Spot Painting to Extend Highway Bridge Coating Life: Volume 2: Research Overview (2018)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

NCHRP Project 14-30 32 CHAPTER 3 Field Spot Painting Work 3.1 Background To address the fourth element of the KTC research, candidate bridges were sought for field spot painting. This work employed processes presented in the guidance document including coating assessments, the SSPC-SP 3 “Power Tool Cleaning” specification (using vacuum shrouded tools to reduce airborne particulate exposure), and conformance with worker safety and environmental regulations. The field work incorporated spot repairs/test patches using the same experimental “repair” coatings/systems employed in the KTC laboratory test program. The field testing was necessary to validate the KTC approach to spot painting that employed dry wiping in place of pressure washing to eliminate the need for waste water collection and disposal, and the extra time needed for washed substrates to dry. The field work was to address typical coating failures on highway bridges that were potential candidates for spot coating repairs. 3.2 Bridge Selection and Field Coating Assessments The Kentucky Transportation Cabinet (KYTC) assisted this research effort by providing KTC researchers access to its bridges for performing field spot painting tests. KTC initially inspected six bridges to find several candidates that might match the KTC test requirements. Three bridges were selected for the spot painting tests, the KY 9002 Bluegrass Parkway Bridge over US 60 and the KY 922 Newtown Pike Twin Bridges over KY 4 both in or near Lexington, Kentucky (Figures 18 and 19). Figure 18. Bluegrass Parkway Bridge over US 60 Near Lexington, KY.

NCHRP Project 14-30 33 3.2.1 KY 9002 Bluegrass Parkway Bridge over US 60 The Bluegrass Parkway Bridge was constructed in the mid-1960s. The bridge was originally painted with the basic lead silico-chromate alkyd primer and leafing and non-leafing aluminum alkyd intermediate and finish (top) coats respectively. The bridge was experimentally overcoated in 1993 using an aluminum-pigmented moisture cure urethane spot primer over hand-tool cleaned rust/mill scale with two full overcoats consisting of a tie-coat of two- component polyurethane and a high-gloss aliphatic polyurethane finish coat. There were leaking deck joints at the abutments and two piers resulting in corrosion of the beams and cross-bracing at the end of the spans. In 1991, prior to the overcoating work, KTC performed extensive adhesion tests on the existing alkyd coatings and found them to be poorly adhered to the mill scale substrate with the lead alkyd primer also prone to low-strength cohesive failures. At that time, the adhesion test values for the alkyd coating system prior to overcoating varied between about 50 to 250 psi. The overcoating remained in very good condition for about 20 years with a few areas of localized coating failures and corrosion re-occurring at the beam ends under the deck joints. KYTC efforts to keep the joints sealed during that period were only partially successful and several joints continued to leak, pouring deck runoff onto the beams (Figure 20). The bridge began to experience localized sheet-type disbonding failures at random areas throughout the structure from 2013-2015 (Figure 21) due to cold winters. At those locations, local failures consisted of the entire coating disbonding down to the mill scale. This bridge was selected as it exhibited: 1) a brittle coating that was not amenable to repeated overcoating, 2) disbonding failures that could possibly be arrested by locking down the existing coating at the failure boundaries, and 3) corrosion under leaking deck joints where significant soluble salt concentrations were possible from deicing salt runoff. While this bridge might be a better choice for total coating removal and replacement, if funds for that work were not immediately available, spot painting could be used as a stopgap treatment to prevent corrosion damage until the bridge could be completely repainted. Figure 19. KY 922 Southbound Bridge over KY 4 in Lexington, KY.

NCHRP Project 14-30 34 3.2.2 KY 922 Twin Bridges over KY 4 The KY 922 twin overpass bridges were continuous multi-span bridges with joints over the abutments. The bridges were built in the 1960s. In 1989, they were blast cleaned and repainted with an inorganic zinc primer with a vinyl topcoat. That system was beginning to deteriorate with rusted patches on the exterior faces of the fascia girders due to salt- contaminated aerosols kicked up by traffic under the bridges. At the abutments, the beam ends and diaphragms were corroding due to leaking deck joints. That deterioration was accentuated by rust bleeding from the deck joints (Figure 22). From the visual inspection, these bridges appeared to be good candidates for spot painting or overcoating (including localized overcoating of the exterior faces of the fascia girders if spot painting was to be used elsewhere). 3.2.3 Field Coating Assessments Prior to that work, initial field assessments of existing coatings on the subject bridges were performed by KTC researchers using procedures identified in the literature review (Figure 23). Those tests included measurement of existing coating thicknesses using a Tooke gage (per ASTM D4138 “Standard Practices for Measurement of Dry Film Thickness of Protective Coating Systems by Destructive, Cross-Section Means” ) and a magnetic gage (per ASTM D7091, “Standard Practice for Nondestructive Measurement of Dry Film Thickness of Nonmagnetic Coatings Applied to Ferrous Metals and Nonmagnetic Nonconductive Coatings Applied to Non- Ferrous Metal,”) , measurement of coating adhesion (per ASTM D3359 “Standard Test Methods for Rating Adhesion by Tape Test” using an X-cut) and (per ASTM D4541, “Standard Test Method for Pull-Off Strength of Coatings Using Portable Adhesion Testers,”), soluble salts (per Figure 20. Corrosion on Beam Ends and Cross- Bracing under Leaking Deck Joint on the Bluegrass Parkway Bridge at Pier 1 (Area 2). Figure 21. Disbonding of Existing Coating on Girder near East Abutment (Area 1). Figure 22. Corrosion and Rust Staining on Beam Ends and Diaphragm at the KY 922 Southbound Bridge South Abutment.

NCHRP Project 14-30 35 SSPC Technology Guide 15: “Field Methods for Extraction and Analysis of Soluble Salts on Steel and Other Nonporous Substrates,” using the patch and sleeve methods - Methods 4.2.2 and 5.2.5 respectively). Heavy metal analysis was performed using a portable dispersive x-ray fluorescence (XRF) spectrometer (Figure 24). On the Bluegrass Parkway Bridge, tests were performed near the east abutment on three beams and on beam ends at the west pier. The test results are provided in Tables 6 and 7. For the KY 922 bridges, eight test locations were designated with Test Area 1 on the northbound bridge at beam ends on the south abutment and Test Areas 2-8 on the southbound bridge at the south abutment. Test Areas 1, 3, and 7 were given coating tests as provided in Table 8. Table 6. Coating Assessment for Beams 1, 3 & 4 Near East Abutment on KY 9002 Bluegrass Parkway Bridge over US 60. Test Method Maximum Test Value Minimum Test Value Existing Coating Thickness (ASTM D4138) Primer-3.5 mils (88 microns), Intermediate-3.5 mils (88 microns), Finish Coat-5 mils (125 microns)* Primer-3 mils (75 microns), Finish Coat-4 mils (100 microns)* Existing Coating Thickness (ASTM D7091) 12.9 mils (323 microns) avg. 9.1 mils (228 microns)avg. Soluble Salt Test (SSPC- Guide 15 Method 4.2.2) 57 µS/cm 46 µS/cm Soluble Salt Test (SSPC- Guide 15 Method 5.2.5) 40 µg/cm2 15 µg/cm2 Tape Adhesion (ASTM D3359) 2A 1A Tensile Adhesion (ASTM D4541) 550 psi (3.79 MPa) Failure location not recorded 450 psi (3.10 MPa) Adhesion between Primer and Topcoat XRF Reading for Heavy Metal Pb 64.46%** *Could not identify individual coats readily **Percent by weight of all elements analyzed. Figure 23. Coating Assessment Area on the Bluegrass Parkway Bridge at Pier 1. Figure 24. XRF Testing the KY 922 Bridge Coating for Lead.

NCHRP Project 14-30 36 Table 7. Coating Assessment for Beams 1, 2 & 3 Over Pier 1 (West) on KY 9002 Bluegrass Parkway Bridge over US 60. Test Method Maximum Test Value Minimum Test Value Existing Coating Thickness (ASTM D4138) Primer-5 mils (125 microns), Finish Coat-5 mils (125 microns)* Primer-1.5 mils (38 microns), Intermediate-3 mils (75 microns), Finish Coat-4 mils (100 microns)* Existing Coating Thickness (ASTM D7091) 12.9 mils (323 microns) avg. 9.1 mils (228 microns) avg. Soluble Salt Test (SSPC- Guide 15 Method 4.2.2) 99 µS/cm 34 µS/cm Soluble Salt Test (SSPC- Guide 15 Method 5.2.5) 40 µg/cm2 0 µg/cm2 Tape Adhesion (ASTM D3359) 1A 1A Tensile Adhesion (ASTM D4541) 656 psi (4.52 MPa) 40% Cohesion-Primer; 60% Adhesion between Primer & Intermediate Coat 448 psi (3.09 MPa) Adhesion between Primer and Finish Coat XRF Reading for Heavy Metal Pb 64.46%** *Could not identify individual coats readily **Percent by weight of all elements analyzed. Table 8. Coating Assessment for Test Areas 1, 3 & 7 Near the South Abutment on KY 922 Bridge over KY 4. Test Method Maximum Test Value Minimum Test Value Existing Coating Thickness (ASTM D4138) 10 mils (254 microns) Total* 8 mils (200 microns) Total* Existing Coating Thickness (ASTM D7091) 11.5 mils () avg. 5.6 mils () avg. Soluble Salt Test (SSPC- Guide 15 Method 4.2.2) 187 µS/cm 46 µS/cm Soluble Salt Test (SSPC- Guide 15 Method 5.2.5) 15 µg/cm2 7 µg/cm2 Tape Adhesion (ASTM D3359) 5A 5A Tensile Adhesion (ASTM D4541) 254 psi (1.75 MPa) 100% Cohesion in the Finish Coat 174 psi (1.20 MPa) 100% Cohesion in the Finish Coat XRF Reading for Heavy Metal Pb 0.20%** 0.11%** *Could not identify individual coats readily **Percent by weight of all elements analyzed. The field coating assessments indicated that the existing coating thicknesses for both bridges were on the order of about 10 mils and excessive coating thicknesses would not be a factor. The tape adhesion tests per ASTM D3359 indicated that the original alkyd primer on the Bluegrass Parkway Bridge was brittle. However, the tensile adhesion testing values were greater than 450 psi indicating that application of the spot coatings would probably not result in disbonding. Adhesion properties on the KY 922 Bridges did not appear to pose a problem for spot painting. Most of the soluble salt readings were relatively low despite the presence of corrosion under the deck joints at the piers and abutments of the two bridges. Based upon KTC tests, it was

NCHRP Project 14-30 37 determined that in most of the spot painting areas, the residual soluble salt levels would be sufficiently low to mirror those used in the KTC laboratory tests. The XRF testing revealed that lead was the only heavy metal present in the three bridges. The Bluegrass Parkway Bridge was known to have a lead primer that might cause the waste paint residue from power tool cleaning to be a hazardous waste. The coatings on the two KY 922 bridges contained lead, probably from the zinc dust used in the inorganic zinc primer applied in 1989. From the XRF tests, the lead concentrations on the KY 922 Bridges were probably too low to result in the creation of hazardous wastes, but the concentration of lead in the coating system would require protection for and monitoring of workers involved in surface preparation activities. Therefore, all three bridges were treated as deleading projects. The KYTC Division of Environmental Analysis was contacted for assistance with coating disposal issues. All three bridges showed signs of localized corrosion in beam end areas under deck joints. However, none of the corrosion appeared to be sufficient to pose concerns about significant section loss of the steel. 3.3 Field Work Protocol Based upon the field assessments, a work protocol was developed for applying the spot painting tests. This protocol was similar to work standards a state highway agency would prepare for in-house spot painting or specifications/special notes for contractor work. It was based on procedures provided in a companion document “Guidelines for Spot Painting to Extend Highway Bridge Coating Life” (Hopwood et al. 2018). KTC technicians were provided with a list of tasks with general guidance on executing them. Additionally, the KTC principal investigator reviewed those items in detail with the technicians and periodically visited the bridges to inspect the work in progress. 3.3.1 Selection of Test Areas KY 9002 Bluegrass Parkway Bridge Tests • Fly over from west US 60 to west BG Parkway, 3 spans, approximately 230 ft. (70.1 m) long with four 48-inch (122 cm) deep beams. • Eight repair locations were selected at the east abutment (Area 1) that addressed the disbonded areas on the web portion of the beams. • Eight repair locations were selected at the west Pier 1 that included a combination of beam ends, gusset plates, and angle bracing of the diaphragms that addressed corroded areas (Area 2). • Each location had a repair area selected with a minimum of 2 ft2 (0.19 m2). KY 922 (Newtown Pike) over KY 4 (New Circle Rd) • Twin bridges, 4 spans, approximately 220 ft. (67.1 m) long with seven 30-inch (76 cm) deep beams per span. • Eight repair locations were selected at the southern abutments of the two bridges that included a portion of the beams and a portion of the diaphragm (Area 3). • Location 1 was at the southern abutment of the north bound bridge. • Location 2 through 8 were at the southern abutment of the southbound bridge. • Each location had a repair area selected with a minimum of 2 ft2 (0.19 m2).

NCHRP Project 14-30 38 3.3.2 KTC Training Relative to the Field Work The KTC technicians assigned to the spot painting tests had the following training: OSHA 10- Hour Construction Training, lead awareness and fall protection training, SSPC PCI Level I coating inspector (one technician) and man-lift training. The principal investigator had taken the SSPC C3 Supervisor/Competent Person Training for Deleading/Hazardous Coating Removal on Industrial and Marine Structures, but had not maintained that certification. KTC researchers contacted the KYTC Division of Environmental Analysis which assisted with the disposal of the coating residues generated on the project. 3.3.3 Field Spot Painting Work The field work was performed as follows: • Proper personal protection equipment was worn by KTC technicians performing the field tests. They had mask-fit testing and were provided with half- and full-face respirators as required. They were provided with disposable full body suits and booties, protective eyewear and hearing protection. Hand washing provisions were provided in the form of a portable sink, a potable water source and hand soap. • Ground tarps were used in the work areas to contain debris removed from the bridge. • All hand-held power tools used to provide the SSPC-SP 3 surface preparation were shrouded and attached to vacuums equipped with HEPA filters. KTC possessed an electric needle gun, a rotary flapper and a 7-inch (17.78 cm) right angle sander and several small 1-gallon (3,785 cm3) capacity electric vacuums equipped with HEPA filters to be used with single shrouded power tools. Several pneumatic and electric power tools were obtained on loan from a manufacturer of power tool equipment. Those tools included: a pneumatic a 2-inch (5.08 cm) grinder, two 7-inch (17.78 cm) right angle sanders (electric and pneumatic), a pneumatic rotary flapper and a pneumatic needle gun. KTC ordered an extra set of needles for both the electric and pneumatic needle guns and extra non-woven Clean ‘N Strip™ type discs for use with the grinder and sanders. • Each area was cleaned to: SSPC-SP 1, “Solvent Cleaning” (where necessary), SSPC- SP2, “Hand-Tool Cleaning” to remove peeling paint, and SSPC-SP 3, “Power Tool Cleaning.” Surfaces to be painted were wiped with burlap prior to painting. A solvent was provided for cleaning but was not needed at the test locations. Existing paint at the boundaries were to be feathered approximately one to two inches using the 2-inch (5.08 cm) grinder equipped with a non-woven pad. • At each general location, ambient conditions were monitored and the coatings were not applied until proper conditions were met, with the surface temperature of the steel at least 5o F (2.8o C) above the dew point. The experimental spot coatings were applied per manufacturer’s recommendations. • KTC researchers took pictures of representative paint-related activities on both bridges KTC researchers also took digital video-recordings of the use of the power tools during surface preparation. • KTC researchers performed wet film tests during application of the six liquid-applied coatings and the grease to ensure proper coating thicknesses and make necessary repairs while the coatings were in a wet state. KTC arranged with KYTC for the disposal of the lead paint residue, used disposable clothes and respirator filters from the Bluegrass Parkway Bridge. Other wastes generated

NCHRP Project 14-30 39 including any solvents were disposed of by KTC as part of its routine in-house painting procedures. KTC researchers determined that testing of the Bluegrass Parkway Bridge at the West Pier 1 required the use of a man-lift. To perform the work, KTC rented a 45-foot (13.72 m) articulating man-lift for $550/week plus a delivery and pick-up charge to the bridge of $310 each way. A 185 CFM (5.24 CMM) @ 100 psi (0.69 MPa) diesel-powered portable air compressor was also rented for the project at $325/week including hoses. The compressor needed to be equipped with a water separator to be used with the power tools. KTC had a 6,500-Watt portable generator that was used to supply electric power for mixing and use of the electric tools and vacuums. Cans, brushes, solvent, disposable suits, trash bags, tarps, air filters, power tool consumables, hand tools and other miscellaneous supplies all came to less than $1,000 for all three bridges. KTC researchers used two pick-up trucks to take the equipment from the KTC laboratory to the job sites daily. The KYTC District Office in Lexington supported the research by providing a parking lane- closure on the westbound traffic lane of US 60 under the Bluegrass Parkway bridge during the work days from about 8:30 am to 3:00 pm allowing sufficient time for surface preparation and painting of Test Area 2. Work on the Bluegrass Parkway Bridge began in early October, 2015 with follow-up work on the KY 922 Twin Bridge worked started the following week. On the Bluegrass Parkway Bridge, the surface preparation work was initiated on beams near the east abutment that had experienced disbonding problems (Area1). The work set up area was located to the side of the road behind the east Pier 2. It was completely blocked from traffic by a guardrail running between the pier and the roadway. Most of the disbonding spots on Area 1 occurred in sheltered portions of beams under the deck near the abutment. The test areas chosen were readily accessed from the embankment without the need for ladders or scaffolding. Electrical cords and pneumatic hoses were run from the truck-mounted generator and towed compressor to power the vacuums and tools to the work sites. KTC technicians fabricated a manifold for the compressor, enabling the crew to operate two pneumatic power tools concurrently. Prior to the onset of surface preparation, ground tarps were laid to collect any paint chips that fell during the surface preparation. Peeling paint was removed with hand scrapers prior to power tool cleaning and the paint chips were collected daily (Figure 25). The test locations were marked to identify where the specific repair coatings were to be used. During surface preparation, KTC encountered a problem with poor adhesion and brittle coating failures that resulted in areas requiring painting that were 3-5 times larger than originally Figure 25. A Ground Tarp Placed under the Bluegrass Parkway Area 1 to Collect Leaded Paint Chips.

NCHRP Project 14-30 40 anticipated. The progress of the power tool cleaning was rapid in these areas as there was little surface rust to be cleaned and the existing coating readily came off in most areas. Most of the corners, stiffeners and edges of the flanges were power-tool cleaned using electric or pneumatic vacuum-shrouded needle guns (Figure 26). The flatwork on the webs was typically power-tool cleaned using pneumatic or electric 7-in (17.78 cm) vacuum shrouded right angle sanders with non-woven pads (Figure 27). A 2-inch (5.08 cm) vacuum shrouded pneumatic grinder was also used for cleaning in smaller areas such as stiffeners and faces and edges of flanges using a non- woven pad. After the cleaning, the boundaries of the existing coating were probed with dull scrapers to ensure that the surface preparation extended into adherent existing coating. Then, the edge of the existing coating was feathered using the 2-inch (5.08 cm) grinder. That proved much quicker than manually feathering the existing coatings with sandpaper. The prepared surfaces were visually inspected to ensure they were properly cleaned. After the laboratory work, KTC technicians had experience in preparing SSPC-SP 3 substrates and did not rely on the VIS 3 standard to determine when the surface preparation was acceptable (Figures 28 and 29). Prior to coating application, each test spot was thoroughly wiped down using a coarse burlap to remove any residual dust or soils. Figure 26. Surface Preparation to SSPC-SP 3 Standard Using a Pneumatic Needle Gun. Figure 27. Surface Preparation to SSPC-SP 3 Standard Using an Electric Sander. Figure 28. Area 2 Test Location on the Bluegrass Parkway Bridge with SSPC-SP 3 Surface Preparation. Figure 29. Test Location on the KY 922 Southbound Bridge with SSPC-SP 3 Surface Preparation.

NCHRP Project 14-30 41 An electronic digital dew point meter was used to measure air temperature, surface temperature of the steel, the relative humidity and the dew point (Figure 30). Once suitable ambient conditions were met, the repair coatings were applied. The liquid-applied coatings were provided in one-gallon kits for the two-component MIO-epoxy, epoxy sealer and two-component polyurethane and one-gallon cans for the single-component alkyd, acrylic and moisture cure urethane coatings. The calcium sulfonate alkyd coating was provided in a five-gallon bucket. It was resealed and coating from the container was used in subsequent applications. The coatings were mixed or agitated at the staging area. None of the coatings were thinned during either the laboratory or field work though a small amount (10%) was recommended for some coatings when brush-applied (Figure 31). The liquid-applied coatings were applied using 3-inch (7.62 cm) brushes. The grease was wiped on by hand using rubber gloves and smoothed out using a scraper (Figure 32). For all of the liquid-applied coatings, WFT measurements were used to ascertain proper film thicknesses during application. Those were to conform to recommended WFT values in the coating manufacturers’ product data sheets. DFT readings were subsequently obtained using a Tooke gage. The grease was applied at an approximate thickness of 20 mils (500 microns). The tape was supplied in a 9-inch (22.86 cm) wide roll. Due in part to the large repair area created by the SSPC-SP 3 surface preparation, KTC technicians ran short on tape. They were unable to get additional material in time for the field work and substituted a similar (green) product made by the same manufacturer (Figure 33). Both tapes were applied in a manner similar to hanging wall paper. They were self-adherent on one face with protective cover that was peeled off prior to application. Figure 30. Dew Point Meter Used to Monitor Ambient Conditions on the Bluegrass Parkway Bridge Prior to Coating Application. Figure 31. KTC Technician Brushing an Alkyd Coating on the KY 922 Southbound Bridge.

NCHRP Project 14-30 42 Measurements were taken to determine the widths of the repair area to be covered with each application of the tapes. The tapes were cut into strips and placed over the repair area lapping the existing coating at the repair boundary by about two inches. Care was taken to completely press down the tape strips to eliminate air pockets between the tapes and the substrates and bond the overlapping strips together. While the tapes did not pose any significant application issues, the beige material picked up grime from handling. The same problem was encountered in applying the material in the laboratory on test specimens. While that detracted from its cosmetic appearance, that did not impact its overall performance in terms of corrosion protection. Problems were encountered in applying the acrylic 2-coat system (Coating 7). Rust bleed occurred through the primer as had been encountered in the laboratory tests during application of that coating system on the test panels (Figure 34). Figure 34. Rust Bleed Problem with the Direct-to-Metal Acrylic Primer. Figure 32. KTC Technician Applying Grease Coating with a Glove before Smoothing It out with a Scraper. Figure 33. Green Tape Used to Supplement Beige Tape Due to a Shortage.

NCHRP Project 14-30 43 In addition to the spot coating repairs performed under the deck, a spot repair was made to a badly peeling area on the external face on a fascia girder near the east abutment (about 20 ft2). Surface preparation was performed using a pneumatic rotary flapper for the flat work and needle gun for the corners. The area was painted using the calcium sulfonate alkyd applied with a small battery-powered airless hand sprayer. The sprayer applied the coating much faster than brushing to achieve the total required film build using wet-on-wet application which was permitted by the coating manufacturer. The resulting total dry film thickness was measured at 10 mils (250 microns) which was within the coating manufacturer’s prescribed tolerance. Area 2 above the west Pier 1 was located on the other end of the bridge from Area 1. The two work areas were separated by US 60, a four-lane divided roadway. The test areas for spot painting chosen by the KTC technicians were typically located on angle cross-bracing between girders under the deck and attachment gussets or at beam webs/stiffeners riveted/bolted to the cross-braces. The test areas were marked with indelible ink. Those areas were more corroded than those in Area 1 at the east abutment. Surface preparation was slow due to the irregular shapes being cleaned. The existing paint was more adherent in Area 2 and the repair areas were smaller. The surface preparation and coating application processes was the same as that performed on Area 1 with the 8 “repair” coatings applied. In that case, the liquid-applied coatings were mixed in the staging area adjacent to Area 1 and hand carried across the road for application on Area 2 in one-gallon cans. Fir field spot painting tests on the KY 922 Bridges, a staging area was located adjacent to the south abutment of the southbound bridge behind a guardrail. As work was performed under the abutments, no traffic control was required. KTC technicians accessed the work areas on paved berms running up to the abutments. There were no problems in routing the electric chords and air hoses from the staging area to the work sites. The spot painting process was identical to that used on the two test areas of the Bluegrass Parkway Bridge. As previously noted, the test locations were on adjacent abutments of the two bridges. Each test location was in a different bay and consisted of a portion of a beam end that was corroded (on both the web and faces of the upper and lower flanges), the face of a stiffener with fasteners, and about a one-foot length of the outer face of a channel diaphragm connecting the beams in each bay. The 7-inch (17.78 cm) vacuum shrouded angle sanders were used for the flat work on the faces of the webs and channels and vacuum shrouded needle guns were used for the corners, edges and bolted connections. The work was completed without incident and the test areas were painted the following two days. For the tape test, the substitute tape from the manufacturer had to be used as the tape employed for the laboratory tests had been consumed and additional material could not be obtained in time for the completion of the field work. Painting of each bridge took a three-person crew about three days to complete (Figures 35-38). No weather delays were encountered. The sites were cleaned of all waste materials. The waste paint residue from the Bluegrass Parkway Bridge was provided to KYTC for disposal with waste paint debris from another larger project. Other wastes generated during the field tests were collected at the jobsites were collected and properly disposed.

NCHRP Project 14-30 44 During the field work, one KTC technician was fitted with an air monitor to determine his respirable lead exposure during an 8-hour period when conducting surface preparation using needle guns and sanders and grinders. He was monitored working on the three bridges while wearing a half-face respirator by a certified industrial hygienist who routinely performed similar tests for paint contractors working in Kentucky. For the Bluegrass Parkway Bridge, the respirable lead exposure level was determined to be 120 µg/m3 for 8 hours which is above the personal exposure level of 50 µg/m3. With the half-face respirator protection factor of 10, his effective exposure level is 12 µg/m3 which is considered acceptable. For the KY 922 Bridges, the respirable lead exposure level was determined to be 13.44 µg/m3 for 8 hours which would be acceptable without a respirator. However, it is good practice to use one to limit inhalation of nonhazardous airborne particulates generated during surface preparation. After the coatings had cured, DFT measurements were taken of the six liquid-applied “repair” coatings in Area 1 of the Bluegrass Parkway Bridge and the KY 992 Bridges. Those are provided in Table 9. In one case, the epoxy sealer/polyurethane on the Bluegrass Parkway Figure 35. Calcium Sulfonate Alkyd Repair Coating (2 Coats) on Area 1 of the Bluegrass Parkway Bridge. Figure 36. MIO-Epoxy/Polyurethane Repair Coating on Area 2 of the Bluegrass Parkway Bridge. Figure 37. A Moisture Cure Urethane Repair Coating on the KY 922 Southbound Bridge. Figure 38. A Grease Repair Coating on the KY 922 Southbound Bridge.

NCHRP Project 14-30 45 Bridge, had been applied at an excessive paint thickness (the polyurethane top coat). On the KY 922 Bridge, the thickness of the MIO-epoxy primer in the MIO-epoxy/polyurethane was several mils under the manufacturer’s recommended thickness. Table 9. Dry Film Thicknesses for Liquid-Applied Coatings on KTC Field Tests (Topcoat/Primer) Test Area – Coating Types (Topcoat/Primer) Product Data Sheet mils (microns) Bluegrass Parkway Bridge Area 1 mils (microns) KY 922 Bridges mils (microns) Area 1 – Alkyd/Alkyd 2-3/2-5 (50-75)/(50-125) 5 (125) Total 7 (175) Total Area 2 – Polyurethane/Epoxy Sealer 3-5/1-2 (75-125)/(25-50) 7/2 (175)/(50) 4/5 (100)/(125) Area 3 – MCU/MIO-Al MCU 2-3/2-3 (50-75)/(50-75) 2/4 (50/100) 3/2 (75)/(50) Area 4 – Calcium Sulfonate Alkyd (2 coats) 5-6/5-6 (125-150)/(125-150) 12 (475) Total 12 (475) Total Area 7 – Acrylic/Acrylic 2.5-4/2-4 (63-100)/(50-200) 5 (125) Total 7 (175) Total Area 8 – Polyurethane/MIO-Epoxy 3-5/5-10 (75-125)/(125-250) 5/10 (125)/(250) 5/3 (125)/(75) 3.4 Follow-up Evaluations off Field Work Approximately 14 months after the “repair” spot coatings were applied on the Bluegrass Parkway and KY 922 Twin Bridges, KTC researchers went to the bridges to perform follow-up evaluations of the coatings. In Area 1 of the Bluegrass Parkway Bridge, the alkyd 2-coat system (Coating 1) generally was performing well with one small disbonding location. No rust was evident on the test area. The epoxy sealer/polyurethane (Coating 2) also appeared to be in good condition with one small rust spot visible on the web-to-flange fillet weld, but there was no disbonding. The moisture cure urethane 2-coat system (Coating 3) also appeared to be in good condition with no sign of rust or disbonding (Figure 39). The calcium sulfonate alkyd 2-coat system (Coating 4) was performing well with no sign of rust or disbonding. The tape (Coating 5) was intact with no signs of rust or disbonding (Figure 40). The grease (Coating 6) was intact with no signs of rust or disbonding. However, the tacky coating had picked up a significant amount of airborne debris. The acrylic 2-coat system (Coating 7) generally was in good condition with some spot rusting starting at the lower flange, but no signs of disbonding. The MIO-epoxy/polyurethane (Coating 8) was in good condition with no sign of rusting or disbonding. The spot repair to the exterior fascia girder was performing well with a small spot of rust near a vertical stiffener.

NCHRP Project 14-30 46 In Area 2 of the Bluegrass Parkway Bridge, alkyd 2-coat system (Coating 1) generally was performing well with one small rust spot at the bottom of the test patch where there had been significant corrosion pitting prior to the spot painting (Figure 41). The epoxy sealer/polyurethane (Coating 2) also appeared to be in fair-to-good condition with several small rust spots visible on fasteners, the edge of the cross-bracing gusset plate, the web-to-flange fillet weld, and random spots on the face of one angle and the gusset plate where the repair surface had been severely pitted by corrosion. The moisture cure urethane 2-coat system (Coating 3) was performing poorly with rust on fasteners and on a cross-bracing angle and gusset plate. The calcium sulfonate alkyd 2-coat system (Coating 4) was performing well with one small rust spot. The tape (Coating 5) was intact with no sign of rust or disbonding. The grease (Coating 6) was intact with no signs of corrosion or disbonding. However, the tacky grease had picked up a significant amount of airborne debris. The acrylic 2-coat system (Coating 7) was performing poorly with several failed and corroded spots and severe rust bleeding through the test patch in many locations (Figure 42). The MIO-epoxy/polyurethane (Coating 8) was in generally performing good condition with a small rust spot at the bottom edge of the test patch, on the edge of two cross-bracing angles and on the gusset plate. No disbonding was observed on any spot coating test patch in Area 2. On the KY 922 Bridges, alkyd 2-coat system (Coating 1) was in fair condition exhibiting spot failures on the upper flange-to-stiffener fillet weld area, the lower flange, and on the painted portion of the diaphragm. The epoxy sealer/polyurethane (Coating 2) was in better condition than Coating 1, but rust was observed on the lower flange adjacent to the stiffener and rust spotting was visible on the face of the diaphragm especially adjacent to the edge of the stiffener which also exhibited rust spotting (Figure 43). The rust spotting appeared to be related to areas where the steel had been severely pitted. The moisture cure urethane 2-coat system (Coating 3) appeared to be in similar condition to Coating 2. The worst corrosion was on the lower flange adjacent to the stiffener. The calcium sulfonate alkyd 2-coat system (Coating 4) was fair condition with rusting on the lower flange adjacent to the stiffener and on the gusset plate (that also exhibited rust bleed from coating failures at the top of the diaphragm). The tape (Coating 5) was intact with two small spot failures and rust bleed at the end of the test patch. Apparently, the rough surface at this location had been pushed through the tape causing it to fail. Otherwise, no problems were observed with the tape. The grease (Coating 6) was in fair condition with corrosion evident on the lower flange adjacent to the stiffener, along the web-to-flange fillet weld, and on the edge of the stiffener with no signs of corrosion or disbonding. Those appeared to be due to Figure 39. The Moisture Cure Urethane Spot Repair in Good Condition on the Bluegrass Parkway Bridge. Figure 40. The Tape Spot Coating in Good Condition on Area 1 of the Bluegrass Parkway Bridge.

NCHRP Project 14-30 47 cleaning problems in tight areas combined with insufficient thickness of the grease coating in the rusted areas. The grease was tacky and had picked up a significant amount of airborne debris. The acrylic 2-coat system (Coating 7) generally was in poor condition with large-scale failure on the face of the stiffener, the lower flange of the girder and the face of the diaphragm. The MIO- epoxy/polyurethane (Coating 8) was in good condition with rusting on the lower flange of the girder adjacent to the stiffener and along the top of the diaphragm (Figure 44). The coating was disbonding adjacent to one fastener. There were no other signs of disbonding on any of the test patches. 3.4.1 Work Times for Spot Painting Tasks After the field work was completed, the KTC technicians provided the following information/judgements concerning the work performed. In terms of manpower requirements, it took a total of two person-days to conduct the preliminary field coating assessments on the two locations on the Bluegrass Parkway Bridge and one person-day on the KY 922 Bridges. The Figure 41. The Alkyd Spot Coating in Area 2 of the Bluegrass Parkway Bridge in Good Condition. Figure 42. The Acrylic Spot Coating in Area 2 of the Bluegrass Parkway Bridge in Poor Condition. Figure 43. The Epoxy Sealer/Polyurethane Repair Coating on the KY 922 Southbound Bridge in Fair-to-Good Condition. Figure 44. The MIO-Epoxy/Polyurethane Repair Coating on the KY 922 Southbound Bridge in Good Condition.

NCHRP Project 14-30 48 Bluegrass Parkway Bridge required the use of a ladder in assess the existing coatings in Area 2 of the bridge steel over Pier 1. The abutment locations on the bridges were accessed at ground level. Mobilization for both bridges took a total of two-person days to make all purchases and arrange for KYTC traffic control on the Bluegrass Parkway Bridge. One-and-a-half person-days were required for deployments at the Bluegrass Parkway Bridge while one-half person day was required on the KY 922 Bridges. That time was minimized by good preparation of the KTC researchers at the shop prior to travelling to the jobsites and aided by technician familiarity with the sites gained during the coating assessments. The SSPC-SP 3 surface preparation at each test spot on all bridges averaged about one person-hour. The larger test areas (approximately 8- 24 ft2) were in Area 1 of the Bluegrass Parkway Bridge. They consisted primarily of flatwork only requiring limited coating removal and substrates that were primarily mill scale with very little corrosion. Those were rapidly cleaned primarily using the 7-inch angle sanders with non-woven pads and needle guns to clean the corners. The 2-inch grinder with a non-woven pad rapidly feathered the edges of the existing coating. The test areas on Area 2 of the Bluegrass Parkway Bridge were smaller with total test areas of about 2-3 ft2 having irregular surfaces that were pitted and corroded which slowed the surface preparation work. The KY 922 Bridges had test areas estimated at about 4-5 ft2 with more flatwork than Area 2 of the Bluegrass Parkway Bridge. However, the KY 922 Bridge test sites had some irregular surfaces (fasteners) and pitted, corroded surfaces that proved difficult to clean. Mixing and applying the liquid-applied coatings took less than one person-hour per coating for Area 1 of the Bluegrass Parkway Bridge and the KY 922 Bridges. It took slightly longer to paint Area 2 of the Bluegrass Parkway Bridge due to the need to transport the coatings to the opposite side of the road after mixing. The tape took two person-hours to apply to a test location with the irregular surfaces requiring more time to properly coat than equivalent flat areas. The grease took about one-half person-hour per location as no mixing/agitation was required prior to application. The clean-up and demobilization took about one-person day each for the Bluegrass Parkway and KY 922 Bridges. 3.4.2 KTC Technician Ratings of Power Tools and Coatings The KTC technicians were asked to provide subjective ratings of the performance of power tools and coatings. The power tools were rated on a simple three-point scale as shown in Table 10. None of the tools presented any problem with set-up or duty cycles. The latter was due to the small test areas cleaned during the field testing. The pneumatic needle gun was considered the best all-around tool for cleaning work being used on corners and smaller steel surfaces such as gussets, stiffeners, and angles. While not normally used as an SSPC-SP 3 power tool, the pneumatic rotary flapper rapidly stripped the fascia girder web of paint and no attempt was made to use it to remove the mill scale or cut a profile in the underlying steel. The 7-inch (17.78 cm) angle sanders required the operator exert force to push them against the work piece as well as support the weight of the tool. The KTC technicians found those units to be fatiguing if used for more than a few minutes. The 2-inch (5.08 cm) grinder was convenient for cleaning tight areas and feathering existing paint. The KTC technician ratings of the coatings also used a simple three-point scale (in additional to a rating of ‘0’ for “not applicable”) as shown in Table 11. The small areas painted and the close proximity of the painting operations to the staging/mixing area did not pose a pot- life/recoating issues with the epoxies due to the generally ideal temperatures encountered during the field work. Mixing/agitation was performed with electric or pneumatic drills with mixing blades and usually did not take more than 10 minutes to complete.

NCHRP Project 14-30 49 Table 10. KTC Ratings of Power Tools Tools \ Rating Category Set Up Ease of Use Duty Cycle Effectiveness Comments Needle Gun (Electric) 1 2 1 1 Unit heavy and difficult to use in tight areas Needle Gun (Pneumatic) 1 1 1 1 Considered the most effective power tool 7-Inch Angle Sander (Electric) 1 2 1 2 Hard to use due to force necessary to use shroud. Could not see the work while using the tool 7-Inch (17.78 cm) Angle Sander (Pneumatic) 1 2 1 2 Same comments for electric grinder 2-Inch (5.08 cm) Grinder (Pneumatic) 1 1 1 1 Easy to use. Good for tight places. Good for feathering existing paint. Rotary-Flapper (Pneumatic) 1 2 1 1 Worked well on flat surfaces. Heavy. Rating: 1 – good, 2 – fair, and 3 – poor Table 11. KTC Rating of Coatings Coating\Category Mixing Application Pot Life Clean Up Comments Acrylic/Acrylic 1 2 0 1 Rust bleed through primer. Difficult to get coating thickness Alkyd/Alkyd 1 1 0 1 Difficult to apply specified coating thickness (primer) Polyurethane/MIO- Epoxy 1 2 1 1 Pot life not an issue. Difficult to get coating thickness (primer) Polyurethane/Epoxy Sealer 1 1 1 1 Pot life not an issue. MCU (2 coats) 1 1 0 1 Easy coating to apply. Calcium Sulfonate Alkyd (2 Coats) 1 1 0 1 Easy coating to use. Difficult to get coating thickness with brush – easier with airless sprayer Grease 0 2 0 1 Messy if care not taken during application Tape 0 3 0 1 The tape is difficult cut to apply properly Rating: 1 – good, 2 – fair, 3 – poor, 0-not applicable 3.5 Field Work Discussion The field testing work followed the general process presented in the NCHRP 14-30 spot painting guidance document beginning with preliminary visual assessments to identify candidate spot painting bridges. That was followed by field coating assessments of candidate bridges to: • Assess site conditions, identify test locations for spot painting tests • Determine the properties of the existing coatings/substrates for spot repairs using SSPC-SP 3, “Power Tool Cleaning” • Identify whether soluble salt treatment was needed • Determine if special conditions existed (e.g., lead in existing coatings) that warranted special worker safety or environmental actions. Thereafter, a test (work) protocol was developed to perform the work. The KYTC Division of Environmental Analysis was engaged to dispose the lead paint residue. KTC technicians

NCHRP Project 14-30 50 performing work had the appropriate worker safety training and fitting of respirators. The necessary equipment and supplies (as previously noted) were obtained. Traffic control required coordination with the KYTC district office. The field work test protocol was followed for both surface preparation and coating application including verification of conditions suitable for painting. Coating thicknesses were addressed as a quality control measure using tooth gages and subsequently verified after curing using a dry film thickness measuring instrument. The work was monitored after completion to determine worker and protocol performance and to identify problems that could be addressed in future spot painting work. Work times, and experiences with power tools and coatings were obtained from field that could prove useful for planning follow-on spot painting projects. The approach to the laboratory and field work was consistent with the type of low-technology spot painting work that would be expected from state highway agency in-house forces. The one-year inspection indicated that the KTC spot painting procedure would work well where soluble salts and pitting were not major problems. In the areas where appreciable corrosion was present, the procedure did not prevent some local failures though rust bleeding made the extent of coating failures appear greater than they really were. Those areas were located under joints on both the Bluegrass Parkway and KY 922 twin Bridges. The MnDOT, “Bridge Maintenance Manual Field Guide-Chapter 8,” (2017) noted that with SSPC-SP 3 surface preparation, one-coat (polyamide epoxy) spot repairs would provide 5 years of service at 3%-5% failure and 7.5 years to 5%-10% failure. Two-coat (polyurethane over polyamide epoxy) would provide 7 years of service at 3%-5% failure and 10.5 years at 5%-10% failure. Those will be yardsticks for measuring the success of the KTC spot painting process. Unfortunately, the Bluegrass Parkway Bridge was completely painted in the fall of 2017 using removal and replacement and those test patches were lost. Only the KY 922 Bridges will remain as test beds for long-term evaluation of the experimental spot painting work.