Click for next page ( 117


The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 116
56 8 QUALITY CONTROL AND INSPECTION 8.1 Introduction Owners should make provisions to ensure the quality of the coating by providing detailed specifications and competent inspection. Wire manufacturers need to have quality control systems in place to ensure that the wire can be applied and will perform as intended. Applicators must ensure that they have the correct experience and equipment and competent applicators to prepare the structure and apply the coating. Inspection is one of the most important aspects of coating. It provides a written record of the details of the coating application, ensures that the coating specifications are met, and finds and ensures the correction of inadequate coating areas before they become failure locations in the future requiring costly correction. Particular attention must be directed toward difficult-to-coat areas, such as the inside surfaces of H-shapes and the interlock knuckles of sheet piling, because these are areas where optimum nozzle/gun angles and distances will be difficult to attain. 8.2 Quality Assurance Functions for Owners 8.2.1 Informed Selection An informed selection should be made of TSMCs taking into account planned use of the coatings and the environment in which they are to be used. 8.2.2 Provide Definitive Specifications Specifications should include, as a minimum and as an addition to contractual provisions, the following: Scope of work, to include the structure to be coated and portions not to be coated; All applicable references; Provisions for payment; Definitions; A list of required submittals; Safety provisions; Requirements for delivery, storage, and handling of materials and supplies; Chemical composition, finish, coil weight, and preparation of metallizing wire; Requirements for sampling and testing thermally sprayed materials and the applied sealer; A job reference standard with a description of appearance and adhesion requirements; Requirements for surface preparation; Metallizing application; Workmanship; Atmospheric and surface conditions;

OCR for page 116
57 Sequence of operations; Approved methods of metallizing; Coverage and metallizing thickness; Progress of metallizing work; Sealing and painting instructions; Metallizing schedule; and Quality control requirements. 8.2.3 Coating Inspector Provide a qualified coating inspector to provide full-time inspection services. This should be a third-party inspector with the power and ability to work out problems with the applicator to achieve the desired coating quality. Section 9.3 lists the necessary qualifications of an inspector. 8.3 Quality Control 8.3.1 Documentation The documentation of inspection activities provides a permanent record of the thermal spray job. Thorough documentation provides a written record of the job in the event of a contract dispute or litigation. Inspection records may also be used to help diagnose a premature coating failure. Future maintenance activities may also be simplified by the existence of complete inspection records. As a minimum, at least one full-time inspector should be used on all thermal spray jobs to ensure adequate inspection and documentation. A qualified third- party inspector from a reputable firm should perform the inspection. As a minimum, the inspector should perform and document the inspection procedures described in this section. Sample documentation forms for industrial coating activities are available through NACE International and the Society for Protective Coatings. The inspector should record the production and quality control information required by the purchaser or the purchasing contract. Among the items that should be recorded are Information about the contractor and purchaser; Surface preparation and abrasive blasting media requirements; Flame or wire-arc spray equipment used; Spraying procedure and parameters used; TSMC requirements; Safety precautions followed; Environmental precautions; Test data taken, including Nature of the test, When conducted, Where conducted,

OCR for page 116
58 Results, and Abnormalities and resolution; and Problems and resolution. The inspector should keep records for the time period required for regulatory compliance and required by the purchasing contract. 8.3.2 Testing Frequency The required frequency of inspection procedures should be documented in the specification. Inspection can be expensive, and care should be taken not to overspecify inspection procedures. Conversely, inspection has an intrinsic value that is sometimes intangible. It is difficult to measure the value added by inspection resulting from the conscientious performance of the contract. Thermal spray can be quite sensitive to the quality of surface preparation, thermal spray equipment setup, and application technique. Therefore, it is important to specify an appropriate level of inspection. Table 8 presents recommended frequencies for various inspection procedures. 8.3.3 Job Reference Standard and Material Samples 8.3.3.1 Material samples. Reference samples of each material used on a thermal spray job should be collected, including clean, unused abrasive blast media; thermal spray wire; sealer; and paint. Samples may be used to evaluate the conformance of materials to any applicable specifications. A 2.2-lb (1-kg) sample of blast media should be collected at the start of the job. The sample may be used to verify the cleanliness, media type, and particle size distribution of the virgin blast media. A 12-in. (30-cm) sample of each lot of thermal spray wire should be collected. The wire sample may be used to confirm that the manufactured wire conforms to the size and compositional requirements of the contract. One-quart (1-liter) samples of all sealers and paints should be collected for compliance testing. TABLE 8 Recommended inspection frequencies for selected procedures Inspection Procedure Recommended Frequency per Unit Area Surface profile 3 per 500 ft2 (45 m2) or less Thermal spray coating thickness 5 per 100 ft2 (9 m2) or less Thermal spray adhesion 2 per 500 ft2 (45 m2) or less Sealer thickness 2 per 500 ft2 (45 m2) or less Paint thickness 2 per 500 ft2 (45 m2) or less Soluble salts 1 per 1,000 ft2 (90 m2) or less

OCR for page 116
59 8.3.3.2 Job reference standard. A thermal spray job reference standard (JRS) should be prepared. The JRS may be used at the initiation of a thermal spray contract to qualify the surface preparation, thermal spray application, and sealing processes. The JRS and the measured values may be used as a visual reference or job standard for surface preparation, thermal spray coating, sealing, and painting, in case of dispute. 8.3.3.3 Preparing the JRS. The JRS should be prepared prior to the onset of production work. To prepare the JRS, a steel plate of the same alloy and thickness to be coated, measuring 2 2 ft (60 60 cm) should be solvent and abrasive blast cleaned in accordance with the requirements of the contract. The abrasive blast equipment and media used for the JRS should be the same as those that will be used on the job. One-quarter of the JRS plate should be masked using sheet metal, and the TSMC should be applied to the unmasked portion of the plate. The TSMC should be applied using the same equipment and spray parameters proposed for use on the job. The gun should be operated in a manner substantially the same as the manner in which it will be used on the job. The approximate traverse speed and standoff distance during spraying should be measured and recorded. Two-thirds of the thermal spraycoated portion of the JRS should be sealed in accordance with the requirements of the contract. One-half of the sealed area should be painted in accordance with the contract if applicable. The sealer and paint should be applied using the same paint spray equipment that will be used for production. The prepared JRS should be preserved and protected in such a manner that it remains dry and free of contaminants for the duration of the contract. The preserved JRS should then be archived for future reference in the event of a dispute or premature coating failure. Once the JRS is qualified, the operating parameters should not be altered by the contractor, except as necessitated by the requirements of the job. Figure 6 depicts a representative JRS. Figure 6. Job reference standard configuration (1 in. = 2.54 cm, 1 ft = 30.48 cm).

OCR for page 116
60 8.3.3.4 Evaluating the JRS. The surface cleanliness; blast profile shape and depth; thermal spray appearance, thickness, and adhesion; and sealer and paint thickness should be determined in accordance with the contract requirements and recorded. 8.3.4 Testing Prior to Surface Preparation 8.3.4.1 Ambient conditions measurement. An assessment of the local atmospheric conditions should be made before surface preparation and thermal spray application begins. Measurement of ambient conditions includes substrate temperature, air temperature, dew point, and relative humidity. A contact thermocouple or infrared pyrometer should be used to measure the substrate temperature. Air temperature should be measured using a sling psychrometer, thermometer, or digital measurement instrument. Dew point should be calculated using the appropriate psychrometric charts. Humidity should be determined in accordance with ASTM E337, "Test Method for Measuring Humidity with a Psychrometer (The Measurement of Wet-Bulb and Dry-Bulb Temperatures)." 8.3.4.2 Inspection preceding surface preparation. Prior to abrasive blasting, inspect the substrate for the presence of contaminants including grease and oil, weld flux and spatter, heat-affected zones, flame-cut edges, pitting, sharp edges, and soluble salts. Grease and oil. Painted surfaces and newly fabricated steel should be visibly inspected for the presence of organic contaminants such as grease and oil, as required by the project specification. Continue degreasing until all visual signs of contamination are removed. Conduct the UV light test, qualitative solvent evaporation test, or the heat test to detect the presence of grease and oil. Use a UV lamp to confirm the absence of oil or grease contamination. The solvent evaporation test should be made by applying several drops or a small splash of a residual-less solvent, such as trichloromethane, on the areas suspected of oil and grease retention (e.g., pitting and crevice corrosion areas and depressed areas, especially those collecting contamination, etc.). An evaporation ring will form if oil or grease contamination is present. The heat test should be made by using a torch to heat the degreased metal to about 225oF (110oC). Residual oil/grease contamination should be drawn to the metal surface and is visually apparent. Weld flux and spatter. A visual inspection for the presence of weld flux and spatter should be performed, as required by the project specification. Weld flux should be removed prior to abrasive blast cleaning using a suitable SSPC-SP 1 "Solvent Cleaning" method. Weld spatter may be removed either before or after abrasive blasting using suitable impact or grinding tools. Areas that are power-tool cleaned of weld spatter should be abrasive blast cleaned.

OCR for page 116
61 Heat-affected zones caused by welding. Heat-affected zones should be identified and marked prior to abrasive blasting as required by the specification. Extra care during surface preparation and extra attention to profile inspection should be given to these areas. Flame-cut edges. Flame-cut edges should be identified and marked prior to abrasive blasting as required by the specification. The demarcated areas should be ground using power tools prior to abrasive blast cleaning. Pitting. Deep pits or pitted areas should be identified and marked prior to abrasive blast cleaning as required by the specification. The demarcated areas should be ground using power tools prior to abrasive blast cleaning. Sharp edges. Sharp edges should be identified and marked prior to abrasive blasting as required by the specification. The demarcated edges should be prepared by grinding to a minimum radius of 1/8 in. (3 mm) prior to blast cleaning. Soluble salts. When soluble salt contamination is suspected, the contract documents should specify a method of retrieving and measuring the salt levels as well as acceptable levels of cleanliness. Salt contamination is prevalent on structures exposed in marine environments and on structures such as parking decks and bridges exposed to deicing salts. Structures that are likely to have soluble salt contamination, including those in marine or severe industrial atmospheres, bridges or other structures exposed to deicing salts, and seawater immersed structures, should be tested. Soluble salt levels should be rechecked for compliance with the specification after solvent cleaning and abrasive blasting have been completed. Common methods for retrieving soluble salts from the substrate include cell retrieval methods and swabbing or washing methods. Various methods are available for assessing the quantity of salts retrieved, including conductivity, commercially available colorimetric kits, and titration. The rate of salt retrieval is dependent on the retrieval method. The retrieval and quantitative methods should be agreed upon in advance. The recommended testing procedure employs the Bresle cell (ISO 8502-6) to extract soluble salts from the substrate. Chloride ion concentration is readily measured in the field using titration strips available from Quantab. The test strip analyzes the collected sample and measures chloride ion concentration in parts per million. The unit area concentration of chloride ions is calculated in micro-grams per centimeter. The lower detection limit for the Bresle/Quantab method is about 2 m/cm2. SSPC-SP-12/NACE #5 describes levels of soluble salt contamination. It is recommended that surfaces cleaned to an SC-2 condition be used for TSMCs. An SC-2 condition is described as having less than 7 m/cm2 of chloride contaminants, less than 10 m/cm2 of soluble ferrous ions, and less than 17 m/cm2 of sulfate contaminants. The number of tests per unit area (e.g., 1 per 1,000 ft2 [90 m2]) should be specified in the contract documents. Also refer to "SSPC Technology Update: Field Methods for Retrieval and Analysis of Soluble Salts on Substrates," and SSPC-91-07. 8.3.5 Testing During Surface Preparation 8.3.5.1 Abrasive cleanliness. Abrasive blast media must be free of oil and salt to prevent contamination of the substrate. Recycled steel grit abrasive should comply with requirements of SSPC-AB-2, "Specification for Cleanliness of Recycled Ferrous Metallic Abrasives." Evaluating for salt in abrasives. Most abrasives used to prepare steel substrates for thermal spraying are unlikely to contain appreciable amounts of soluble salts. However,

OCR for page 116
62 slag abrasives used for strip blasting may sometimes contain measurable quantities of salts. Slag abrasives should be evaluated in accordance with ASTM D4940, "Test Method for Conductimetric Analysis of Water Soluble Ionic Contamination of Blasting Abrasives." Testing for oil in abrasives. To test for oil in abrasives, a clear glass container should be half filled with unused abrasive, and then distilled or deionized water should be added to fill the container. The resulting slurry mixture should be stirred or shaken and allowed to settle. The water should then be examined for the presence of an oil sheen. If a sheen is present, the media should not be used, and the source of contamination should be identified and corrected. 8.3.5.2 Air cleanliness. The following guidelines apply to air cleanliness. The compressed air used for abrasive blasting, thermal spraying, sealing, and painting should be clean and dry. Oil or water in the blasting air supply may contaminate or corrode the surface being cleaned. Oil or water in the thermal spray, sealing, or painting air supply may result in poor coating quality or reduced adhesion. Compressed air cleanliness should be checked in accordance with ASTM D4285, "Method for Indicating Water or Oil in Compressed Air." The air compressor should be allowed to warm up, and air should be discharged under normal operating conditions to allow accumulated moisture to be purged. An absorbent clean white cloth should be held in the stream of compressed air not more than 24 in. (60 cm) from the point of discharge for a minimum of 1 minute. The air should be checked as near as possible to the point of use and always after the position of the in-line oil and water separators. The cloth should then be inspected for moisture or staining. If moisture or contamination is detected, the deficiency should be corrected before going further. 8.3.5.3 Blast air pressure. The contractor should periodically measure and record the air pressure at the blast nozzle. The measurement should be performed at least once per shift and should be performed on each blast nozzle. Measurements should be repeated whenever work conditions are altered such that the pressure may change. Pressures should be checked concurrently with the operation of all blast nozzles. The method employs a hypodermic needle attached to a pressure gauge. The needle is inserted into the blast hose at a 45-degree angle toward and as close to the nozzle as possible. The blast pressure is read directly from the gauge. 8.3.5.4 Blast nozzle orifice. The contractor should visually inspect the blast nozzle periodically for wear or other damage. Gauges are available that insert into the end of the nozzle and measure the orifice diameter. Nozzles with visible damage or nozzles that have increased one size should be replaced. Worn nozzles are inefficient and may not produce the desired blast profile. Damaged nozzles may be dangerous. 8.3.5.5 Surface cleanliness. The following applies to surface cleanliness. Blast Cleanliness. The final appearance of the abrasive cleaned surface should be inspected for conformance with the requirements of SSPC-SP-5. An SP-5 surface is defined as free of all visible oil, grease, dirt, dust, mill scale, rust, paint, oxides, corrosion products, and

OCR for page 116
63 other foreign matter. The appearance of SP-5 surfaces is dependent on the initial condition of the steel being cleaned. SSPC-VIS-1 may be used to interpret the cleanliness of various blast-cleaned substrates based on the initial condition of the steel and the type of abrasive used. Initial conditions depicted include the following: Rust Grade A--a steel surface completely covered with adherent mill scale with little or no rust visible; Rust Grade B--a steel surface covered with both mill scale and rust; Rust Grade C--a steel surface completely covered with rust with little or no pitting; and Rust Grade D--a steel surface completely covered with rust with visible pitting. The inspector should determine the initial substrate condition or conditions. The final appearance of the surfaces should then be compared with the appropriate photograph. No stains should remain on the SP-5 surface. However, the appearance of the surface may also vary somewhat, depending on the type of steel, presence of roller or other fabrication marks, annealing, welds, and other differences in the original condition of the steel. The job reference standard should be used as the basis for judging the surface cleanliness. Dust. Abrasive blasting and overspray from painting or metallizing can leave a deposit of dust on a cleaned substrate. The dust may interfere with the adhesion of the TSMC. Residual dust may be detected by applying a strip of clear tape to the substrate. The tape is removed and examined for adherent particles. Alternatively, a clean white cloth may be wrapped around a finger and wiped across the surface. The cloth and substrate are then examined for signs of dust. The preferred method of removing residual dust is by vacuuming. Alternatively, the surface may be blown down with clean, dry compressed air. 8.3.5.6 Surface profile. The following applies to surface profile. ASTM D4417, "Test Methods for Field Measurements of Surface Profile on Blast Cleaned Steel," provides test methods for surface profile measurement. Methods A and B use either a needle depth micrometer to measure the depth of the valleys in the steel or comparator charts. Method C is the recommended method for measuring the surface profile depth. Methods A and B may provide unreliable measures of the blast profile. Method C employs replica tape and a spring gauge micrometer to measure the surface profile. With the wax paper backing removed, the replica tape is placed face down against the substrate, and a burnishing tool is used to rub the circular cutout until a uniform gray appearance develops. The replica tape thickness (compressible foam plus plastic backing) is then measured using the spring micrometer. The profile is determined by subtracting the thickness of the plastic backing material, 0.002 in. (50 m), from the measured value. Three readings should be taken within a 16-in.2 (100-cm2) area, and the surface profile at that location should be reported as the mean value of the readings. The number of measurements per unit area (e.g., 3 per 500 ft2 [45 m2]) should be specified in the contract document. Two types of replica tape are available, coarse (0.0008 to 0.002 in. [20 to 50 m]) and X-coarse (0.0015 to 0.0045 in. [37.5 to 112.5 m]). In most cases, the X-coarse tape will be used to measure profile. It may be possible to measure profiles as high as 0.006 in. (150 m) using the X-coarse tape.

OCR for page 116
64 The inspector must be aware that TSMCs create safety and health risks in the form of hot surfaces, fumes, ultraviolet light, and noise. Precautions must be taken to avoid these hazards--refer to Section 2. 8.3.6 Testing During and After Coating Application 8.3.6.1 Coating thickness. The following applies to coating thickness. Coating thickness is measured in accordance with SSPC-PA-2 using a Type 2 gauge. Calibrate the instrument using a calibration wedge that is close to the contract-specified thickness placed over a representative sample of the contract-specified abrasive-blasted steel or a prepared bend coupon, or both. Thickness readings should be made either in a straight line with individual readings taken at 1-in. (2.5-cm) intervals or spaced randomly within a 2-in. (5-cm) diameter area. Line measurements should be used for large flat areas, and area measurements should be used on complex surface geometry and surface transitions such as corners. The average of five readings constitutes one thickness measurement. A given number of measurements per unit area (e.g., five per 100 ft2 [9 m2]) should be specified in the contract documents. Figure 7 illustrates this method. Measure thickness according to ASTM D4138, "Test Methods for Measurement of Dry Film Thickness of Protective Coating Systems by Destructive Means," Test Method A. This method uses a tungsten carbide-tipped instrument to scribe through the sealer and paint, leaving a V-shaped cut. A heavy dark-colored marking pen is first used to mark the coated surface. The scribing instrument is then drawn across the mark. This process sharply delineates the edges of the scribe. A reticle-equipped microscope is used to read the film thickness. A total of three thickness readings should be performed in a 16-in.2 (100-cm2) area, with the average of the three tests reported as a single measurement. The number of measurements per unit area (e.g., 1 per 500 ft2 [45 m2]) should be specified in the contract documents. Thickness testing using this method should be minimized because the test method destroys the sealer and paint. Areas damaged by adhesion testing must be repaired by 5 in line at about 1 in. [2.5 cm] 5 in a spot of about 2 in. dia (5 cm) Figure 7. Methods of taking coating thickness measurements.

OCR for page 116
65 touch-up with sealer or paint using a brush or spray gun. Thickness testing should be performed in a small area (16 in.2 [100 cm2]) to limit the area that must be repaired. 8.3.6.2 Adhesion tests. The following applies to adhesion tests. Bend Test. The bend test (180-degree bend on a mandrel) is used as a qualitative test for verifying proper surface preparation, equipment setup, and spray parameters. The bend test puts the TSMC in tension. The mandrel diameter for the threshold of cracking depends on substrate thickness and coating thickness. Table 9 summarizes a very limited bend-test cracking threshold for arc-sprayed zinc TSMC thickness versus mandrel diameter for steel coupons 0.050 in. (13 mm) thick. Test panels--the test panels should be a cold-rolled steel measuring 3 6 0.05 in. (7.5 15 1.25 cm). The panels should be cleaned and blasted in the same fashion in which the panels will be cleaned and blasted for the job. Application of thermal spray--the TSMC should be applied to five test panels using the identical spray parameters and average specified thickness that will be used on the job. The coating should be applied in a cross-hatch pattern using the same number of overlapping spray passes as used to prepare the job reference standard. The coating thickness should be measured to confirm that it is within the specified range. Conduct bend test--test panels should be bent 180 degrees around a steel mandrel of a specified diameter, as shown in Figure 5. Pneumatic and manual mechanical bend test apparatus may be used to bend the test panels. Examine bend test panels--test panels should be examined visually without magnification. The bend test is acceptable if the coating shows no cracks or exhibits only minor cracking with no lifting of the coating from the substrate. If the coating cracks and lifts from the substrate, the results of the bend test are unacceptable. TSMCs should not be applied if the bend test fails, and corrective measures must be taken. Figure 5 depicts representative bend test results. A knife blade can be used to facilitate the evaluation. Apply moderate pressure to the knife blade and if the coating cannot be dislodged, the adhesion can be considered satisfactory. Tensile Adhesion. Field Measurement--Evaluate the adhesion of the TSMC with the specification in accordance with ASTM D4541, "Test Method for Pull-Off Strength of Coatings Using Portable Adhesion Testers." A self-aligning Type IV tester, described in Annex A4 of ASTM D4541, should be used. A total of three adhesion tests should be performed in a 16-in2 (100-cm2) area, and the average of the three tests should be TABLE 9 Bend-test mandrel diameter versus zinc thermal spray coating thickness (for steel coupons 0.050 in. [13 mm] thick) TSMC Thickness (mils) 10 (254 m) 15 (381 m) 25 (635 m) Mandrel Diameter 1/2 in. (1.27 cm) 5/8 in. (1.59 cm) <1 in. (2.54 cm)

OCR for page 116
66 reported as a single measurement. Portable instruments with large-diameter test specimens, for instance, 2-in. versus 1-in. (50-mm versus 25-mm) diameter, produce better statistical results. The number of measurements per unit area (e.g., 1 per 500 ft2 [45 m2]) should be specified in the contract documents. Areas of deficient adhesion should be abrasive blasted, and the coating should be reapplied. Additional testing will probably be necessary to determine the extent of the area exhibiting poor adhesion. Adhesion testing should be minimized because the test method destroys the coating. Areas damaged by adhesion testing must be repaired by abrasive blasting and reapplication of the metallic coating. Adhesion testing is performed in a small area (16 in2 [100 cm2]) to limit the area that must be repaired. As an alternative to testing adhesion to the failure point, the tests can be interrupted when the minimum specified adhesion value is achieved. This method precludes the need to repair coatings damaged by the test. The adherent pull stubs can then be removed by heating to soften the glue or by firmly striking the side of the stub. Table 10 lists the recommended adhesion requirements for field- or shop-applied thermal spray coatings of zinc, aluminum, and 8515 wt% zinc/aluminum. As a caution in performing this type of test, the inspector must be aware that since the coating does contain some porosity, a low-viscosity adhesive might penetrate the coating and reach the substrate. If this occurs, the measured adhesion value will be influenced by the adhesion between the glue and the substrate. It is best to avoid low- viscosity liquid adhesives in favor of high-viscosity pastes. If there is any doubt, comparison tests should be performed to select the appropriate adhesive. Laboratory Measurement--Tensile-bond test specimens should be carbon steel, 1 in. (2.54 cm) in diameter and 1 in. (2.54 cm) in length, threaded per ASTM C 633, "Adhesion or Cohesive Strength of Flame-Sprayed Coatings." Cut Test. The thermal spray coating cut test consists of a single cut, 1.5 in. [40 mm] long, through the coating to the substrate without severely cutting into the substrate. All cuts should be made using sharp-edge tools. The chisel cut should be made at a shallow angle. The bond should be considered unsatisfactory if any part of the TSMC along the cut lifts TABLE 10 Typical adhesion of field- and shop-applied thermally sprayed metal coatings measured by pull-off testing Thermal Spray Tensile Adhesion Material psi [MPa] Zinc 500 [3.45] Aluminum 1,000 [6.89] 85/15 Zinc-Aluminum 700 [4.83] 90/10 Aluminum Oxide 1,000 [6.89]

OCR for page 116
67 from the substrate. The cutting tool used (knife, hammer and chisel, or other tool) should be specified in the contract. Perform an adhesion test every 100 ft2 (9.3 m2). The tested area and coated surfaces that have been rejected for poor adhesion shall be blast cleaned and recoated. 8.3.7 Appearance The coating should be free of blisters, cracks, chips or loosely adhering particles, oil, pits exposing the substrate, and nodules. A very rough coating might indicate that the coating was applied with the gun at too great an angle or too far from the surface. Evaluate coatings that appear powdery or oxidized by scraping. If scraping does not produce a silvery metallic appearance, the coating is defective and must be replaced. 8.3.8 Coating Morphology Metallographic examination may be used for qualifying spraying parameters, but it is not normally used for process control for corrosion control applications. Parameters included in this examination include percent porosity, percent unmelted particles, percent oxides, and the presence and amount of interface contamination.