? The primary reason is that a specification may require it as one way to help ensure the coating was applied correctly to a properly prepared surface. But one could argue that if in-process QC and QA inspection was performed by knowledgeable, trained, and certified inspectors then there is little need for this type of verification. Nonetheless it may be contractual. However, in the case of thermal spray coatings (TSC) applied by metallizing, in-process tensile (pull-off) adhesion is frequently required (in addition to companion coupon bend tests), since there are many application variables that can influence the bond of the TSC to the prepared substrate. More on that a little later. Adhesion testing is an excellent way to determine the integrity of an existing coating system that is being considered for overcoating to extend its service life. In fact, overcoating an aged coating system without evaluating the adhesive/cohesive properties of the existing system can be high-risk and can result in catastrophic disbonding if the existing system has low adhesive/cohesive strength.  Finally, adhesion testing is often performed during a coating failure investigation, but that is outside the scope of this article. 

While adhesion gives us a “warm and fuzzy feeling” that the coating system is well-adhered, it is not necessarily a predictor of long-term performance. That is, a coating system that displays a pull-off value of 1,000 psi is not going to perform twice as long as a coating that displays a pull-off value of only 500 psi. If the substrate is protected and the coating remains attached, the relative strength of the coating bond to the substrate (and to underlying coating layers) may be somewhat moot. And even if a coating possesses excellent adhesion, corrosion protection can be compromised for other reasons, such as insufficient thickness.  

Selecting an Adhesion Testing Method 

There are four common methods for testing the adhesion of protective coatings and linings, listed in the table below. 

ASTM Standard Reference Title of Method 
D3359 Standard Test Methods for Rating Adhesion by Tape Test 
D6677 Standard Test Method for Evaluating Adhesion by Knife 
D4541 Standard Test Method for Pull-off Strength of Coatings Using Portable Adhesion Testers 
D7234 Standard Test Method for Pull-off Strength of Coatings on Concrete Using Portable Adhesion Testers 

The first two methods are used to evaluate the shear, peel-back, or peel strength of a coating, while the latter two methods are used to evaluate the tensile or pull-off strength of a coating. Since the testing mechanisms differ (peel-back versus pull-off) the data generated should not be compared. That is, a coating can exhibit a high pull-off value, but a low peel-back rating, and vice-versa. If adhesion testing of in-process work is to be performed, the specific test method(s) should be specified. If testing is not specified but is mandated by the engineer or facility owner after-the-fact, the purpose of the testing will help them to select the method(s) to be used.  For example, if there is a dispute on workmanship, or a question arises as to whether some of the existing coating on a structure can remain and be overcoated, both peel-back and pull-off data may be beneficial. 

Measuring Adhesion by the Tape Test (ASTM D3359) 

The procedures described in ASTM D3359 are designed for use on metallic substrates. The methods do require knowledge of the applied coating thickness, which can be measured nondestructively (ASTM D70911). Method A (X-cut) is used on coatings/systems exceeding 5 mils, while Method B (Crosscut) is used on coatings up to and including 5 mils.  The distance between the cuts creating the crosscut pattern is 2 mm for coatings between 2 and 5 mils in thickness, or 1 mm for coatings less than 2 mils. Method B can be used on coatings thicker than 5 mils if the spacing between the cuts is wider than 2 mm, but the specific spacing is not defined. Although not defined, a spacing of 5 mm has been used for heavier coatings. 

The required equipment includes a straight edge, a razor knife or scalpel, a brush, tape with very specific adhesive properties, and a magnifier. A cross-cut guide or X-cut guide can be used instead of a straight edge, and a utility knife can be used in place of a razor knife/scalpel. A cross hatch cutter is also available as an alternative to the straight edge or cross-cut guide for Method B. The cross-hatch cutter makes all of the cuts at the same time, rather than individually as is the case with the straight edge or cross-cut guide. The cross-cut guide is only available with 2mm spacing (for testing coatings from 2-5 mils thick). If the coating is less than 2 mils thick, the cross-hatch cutter is the best alternative to making individual cuts with 1mm spacing. 

To perform Method A (X-cut), make a 1.5” cut through the coating system in a single pass. If the cut does not go all the way to the substrate, do not recut the incision. Move to a new location and make a new cut. Make a second cut to form an “X” with a 30-45° angle at the intersection of the “X.” Inspect the cuts with the magnifier to verify the incisions penetrate down to the metal surface. Remove the cutting debris with the brush and remove 2 laps of tape (one time for each day of testing) and discard. Apply a 3” piece of tape to the X-cut and rub it firmly on the surface to ensure good contact.  

Remove the tape within 90 +/- 30 seconds 180° back across the X-cut (do not pull upwards). Repeat the test at two more spots at each test location (triplicate testing). Evaluate the X-cuts for coating detachment and rate the adhesion from 5A to 0A (rating chart shown below).

To perform Method B (crosscut), conduct the test in a similar fashion to that described for Method A, except that a grid is cut into the coating. The grid consists of a series of equally spaced parallel cuts, with a second series of parallel cuts made perpendicular to the first. The cuts are kept straight using a straight edge, the cross-cut guide, or the cross-hatch cutter. Eleven cuts, 1 mm apart are made in each direction for coatings up to 2 mils in thickness, and six cuts, 2 mm apart are made in each direction for coatings from 2 to 5 mils. Triplicate testing is required. Evaluate the crosscut for coating detachment and rate the adhesion from 5B to 0B (rating chart shown below). 

The pressure-sensitive tape that is used for both Methods A and B must have an adhesive peel strength between 58 and 64 oz./in. when tested according to ASTM D33302. This range is based on the properties of Permacel P-99 tape (discontinued in 2009). The SEMicro CCT conforms to this requirement (and comes with a Certificate of Conformance); other tapes can be used if the contracting parties agree. 

Here are a few things to remember when performing tape adhesion testing: 

  • Dull knife blades/cutter bundles can produce erroneous results since the coating may chip when the cuts are made.  
  • Maintaining proper spacing between incisions when employing the cross-cut tape adhesion method is critical. 
  • When employing the X-cut method, maintaining the proper angle at intersection of “X” is critical. The legs of the “X” should be ~ 1” apart. 
  • The tape should be stored properly. The tape typically has a one-year shelf life. 
  • The name of the tape used should be included in the report. 
  • The tape may not be effective in cold temperatures. 
X Cut Guide Order

Evaluating Adhesion by the Knife Test (ASTM D6677) 

The procedure described in ASTM D6677 is very similar to Method A in ASTM D3359 except that no tape is used. Instead, the tip of the knife blade is used to probe at the intersection of the “X” and the adhesion rated on a scale of 10 to 0 (rating chart shown below). 

Rating Description 
10 Coating is extremely difficult to remove. Fragments no larger than 1/32” x 1/32” removed with great difficulty 
Coating is difficult to remove. Coating chips from 1/16” x 1/16” to 1/8” x 1/8” removed with difficulty 
Coating is somewhat difficult to remove. Coating chips from 1/8” x 1/8” to 1/4” x 1/4” removed with slight difficulty 
Coating is somewhat difficult to remove. Coating chips in excess of 1/4” x 1/4” removed using light pressure with knife blade 
Coating is easy to remove. Once started with the knife blade, the coating can be grasped with fingers and easily peeled to a length of at least 1/4” 
Coating is easy to remove. Once started with the knife blade, the coating can be grasped with fingers and easily peeled to a length greater than 1/4” 

Measuring Pull-off Adhesion of Coatings on Steel (ASTM D4541) 

Type II (fixed alignment)
Type V (self-aligning hydraulic)

There are five different pull-off adhesion testers referenced in ASTM D4541, including one mechanically operated tester (fixed alignment), one pneumatically operated tester (self-aligning), and three hydraulically operated testers (self-aligning). Fixed alignment testers cannot compensate for misalignment of the loading fixture (described below) to the coated surface, while self-aligning testers can. Since different testers yield different pull-off values (due in part to fixed versus self-alignment, consistency of the applied loading rate, and other factors), it is important to decide which type of tester will be used prior to testing. This point cannot be overemphasized. The results obtained from each of the testers listed in ASTM D4541 will not be identical, so it is critical that the specific test device be identified before the work begins. Information on the variation in pull-off values of several coating types using the different testers (based on an interlaboratory study used to prepare the precision and bias statement) is provided in ASTM D4541.  

To perform the testing, the coated surface and the base of the loading fixture (occasionally referred to as a dolly) are roughened to help ensure a good bond. The loading fixture is the “plug” that is attached to the coating and removed with the tester. The design and diameter of the loading fixtures varies depending on the type of tester. Loading fixture diameters for coatings on steel can be as small as 10mm (approx. 0.5-inch); typically, 20mm (approx. 1-inch) loading fixtures are used.  

The loading fixture is bonded to the coated surface using an adhesive (glue). The glue type can vary from epoxy to cyanoacrylate gel. After the glue cures, the adhesion testing device is attached to the loading fixture and perpendicular force is applied (mechanically, pneumatically, or hydraulically) until the loading fixture is detached, the specified load is applied without detachment of the fixture, or the maximum load of the instrument is reached. The pressure loading rate cannot exceed 150 psi/second and scoring the coating to the substrate prior to applying the pressure load is not performed unless agreed upon by the contracting parties. Note that certain adhesion testers either monitor the pressure loading rate or control the pressure loading rate automatically (red-circled area on display image). If not, the operator can use a stopwatch to verify the maximum rate is not exceeded. 

The pressure required to detach the loading fixture (or the pressure applied if the maximum load is reached) is recorded, along with the location of break, which is defined as follows: 

  • Adhesive Break: A break between coating layers or between the substrate and first coating layer 
  • Cohesive Break: A break within an individual coating layer 
  • Glue Break: Coating adhesive and/or cohesive strength exceeds bonding strength of the glue used to attach the loading fixture. 

If multiple locations of break occur, the percentage of each (e.g., 75% cohesive within primer; 25% adhesive between primer and topcoat) is estimated. 

Measuring Pull-off Adhesion of Coatings on Concrete (ASTM D7234) 

Evaluating the tensile strength of coatings applied to concrete is similar to testing coatings applied to steel, with the following testing variations.   

Scoring the coating 
  • Thick film coatings (>20 mils) have lateral bond strength that may produce misleading pull-off values (false, high), so scoring the coating is required. 
  • Scoring through the coating (down to the concrete surface) is typically performed prior to attachment of loading fixture (except if the loading fixture is square) 
  • Scoring coatings <20 mils may also be required by the specification, or as agreed upon by the contracting parties. 
  • When attaching the loading fixture, it is important to prevent the glue from flowing into the groove created by scoring. 
Loading Fixture Size and Loading Rate 
  • Unlike steel, concrete is non-uniform and contains cement, aggregate, and additives, therefore the loading fixture diameter should be larger – 50mm (approx. 2-inch) to ensure the test is performed over a representative area. 
  • Square loading fixtures are used with some testers. 
  • The pressure loading rate should be 30 psi/second; the test should be completed in 5-30 seconds. Again, certain adhesion testers either monitor the pressure loading rate or control the pressure loading rate automatically. If not, the operator can use a stopwatch to monitor the rate of loading. 
Compressive Vs. Tensile Strength  
  • “Typical” concrete has high compressive strength (e.g., 4000 psi), but low tensile strength (e.g., 400 psi). High strength concrete has higher tensile strength. 
  • Low pull-off values are common, and the plane of fracture is commonly within the concrete. 
  • When fracture occurs in the concrete, the minimum coating tensile adhesion properties can be reported; however, the actual coating tensile strength remains unknown. 

Unlike ASTM D4541, the various instruments are not described in Annexes in ASTM D7234. An interlaboratory study (ILS) published by ASTM involving two coatings applied to concrete and seven instruments revealed insignificant variation in pull-off values between the various tester types. Accordingly, the results between the instruments are more comparable. 

Here are a few things to remember when performing tensile (pull-off) adhesion testing: 

  • Determine which tester to use in advance. Different testers generate different pull-off values on the same coating when applied to steel. 
  • Verify the testing equipment has a current calibration certificate. 
  • Perform testing in triplicate at each test location. 
  • The rigidity of the substrate can affect the results. Testing on steel less than 1/8” thick can result in lower pull-off values because the steel can flex. 
  • Clean the coating surface and the loading fixtures to reduce the frequency of glue failures and use care when abrading the surface to prevent the introduction of surface flaws. 
  • Verify the glue is compatible with the coating. 
  • Allow the glue to cure. Conducting pull tests before the glue is cured often results in glue breaks. 
  • Verify perpendicular pull. 
  • Support the pull-off device when testing coatings on vertical surfaces. 

Reporting Tensile (pull-off) Adhesion Test Data 

The flow diagram shown below was excerpted from ASTM D4541 and contains the reporting requirements based on two protocols. Protocol 1 (Test to Fracture) is used when the pull-off value is unknown or unspecified and you want to know the pull-off strength; Protocol 2 (Pass/Fail) is used when the pull-off value is known or specified. Note that for Protocol 1, glue failures exceeding ¼ of the loading fixture contact surface are considered invalid tests and the results are discarded. For Protocol 2, if the glue fails prior to achieving the predetermined pull-off value (i.e., passing value), retesting is typically required. 

In-process Testing of Thermal Spray Coatings (TSC) 

It is unusual to test the adhesion of coatings in the process of application since coating cure is critical to achieving proper adhesion. However, in the case of thermal spray coatings (TSC) that are neither liquid applied nor require cure time, in-process adhesion testing is frequently specified. Surface profile depth, angularity, surface cleanliness, and the wire quality, amperage, voltage, spray distance, and spray angle can all impact the adhesion of the applied TSC. Therefore, specifications that invoke SSPC-CS 23.00/NACE No. 123 frequently require tensile (pull-off) adhesion testing on the coated part or on companion panels using a self-aligning adhesion tester.  

This requirement can be tricky since production is stopped until pull-off values are known and acceptable. Waiting for glue to cure can be tedious. The manufacturer of the pneumatic pull-off tester makes special loading fixtures with a transparent base. A light (UV) curable glue such as Loctite AA 3554 is applied to the base of the loading fixture and a UV light source is used to cure the glue in a few seconds, enabling the inspector to perform the adhesion testing with minimal interruption to production. As indicated above, some specifications allow for the preparation of companion panels at the same time the production work is being performed, with the coating on the panels, rather than the coating on the structure, being tested.  


This edition of the Certified Coating Inspector Forum focused on post-application adhesion testing of liquid-applied protective coating systems and in-process adhesion testing of thermal spray coatings. Four methods of adhesion testing were described. Independent of the method used, adhesion testing is destructive to the coating film (unless the loading fixture is not detached during tensile [pull-off] testing and is allowed to remain attached) and should not be performed by an inspector unless required by the project specification. In-process adhesion testing is commonly required for TSC and may be performed on the actual coated steel or on companion panels prepared to the same degree as the structural steel. 

[1] Standard Practice for Nondestructive Measurement of Dry Film Thickness of Nonmagnetic Coatings Applied to Ferrous Metals and Nonmagnetic, Nonconductive Coatings Applied to Non-Ferrous Metals

[2] Standard Test Method for Peel Adhesion of Pressure-Sensitive Tape

[3] Specification for the Application of Thermal Spray Coatings (Metallizing) of Aluminum, Zinc, and Their Alloys and Composites for the Corrosion Protection of Steel

5 thoughts on “Let’s Talk About In-Process and Post-Application Adhesion Testing”

  1. Thank you for the very detailed and informative article
    Adhesion is a primary function of protective coatings
    A liquid-applied protective coating system will need to possess excellent adhesion, both of primers to steel substrates and of coatings to each other

Leave a Comment

Your email address will not be published. Required fields are marked *