National Academies Press: OpenBook
« Previous: Chapter 3 - Factors Contributing to Weldment Cracking Due to the Galvanizing Process
Page 16
Suggested Citation:"Chapter 4 - Experimental Test Program." National Academies of Sciences, Engineering, and Medicine. 2021. Mitigation of Weldment Cracking in Steel Highway Structures Due to the Galvanizing Process. Washington, DC: The National Academies Press. doi: 10.17226/26223.
×
Page 16
Page 17
Suggested Citation:"Chapter 4 - Experimental Test Program." National Academies of Sciences, Engineering, and Medicine. 2021. Mitigation of Weldment Cracking in Steel Highway Structures Due to the Galvanizing Process. Washington, DC: The National Academies Press. doi: 10.17226/26223.
×
Page 17
Page 18
Suggested Citation:"Chapter 4 - Experimental Test Program." National Academies of Sciences, Engineering, and Medicine. 2021. Mitigation of Weldment Cracking in Steel Highway Structures Due to the Galvanizing Process. Washington, DC: The National Academies Press. doi: 10.17226/26223.
×
Page 18
Page 19
Suggested Citation:"Chapter 4 - Experimental Test Program." National Academies of Sciences, Engineering, and Medicine. 2021. Mitigation of Weldment Cracking in Steel Highway Structures Due to the Galvanizing Process. Washington, DC: The National Academies Press. doi: 10.17226/26223.
×
Page 19
Page 20
Suggested Citation:"Chapter 4 - Experimental Test Program." National Academies of Sciences, Engineering, and Medicine. 2021. Mitigation of Weldment Cracking in Steel Highway Structures Due to the Galvanizing Process. Washington, DC: The National Academies Press. doi: 10.17226/26223.
×
Page 20
Page 21
Suggested Citation:"Chapter 4 - Experimental Test Program." National Academies of Sciences, Engineering, and Medicine. 2021. Mitigation of Weldment Cracking in Steel Highway Structures Due to the Galvanizing Process. Washington, DC: The National Academies Press. doi: 10.17226/26223.
×
Page 21
Page 22
Suggested Citation:"Chapter 4 - Experimental Test Program." National Academies of Sciences, Engineering, and Medicine. 2021. Mitigation of Weldment Cracking in Steel Highway Structures Due to the Galvanizing Process. Washington, DC: The National Academies Press. doi: 10.17226/26223.
×
Page 22
Page 23
Suggested Citation:"Chapter 4 - Experimental Test Program." National Academies of Sciences, Engineering, and Medicine. 2021. Mitigation of Weldment Cracking in Steel Highway Structures Due to the Galvanizing Process. Washington, DC: The National Academies Press. doi: 10.17226/26223.
×
Page 23
Page 24
Suggested Citation:"Chapter 4 - Experimental Test Program." National Academies of Sciences, Engineering, and Medicine. 2021. Mitigation of Weldment Cracking in Steel Highway Structures Due to the Galvanizing Process. Washington, DC: The National Academies Press. doi: 10.17226/26223.
×
Page 24
Page 25
Suggested Citation:"Chapter 4 - Experimental Test Program." National Academies of Sciences, Engineering, and Medicine. 2021. Mitigation of Weldment Cracking in Steel Highway Structures Due to the Galvanizing Process. Washington, DC: The National Academies Press. doi: 10.17226/26223.
×
Page 25
Page 26
Suggested Citation:"Chapter 4 - Experimental Test Program." National Academies of Sciences, Engineering, and Medicine. 2021. Mitigation of Weldment Cracking in Steel Highway Structures Due to the Galvanizing Process. Washington, DC: The National Academies Press. doi: 10.17226/26223.
×
Page 26
Page 27
Suggested Citation:"Chapter 4 - Experimental Test Program." National Academies of Sciences, Engineering, and Medicine. 2021. Mitigation of Weldment Cracking in Steel Highway Structures Due to the Galvanizing Process. Washington, DC: The National Academies Press. doi: 10.17226/26223.
×
Page 27
Page 28
Suggested Citation:"Chapter 4 - Experimental Test Program." National Academies of Sciences, Engineering, and Medicine. 2021. Mitigation of Weldment Cracking in Steel Highway Structures Due to the Galvanizing Process. Washington, DC: The National Academies Press. doi: 10.17226/26223.
×
Page 28
Page 29
Suggested Citation:"Chapter 4 - Experimental Test Program." National Academies of Sciences, Engineering, and Medicine. 2021. Mitigation of Weldment Cracking in Steel Highway Structures Due to the Galvanizing Process. Washington, DC: The National Academies Press. doi: 10.17226/26223.
×
Page 29
Page 30
Suggested Citation:"Chapter 4 - Experimental Test Program." National Academies of Sciences, Engineering, and Medicine. 2021. Mitigation of Weldment Cracking in Steel Highway Structures Due to the Galvanizing Process. Washington, DC: The National Academies Press. doi: 10.17226/26223.
×
Page 30
Page 31
Suggested Citation:"Chapter 4 - Experimental Test Program." National Academies of Sciences, Engineering, and Medicine. 2021. Mitigation of Weldment Cracking in Steel Highway Structures Due to the Galvanizing Process. Washington, DC: The National Academies Press. doi: 10.17226/26223.
×
Page 31
Page 32
Suggested Citation:"Chapter 4 - Experimental Test Program." National Academies of Sciences, Engineering, and Medicine. 2021. Mitigation of Weldment Cracking in Steel Highway Structures Due to the Galvanizing Process. Washington, DC: The National Academies Press. doi: 10.17226/26223.
×
Page 32
Page 33
Suggested Citation:"Chapter 4 - Experimental Test Program." National Academies of Sciences, Engineering, and Medicine. 2021. Mitigation of Weldment Cracking in Steel Highway Structures Due to the Galvanizing Process. Washington, DC: The National Academies Press. doi: 10.17226/26223.
×
Page 33
Page 34
Suggested Citation:"Chapter 4 - Experimental Test Program." National Academies of Sciences, Engineering, and Medicine. 2021. Mitigation of Weldment Cracking in Steel Highway Structures Due to the Galvanizing Process. Washington, DC: The National Academies Press. doi: 10.17226/26223.
×
Page 34
Page 35
Suggested Citation:"Chapter 4 - Experimental Test Program." National Academies of Sciences, Engineering, and Medicine. 2021. Mitigation of Weldment Cracking in Steel Highway Structures Due to the Galvanizing Process. Washington, DC: The National Academies Press. doi: 10.17226/26223.
×
Page 35
Page 36
Suggested Citation:"Chapter 4 - Experimental Test Program." National Academies of Sciences, Engineering, and Medicine. 2021. Mitigation of Weldment Cracking in Steel Highway Structures Due to the Galvanizing Process. Washington, DC: The National Academies Press. doi: 10.17226/26223.
×
Page 36
Page 37
Suggested Citation:"Chapter 4 - Experimental Test Program." National Academies of Sciences, Engineering, and Medicine. 2021. Mitigation of Weldment Cracking in Steel Highway Structures Due to the Galvanizing Process. Washington, DC: The National Academies Press. doi: 10.17226/26223.
×
Page 37
Page 38
Suggested Citation:"Chapter 4 - Experimental Test Program." National Academies of Sciences, Engineering, and Medicine. 2021. Mitigation of Weldment Cracking in Steel Highway Structures Due to the Galvanizing Process. Washington, DC: The National Academies Press. doi: 10.17226/26223.
×
Page 38
Page 39
Suggested Citation:"Chapter 4 - Experimental Test Program." National Academies of Sciences, Engineering, and Medicine. 2021. Mitigation of Weldment Cracking in Steel Highway Structures Due to the Galvanizing Process. Washington, DC: The National Academies Press. doi: 10.17226/26223.
×
Page 39
Page 40
Suggested Citation:"Chapter 4 - Experimental Test Program." National Academies of Sciences, Engineering, and Medicine. 2021. Mitigation of Weldment Cracking in Steel Highway Structures Due to the Galvanizing Process. Washington, DC: The National Academies Press. doi: 10.17226/26223.
×
Page 40
Page 41
Suggested Citation:"Chapter 4 - Experimental Test Program." National Academies of Sciences, Engineering, and Medicine. 2021. Mitigation of Weldment Cracking in Steel Highway Structures Due to the Galvanizing Process. Washington, DC: The National Academies Press. doi: 10.17226/26223.
×
Page 41
Page 42
Suggested Citation:"Chapter 4 - Experimental Test Program." National Academies of Sciences, Engineering, and Medicine. 2021. Mitigation of Weldment Cracking in Steel Highway Structures Due to the Galvanizing Process. Washington, DC: The National Academies Press. doi: 10.17226/26223.
×
Page 42
Page 43
Suggested Citation:"Chapter 4 - Experimental Test Program." National Academies of Sciences, Engineering, and Medicine. 2021. Mitigation of Weldment Cracking in Steel Highway Structures Due to the Galvanizing Process. Washington, DC: The National Academies Press. doi: 10.17226/26223.
×
Page 43
Page 44
Suggested Citation:"Chapter 4 - Experimental Test Program." National Academies of Sciences, Engineering, and Medicine. 2021. Mitigation of Weldment Cracking in Steel Highway Structures Due to the Galvanizing Process. Washington, DC: The National Academies Press. doi: 10.17226/26223.
×
Page 44
Page 45
Suggested Citation:"Chapter 4 - Experimental Test Program." National Academies of Sciences, Engineering, and Medicine. 2021. Mitigation of Weldment Cracking in Steel Highway Structures Due to the Galvanizing Process. Washington, DC: The National Academies Press. doi: 10.17226/26223.
×
Page 45
Page 46
Suggested Citation:"Chapter 4 - Experimental Test Program." National Academies of Sciences, Engineering, and Medicine. 2021. Mitigation of Weldment Cracking in Steel Highway Structures Due to the Galvanizing Process. Washington, DC: The National Academies Press. doi: 10.17226/26223.
×
Page 46
Page 47
Suggested Citation:"Chapter 4 - Experimental Test Program." National Academies of Sciences, Engineering, and Medicine. 2021. Mitigation of Weldment Cracking in Steel Highway Structures Due to the Galvanizing Process. Washington, DC: The National Academies Press. doi: 10.17226/26223.
×
Page 47
Page 48
Suggested Citation:"Chapter 4 - Experimental Test Program." National Academies of Sciences, Engineering, and Medicine. 2021. Mitigation of Weldment Cracking in Steel Highway Structures Due to the Galvanizing Process. Washington, DC: The National Academies Press. doi: 10.17226/26223.
×
Page 48
Page 49
Suggested Citation:"Chapter 4 - Experimental Test Program." National Academies of Sciences, Engineering, and Medicine. 2021. Mitigation of Weldment Cracking in Steel Highway Structures Due to the Galvanizing Process. Washington, DC: The National Academies Press. doi: 10.17226/26223.
×
Page 49
Page 50
Suggested Citation:"Chapter 4 - Experimental Test Program." National Academies of Sciences, Engineering, and Medicine. 2021. Mitigation of Weldment Cracking in Steel Highway Structures Due to the Galvanizing Process. Washington, DC: The National Academies Press. doi: 10.17226/26223.
×
Page 50
Page 51
Suggested Citation:"Chapter 4 - Experimental Test Program." National Academies of Sciences, Engineering, and Medicine. 2021. Mitigation of Weldment Cracking in Steel Highway Structures Due to the Galvanizing Process. Washington, DC: The National Academies Press. doi: 10.17226/26223.
×
Page 51
Page 52
Suggested Citation:"Chapter 4 - Experimental Test Program." National Academies of Sciences, Engineering, and Medicine. 2021. Mitigation of Weldment Cracking in Steel Highway Structures Due to the Galvanizing Process. Washington, DC: The National Academies Press. doi: 10.17226/26223.
×
Page 52
Page 53
Suggested Citation:"Chapter 4 - Experimental Test Program." National Academies of Sciences, Engineering, and Medicine. 2021. Mitigation of Weldment Cracking in Steel Highway Structures Due to the Galvanizing Process. Washington, DC: The National Academies Press. doi: 10.17226/26223.
×
Page 53

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.

16 The experimental test program included tests of steel tension specimens in air (RT and ET and in different galvanizing baths) and steel WOL specimens tested in the loaded condition in the galvanizing bath. Hardness measurements and microstructural analyses were also performed to characterize the influence of some factors included in the study. Tension Specimen Preparation The tension specimens were standard 505-type (Figure 4-1) and 252-type steel specimens (Figure 4-2) fabricated from A36, A572-50, A572-65, and HPS100W steel plates according to the specifications in the standard ASTM E8 (ASTM, 2013b). Each specimen blank was extracted from the plate material using a waterjet table. The A36, A572-50, and A572-65 plates were 1 in. thick, and the HPS100W material was 2 in. thick. Each grade of plate material was from a single heat (i.e., a single batch of steel chemistry). Specimens were extracted from plate mid-thickness since the diameter of the 505-type specimens in the grip region was 0.75 in. Figure 4-3 shows a photograph of completed 505- and 252-type specimens. Table 4-1 lists the types of tension specimens used together with the testing environment (e.g., in zinc, in air), and specimen condition (e.g., bare steel, cold-worked, stress concentration). Table  4-2, Table  4-3, and Table  4-4 show the mill-certified chemical composition and mechanical properties for grades A36, A572-50, and A572-65 material, respectively, as well as chemical compositions determined through laboratory analysis. Table 4-5 shows the chemical composition for HPS100W as determined from laboratory analysis (mill-certified values were unavailable). According to ASTM (2015), steel is considered to be “reactive” to galvanizing when its silicon content ranges from 0.04% to 0.15% or above 0.22%. The silicon content of HPS100W steel was 0.29% and thus considered reactive to galvanizing. The silicon content of the A36 steel was also outside of this range but with less deviation than the HPS100W material. The chemical compositions provided in the mill certificates were used to estimate CEZ accord- ing to the formula contained in the Japanese Industrial Standard JIS G 3129 (JIS, 2005), which suggests limiting CEZ values to 0.44% would control LMAC; results of this analysis are listed in Table 4-6. The computed CEZ values for A36, A572-50, and A572-65 grades were higher than the suggested limit, with A572-50 and A572-65 showing the largest deviations and thus suggest- ing the greatest susceptibility to LMAC. Table 4-7 presents the CEZ computation for the four grades of steel based on laboratory analysis. The laboratory analysis did not report data for vanadium, niobium, titanium, or boron; C H A P T E R 4 Experimental Test Program

Experimental Test Program 17   Figure 4-1. Standard 505-type tension specimen. Figure 4-2. Details of 252-type tension specimen.

18 Mitigation of Weldment Cracking in Steel Highway Structures Due to the Galvanizing Process Figure 4-3. 505- and 252-type specimens. Test Condition Steel Grade Tension Specimens 505-type 252-type Bare Steel Cold-Work Stress Concentration Bare Steel Heat-Affected Zone* 6% 8% 9% 90° bend 1.5 2.5 Fast CR Medium CR Slow CR In A ir Room Temp. A36 2 2 2 3 3 2 2 2 2 A572-50 2 3 2 2 3 3 2 2 2 2 A572-65 2 2 HPS100W 2 2 2 2 2 2 1 1 1 Room Temp. + Acid Dip A36 2 2 2 2 A572-50 2 2 2 2 A572-65 HPS100W 1 1 1 1 Elevated Temp. A36 2 1 1 2 A572-50 2 1 1 2 A572-65 2 2 HPS100W 2 1 1 2 In Z in c SHG Zinc A36 3 2 2 2 2 2 2 2 A572-50 3 3 2 2 2 2 2 2 2 A572-65 2 HPS100W 3 2 1 1 1 1 1 1 SHG Zinc + 0.1% Bi A36 2 A572-50 3 A572-65 2 HPS100W 2 SHG Zinc + 1% Pb A36 8 A572-50 8 A572-65 3 HPS100W 8 SHG Zinc + 1% Pb + 0.1% Bi A36 17 3 3 2 2 2 2 2 A572-50 12 3 3 2 2 2 2 2 2 A572-65 4 HPS100W 7 1 1 1 1 1 1 1 * Note: The 252-type specimens tested in zinc were notched with SCF = 2.5. Table 4-1. Test matrix for the tension specimens.

Mechanical properties from mill cert: Fy = 40.9 ksi Fu = 68.7 ksi Percent elongation over 8 in. = 27.2% C Mn P S Si Cu Ni Cr Mo Cb V Al Ti N B Ca Chemical composition from mill cert: 0.19 0.85 0.01 0.008 0.06 0.18 0.06 0.08 0.016 0.002 0.004 0.03 0.001 0.004 0.0002 0.0021 Chemical composition from laboratory testing: 0.18 0.81 0.006 0.008 0.05 0.15 0.07 0.07 0.01 -- -- -- -- -- -- -- Table 4-2. Mechanical properties and chemical composition of A36. Mechanical properties from mill cert: Fy = 50.4 ksi Fu = 79.5 ksi Percent elongation over 2 in. = 25.3% C Mn P S Si Cu Ni Cr Mo Cb V Al Ti N B Ca Chemical composition from mill cert: 0.19 1.16 0.0090 0.003 0.19 0.17 0.06 0.07 0.0180 0.003 0.0720 0.026 0.001 0.006 0.0003 0.0028 Chemical composition from laboratory testing: 0.14 1.32 0.008 0.005 0.03 0.16 0.08 -- 0.01 -- -- -- -- -- -- -- Table 4-3. Mechanical properties and chemical composition of A572-50. Mechanical properties from mill cert: Fy = 65.6 ksi Fu = 86.6 ksi Percent elongation over 8 in. = 25.4% C Mn P S Si Cu Ni Cr Mo Al V Nb Ti N Ca B Sn Ceq Pcm Chemical composition from mill cert: 0.15 1.38 0.009 0.001 0.04 0.18 0.12 0.08 0.01 0.028 0.058 0.043 0.002 0.0109 0.0025 0.0002 0.009 0.43 0.24 Chemical composition from laboratory testing: 0.14 1.33 0.008 0.005 0.03 0.16 0.14 0.08 0.01 -- -- -- -- -- -- -- -- -- -- Table 4-4. Mechanical properties and chemical composition of A572-65. C Mn P S Si Cu Ni Cr Mo 0.1 0.85 0.014 0.005 0.29 0.29 0.87 0.53 0.5 Table 4-5. Chemical composition of HPS100W. C + Si/17 + Mn/7.5 + Cu/13 + Ni/17 + Cr/4.5 + Mo/3 + V/1.5 + Nb/2 + Ti/4.5 + 420(B) = CEZ (%) A36 0.19 0.06 0.85 0.18 0.06 0.08 0.016 0.004 0.001 0.0002 0.434 A572-50 0.19 0.19 1.16 0.17 0.16 0.07 0.018 0.072 0.001 0.0003 0.574 A572-65 0.15 0.04 1.38 0.18 0.12 0.08 0.01 0.058 0.043 0.002 0.0002 0.523 -- -- Table 4-6. Chemical composition and CEZ values (CEZ, %) computed from mill-supplied chemistry. Table 4-7. Chemical composition and CEZ values (CEZ, %) computed from laboratory chemical analysis. C + Si /17 + Mn/7.5 + Cu/13 + Ni/17 + Cr/4.5 + Mo/3 + V/1.5 + Nb/2 + Ti/4.5 + 420(B) = A36 0.18 0.05 0.81 0.15 0.07 0.07 0.01 0 0 0 0 A572-50 0.14 0.03 1.32 0.16 0.08 0 0.01 0 0 0 0 A572-65 0.14 0.03 1.33 0.16 0.14 0.08 0.01 0 0 0 0 HPS100W 0.1 0.29 0.85 0.29 0.87 0.53 0.5 0 0 0 0 CEZ (%) 0.325 0.338 0.361 0.588

20 Mitigation of Weldment Cracking in Steel Highway Structures Due to the Galvanizing Process these values were assumed to be zero, resulting in lower CEZ values than those computed based on the mill certificate. However, the data show a similar trend with A572-50 and A572-65 being more susceptible to LMAC than A36. The CEZ value computed for HPS100W is substantially greater than those computed for A36, A572-50, and A572-65 grades, indicating a higher sus- ceptibility to LMAC. The base metal hardness was measured and found to be 145 HV, 200 HV, and 300 HV for A36, A572-50, and HPS100W steels, respectively. These values suggest that HPS100W steel is more susceptible to LMAC than the other steel grades when considering the threshold value of 270 HV proposed by Elboujdaini et al. (2004) as an indicator of susceptibility to LMAC. Treatment of 505-Type Specimens Some specimens were modified to investigate the influence of additional factors, such as stress concentration, cold-working, and HAZ effects. Procedures used for these investigations are discussed as follows. Stress Concentration To investigate the effect of stress concentration, 42 505-type specimens were modified by fabricating a notch that circumscribed the diameter of the tension specimen using a fully pro- grammable computer numerical control (CNC) machine (see Figure 4-4). Two SCFs, SC1 = 1.5 (low-medium severity) and SC2 = 2.5 (medium-high severity), were considered. Uniform Cold-Working Through Tension To investigate the effect of the cold-working level, two levels of uniform cold-working— 6% and 8%—were included. The 8% value was selected for all grades of steel being tested to ensure uniform cold-working at the test section and that no severe plastic deformation (necking) will occur in the specimens. The cold-work (see Figure 4-5) was achieved by pre-deforming the tension specimens using the uniaxial test frame at RT. After the permanent deformation was induced, the specimens were tested in the galvanizing bath. To cold-work the specimens, a gauge length of 2 in. centered around the mid-length point of the reduced section was measured and marked. The specimens were then cold-worked to the desired level of elongation as measured using an extensometer during loading (the setup is shown in Figure 4-6). SC1=1.5 SC2=2.5 (A) Un-notched (B) Notched-SC1 (C) Notched-SC2 Figure 4-4. Un-notched and notched 505-type specimens.

Experimental Test Program 21   Hardness values obtained for the 6% and 8% cold-worked specimens are shown in Figure 4-7. Material hardness increased for the 6% cold-worked specimens over the base metal hardness, and the 8% cold-worked specimens exhibited a higher average hardness than the 6% cold- worked specimens. The 270 HV suggested by Elboujdaini et al. (2004) to correlate with LMAC is indicated with a dashed horizontal line in the figure; this value is exceeded for HPS100W specimens. Non-uniform Cold-Working Through 90ç Cold-Bend The majority of the cold-worked specimens were cold-deformed under tension. For com- parison, six additional tension specimens were fabricated from ¾-in. thick A572-50 steel plates that Step 1: Start with an un-treated, un-notched 505-type specimen. Step 2: Using a universal test frame, load the specimen past the point of yield to induce locked-in plastic deformations (either 3% or 6%) at room temperature. Lo 505 505 Step 3: Remove the external load. Step 4: After cold-working, test each 505 specimen to failure while submerged in the galvanizing bath. 505 L 505 0%, 6%, or 8% cold-work 0%, 6%, or 8% cold-work Figure 4-5. Process of cold-working the 505-type specimens. Figure 4-6. Test setup for cold-working.

22 Mitigation of Weldment Cracking in Steel Highway Structures Due to the Galvanizing Process were deformed by a 90° cold-bend; the process for fabricating these specimens is illustrated in Figure 4-8. First, a ¾-in. thick A572-50 flat plate was deformed into an “L” shape with a 90° bend with a radius of 4t (3 in.), which corresponds to a theoretical value of approximately 14% cold-work at the bend. Next, the shaded portion shown in Figure 4-8(c) was removed from the plate, and a 505-type tension specimen was fabricated from the center of the cold-worked strip. The strain at the outer surface of the tension specimen’s reduced section was about 9%. The tension specimens oriented in this direction are appropriate to provide information regard- ing the cracking sensitivity of steel tubes near a base plate weld. These specimens were tested to failure while submerged in the galvanizing bath using the protocol used for the other tension 0 50 100 150 200 250 300 350 400 450 A36 Base A36 CW6 A36 CW8 A572-50 Base A572-50 CW6 A572-50 CW8 HPS 100W Base HPS 100W CW6 HPS 100W CW8 Ha rd ne ss , V m h Figure 4-7. Hardness values for cold-worked specimens. Region of cold-working from bending process Remove material along center of bend line and fabricate 505 tension specimen. Top View Side View Steel plate, before deformation (a) (b) (c) Figure 4-8. Fabrication of 505-type specimens from cold-bent plates.

Experimental Test Program 23   specimens. These specimens were included in the test program to provide additional informa- tion beyond that obtained from specimens that were cold-worked through tensile loading. HAZ-Simulated 252-Type Specimens To simulate the effect of different welding processes, the 252-type specimens were treated in a Gleeble simulator to produce the microstructures of HAZ that will result from different cooling rates (Adonyi, 2016). The specimens were first preheated to a peak temperature (1300°C) and then cooled at three different cooling rates in the 800°C-500°C range. The cooling rates were 12°C/sec (slow cooling rate or SCR), 30°C/sec (medium cooling rate or MCR), and 60°C/sec (fast cooling rate or FCR) to simulate welding heat inputs of 60 kJ/in. and 30 kJ/in. and an FCR for oxyfuel flame cutting, respectively. Figure 4-9 shows a 252-type specimen during treatment. Metallographic analysis was performed on initial steel samples to establish the basic parameters for 30 kJ/in. and 60 kJ/in. heat input welding. This analysis determined that the cooling rate of 60°C/sec in the 800°C-500°C range best represented low heat input welding/thermal cutting; the following cooling rates were used to control the HAZ-simulated specimens: 1. SCR: 12°C/sec cooling rate, 60 kJ/in. heat input, coarse grain heat-affected zone (CGHAZ) simulations 2. MCR: 30°C/sec cooling rate, 30 kJ/in. heat input, CGHAZ simulations 3. FCR: 60°C/sec cooling rate, oxyfuel flame cutting, 2 in. plate edge simulations Microscope images of the three treated grades of steel before and after treatment are shown in Figures 4-10, 4-11, 4-12, and 4-13. Hardness results for 12 samples are shown in Figure 4-14. In eight of the nine cases, hardness was approximately equal to or greater than 270 HV (marked as the dashed horizontal line in the figure), which is suggested by Elboujdaini et al. (2004) to correlate with LMAC. Notching HAZ-Simulated Specimens It was determined later that the HAZ-simulation treatment process introduced a non-uniform temperature gradient along the length of the reduced section of the 252-type specimens. The mid- length point experienced the target cooling rates, but the cooling rate decreased moving away from the mid-length point. As a result, after the HAZ-simulation treatment, the mid-length point Figure 4-9. A 252-type specimen at 1300çC with copper tubing using helium to cool the specimen.

(a) Gr. A36; 29.9 microns average grain size, 142 HV 100 grams (g) average hardness (b) Gr. A572-50; 15 microns average grain size, 200 HV 100 g average hardness (c) Gr. HPS100W; 23 microns average grain size, 303 HV 100 g average hardness Figure 4-10. Base metal microstructure, nital etch, 50x magnification. (a) Gr. A36; 12°C/sec cooling rate (SCR) (b) Gr. A572-50; 12°C/sec cooling rate (SCR) (c) HPS100W; 12°C/sec cooling rate (SCR) Figure 4-11. Microscopy of simulated CGHAZ, heated to 1300çC, cooled at 12çC/sec in the 800°C-500çC range. 100x magnification: (a) Gr. A36; (b) Gr. A572-50; and (c) HPS100W. (a) Gr. A36; 30°C/sec cooling rate (MCR) (b) Gr. A572-50; 30°C/sec cooling rate (MCR) (c) HPS100W; 30°C/sec cooling rate (MCR) Figure 4-12. Microscopy of simulated CGHAZ, heated to 1300çC, cooled at 30çC/sec in the 800çC-500çC range. 100x magnification: (a) Gr. A36; (b) Gr. A572-50; and (c) HPS100W.

Experimental Test Program 25   had greater strength than the two ends, and initial tests performed on the 252-type specimens resulted in failures at the end of the reduced sections. To ensure failure at the mid-length point for the 252-type specimens, the specimens were modified by fabricating a notch that circumscribed the diameter of the tension specimen, pre- senting an SCF of 2.5. Because this modification introduced another factor (i.e., stress concen- tration) to cooling rates, interpretation of data was based on comparative analysis performed within the 252-type specimen group. (a) Gr. A36; 60°C/sec cooling rate (FCR) (b) Gr. A572-50; 60°C/sec cooling rate (FCR) (c) Gr. HPS100W; 60°C/sec cooling rate (FCR) Figure 4-13. Microscopy of simulated CGHAZ, heated to 1300çC, cooled at 60çC/sec in the 800çC-500çC range. 100x magnification: (a) Gr. A36; (b) Gr. A572-50; and (c) HPS100W. Ha rd ne ss , V m h Figure 4-14. Average hardness values in the HAZ-simulated samples at different cooling rates.

26 Mitigation of Weldment Cracking in Steel Highway Structures Due to the Galvanizing Process Tension Test Setup and Testing Protocols The tension specimens were loaded until failure under four different environments: while submerged in a galvanizing bath, while cooling after being submerged in a galvanizing bath, in air at ET, and in air at RT. Testing protocols for the different test environments are described as follows. Submerged in-Zinc Tests Tension tests were performed using a self-reacting load frame submerged in an 840°F zinc galvanizing bath contained in an electric furnace with a removable steel crucible. Load was applied to the specimen using a hydraulic jack that was outfitted with a remote servo-valve, load cell, and high-temperature LVDT (linear variable differential transformer) and controllable by an MTS TestStar controller in load and displacement control modes, as shown schematically in Figure 4-15. The test procedure is summarized as follows: 1. Prepare the tension specimen by degreasing, rinsing, acid pickling in hydrochloric acid, rins- ing, and then fluxing. 2. Install the tension specimen in the load frame. 3. Remove the lid from the electric crucible furnace. 4. Properly rig and then lift the load frame using the overhead crane, and move the frame over top of the electric crucible furnace. 5. Lower the load frame until it rests completely on the support frame (crane rigging is slack) and the specimen is submerged in zinc. 6. Allow the specimen to dwell in the galvanizing bath for 400 sec. 7. Perform tension test to failure using the programmed loading protocol in the MTS station manager. 8. Lift the load frame using the overhead crane and move it back to the first support frame. 9. Allow the load frame and specimen to cool for at least four hours before handling, and then remove the specimen from the setup. Figure 4-16 shows the test setup. Two support frames were used. One support frame was used for test preparation and the other for supporting the load frame during testing. The figure also shows a specimen installed in the load frame and the setup during the tension test with the specimen submerged in zinc. Loading was commenced after the specimen was submerged in the galvanizing bath for 400 sec. The 505-type specimens were first pre-loaded to 3,000 pounds of force (lbf ) at a force control rate of 10 lbf/sec, then unloaded to 2,000 lbf at a rate of 200 lbf/sec to establish a repeat- able pre-loading protocol for each specimen, and finally loaded to failure using displacement control at a rate of 0.053 inch per minute (in./min.). The 252-type specimens were first pre- loaded to 1,500 lbf at a force control rate of 100 lbf/sec, then unloaded to 1,000 lbf at a rate of 100 lbf/sec, and finally loaded to failure using displacement control at a rate of 0.015 in./min. Time-histories of load and actuator displacement were both recorded during the tests at a sampling rate of 5 hertz. To compensate for system compliance during the tension coupon tests, compliance speci- mens were developed and loaded to measure test frame and specimen grip deformation. Fig- ure 4-17 shows the compliance specimens for 505 and 252 tension tests. These specimens were designed such that the grip portions of the specimens exactly replicated the geometry (length, diameter, and threads) of the 505- and 252-type specimens, but the mid-portion that corre- sponded with the reduced section in the specimens was assigned a very large diameter in the compliance specimens (i.e., the mid-portion of the compliance specimens was very stiff). Thus,

Experimental Test Program 27   To MTS hydraulic pump Threaded coupler Steel tensile coupon Galvanizing bath, 450C/840F Enerpac hydraulic cylinder Load Cell Insulation and external support framing LVDT Remote Servovalve Digital signal to/from MTS Controller Plate Western Technologies Electric Crucible Furnace Figure 4-15. Schematic of setup for tension tests in molten zinc environment.

Kettle Lid Load frame Enclosure Load frame Actuator Support frame for preparation Support frame for testing Kettle Load frame Actuator Tension specimen Figure 4-16. Test setup for tension tests in molten zinc environment. Figure 4-17. Specimens used to determine test setup compliance.

Experimental Test Program 29   load-deformation behavior recorded while loading a compliance specimen would reasonably indicate only deformations that were occurring in the grip region and in the frame, allowing those system compliance deformations to be “backed out” from actual tests on 505- and 252-type specimens. In this manner, adjusted load-deformation curves for the 252- and 505-type spec- imens that represent deformations occurring over their respective reduced lengths, similar to what would be obtained with an extensometer, could be developed. This approach allowed developing separate compliance curves for 505- and 252-type speci- mens (which had different grip lengths), as well as for in-air and in-zinc test environments (the in-zinc environment corresponded to significantly more system compliance due to its high temperature). This approach also enabled the derivation of high-quality stress-strain curves from the load-displacement data and accurate calculation of Young’s modulus of the steel specimens during in-air tests. While-Cooling Tests Prior research has reported concentrations of heavy metal at the tips of cracks after galva- nizing (Kinstler, 2005). The combination of some heavy metals can form a eutectic, such that the melting point of the metal combination is lower than the melting point of any of the indi- vidual metals. An eutectic for bismuth and lead that forms during the cooling phase or after removal from the galvanizing bath could possibly amplify the formation of LMAC. Therefore, tension tests were also performed on specimens while they were cooling after being sub- merged. These tests were performed in each bath combination to determine if while-cooling is the most critical stage for tension specimens during the galvanizing process. For these tests, the frame with a specimen installed was submerged for 600 sec. The load frame was then raised to a height such that the specimen was exposed to air, but the bottom portion of the load frame remained submerged in the liquid zinc to prevent the zinc from freezing at the bottom of the frame and changing the boundary condition of the specimen. The 600 sec dwell time (which is longer than the 400 sec used for the submerged in-zinc tests) was necessary to ensure that all frozen zinc around the specimen (due to the thermal mass of the load frame) was fully molten before conducting the while-cooling test. Loading was started immediately after the specimen and frame were in this position, and it was conducted following the loading protocol described for submerged in-zinc tests. Figure 4-18 illustrates the test setups for both “submerged” and “while-cooling” in-zinc tests. (a) Setup for testing specimens while submerged in zinc (b) Setup for testing specimens while cooling Note that the bottom part of the load frame was kept in the zinc. Zinc level Zinc level Figure 4-18. Setups for (a) “submerged” and (b) “while-cooling” cases.

30 Mitigation of Weldment Cracking in Steel Highway Structures Due to the Galvanizing Process In-Air ET Tests Tension tests were also performed in air at an ET of 840°F. A test frame equipped with a tem- perature chamber was used to perform these tests (see Figure 4-19). The loading protocol applied for the in-air ET tests was the same protocol used for the submerged in-zinc and RT in-air tests. The ET in-air tests were conducted in a traditional closed-loop servo-controlled load frame, and an extensometer was used until significant plasticity was noted. However, since the com- pliance of this load frame and setup can be expected to differ from that used for the in-zinc tests, load-deformation data were recorded while loading the compliance specimen in order to account for compliance. Thus, the extensometer-based data served as an external check, and the stress-strain curves were generated in the same manner as for all the tests performed using con- tinuous (non-spliced) load-deformation data and a compliance curve developed for the frame and setup. In-Air RT Tests A series of tension tests were performed in air at RT to serve as a benchmark for tests per- formed in zinc. These tests were conducted in the same loading frame used for in-zinc tests, using the same loading protocol. Load-deformation data measured for 505- and 252-type ten- sion specimens were adjusted using compliance curves developed in air at RT. Tension Testing Results Data Processing Test data from the compliance specimens were used to account for the load frame and sys- tem deformation in the tension specimen test data. Figures 4-20 and 4-21 show example load- displacement curves for 505-type specimens tested in air at RT before and after accounting for the load frame compliance, respectively. Figure 4-22 shows the stress-strain curves converted from the adjusted load-displacement curves, in which the strain was obtained by dividing the adjusted displacement by the gauge length of the tension specimen. Figure 4-19. ET test setup showing a tension specimen in a temperature chamber.

Experimental Test Program 31   HPS-100W A36 A572-65 A572-65 Figure 4-20. Unadjusted load-displacement curves. HPS-100W A36 A572-65 A572-65 Figure 4-22. Adjusted stress-strain curves. HPS-100W A36 A572-65 A572-65 Figure 4-21. Adjusted load-displacement curves.

32 Mitigation of Weldment Cracking in Steel Highway Structures Due to the Galvanizing Process Tension Test Results The strain value at failure (εf) for each specimen tested in liquid zinc is defined as the last data point captured on the stress-strain curve before the LVDT data became unstable due to speci- men failure; this strain value is a direct indicator of susceptibility to cracking of the specimens tested while either in air or fully submerged in zinc. The specimens tested while cooling in air after dipping in zinc always failed at the interface with the liquid zinc surface, not within the reduced section of the specimen, indicating the severity of LMAC in the in-zinc condition. Submerged, In-Zinc Tests Versus While-Cooling Tests Figure 4-23 shows the stress-strain curves for the two zinc test environments (submerged and while cooling) for the four grades of steel. The tests of specimens submerged in zinc provided clean and consistent data, while data from testing of specimens while cooling indicated more scatter, especially for A572-50 and A572-65 steel. The scatter could be attributed to the exposure of the specimens to both liquid zinc and air at the same time during cooling. Also, in spite of attempts to maintain the same location of the specimen relative to the zinc surface, small varia- tions existed between tests. An important observation from the while-cooling tests was the failure location. As shown in Figure 4-24, specimens tested while submerged in zinc always failed at the mid-point of the (a) A36 (b) A572-50 (c) A572-65 (d) HPS100W 0 10 20 30 40 50 60 70 80 0 0.03 0.06 0.09 0.12 0.15 0.18 0.21 0.24 0.27 St re ss (k si ) Strain (in./in.) While cooling Submerged 0 10 20 30 40 50 60 70 80 90 0 0.03 0.06 0.09 0.12 0.15 0.18 0.21 0.24 0.27 St re ss (k si ) Strain (in./in.) Submerged While cooling 0 10 20 30 40 50 60 70 80 90 0 0.03 0.06 0.09 0.12 0.15 0.18 0.21 0.24 0.27 St re ss (k si ) Strain (in./in.) While cooling Submerged 0 20 40 60 80 100 120 140 0 0.03 0.06 0.09 0.12 0.15 0.18 0.21 0.24 0.27 St re ss (k si ) Strain (in./in.) While cooling Submerged Figure 4-23. Stress-strain curves of 505-type specimens tested while submerged in zinc and while cooling.

Experimental Test Program 33   reduced section. However, failure of the specimens tested while cooling in air always occurred at the zinc surface, which indicates that having direct contact with liquid zinc was more critical to failure. These observations indicate that the specimens tested while fully submerged were subjected to the most critical galvanizing environment because (1) failure of the specimens subjected to both liquid zinc and air during the while-cooling test always occurred at the liquid zinc level and not in air; and (2) tests on fully submerged specimens produced more consistent data with less scatter. Therefore, the report focuses on the data for the tests on the specimens fully submerged in zinc. 90ç Cold-Bend Specimens Figure 4-25 shows the stress-strain curves for A572-50 bare steel and 90° cold-bend specimens tested in air at RT and in SHG zinc. A general observation is that the cold-bend process did not seem to affect the mechanical behavior of the specimen—either in air or in zinc—possibly because the cold-bend process only introduced a 14% strain at the outermost fiber of the steel plate, while through-thickness fibers experienced lower levels of cold-work. In addition, because the tension specimen was extracted from the center of the steel plate, it further reduced the level of strain in the specimen (9% of cold-work is introduced to the outer surface of the specimen’s reduced section). However, the 90° cold-bend specimens, which fairly represent the situation that exists in a polygonal tube or pole shape, showed no influence from cold-working. The reported results and discussion focus on specimens cold-worked through tension because they appeared to capture some influence of cold-work (although minor). Stress-Strain Curves Figure 4-26 shows sample stress-strain curves for 505-type tension specimens of the four steel grades tested in air at RT, in air at ET (840°F), and in SHG zinc. The specimens of all steel grades experienced lower strains in SHG zinc than in air (both at RT and at ET). Steel Composition and Galvanizing Composition Figure 4-27 shows the strain-at-failure data for the four steel grades (A36, A572-50, A572-65, and HPS100W) tested in air and in SHG zinc (also known as pure zinc). All steel grades showed significantly lower strains when tested in the SHG zinc than in air; the reductions were Failure location Failure location (a) Test while submerged; failure occurs at the middle section (A65) (b) Test while cooling; failure occurs at the zinc surface (A65) Figure 4-24. Failure locations of specimens for the two galvanizing test environments (in zinc and while cooling).

0 25 50 75 100 125 0 0.05 0.1 0.15 0.2 0.25 St re ss (k si ) Strain (in./in.) Bare steel, in air at room temp 90º cold bend, in air at room temp 90º cold bend, in SHG Zinc Bare steel, in SHG Zinc Figure 4-25. Stress-strain curves for A572-50 steel (bare steel specimens and cold-worked specimens through 90ç cold-bend) in air at RT and in SHG zinc. (a) A36 (b) A572-50 (c) A572-65 (d) HPS100W 0 25 50 75 100 125 0 0.03 0.06 0.09 0.12 0.15 0.18 0.21 0.24 0.27 St re ss (k si ) Strain (in./in.) Room temp SHG Zinc Elevated temp 0 25 50 75 100 125 0 0.03 0.06 0.09 0.12 0.15 0.18 0.21 0.24 0.27 St re ss (k si ) Strain (in./in.) SHG Zinc Room temp Elevated temp 0 25 50 75 100 125 0 0.03 0.06 0.09 0.12 0.15 0.18 0.21 0.24 0.27 St re ss (k si ) Strain (in./in.) Room temp Elevated tempSHG Zinc 0 25 50 75 100 125 0 0.03 0.06 0.09 0.12 0.15 0.18 0.21 0.24 0.27 St re ss (k si ) Strain (in./in.) Elevated temp Room temp SHG Zinc Figure 4-26. Stress-strain curves for bare 505-type tension specimens tested in air at RT, at ET, and in SHG zinc.

Experimental Test Program 35   65% (A36), 57% (A572-50), 58% (A572-65), and 70% (HPS100W). Strain magnitudes at failure were similar for A36, A572-50, and A572-65 specimens tested in SHG zinc but lower for the HPS100W specimens, indicating a higher susceptibility to cracking during galvanizing. Figure 4-28 shows the strain-at-failure data for the four steel grades (A36, A572-50, A572-65, and HPS100W) tested in air and in zinc with 0.1% bismuth (SHG + 0.1% Bi). The reductions in strain at failure for all steel grades for SHG + 0.1% Bi were similar to those obtained for tests in pure SHG zinc bath: 66% versus 65% for A36; 54% versus 57% for A572-50; 54% versus 58% for A572-65; and 74% versus 70% for HPS100W for tests in SHG + 0.1% Bi versus SHG, respec- tively. Also, the average strains at failure for A36, A572-50, A572-65, and HPS100W specimens tested in SHG + 0.1% Bi were very similar to those for specimens tested in SHG, indicating no effect of the addition of bismuth on cracking propensity of all steel grades. Figure 4-29 shows a comparison of strain-at-failure data for the four steel grades (A36, A572-50, A572-65, and HPS100W) tested in air and in SHG + 1% Pb. The reductions in strains at failure for all steel grades for SHG + 1% Pb were similar to those obtained for tests in pure SHG zinc bath: 67% versus 65% for A36; 58% versus 57% for A572-50; 58% versus 58% for A572-65; and 72% versus 70% for HPS100W for tests in SHG + 1% Pb versus SHG, respectively. Figure 4-30 shows the strain-at-failure data for all steel grades (A36, A572-50, A572-65, and HPS100W) while tested in air and in SHG + 0.1% Bi + 1% Pb. The reductions in strains at failure for all steel grades for SHG + 0.1% Bi + 0.1% Pb were similar to those obtained for tests in pure SHG zinc bath: 61% versus 65% for A36; 54% versus 57% for A572-50; 52% versus 58% for A572-65; and 69% versus 70% for HPS100W for tests in SHG + 0.1% Bi + 1% Pb versus SHG, respectively. 0 0.05 0.1 0.15 0.2 0.25 St ra in a t F ai lu re (i n. /i n. ) 65% reduction 57% reduction 58% reduction 70% reduction In air In SHG zinc A36 A572-50 A572-65 HPS100W A36 A572-50 A572-65 HPS100W Figure 4-27. Strain at failure for in-air and SHG zinc tests.

36 Mitigation of Weldment Cracking in Steel Highway Structures Due to the Galvanizing Process 0 0.05 0.1 0.15 0.2 0.25 St ra in a t F ai lu re (i n. /i n. ) 66% reduction 54% reduction 54% reduction 74% reductionIn air In SHG + 0.1% Bi A36 A572-50 A572-65 HPS100W A36 A572-50 A572-65 HPS100W Figure 4-28. Strain at failure for in-air and SHG + 0.1% Bi tests. 0 0.05 0.1 0.15 0.2 0.25 St ra in a t F ai lu re (i n. /i n. ) 67% reduction 58% reduction 58% reduction 72% reduction In air In SHG + 1% Pb A36 A572- 50 A572- 65 HPS100W A36 A572-50 A572-65 HPS100W Figure 4-29. Strain at failure for in-air and SHG + 1% Pb tests.

Experimental Test Program 37   Stress Concentration Figure 4-31 shows the strain-at-failure data for un-notched and notched A36 steel specimens tested in two in-zinc environments: SHG and SHG + 0.1% Bi + 1% Pb. Two SCFs were consid- ered: 1.5 and 2.5. The main observations were as follows: 1. Stress concentration had a large effect on total strain, both in air and in zinc. The A36 specimens exhibited a reduction in total strain of 73.1% in air due to the introduction of a SCF = 1.5 and an additional 62.6% reduction due to exposure to SHG zinc. For SCF = 2.5, the reduction in total strain in air was 70.7% with an additional 78.3% due to exposure to SHG zinc. 2. The notched specimens with SCF = 1.5 experienced similar strain loss in SHG zinc to that of the un-notched specimens (62.6% for notched versus 65% for un-notched specimens). However, the notched specimens with SCF = 2.5 experienced greater strain loss in SHG zinc (78.3% for notched versus 65% for un-notched specimens). Thus, strain loss for notched specimens in zinc was much larger than for similar specimens in air, indicating an effect of stress concentration when the specimen is exposed to liquid zinc. 3. For A36 steel, the presence of 0.1% bismuth and 1% lead led to a slightly and consistently larger increase in strain loss than SHG zinc. For SCF = 1.5, the strain loss was 62.6% for SHG and 64.2% for SHG + 0.1% Bi + 1% Pb; for SCF = 2.5, the strain loss was 78.3% for SHG and 78.4% for SHG + 0.1% Bi + 1% Pb. Figure 4-32 shows the strain-at-failure data for un-notched and notched A572-50 steel specimens tested in both SHG and SHG + 0.1% Bi + 1% Pb. The main observations were as follows: 1. Stress concentration had a large effect on ductility, both in air and in zinc. The A572-50 speci- mens with SCF = 1.5 experienced a 70.5% reduction in total strain in air because of the notch 0 0.05 0.1 0.15 0.2 0.25 St ra in a t F ai lu re (i n. /i n. ) 61% reduction 54% reduction 52% reduction 69% reduction In air In SHG + 0.1% Bi + 1% Pb A36 A572- 50 A572- 65 HPS100W A36 A572-50 A572-65 HPS100W Figure 4-30. Strain at failure for in-air and SHG + 0.1% Bi + 1% Pb tests.

0 0.05 0.1 0.15 0.2 0.25 St ra in a t F ai lu re (i n. /i n. ) 73.1% reduction 62.6% reduction 70.7% reduction 78.3% reduction 65% reduction 64.2% reduction 78.4% reduction Un-notched Notched: SCF = 1.5 Notched: SCF = 2.5 In air SHG In air SHG SHG+ Bi+Pb In air SHG SHG+Bi+Pb Figure 4-31. Strain at failure for A36 steel for different stress concentrations and zinc chemistries. 0 0.05 0.1 0.15 0.2 0.25 St ra in a t F ai lu re (i n. /i n. ) 70.5% reduction 67.1% reduction 68.5% reduction 79.1% reduction 57% reduction 70.4% reduction 81% reduction In air SHG In air SHG SHG+Bi+Pb In air SHG SHG+Bi+Pb Un-notched Notched: SCF = 1.5 Notched: SCF = 2.5 Figure 4-32. Strain at failure for A572-50 steel for stress concentration and different zinc chemistries.

Experimental Test Program 39   and another reduction of 67.1% due to exposure to SHG zinc. For SCF = 2.5, a reduction of 68.5% was introduced in air because of the notch and another 79.1% due to exposure to SHG. 2. Notched specimens experienced a higher percentage of total strain loss than un-notched specimens in SHG zinc (67.1% for notched specimens with SCF = 1.5; 79.1% for notched specimens with SCF = 2.5; and 57% for un-notched specimens). The strain loss for notched specimens was substantially higher in zinc than in air, indicating an effect of stress concentration when the specimen is exposed to liquid zinc. 3. The presence of 0.1% bismuth and 1% lead slightly but consistently amplified the strain loss compared with SHG zinc. For SCF = 1.5, the strain reduction was 67.1% and 70.4% for SHG and SHG + 0.1% Bi + 1% Pb, respectively. For SCF = 2.5, the strain reduction was 79.1% and 81.0% for SHG and SHG + 0.1% Bi + 1% Pb, respectively. These data show that bath chemistry has a slightly larger effect on notched A572-50 specimens than on notched A36 specimens. Figure 4-33 shows the strain-at-failure data for HPS100W steel for un-notched and notched specimens tested in both SHG and SHG + 0.1% Bi + 1% Pb. The main observations were as follows: 1. Stress concentration had a large effect on total strain of HPS100W specimens, both in air and in zinc. For SCF = 1.5, the HPS100W specimens experienced a reduction of 76.6% in total strain in air because of the notch and another 86.1% reduction due to SHG zinc exposure. For SCF = 2.5, a reduction in air of 80.9% was introduced because of the notch, and there was another reduction of 82.5% due to SHG zinc. In general, notched HPS100W specimens exhibited a relatively low strain at failure when tested in zinc (1%-1.8%). 0 0.05 0.1 0.15 0.2 0.25 St ra in a t F ai lu re (i n. /i n. ) 76.6% reduction 86.1% reduction 80.9% reduction 82.5% reduction 70% reduction 91.1% reduction 76.7% reductionIn air SHG In air SHG SHG+Bi+Pb In air SHG Un-notched Notched: SCF = 1.5 Notched: SCF = 2.5 SHG+Bi+Pb Figure 4-33. Strains at failure (df) for HPS100W steel for stress concentration and different zinc chemistries.

40 Mitigation of Weldment Cracking in Steel Highway Structures Due to the Galvanizing Process 2. The notched specimens experienced a higher strain reduction than un-notched specimens in SHG zinc (86.1% for notched specimens with SCF = 1.5; 82.5% for notched specimens with SCF = 2.5; and 70% for un-notched specimens). The strain loss for notched specimens was substantially higher in zinc than in air, indicating an effect of stress concentration when the specimen is exposed to liquid zinc. 3. The presence of 0.1% Bi + 1% Pb increased strain reduction compared with SHG zinc from 86.1% (SHG) to 91.1% (SHG + 0.1% Bi + 1% Pb) for SCF = 1.5 but decreased the strain reduc- tion for SCF = 2.5 from 82.5% (SHG) to 76.7% (SHG + 0.1% Bi + 1% Pb). HAZ Properties To evaluate the effect of heat inputs that would result from various welding processes, the 252-type specimens were preheated to the same peak temperature (1300°C) and cooled at three different rates in the 800°C-500°C range. Cooling rates were 12°C/sec (SCR), 30°C/sec (MCR), and 60°C/sec (FCR) to simulate welding heat inputs of 60 kJ/in. and 30 kJ/in. and an FCR of oxyfuel flame cutting, respectively. The specimens were notched to ensure that failure would occur at mid-length. Figure 4-34 shows the strain-at-failure data for A36 252-type specimens tested in air, in air after acid dip, and in zinc (SHG and SHG + 0.1% Bi + 1% Pb). The main observations were as follows: 1. The absolute strain values at failure were very small. 2. Acid dipping did not significantly affect strain at failure for in-air tests. 3. A36 specimens in air exhibited greater strains when cooled at FCR than at MCR or SCR, but specimens in SHG + 0.1% Bi + 1% Pb exhibited similar strains for FCR, MCR, and SCR. The largest strain loss occurred for specimens tested in SHG + 0.1% Bi + 1% Pb and cooled at FCR. 0 0.05 0.1 0.15 0.2 0.25 St ra in a t F ai lu re (i n. /i n. ) FCR MCR SCR SCRSCR SCRSCRSCRMCR MCR MCR MCR MCR MCRFCR 3% reduction 21% reduction 79% reduction 38% reduction 41% reduction 23% reduction 0% reduction In air SHG + 0.1% Bi + 1% Pb Acid Dip SHG Figure 4-34. Strains at failure (df) for A36 steel treated under different cooling rates.

Experimental Test Program 41   4. A36 specimens tested in SHG + 0.1% Bi + 1% Pb exhibited lower strains than those tested in SHG, indicating some effects of HAZ, notching, and bath chemistry on this grade of steel. Figure 4-35 shows the strain-at-failure data for A575-50 specimens tested in air, in air after acid dip, and in zinc (SHG and SHG + 0.1% Bi + 1% Pb). The main observations include: 1. Acid dipping did not show a significant or consistent effect on in-air test results (similar to that for A36 steel). 2. After submersion in the SHG + 0.1% Bi + 1.0% Pb bath for several minutes, the A572-50 steel specimens cooled at FCR failed after applying a very small initial pre-load and before start- ing the tension test. This behavior indicates a significant effect of the heat treatment (flame cutting or similar processes) on strain loss of A572-50 steel in zinc, possibly related to an increased hardness caused by the heat treatment. Because of this failure, the failure strains for the FCR specimens were recorded as zero (i.e., 100% strain loss). 3. All specimens tested in SHG + 0.1% Bi + 1% Pb exhibited lower strains than those tested in SHG, indicating effects of HAZ, notching, and bath chemistry on this grade of steel. Figure 4-36 shows the strain-at-failure data for HPS100W specimens tested in air, in air after acid dip, and in zinc (SHG and SHG + 0.1% Bi + 1% Pb). The main observations were as follows: 1. Acid dipping did not have a significant effect on in-air strains at failure (similar to that for A36 and A575-50 steels). 2. All specimens tested in SHG + 0.1% Bi + 1% Pb and cooled at any cooling rate (FCR, MCR, or SCR) exhibited premature failure in zinc before the tension tests could be conducted. This result indicates an effect of cooling rate and bath chemistry on this steel grade. Because of this failure, the failure strains were recorded as zero (i.e., 100% strain loss). 3. None of the specimens in the SHG bath failed before loading, indicating an effect of HAZ, notching, and bath chemistry on this grade of steel. 0 0.05 0.1 0.15 0.2 0.25 St ra in a t F ai lu re (i n. /i n. ) FCR FCR MCR SCR SCRMCR MCRMCR FCR FCR FCRFCR 13% reduction 14% increase 28% increase 100 % reduction 55% reduction 54% reduction 3 % increase In air SHG + 0.1% Bi + 1% Pb Acid Dip SHG SCRSCRMCRMCR Figure 4-35. Strains at failure (df) for A575-50 steel treated under different cooling rates.

42 Mitigation of Weldment Cracking in Steel Highway Structures Due to the Galvanizing Process Cold-Working Figure 4-37 shows the strain values at failure for non-cold-worked and cold-worked A36 steel specimens tested in SHG and SHG + 0.1% Bi + 1% Pb. Results for the two cold-work levels applied in uniaxial tension (6% and 8%, or CW6 and CW8) indicate the following: 1. Cold-worked specimens exhibited lower strains at failure in air than non-cold-worked speci- mens. In general, higher cold-work levels resulted in lower strains in air (εf = 0.203 for CW6 and 0.162 for CW8 A36 steel). 2. Despite loss in total strain after cold-working, the strain at failure in zinc was similar to that for non-cold-worked specimens, indicating no meaningful effect of cold-working on A36 in SHG zinc. For example, the strain at failure in SHG zinc was εf = 0.086 for non-cold-worked, 0.073 for CW6, and 0.084 for CW8 specimens. This explains the lower loss due to zinc for cold-worked specimens (64% for CW6 and 48% for CW8) compared to 65% for non-cold- worked specimens. 3. The presence of 0.1% bismuth and 1% lead showed no effect on strain for CW6 specimens (εf = 0.075 in SHG + 0.1% Bi + 1% Pb versus εf = 0.0729 in SHG) but a slight negative effect on strain for CW8 specimens. Figure 4-38 shows the strain-at-failure values for non-cold-worked and cold-worked A572-50 steel specimens tested in both SHG and SHG + 0.1% Bi + 1% Pb. The main observations were as follows: 1. Cold-worked A572-50 specimens showed lower strains at failure in air than non-cold-worked specimens (εf = 0.118 for CW6 specimens and 0.134 for CW8 specimens). 0 0.05 0.1 0.15 0.2 0.25 St ra in a t F ai lu re (i n. /i n. ) 29% reduction 13% reduction 14% reduction 100% reduction 62% reduction 23% reduction 0% reduction FCR FCR FCRMCR MCR MCR MCR MCRSCR SCR SCR SCR In air SHG + 0.1% Bi + 1% Pb Acid Dip SHG FCR FCR Figure 4-36. Strains at failure (df) for HPS100W steel treated under different cooling rates.

Experimental Test Program 43   0 0.05 0.1 0.15 0.2 0.25 St ra in a t F ai lu re (i n. /i n. ) 30% reduction 37% reduction 34% reduction 48% reduction In air SHG In air SHG SHG + 0.1% Bi + 1% Pb No cold-work 6% cold-work 8% cold-work SHG + 0.1% Bi + 1% Pb Figure 4-38. Strains at failure (df) for A572-50 steel for cold-working and different zinc chemistries. 0 0.05 0.1 0.15 0.2 0.25 St ra in a t F ai lu re (i n. /i n. ) 64% reduction 48% reduction63% reduction 52% reduction In air SHG In air SHGSHG + 0.1% Bi + 1% Pb No cold-work 6% cold-work 8% cold-work SHG + 0.1% Bi + 1% Pb Figure 4-37. Strains at failure (df) for A36 steel for cold-working and different zinc chemistries.

44 Mitigation of Weldment Cracking in Steel Highway Structures Due to the Galvanizing Process 2. Despite the loss in total strain after cold-working, strains at failure in zinc were at a level simi- lar to those for non-cold-worked specimens, indicating no effect of cold-working on A572-50 steel in SHG zinc. For example, the strain at failure in zinc was 0.082 for non-cold-worked specimens (Figure 4-27), 0.083 for CW6, and 0.085 for CW8 specimens. This explains the lower percentage strain loss due to zinc for cold-worked specimens (30% for CW6 and 37% for CW8) compared with non-cold-worked specimens (57%). 3. The presence of 0.1% bismuth and 1% lead showed no effect on the strain for CW6 specimens (εf = 0.0877 for SHG + 0.1% Bi + 1% Pb versus εf = 0.0825 for SHG) but a slight negative trend for the 8% cold-worked specimens. Figure 4-39 shows the strain-at-failure data for non-cold-worked and cold-worked HPS100W steel specimens tested in both SHG and SHG + 0.1% Bi + 1% Pb. The main observations include: 1. Cold-worked specimens showed lower total strains at failure in air compared with non-cold- worked specimens. For example, εf = 0.097 for CW6 and 0.085 for CW8 specimens. 2. Despite the elongation loss after cold-working, strains at failure in zinc were similar to those for non-cold-worked specimens, indicating no effect of cold-working on HPS100W steel in SHG zinc. For example, the strain at failure in zinc was εf = 0.043 for non-cold-worked specimens (Figure 4-27), εf = 0.041 for CW6, and εf = 0.053 for CW8. This explains the lower percentage strain loss due to zinc for cold-worked specimens (58% for CW6 and 37% for CW8) compared with non-cold-worked specimens (70%). 3. The presence of 0.1% bismuth and 1% lead had no effect on strain for CW6 specimens (εf = 0.041 for SHG versus εf = 0.045 for SHG + 0.1% Bi + 1% Pb) but a slight negative effect on the CW8 specimens. 0 0.05 0.1 0.15 0.2 0.25 St ra in a t F ai lu re (i n. /i n. ) 58% reduction 37% reduction55% reduction 53% reduction In air SHG In air SHG SHG + 0.1% Bi + 1% Pb SHG + 0.1% Bi + 1% Pb No cold-work 6% cold-work 8% cold-work Figure 4-39. Strains at failure (df) for HPS100W steel for cold-working and different zinc chemistries.

Experimental Test Program 45   Relative Effect of Factors Figure 4-40 shows average strain-at-failure data for specimens tested both in air and in SHG zinc. The total height of each bar shows the average strain at failure in air, and the height of the solid portion indicates the average strain at failure in SHG zinc. Thus, the portion of each bar filled with the horizontal line patterns depicts the absolute reduction of strain at failure from in air to in SHG zinc. The percentage of reduction is shown numerically on top of each bar (the short line superimposed on each bar denotes 50% of strain capacity in air, as a visual benchmark). The bars associated with the different steel grades (i.e., A36, A572-50, A572-65, and HPS100W) show average results from bare tension specimens without cold-work, stress concentration, or heat treatment. The bars associated with other factors show an average of three steel grades, excluding HPS100W steel because of its substantially lower strain at failure. Comparing the data for A572-50, A572-65, and A36 (three rightmost bars in Figure 4-40), it is apparent that the three steel grades exhibit different strains at failure in air but similar average strains in zinc, indicating that a higher strain at failure in air does not necessarily result in a higher strain at failure in zinc. Regarding cold-worked specimens, CW6 and CW8 resulted in different strains at failure in air but capacities in zinc that were similar to those for specimens without cold-work. This indicates that 6% and 8% cold-work levels alone do not contribute to strain loss. Other factors, including stress concentration and heat treatment cooling rate, showed apparent interactive effects with other factors that influence strain at failure (and thus influence the 0.0000 0.0500 0.1000 0.1500 0.2000 0.2500 SCR MCR FCR SC 1.5 SC 2.5 CW 8 HPS100W CW 6 A50 A65 A36 Av er ag e St ra in a t F ai lu re (i n. /i n. ) Factor Note: SC, CW, and Cooling exclude HPS100W. 12.9% reduction 13.8% reduction 94.4% reduction 64.6% reduction 78.7% reduction 46.6% reduction 69.5% reduction 50.8% reduction 57.6% reduction 55.0% reduction 65.2% reduction Total bar: average strain at failure in air Solid bar: average strain at failure in SHG zinc Figure 4-40. Average strain at failure associated with different factors (in air and in SHG zinc).

46 Mitigation of Weldment Cracking in Steel Highway Structures Due to the Galvanizing Process occurrence of LMAC in steel structures during galvanizing). The HAZ specimens had a small diameter (0.252 in.) and stress concentration (SCF = 2.5); the strains at failure for these speci- mens (SCR, MCR, and FCR) were affected by both heat treatment and stress concentration. For this reason, the comparison was limited to the effect of the cooling rates on the HAZ specimens (the strains at failure were very low, below εf = 0.050). A distinct observation among the SCR, MCR, and FCR data is that the FCR (which corre- sponds to thermal cutting) resulted in a 100% reduction of strain at failure in SHG zinc when combined with stress concentration. In practical applications, thermal cutting imposes fast cool- ing and introduces surface defects (stress concentration) to the steel (Wood, 1994). On the other hand, specimens subjected to MCRs and SCRs—which correspond to single- pass and large multi-pass welds, respectively—showed much smaller reductions in strain at failure in SHG zinc than in air when combined with the same stress concentration. Thus, MCRs and SCRs result in strain reduction in air but very small reduction in SHG zinc. Stress concen- tration, in this case, is likely to have a more significant effect on the strain at failure in SHG zinc. Thus, the occurrence of cracks around weld toes after galvanizing could be attributed to stress concentrations associated with the non-smooth geometry at the weld toes. Stress concentration (SC1 = 1.5 and SC2 = 2.5) was found to influence strain at failure. A higher SCF leads to a larger reduction in strain at failure and results in a lower strain at failure in zinc than in air. The strain-at-failure data obtained from the tests performed on specimens subjected to differ- ent levels of cooling, stress concentration, and cold-working indicated that cooling rate has the largest effect on strain at failure, followed by stress concentration, and then by cold-working. Figure 4-41 shows the data for specimens in SHG + 0.1% Bi + 1% Pb. These data indicate small differences from those obtained for SHG zinc. Figure 4-42 shows averages of the data for both zinc baths, indicating similar conclusions regarding the relative effect of the different variables. WOL Specimens The effect of severe flaws on susceptibility to cracking during galvanizing was investigated using a series of WOL specimens. This type of specimen, shown in Figure 4-43, has been used by other researchers to analyze the crack propagation behavior of steel subjected to different environmental conditions and to study stress corrosion in salt water, as well as susceptibility to cracking of steels used in pressure vessels in the chemical processing industry (Barsom, 1971; Novak & Rolfe, 1969). The tests were performed to capture the effect of a sharp crack in a structure during galvanizing. The KISCC—a measure of susceptibility to sub-critical crack- ing developed to measure the susceptibility of a material to an environment, stress-corrosion cracking, or hydrogen embrittlement (Barsom, 1971; Novak & Rolfe, 1969)—was used in this evaluation. Test Setup and Methods The WOL is loaded to various levels of applied stress in the presence of a crack (K1) and submerged in the galvanizing bath. The WOL specimens included a sharp crack in order to simulate a crack formed in a structure during the fabrication process and to study crack growth caused by the galvanizing process. The change in strain experienced by the specimens upon immersion in the galvanizing bath was assumed to represent that of a structure with sharp, crack-like defects.

Experimental Test Program 47   0.0000 0.0500 0.1000 0.1500 0.2000 0.2500 SCR MCR FCR SC 1.5 SC 2.5 HPS100W CW 8 CW 6 A50 A65 A36 Av er ag e St ra in a t F ai lu re (i n. /i n. ) Factor Note: SC, CW, and Cooling exclude HPS100W. 48.6% reduction 42.6% reduction 88.8% reduction 67.0% reduction 79.6% reduction 68.5% reduction 50.3% reduction 47.8% reduction 57.9% reduction 52.4% reduction 65.0% reduction Total bar: average strain at failure in air Solid bar: average strain at failure in SHG zinc + 0.1% Bi + 1% Pb Figure 4-41. Average strain at failure associated with different factors (in air and in SHG zinc + 0.1% Bi + 1% Pb). A 0.15-in.-long fatigue crack was introduced in each WOL specimen before testing in the galvanizing bath (see Figure 4-44). The crack was formed using a uniaxial loading system that controls the fatigue loading cycles to automatically form a crack with a specified length. A crack opening displacement (COD) gauge installed at the mouth of the notch measured mouth open- ing for crack length calculation. All cracks were introduced under a fatigue load with a stress intensity factor of K = 20 ksi • √in. to avoid the development of a plastic zone at the crack tip. After the cracks were introduced, each specimen was loaded using the threaded bolt to achieve a specific stress intensity K1 or K2. K1 is obtained from Equation 4-1, based on the requirement in ASTM E399 to limit crack tip plasticity, as follows: = σ 2.5 Equation 4-11 0K by where b0 is the distance from the crack tip to the back of the WOL specimen and σy is the yield strength of steel. K2 = 91 ksi • √in., a median stress intensity value determined according to ASTM E1921. The resulting K1 values were 25.1 ksi • √in for A36, 34.8 ksi • √in for A572-50, 45.2 ksi • √in for A572-65, and 69.6 ksi • √in for HPS100W. These values were then translated to crack mouth opening displacements according to ASTM E1820 provisions and introduced in the test speci- mens. Subsequently, the specimens were submerged in the galvanizing bath with SHG + 0.1% Bi + 1% Pb for 0.5 hour or 24 hours. A small amount of acid was applied directly to the crack

48 Mitigation of Weldment Cracking in Steel Highway Structures Due to the Galvanizing Process 0.0000 0.0500 0.1000 0.1500 0.2000 0.2500 SCR MCR FCR SC 1.5 SC 2.5 CW 8 HPS100W CW 6 A50 A65 A36 Av er ag e St ra in a t F ai lu re (i n. /i n. ) Factor Note: SC, CW, and Cooling exclude HPS100W. 12.9% reduction 13.8% reduction 94.4% reduction 64.6% reduction 78.7% reduction 46.6% reduction 69.5% reduction 50.8% reduction 57.6% reduction 55.0% reduction 65.2% reductionTotal bar: average strain at failure in air Solid bar: average strain at failure in both zinc baths: SHG zinc and SHG + 0.1% Bi + 1% Pb Figure 4-42. Average strain at failure associated with different factors (in air and in both zinc baths: SHG zinc and SHG + 0.1% Bi + 1% Pb). Machined notch Fine-thread bolt Tool-steel loading surface for bolt Steel material being tested Figure 4-43. Modified WOL specimen.

Experimental Test Program 49   tip of some WOL specimens before they were submerged in zinc to investigate the potential of cracking due to hydrogen embrittlement caused by the pickling process. After testing in zinc, the specimens were retrieved and inspected to determine propagation of the initial fatigue crack. Table 4-8 shows the test matrix for the WOL specimens. Test Results A WOL specimen made of the A572-65 steel (Specimen #3) was tested in SHG + 0.1% Bi + 1% Pb for 30 minutes, then retrieved and polished to remove the zinc coating, and acid etched. The specimen was then loaded with the threaded bolt to open up the specimen for inspec- tion. The measurement of crack length (Figure 4-45) indicated no crack growth; crack length remained at 0.15 inch. Pre -crack COD gauge Figure 4-44. Pre-cracking test setup showing the COD gauge and a WOL specimen with a 0.15-in. pre-crack (A572-65 steel). Test Groups Specimen # Steel Grade K at 70°F (ksi·√in) ΔK (ksi·√in) for ΔT = 770°F Bolt Length at 70°F, L0 (in.) Acid Before dipping? Dipping Time (hours) Zinc Bath 1 1 A36 25.1* 0.4711 0.0688 No 0.5 SH G + 0. 1% Bi + 1% Pb 2 A572-50 34.8* 0.5161 0.0712 3 A572-65 45.2 0.5643 0.0738 4 HPS100W 69.6 0.6768 0.0799 2 5 A36 25.1 0.4711 0.0688 246 A572-50 34.8 0.5161 0.0712 7 A572-65 45.2 0.5643 0.0738 3 8 A36 25.1 0.4711 0.0688 Yes 0.5 9 A572-50 34.8 0.5161 0.0712 10 A572-65 45.2 0.5643 0.0738 4 11 A36 91.0** 0.7758 0.0852 No 12 A572-50 5 13 A36 Yes 14 A572-50 * = K1, ** = K2 Table 4-8. Test matrix for the WOL specimens.

50 Mitigation of Weldment Cracking in Steel Highway Structures Due to the Galvanizing Process The other specimens were subsequently tested; Figure 4-46 shows photos of the specimens after testing and polishing for inspection. The inspection indicated no crack propagation resulted from immersion in zinc. The original 0.15 in. crack in Specimen #11 propagated during self-loading due to the higher stress intensity factor, but no further change occurred during the test in zinc. Finite Element Analysis The temperature to which each WOL was exposed during galvanizing was expected to intro- duce non-uniform thermal expansion between the WOL specimen and the loading bolt that would influence the stress intensity factor, K. The increased temperature causes expansion of both the WOL and the loading bolt, resulting in increased crack opening. If the increase of crack opening due to WOL expansion is larger than the elongation of the loading bolt, force relax- ation will occur at the crack tip, leading to a lower stress intensity factor. However, if the bolt elongation is larger than the increase of crack opening due to WOL expansion, the net force will increase at the loading bolt and lead to a higher stress intensity factor. Since K is directly related to crack opening width, the thermal elongation of the loading bolt and the increase of crack opening due to WOL expansion were calculated first to obtain the change in K due to galvanizing temperature. Bolt elongation was calculated using Equation 4-2: ∆ = α∆ Equation 4-20L TLbolt where α is the thermal expansion coefficient of steel (6.0 × 10−6 in./in./F), ΔT is the tempera- ture increase from RT (70°F) to galvanizing temperature (840°F), and L0 is the initial length of the loading bolt between the gap at 70°F that is subject to thermal expansion (see Table 4-8). The calculated ΔLbolt values ranged from 0.00032 in. to 0.00039 in. Detailed dimensions of the WOL specimen are shown in Figure 4-47. The increase of the crack opening due to thermal expansion of the WOL was obtained through finite element (FE) analysis. Properties of structural steel (modulus of elasticity of 29,000 ksi, and a density of 0.284 pound per cubic inch [lb/in.3]) and thermal properties (coefficient of thermal expansion Figure 4-45. Measuring the crack length after submerging the WOL in SHG + 0.1% Bi + 1% Pb for 30 minutes (A572-65 steel).

Experimental Test Program 51   Specimen #1: A36 Specimen #3: A572-65 Specimen #2: A572-50 Specimen #4: HPS100W Specimen #5: A36 Specimen #6: A572-50 Specimen #7: A572-65 Specimen #8: A36 Figure 4-46. WOL specimens after being tested in zinc and then polished. (continued on next page)

52 Mitigation of Weldment Cracking in Steel Highway Structures Due to the Galvanizing Process of 6.0 × 10−6 in./in./F and a specific heat of 0.11 British thermal unit per pound per degree Fahrenheit [Btu/lb/°F]) were assigned to the material being simulated. The crack was modeled by creating a 0.15 in. by 1.1 in. rectangular shell element in conjunction with the extended finite element method (XFEM). Figure 4-48 shows the FE model and the vertical displacement field due to thermal expansion under 770°F. The increase in crack opening at the location of the bolt was ΔLWOL = 0.0002 in., which was less than the bolt elongation ΔLbolt. Thus, the exposed length of the loading bolt expanded more than the WOL specimen at the bolt location, resulting in an increased stress intensity at the crack tip during the galvanizing process. The net increase in stress intensity factor, ΔK, for each specimen case was calculated from the difference between ΔLbolt and ΔLWOL according to ASTM E1820. The K value for each loaded specimen at 840°F was obtained by adding ΔK to the K value computed for the RT case; Table 4-8 shows both K at 70°F and ΔK values for each specimen; these values indicate small increases for each specimen during galvanizing. Specimen #9: A572-50 Specimen #10: A572-65 Specimen #11: A36 Specimen #12: A572-50 Specimen #13: A36 Specimen #14: A572-50 Figure 4-46. (Continued).

Experimental Test Program 53   Figure 4-48. FE model of the WOL (left) and the vertical displacement field under thermal loading (right). Figure 4-47. Detailed dimensions of the WOL specimen.

Next: Chapter 5 - Computational Simulations High Mast Illumination Poles »
Mitigation of Weldment Cracking in Steel Highway Structures Due to the Galvanizing Process Get This Book
×
MyNAP members save 10% online.
Login or Register to save!
Download Free PDF

Structural supports for signs, luminaires, and traffic signals and other steel highway structures are generally galvanized to prevent corrosion and provide a long service life. However, recent investigations have revealed incidents of cracking in weldments of galvanized structures that appear to be induced during the galvanizing process.

The TRB National Cooperative Highway Research Program's NCHRP Research Report 965: Mitigation of Weldment Cracking in Steel Highway Structures Due to the Galvanizing Process proposes improved design, materials, and construction specifications of galvanized steel highway structures to mitigate weldment cracking caused by the galvanizing process.

Supplemental materials to the report are appendices that provide details of the work performed in the project.

  1. ×

    Welcome to OpenBook!

    You're looking at OpenBook, NAP.edu's online reading room since 1999. Based on feedback from you, our users, we've made some improvements that make it easier than ever to read thousands of publications on our website.

    Do you want to take a quick tour of the OpenBook's features?

    No Thanks Take a Tour »
  2. ×

    Show this book's table of contents, where you can jump to any chapter by name.

    « Back Next »
  3. ×

    ...or use these buttons to go back to the previous chapter or skip to the next one.

    « Back Next »
  4. ×

    Jump up to the previous page or down to the next one. Also, you can type in a page number and press Enter to go directly to that page in the book.

    « Back Next »
  5. ×

    To search the entire text of this book, type in your search term here and press Enter.

    « Back Next »
  6. ×

    Share a link to this book page on your preferred social network or via email.

    « Back Next »
  7. ×

    View our suggested citation for this chapter.

    « Back Next »
  8. ×

    Ready to take your reading offline? Click here to buy this book in print or download it as a free PDF, if available.

    « Back Next »
Stay Connected!