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.
10 Organization of Factors Contributing to Liquid Metal Assisted Cracking The literature review led to the identification of a number of factors contributing to cracking associated with the galvanizing process. Two cracking phenomena associated with galvanizing were identified; these were referred to as Type 1 and Type 2 cracking. ⢠Type 1 cracking occurs during galvanizing (i.e., during dipping, dwelling, extracting, or cool- ing). The majority of Type 1 cracking appears to be related to LMAC; literature describing the effects of contributing factors is discussed in Appendices A and B. ⢠Type 2 cracking is premature fatigue cracking of in-service galvanized steel highway struc- tures (as compared to black steel structures); limited literature addresses this type of cracking (Goyal et al., 2012; Magenes, 2011; Ocel, 2014; Pool, 2010; Stam et al., 2011). Figure 3-1 shows the factors associated with galvanizing welded steel highway structures. Contributing factors have been differentiated based on whether cracks can be identified imme- diately after galvanizing in the galvanizing facility (Type 1 cracking) and premature fatigue cracking that occurs in the structure after it has been placed in service (Type 2 cracking). Factors identified as affecting cracking sensitivity during or immediately after galvanizing (Type 1 cracking) were grouped into three categories, according to the cause of cracking: ther- mal effects, LMAC, and embrittlement. The literature (summarized in Appendix B) examined these factors separately, providing no information on the extent of interactions between the contributing factors or their relative influence. Type 1 Cracking The factors contributing to the cause of Type I cracking, organized into three categories, are described in Figure 3-1. Thermal Effects Thermal effects were found to be a key contributor to cracking during galvanizing welded steel structures by influencing stresses and strains in a steel part (Feldmann et al., 2009; Kinstler, 2005; Kleineck, 2011; Rudd et al., 2008; Vervisch, 2009). Galvanizing inherently subjects steel components to a high-temperature environment (â¼840°F) that can produce significant thermal strains and stresses in the structure, particularly welded structures that typically include components of various dimensions and masses. Differences in geometry and mass within a structure lead to differential thermal expansion and contraction during galvanizing, resulting in C H A P T E R 3 Factors Contributing to Weldment Cracking Due to the Galvanizing Process
Factors Contributing to Weldment Cracking Due to the Galvanizing Process 11  thermally induced stresses and strains that can be significant. Locations of geometric transition also result in localized stress concentrations that amplify thermally induced stresses and strains during galvanizing. Consider as an example an HMIP, which is a relatively simple structure consisting of a thin steel pole with a relatively thick base plate welded to the pole. As the HMIP structure is dipped in the galvanizing bath, the thin pole element tends to expand rapidly while restrained at one end by a much thicker base plate elementâwhose temperature rises slowlyâsubjecting the weld connecting the two elements to large stresses/strains during the dipping and cooling stages of the galvanizing cycle. This thermal response is further amplified by the stress concentrations at the weld detail and at other geometric discontinuities (e.g., corner bends of poles with polygonal cross-sections). Details of the galvanizing process such as dipping angle, dipping speed, and dwell time also influence the temperature distribution in the structure being galvanized, which in turn influence the localized stresses and strains in the structure. Liquid Metal Assisted Cracking As introduced in Chapter 2, LMAC is a primary contributor to cracking during galvanizing. The following sections discuss the factors that influence LMAC in steel structures during galva- nizing (detailed discussions are provided in Appendices A and B). Steel Chemistry Certain elements in steel (silicon in particular) have shown to increase the likelihood of cha- otic growth of intermetallic layers between the zinc and steel substrate (Maass & Peissker, 2011), which could influence the propensity of a structural member to experience LMAC. A study indicated that the presence of phosphorus and boron increases the susceptibility of steels to LMAC, and it suggested controlling steel composition as a means of mitigating galvanizing- induced cracking (Tomoe Corporation, 2001). This study also recommended a formula for carbon equivalent zinc (CEZ) to evaluate steel composition. Galvanizing Bath Chemistry Several studies focused on the effects of alloying elements used in the galvanizing bath (Feldmann et al., 2009; Judd, 2006; Katiforis & Papadimitriou, 1996; Kinstler, 2005; Maass & Peissker, 2011; Type 1 Cracking: During Galvanizing (In the galv. shop) Type 2 Cracking: Premature Fatigue Cracking (In the field) Life of Structure Thermal Effects LMAC Embrittlement ⢠Galvanizing bath chemistry ⢠Steel chemistry ⢠Residual stresses ⢠Geometry ⢠Cold-working ⢠Hydrogen embrittlement ⢠Strain-age embrittlement Undetected Type 1 cracking or micro-cracking Intermetallics ⢠Structure of intermetallic layers as potential initial flaws ⢠Small cracks in galvanizing coating propagating into base metal ⢠Differential rate of expansion between welded components Figure 3-1. Sources of cracking in galvanized structures and influential variables.
12 Mitigation of Weldment Cracking in Steel Highway Structures Due to the Galvanizing Process Mraz & Lesay, 2009; Poag & Zervoudis, 2003; Rudd et al., 2008). Alloying elements such as tin, bismuth, and lead were found to influence the propensity of a steel structure to undergo LMAC during the galvanizing process, with tin having the most pronounced effect. Residual Stresses Residual stresses/strains have been found to increase susceptibility to LMAC; the potential for cracking increases as the magnitude of tensile residual stresses and strains in a member increase. However, other factors need to be present in order for cracking to occur during gal- vanizing (Beal et al., 2012; Beldyk, 2004; Cristol et al., 2012; Džupon et al., 2013; Elboujdaini et al., 2004; European General Galvanizers Association, 2014; Feldmann et al., 2006; Katzung & Schulz, 2005; Kinstler, 2005; Kuklik, 2012; Mraz & Lesay, 2009; Poag & Zervoudis, 2003; Rudd et al., 2008; Vermeersch et al., 2011). Cold-Working The effect of cold-working on LMAC has been considered by a number of researchers (Clegg & Jones, 1998; Džupon et al., 2013; James, 2008; Kinstler, 2005; Sandelin, 1954; Vervisch, 2009). In general, higher levels of cold-working increase the risk of LMAC, particularly where surface flaws are induced by cold-working. However, many of the cases considered in the literature combine cold-working with other factors, such as stress concentrations occurring at corner radii (Poag & Zervoudis, 2003). Geometry Several researchers have investigated the influence of geometric details and flaws on the susceptibility of a steel part to LMAC. Both geometric details and flaws contribute to stress concentrations in the member; more severe stress concentrations increase the susceptibility of the member to LMAC (Feldmann et al., 2009; Poag & Zervoudis, 2003; Rudd et al., 2008). Welding and Thermal Cutting Welding and thermal cutting processes produce HAZ in the steel substrate, which have dif- ferent microstructural properties than the base metal. These fabrication processes also introduce flaws and residual stresses that contribute to LMAC. McDonald (1975) determined that for most grades of structural steel, high residual/applied stress and the presence of flaws are more relevant to LMAC than HAZ properties; this finding has been supported by other researchers (Aichinger & Higgins, 2006; Rudd et al., 2008). Time in Galvanizing Bath The amount of dwell time in the galvanizing kettle has shown to influence the susceptibility of the steel member to LMAC. Increased exposure time of a steel part increases the likelihood of LMAC occurrence (Feldmann et al., 2009; Kinstler, 2005). Embrittlement Two modes of embrittlement are relevant to the occurrence of cracking associated with the galvanizing process: hydrogen embrittlement and strain-age embrittlement. However, the literature has shown that embrittlement is not a major contributor to cracking associated with the galvanizing process for the grades of steel commonly used in highway structures, where yield strength is usually less than 100 kilopounds per square inch (ksi) (Carpio et al., 2010; Rudd et al., 2008; Vervisch, 2009). Factors influencing the susceptibility of steel to hydrogen embrittlement include the steel properties (particularly tensile strength); hydrogen introduced through steel manufacturing,
Factors Contributing to Weldment Cracking Due to the Galvanizing Process 13  welding, and acid pickling (Vervisch, 2009); and applied stress. While hydrogen embrittlement has not been of concern for grades of steel used in highway structures, it may be an aggravator to LMAC when combined with other factors. ASTM (2007), British Constructional Steelwork Association (BCSA) (2005), and British Standard (BS) (2009b) provide guidance on reducing hydrogen in parts to be galvanized. Factors influencing susceptibility to strain-age embrittlement include dwell time in the kettle (high heat increases the rate of strain-age embrittlement dramatically) and levels of available nitrogen (Vervisch, 2009). The literature indicated that strain aging is not a major contributor to LMAC during galvanizing steel structures (BCSA, 2005; BS, 2009a; Džupon et al., 2013; James, 2008). Type 2 Cracking Premature fatigue cracking occurring in structures (Type 2 cracking) is not well documented, although studies have demonstrated lower fatigue resistance of galvanized specimens compared with non-galvanized specimens (Ocel, 2014; Pool, 2010). The characteristics of the intermetallic layers formed during galvanizing were found to be highly variable and dependent upon the galvanizing bath chemistry, temperature, and steel composition. Some intermetallics are homogeneous and others are more chaotic (Katiforis & Papadimitriou, 1996; Maass & Peissker, 2011); a less uniform intermetallic matrix is expected to perform more poorly under fatigue loading than a well-ordered intermetallic matrix. Vogt et al. (2001) showed that the microscopic flaws in a galvanized coating, especially those that extend through the thickness of the coating, could act as crack initiators in the steel base metal. Thus, minimizing initial flaws would reduce the incidence of Type 2 cracking, and controlling the factors contributing to Type 1 cracking would reduce flaws in the galvanized structure and decrease susceptibility to Type 2 cracking. Development of Test Plan Several factors emerged from the literature review as consistently having detrimental effects on cracking during galvanizing. These were thermal effects; galvanizing bath chemistry; steel material; geometric effects; level of cold-working; magnitude of hardness (which can be influ- enced by cooling rate after welding or cutting processes); and magnitude of residual stresses. Several issues associated with these factors were identified. ⢠In regards to galvanizing bath chemistry, there was no consensus in the literature on the effect of combining some common galvanizing bath alloying elements, such as lead and bismuth. However, the detrimental effects of tin in the zinc bath have been well established and reported. Therefore, lead, bismuth, and lead-bismuth were included in the testing plan as alloying elements. ⢠The effects of combinations of influencing factors such as galvanizing bath chemistry, steel chemistry, cold-working, geometric effects, and HAZ properties have not been well inves- tigated or reported in the literature. ⢠The effects of geometric detailing on thermal stresses have been studied. However, these studies lacked measurement of strain data for elements being galvanized and have not resulted in conclusive findings. As the microstructure of the intermetallic layers is expected to play a role in the susceptibility of a galvanized structure to fatigue cracking, minimizing defects during the galvanizing process will likely decrease the occurrence of premature fatigue cracking for in situ structures.
14 Mitigation of Weldment Cracking in Steel Highway Structures Due to the Galvanizing Process Research Approach A research plan that included an experimental investigation and a computational study was developed to address the identified gaps in the body of knowledge. The experimental investiga- tion addressed the effect of different factors on the occurrence of LMAC during galvanizing. It included tension tests of steel specimens submerged in a galvanizing bath and wedge-opening loaded (WOL) specimens (intended to simulate severe flaws in the parts being galvanized) sub- merged in a galvanizing bath. The tension tests formed the core of the experimental study and enabled examination of the relative influence of the factors contributing to LMAC, including steel material; galvanizing bath chemistry; stress concentration (a proxy for both geometric detailing and flaws); cold-working; and HAZ properties. Kikuchi and Iezawa (1982) investigated the mechanical behavior of steel while immersed in liquid zinc and found a large decrease in total specimen elongation in the presence of liquid zinc. This finding was reproduced by Kinstler (1991) and considered in a large test program conducted by the Japanese company Tomoe Corporation (2001). This program resulted in the development of a CEZ relationship between the chemical composition of the steel and LMAC susceptibility. This approach was adopted in the research plan by devising a test setup to tension 505- and 252-type steel specimens to failure and measuring the strain at failure while the samples were submerged in a galvanizing bath. The strain at failure was used as a measure of the effect of the different factors on performance. Also, the effect of some factors on hard- ness was measured; these data were used as a proxy for LMAC susceptibility, as suggested by Elboujdaini et al. (2004). Factors Included in the Experimental Test Program Four grades of steel were included in the test program to cover a large range of strengths and steel chemistries: A36, A572-50, A572-65, and HPS100W. Steel plates for each grade were sourced from the same heat to ensure similar chemistry. The galvanizing bath was varied to study the influence of different bath compositions. The starting bath composition was special high grade (SHG) zinc, which is essentially pure zinc. Lead and bismuth were considered as alloying elements alone with SHG (i.e., SHG + 0.1% bismuth [Bi] and SHG + 1% lead [Pb]) and in combination (i.e., SHG + 0.1% Bi + 1% Pb). The influence of stress concentration was considered in tension specimens in which notches of two severities were introduced: one with SCF = 1.5 and another with SCF = 2.5. Cold-working was considered by pre-straining tension specimens to either 6% strain or 8% strain and then conducting the tension tests in zinc. In addition, a few tension specimens were extracted from the 90° bend of a plate that was deformed in a large brake press that gen- erated 14% cold-work at the extreme fibers of the bend (this corresponds to about 9% cold- work at the outer surface of the tension specimenâs reduced section). The influence of HAZ properties was studied on HAZ-simulated tension specimens that were heat-treated in a Gleeble⢠simulator to develop microstructures similar to those caused by welding/cutting at different cooling rates: 12°C/sec (degrees Celsius per second, which cor- responds to 60 kilojoules per inch [kJ/in.] of heat input), 30°C/sec (corresponds to 30 kJ/in. of heat input), and 60°C/sec (corresponds to oxyfuel flame cutting on a 2 in. plate). The effect of a very sharp crack was investigated using WOL specimens in which fatigue- generated cracks of similar length were introduced in each specimen and then submerged in the galvanizing bath in a loaded condition. Chapter 4 provides a detailed description of the experimental test program.
Factors Contributing to Weldment Cracking Due to the Galvanizing Process 15  Factors Included in the Computational Study The experimental test plan was designed to evaluate the effects of the factors affecting LMAC during galvanizing but not to address the thermal effects that are known to increase stresses/ strains and the likelihood for LMAC in full-scale structures during galvanizing. This included a computational study of the behavior of HMIPs of different geometries during the different stages of galvanizing. The study considered the poleâs cross-section (round versus polygonal) and included 8-, 10-, and 12-sided polygonal sections and base plate thickness (varied from 1.5 in. to 3.5 in.); a detailed description is provided in Chapter 5.