Click for next page ( 12


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 11
12 sealers and sealing techniques. This investigation should Testing of the effects of edge geometry on coating reten- include the use of 100-percent solids materials with report- tion and TSMC performance. edly higher "pore-penetrating" abilities. This includes the Testing of the effects of coating defects on TSMC newer classes of maintenance painting primers designed performance. to penetrate cracked and aged organic coatings. Testing of the effects of surface contamination (chlo- The study should examine the effect of abrasive mixes rides) on coating performance. on TSMC performance. The program should also focus on those application para- meters that may affect the occurrence of critical pore Test Panel Preparation sizes, pore geometry, and alloy micro-segregation or inter-coat oxide formation that will impact performance. Test panels to evaluate adhesion, sealers, different abra- The program should focus on wire-arc spray applications sive mixes, edge effects, and application parameters were of the materials to obtain optimal application rates. The prepared by CSI Coatings in Nisku, Canada, using Thermion program should include some focused study of the effects Bridgemaster equipment. The steel used for the corrosion of increased productivity on material performance. tests met the requirements of AASHTO M-270 Grade 36 or ASTM A-328. M/020 steel was used for the complex corro- sion test panels and had a nominal analysis of 0.17 to 0.24 car- PROCEDURES USED IN LABORATORY TESTS bon (C), 0.25 to 0.56 manganese (Mn), 0.04 max phosphorus Laboratory testing was designed to improve the usefulness (P), and 0.05 max sulfur (S). Panels for the impact test were of the guide to TSMCs in several basic areas. These were the A36 steel with an actual analysis of 0.16 C, 0.84 Mn, 0.004 S, following: 0.010 P, 0.04 silicon (Si), 0.29 copper (Cu), 0.11 nickel (Ni), 0.08 chromium (Cr), 0.005 vanadium (V), 0.002 cobalt Tests on sealer materials, including high solids, high- (Cb), 0.030 molybdenum (Mo), 0.032 aluminum (Al), and penetration epoxies, and urethanes. 0.034 titanium (Ti). Other test panels were made from ASTM Testing of the effects of different abrasive mixes on the A569 steel having an actual analysis of 0.07 C, 0.46 Mn, performance of TSMC to examine the effects of angu- 0.007 P, 0.004 S, and 0.02 Cu. larity on performance and to examine methods to mea- Test panels were prepared using a Metco wire-arc appa- sure angularity in the field. ratus to evaluate surface contamination and alloy and hard- Evaluation of the spray parameters of standoff dis- ness effects at Corrpro's Ocean City, New Jersey (OC) lab- tance and application angle on coating microstructure oratory facility. Grade A36 steel was used for most of the and performance. testing, and ASTM A572 Grade 50 steel was used in the Testing of the effects of the hardness of the steel sub- hardness comparison tests between A36 and Grade 50. strate on surface preparation requirements. Figures 3 and 4 show the panels being prepared at both Testing of the effects of high-strength, low-alloy steel facilities. Figure 5 shows the complex test panel used in versus carbon steel substrate on TSMC performance. the corrosion testing. Aluminum TSMC Zinc TSMC Figure 3. TSMC application at CSI.

OCR for page 11
13 Aluminum TSMC Zinc TSMC Figure 4. TSMC application at Corrpro's laboratory. Surface Preparation gated the impact of such practices by metallizing over various surface roughness conditions. The most common technique Unless otherwise specified, the test panels for this study for determining angularity compares magnified images of were prepared using 100-percent G-16 steel grit. In all cases, the surface to standard photomicrographs. Part of this work the surface finish was SSPC-SP-5 white metal with a target included tests to try to identify a field-friendly method of quan- profile of 3 mils (76m). titatively measuring angularity. Table 4 lists the abrasive This study also explored the effects of grit-to-shot ratio on mixes tested using aluminum and zinc thermally sprayed coat- surface profile and coating performance using aluminum and ings. The surfaces were prepared to an SSPC-SP-5 white zinc TSMCs. This is important because most current stan- metal finish. dards and guidance documents specify the use of "angular" abrasives to obtain the required surface profile; thus an angu- lar profile is expected. A high degree of angularity is impor- TSMC Application tant because most of the debonding stresses acting on the coating are shear forces, and an angular surface provides more TSMC application was performed with wire-arc spray surface area for the coating to adhere to. However, many steel equipment using standard parameters for the application fabricators employ recycled steel shot as the preferred (and equipment. Test panels to evaluate adhesion, sealers, surface economical) method of surface preparation and often use preparation parameters, edge effects, and application para- mixed shot and grit in order to reduce equipment wear. Vary- meters were prepared by CSI Coatings. Test panels prepared ing levels of angular profile may result. This work investi- to evaluate surface contamination and alloy and hardness effects were prepared at Corrpro's laboratory facility. Dur- ing this application, a target film thickness of 10 to 12 mils (254 to 305 m) was specified. The standoff was nominally BOLT & WASHER 8 to 10 in. (20 to 25 cm), and the gun angle was 90 degrees (COMPLEX SHAPE, from the sample. EDGES, CREVICE) Testing Program WELD SCRIBE Table 5 shows the test plan and Table 6 shows the tests applied to the various objectives of the test program. CHANNEL, WELDED TO PANEL Quality Assurance Testing EDGE TREATMENTS (SHARP + 3) After surface preparation, quality assurance testing was Figure 5. Test coupon used for corrosion testing. conducted on representative samples. This included visual

OCR for page 11
14 TABLE 4 Blast procedures investigated Grit/Shot Steel Shot Steel Grit Rationale Shot Blast 100% S-280 Negative Control. Grit Blast 100% G-16 Positive Control. Observed in shop for TSMC project Shot/Grit Mix 33% S-280 67% G-16 in North Carolina. Alternate Shot/Grit Test the profile provided by a "low- 70% S-280 30% G-16 Mix grit" mixture. TABLE 5 General test matrix Specification Test Coupon Size1 Comments1 Reference Microstructure 1 in. x 3 in. x ANSI/AWS A5.33- Porosity, segregation, oxides, and Analysis 0.125 in.2 98 sealer penetration. ASTM D4541 Tensile (Pull- "Pull-Off Strength Requires a 1520 mil (375500 Off) Adhesion 4 in. x 6 in. x 0.125 in. of Coatings Using m) TSMC thickness to prevent Portable Adhesion adhesive from reaching substrate. Testers" ANSI/AWS C2.18- Coated coupons are deformed 180 Bend Adhesion 2 in. x 4 in. x 93 around a 0.5-in. (13-mm) mandrel 0.063 in. MIL-STD-2138A and inspected for cracking and delamination. Alternate Wet/Dry 4 in. x 6 in. x Representative of splash and tidal Seawater 0.125 in.3 zone exposure. Immersion Full Seawater 4 in. x 6 in. x Representative of complete Immersion 0.125 in. immersion conditions. Modified ASTM Drop Weight D2794, "Resistance A 16-lb spherical weight is Impact of Organic dropped onto the panel. Height 6 in. x 12 in. x will be increased as necessary Coatings to the 0.25 in. using a longer guide tube. This Effects of Rapid Deformation method has higher impact energy (Impact)" than ASTM D2794 provides. NOTE: 1 in. = 2.54 cm. 1 The tests will be replicated three times except for the corrosion tests, which will be replicated two times. 2 One sample from the beginning of the coating application run, one from the middle, and one from the end. 3 Special panel containing crevice, scribe, fastener, and edge treatments--see Figure 5. TABLE 6 Tests applied to the objectives TEST Thickness Corrosion adhesion adhesion structure Testing Tensile Micro- impact Profile weight Bend Drop Objective Sealers X X X X X X X Abrasive mixes X X X X X X Spray parameters X X X X X Effect of steel hardness X X X X HSLA1 steel v. carbon steel X X2 Edge geometry effects X X X X Coating defects X X X Surface contamination X X X X 1 HSLA = high-strength, low-alloy. 2 Existing coupons from a previous study and laboratory polarization tests used to evaluate.

OCR for page 11
15 inspection, surface profile evaluation, and chloride contami- nation. Methods for these evaluations are discussed below. Visual Inspection for Surface Quality. Visual inspection of the surface was made in accordance with the Society for Pro- tective Coatings (SSPC) Standard VIS-1-89. The appearance of the prepared surface was compared to the visual standards to determine if it conformed to an SSPC-SP-5, "White Metal Blast Cleaning," condition. Surface Profile Evaluation. The target surface profile was 3 mils (76 microns). Profile evaluation was performed on all samples for the 100-percent shot, 70/30-percent shot/grit, and 33/66-percent shot/grit abrasives. Selected 100-percent- grit abrasive samples were tested. Surface profile was evalu- ated using two methods. Initial measurements were made using Testex brand replica tape. This tape is placed over the Figure 6. Testex tape to evaluate surface profile. blasted substrate and rubbed in place to create an impression of the surface profile. A micrometer is then used to determine Coating Thickness Measurements. DFT measurements were the overall profile (peak-to-valley height) of the surface. This made on samples after preparation and cure (after cooling is the most commonly used field technique to evaluate sur- for TSMC samples and a minimum of 7 days after sealer face profile. Figure 6 shows this measurement. coats were applied). Coating thickness measurements were The second method was the use of a surface profile gauge made using an Elcometer 345 eddy current thickness gauge to determine the profile of the blasted surface. Two gauges (SSPC-PA [paint application] Type 2 gauge). Before thick- were used on the basis of their availability during sample ness measurements were made, the Elcometer 345 thickness preparation. Samples prepared by CSI Coatings were evalu- gauge was calibrated for measurement over a blasted surface. ated using a Perthometer MP4 150 profilometer. Samples Using a representative steel panel blasted to an SSPC-SP-5 prepared at the Ocean City laboratory were evaluated using condition and a 3-mil (76.2-m) surface profile, calibration a Mitutoyo SJ-201 surface roughness gauge. Both models are was performed using standard plastic "shims" of known field usable and capable of measuring various aspects of the thickness that bracketed the expected coating thickness. This profile, which are shown in Table 7. Figure 7 shows the two calibration was performed daily. gauges used to evaluate the profile of these samples. Calibration thickness measurements were made on each Both surface profile gauges use a stylus on a linearly dis- test sample. Typically five measurements per side were placed moving head to measure surface profile characteris- made on all test samples with the exception of the 4- 6-in. tics. This stylus follows the contour of the substrate, mea- (10.2- 15.2-cm) complex samples (8 measurements per suring peak height, valley depth, and the variations of these. side and 16 measurements in total were made on these pan- Both profilometers were calibrated before use, and the same els). Measurements were taken at consistent locations with individual performed the profile measurements at both loca- each type of panel. tions. Both instruments are relatively operator independent. The thickness ranges of the TSMCs applied by CSI coat- These measurements and their statistical manipulation are ings were 12.9 to 20.8 mils (327 to 528 m) for zinc, 14.8 to used to calculate the values shown in Table 7. 22.9 mils (376 to 582 m) for aluminum, and 14.7 to 19.5 mils TABLE 7 Surface profile characteristics NAME ABBREVIATION DESCRIPTION Arithmetic mean The average of the absolute value of the RA deviation height or depth for all measurements. Root-mean-square The square root of the average of the RQ deviation squared absolute height or depth value. The sum of maximum height and depth Maximum profile height RY over a given area. The sum of the mean of the five highest 10-point height RZ peaks and five lowest valleys over a given irregularities area. The number of peaks above a specified Peak count RPC threshold limit from the mean.

OCR for page 11
16 Perthometer Mitutoyo SJ-201 Figure 7. Surface profile gauges. (373 to 495 m) for zinc/aluminum. The thickness ranges of (0.25-MPa) standard deviation for zinc and 1,514 psi (10.4 the TSMCs applied to the A36 and Grade 50 panels at Ocean MPa) with a 263-psi (1.81-MPa) standard deviation. City were 9.9 to 11.8 mils (251 to 300 m) for zinc and 12.3 to 14.2 mils (312 to 361 m) for aluminum. Mandrel Bend Adhesion Test. Mandrel bend testing is used to determine the flexibility and adhesion of a coating mater- Tensile (Pull-Off) Adhesion. Pull-off adhesion testing was ial. For liquid coatings, mandrel bend results for a "good" performed on selected test samples. Adhesion is commonly coating typically have minimal or no cracking because of used to monitor coating quality. MIL-STD-2138A specifies their inherent flexibility. However, TSMCs are more rigid. that a "good" aluminum TSMC should have a minimum adhe- Because of this, mandrel bend test requirements are less sion strength of 1,500 psi (10.3 MPa) for individual samples stringent and allow some cracking of the coating. Figure 9 and 2,000-psi (13.8-MPa) average (MIL STD paragraph shows examples of passing and failure conditions for TSMCs. 5.3.3.3) (4). Adhesion testing measures the bond strength at Mandrel bend testing was performed on the samples indi- the weakest point in a coating system, with both strength cated in Table 5. Testing was conducted on the 2- 4- (stress per unit test area) and failure location reported. 1/16-in. (5- 10- 0.16-cm) samples, which were bent around Adhesion testing was performed in accordance with ASTM a 1/2-in (1.27-cm)-diameter cylindrical mandrel. Samples D4541, "Standard Test Method for Pull-Off Strength of Coat- were bent approximately 180 degrees around this mandrel, ings Using Portable Adhesion Testers." Testing was per- creating a "U" shape similar to that shown in Figure 9. Imme- formed using a Patti Jr. pneumatic adhesion tester. Aluminum diately after testing, samples were evaluated for cracking and pull stubs (dollies) were adhered to the topcoat surface (TSMC disbondment. No disbondment was observed on zinc or alu- or sealer coat) using a two-part epoxy adhesive. Following minum TSMCs prepared by grit blasting. complete cure (24 hours after adhesive application), the pull stub was mounted in the tester, and air pressure was used to disbond the stub from the test sample. This system uses an air bladder to apply an upward (nor- Other Tests mal) force to the pull stub until disbondment or the limit of the apparatus is reached. This apparatus uses an approximate Additional tests were conducted in order to provide input 401 ratio to apply a maximum upward force (normalized to to the objectives of the laboratory program. The specific tests pull-stub contact area) of approximately 4,000 psi (28 MPa) used are listed below. from a 100-psi (0.7-MPa) air source. Figure 8 shows a dia- gram of this apparatus. Once disbondment occurs, the air pressure is recorded along with the location of failure for Falling Weight Impact comparative analysis. The air pressure is then converted into adhesion strength from tabulated values. The falling weight impact test was performed to determine The tensile adhesion strength of TSMC on a 100-percent the ability of the coating to resist damage caused by rapid grit-prepared surface was 895 psi (6.2 MPa) with a 36-psi deformation (impact). Testing was performed both with and

OCR for page 11
17 Figure 8. Pneumatic adhesion test apparatus. without a sealer on aluminum and zinc TSMCs. For the test, Microstructure Analysis (Metallography) a 12.5-lb (5.67-kg) steel ball (weight) was dropped from suc- cessive heights under natural gravitational acceleration at sea Microstructure analysis was performed on untested and level (32.2 ft/s [9.81 m/s]), through a 15-ft (4.6-m) guide post-simulation test samples as identified in Table 5. This tube, onto the test panel, which was place horizontally. analysis was performed using visual microscopy (metallog- During this test, the 12.5-lb (5.67-kg) weight was dropped raphy) to determine from varying heights. After each impact, the panel was inspected for signs of coating (TSMC and/or sealer) penetra- Porosity--size, distribution, geometry, and intercon- tion. Testing was continued until the height of the drop (to nection; the nearest 6 in. [15.2 cm]) at which the coating just resisted Compositional phases--number present; Oxide inclusions--number, size, and distribution; and penetration by the weight was determined. Five replicate Coating substrate interface characteristics--trapped grit, tests were performed at this height to confirm the failure end point. The total energy ft-lb (N-m) that the coating could disbondment, and so forth (general evaluation on untested withstand without penetration was reported. Figure 10 shows samples, local characteristics adjacent to the scribe on post-test samples). this test apparatus. Lehigh Testing Laboratories, Inc. (New Castle, Delaware) performed the microstructure analysis. Data were generated using photomicrographs, visual observations, and the point- count method for porosity and oxide distributions. The samples were examined in their unetched state for porosity and oxide evaluation and in their chemically etched state to show the pres- ence of multiple phases. This did not identify the chemistry of such phases, but showed whether one or more different phases were present. Two samples from each test were examined. Corrosion Tests Laboratory tests consisting of alternate wet-dry seawater Figure 9. TSMC mandrel bend pass/fail examples. exposure and constant immersion were performed to evaluate

OCR for page 11
18 Sketch Test Apparatus Figure 10. Falling weight impact test apparatus. the sealers, surface preparation, and application variables in seawater in this tank was continually refreshed using a trickle this program. On the basis of results from previous studies, (quiescent) flow from the intake system. it is recognized that a short-term exposure test may be inad- During this test, periodic inspections (nominally every 3 equate to differentiate the performance of TSMC/sealer sys- months) were made to evaluate performance. This included tems. Thermally sprayed coating systems may be exposed to evaluations for substrate corrosion (rusting) in accordance harsh environments for several years without exhibiting sig- with ASTM D610, coating blistering in accordance with nificant levels of corrosion. ASTM D714, formation of corrosion products on the samples, Natural seawater immersion testing was used to evaluate the and visible cutback from the intentional holidays. The test performance of TSMC and other preparation variables in nat- methods used for these evaluations are presented in Table 8. ural waters. Testing was conducted at Corrpro's Ocean City, For analytical purposes, the ASTM D714 rating is converted New Jersey, facility using natural seawater obtained from the to a composite blistering rating. On the basis of the size and Inland Intracoastal Waterway adjacent to Corrpro's facility. density of the blisters, a numerical rating from 0 to 10 is given Seawater is pumped through this facility in an open-loop, to the sample. Table 9 shows this composite blister index. once-through system. There are provisions for the filtration of Figure 5 illustrates the type of panel used. The scribe was large debris and biological growth; otherwise, the seawater a diagonal line cut through the metallized coating with a contains all chemicals naturally found at this location. hardened steel tool with a sharp point to ensure that the steel Test samples were placed in a non-metallic (plastic) tank substrate was exposed. The panel edges were used to exam- and held in place with plastic holders. Samples were oriented ine the effect of different edge treatments. at 90 degrees from horizontal and completely submerged in After sample preparation and sealer cure, an intentional the natural seawater environment. To avoid stagnation, the scribe (removal of all coating materials to the steel substrate) TABLE 8 Coating deterioration inspection techniques Evaluation Test Method Description Evaluation of percent corrosion on a test sample by Substrate corrosion ASTM D610 comparison with visual standards (0 to 10 scale, 10 = no corrosion). Evaluation of blister size and frequency on a test sample by comparison with visual standards (0 to 10 for size, 10 Coating blistering ASTM D714 = no blistering; for frequency, F = few, M = medium, MD = medium dense, D = dense). Visual observation for corrosion at the intentional Corrosion products N/A scribes, along edges, in crevices, at welds, general deterioration and other observations. Measurement of visible coating (TSMC or sealer) Cutback from Modified disbondment from intentional holidays evidenced by holidays ASTM D1654 disbondment, blistering, or rusting. Measurements made in millimeters.