Not Quite Your Grandfather's Steel

High-strength A913 in today's green construction
[ Page 2 of 4 ]  previous page Page 1 Page 2 Page 3 Page 4 next page
Sponsored by The Steel Institute of New York
By William B. Millard, PhD

Image courtesy of Goettsch Partners

Union Station Tower at 320 South Canal in Chicago, designed by Goettsch Partners, incorporates high-strength steel members in a way that eliminates the use of corner columns, increasing occupant views.

Stine also dismisses concerns that A913 might require special treatment. “This is not a special thing that you need to go, 'Oh, well, I'm using high-strength steel, which means my specs are going to change; I need to treat it differently; I need to think about ordering it differently.' None of that applies, and there are no surface condition constraints that mean you have to do a special coating on your steel; there's nothing like that. You treat it like traditional steel.”

Weldability: Perspectives From Engineers, Manufacturers, And Welders

The benefits of A913 recurrently stressed by the manufacturers include ease of welding, a substantial advantage given the costs associated with onsite welding. Professionals with experience in this component of construction emphasize that it is a nuanced subject requiring awareness of contexts: weldability is a matter of degree, and experiences with different steels and settings can differ. While regarding A913 as highly weldable, some welders note that in practice the process calls for special attention and high expertise.

Under most circumstances, preheating has been a best practice in welding, releasing hydrogen from the metal and reducing the risk of hydrogen-induced cracking, either in the heat-affected zone or the weld metal. Grades 50 and 65 of A913 are weldable without preheating at temperatures above 32°F, provided filler metals with low hydrogen (H8 or lower, producing a weld with a maximum of 8 ml of hydrogen per 100 g of weld deposit) are used, according to American Welding Society specifications (AWS D1.1). Selected A913 grades are in Category D for minimum preheat and interpass temperatures, where 32°F is the minimum preheat temperature (AWS D1.1/D1.1M-2020, Table 5.8 and Clause 6.8.4), allowing for conditions of restraint, hydrogen level, and other factors. Commentators caution that blanket statements of exemption from preheating are not accurate.

“Any time your base metal that you're welding against is too thick, in general, you have to preheat that steel,” says Stine. “It's dictated by the American Welding Society. So AWS has standards that say, for this grade of steel, for this thickness, you have to preheat the steel to make the weld. Well, the biggest benefit of A913 is, it doesn't matter how thick it is, you don't have to preheat any more, so imagine all the hours in a shop or hanging in the air that steel is being welded together; that may be a safety constraint, it may be hundreds or thousands of man-hours on a large job. You can remove that piece of the equation, bring more safety to your workers, and actually as soon as you butt those two materials together, you can weld them together.”

One manufacturer offers somewhat more qualified recommendations about A913's preheat requirements, referencing European Union welding codes:

“Provided that the general rules of welding and fabrication are respected (see EN1090-2, EN1011-2 or local codes), [the product’s] grades also offer good weldability for all manual and automatic processes. Due to their low carbon equivalent content, it is generally not necessary to preheat under the following conditions:

  • Heat input Q ranges 10-60 kJ/cm
  • Temperature of the product is > 5°C [41°F]
  • Electrodes with low carbon equivalent and low hydrogen content, typically with a diffusible hydrogen content ≤ H10 … are used. The welding of a Jumbo beam of 140 mm flange thickness… was welded without preheating with a filler metal with low hydrogen content ≤ H5.

A discussion of A913's suitability for sites with seismic risks includes an overview of weldability evaluations with jumbo shapes (W14x730) of grade 65 and concludes that the material's low carbon equivalent does allow welding without preheating, even with such large members (Axmann).

Exemption from preheating, if feasible, would be a rare and beneficial property of any steel. A913, says Finnigan, is “one of the only specifications that has been approved by AWS to be welded without preheat in certain applications, so that can be really beneficial in cold-weather applications, because as long as the material is above freezing, you don't have to hit really stringent preheat requirements with the material, which can be onerous for the fabrication procedures and the on-site installation procedures.” Preheating uses nontrivial amounts of energy and time; it also requires considerable expertise both to ensure safety and to control variables that can lead to cracking.

Image courtesy of Neoscape & BIG – Bjarke Ingels Group

Located at the intersection of the High Line and the newly developed Hudson Boulevard Park on Manhattan’s evolving West Side, BIG's Spiral intertwines a continuous green pathway with workspaces on every level.

WSP's Smilow notes that a lighter, stronger, more ductile material like A913 conveys multiple benefits. “Because you have better material than conventional grade 50, your welding improves dramatically, because you have so much less welding; that's a big plus. . . . If we're talking about welding plates together, it's more likely than not you're going to be dealing with full-penetration welds, so [with] even a small reduction, there's a big savings in the amount of welds, and then there's the mushroom effect: the time to do a weld, the potential for cracking of the steel. The more, the thicker, the greater the potential for cracking, because of residual stresses.”

Cracking is more common than one might recognize. “I'm always bumping into the cracking issue,” Smilow reports, recalling one large and ill-fated tourism-oriented project–repeatedly delayed and eventually canceled–where a large base with over 100 anchor bolts was shipped to the site, only for the engineering/construction team to discover cracks proliferating throughout the material. (The component was fixed, but this was one of numerous problems with the project, culminating with engineers from another major firm walking off the site.)

Questions of weldability inevitably raise questions of how much onsite welding is feasible or desirable at all, given its costs. “Field welding is a major issue when you're dealing with very, very thick, significantly sized welds,” Smilow continues, “because once you start some full-penetration welds, you're not permitted to stop in the middle, and you have welds here that have to be preheated, and the process could take literally days to complete. You can imagine how would that work in the field at a construction site, versus a plant where you can lay things down and set up and control what happens with three shifts of laborers. So bolting, by far, is the way they want to proceed, even if it means welding on, in the shop, significant extensions to flanges to enable larger amounts of bolts. And again, the A913s are of such high quality that they enable the welding to take place with less trouble. . . . You don't have to preheat as much as you would on the A992 grade 50 material, even though it's good practice to preheat appropriately, but it all depends on the thickness of the material.”

Jason Chadee, quality assurance manager of New York City Union Iron Workers Locals 40 and 361 Joint Apprentice Committee, has extensive experience on major local projects including the Brooklyn-Queens Expressway, the Verrazano-Narrows Bridge, World Trade Center Tower 1, and Hudson Yards; he also consults with prominent steel erectors and trains fellow ironworkers in welding A913 and other types of structural steel. He finds that A913 has impressive toughness and ductility, speculating that it may also offer greater resistance to hydrogen cracking. With high-strength steels in general, however, “the mechanical properties are improved with the toughness and ductility of it, but as far as weldability, it's a little more difficult to weld.”

Appropriate filler metals with grade 65 must be nickel-based, Chadee says; “the technique is a little more difficult to weld, the travel speed is slower . . . you have to have a lower heat input, also.” If amperage and voltage are too high, “the nickel balloons out faster than the iron, and that affects the chemical composition of the steel,” he has observed. “Because of all these things, it's a slower process, so if you're welding a vertical weld, it's going to be about 40 percent slower than a flat, because of your heat input and travel speed.” He suggests that on projects with extensive high-strength steel, slip-critical bolted connections can replace welded connections.

Decisions between bolted and welded connections depend on variables of cost (welding expertise entails higher labor costs than bolt-tightening), time, and strength considerations (welds, provided cracking does not occur, are stronger). “There are a couple of ways that the good weldability of material like A913 can factor into an equation,” says Finnigan. “There are some conditions where it might just be unavoidable to have site welding: for example, in high seismic areas, or with some elements that are part of the redundancy system or the robustness system of the building, it might be necessary for column splices, for example, to be CJP [complete joint penetration] or even PJP [partial joint penetration] welded. . . . In those conditions, bolted connections might be an alternative, but sometimes you just get to a point where using bolts to achieve the full connectivity of the elements between one another can become such a cumbersome detail that CJP welding is completely necessary. And so in a condition like that, where you're going to be doing it on-site, having a material that's much easier to weld can be super-beneficial for the installation and for the project overall.”

David Tarabji, structural engineer with Magnusson Klemencic Associates (MKA) in Chicago, recommends that fabricators and onsite welders should develop specific familiarity with the highest-grade steels. With grade 80, “fabricators have to get their internal processes up to being able to have all their welding procedures for 80 ksi, and I've heard anecdotally that it wears out the equipment a little bit more; you go through more saw blades and more drill bits. It's a harder material, and so I think as it gets more exposure in the marketplace, more fabricators, as they get accustomed to it, they'll start wanting to see it more, but it has to flood the marketplace first.” On his own firm's major project using grade 80 (see Case Studies below), the fabricator was up to speed on the material and there were no problems, “but it's something that probably not every fabricator is prepared to take on just yet.”

Anthony N. Gopaul, a professional welder with NYC Constructors and a member of Locals 40 and 361, concurs with Chadee's caveats about speed with A913. “The weldability is good,” he observes, but with the high nickel content in the electrodes he and his colleagues use, the process is slower and should not be attempted by inexperienced personnel. “If you really don't know what you're doing, it will crack, because the nickel makes it a little bit more brittle. . . . In terms of A913 steel, it is very difficult to weld in a vertical position. But any welder that really is very experienced with this, they've got no problems.” He dissents, however, from the claim that preheating is unnecessary. “Maybe the manufacturer claims that, but when we're welding, we always preheat. . . . I don't know where that came from, that you do not need preheat. If you don't need preheat, then it's got to be probably very thin steel. But if you're talking about heavy material, no, you need the heat.” Noting differences between official documentation and experience in the field, he says, “It all comes down to what's going to work, what won't work, and what's needed for it to work.”


[ Page 2 of 4 ]  previous page Page 1 Page 2 Page 3 Page 4 next page
Originally published in Architectural Record
Originally published in February 2022