January 15, 2010
Hydrogen cracking can send a project off schedule in a hurry. Here are ways to prevent it.
There it is. It could be plain as day. It could be exposed via a die penetrant test, an ultrasonic test, or perhaps a magnetic particle inspection. It could be anywhere, but when found, it sticks out like a rude, uninvited guest—only very late.
Call it hydrogen cracking. Call it delayed cracking or underbead cracking. Call it a cold crack. It doesn't matter. The assembly isn't passing muster, and rework is in order. Costs mount. Deadlines are missed. Nobody's happy.
Quality can't be inspected into a part, of course, so preventing such cracks should be part of the basic modus operandi of any welding operation.
To form, hydrogen cracks require three basic things: a susceptible microstructure (typically a hardness exceeding 35 Rockwell C [HRC]), hydrogen, and stress. They also often occur when the welded structure is close to ambient temperatures.
Unique among weld defects, hydrogen cracks may take days to form, sometimes longer. Although they can occur anywhere in the weldment, they often start in the coarse-grained region of the heat-affected zone (HAZ), adjacent to the fusion line.
"This region is reheated to a high temperature, above the transformation temperature, and then experiences a relatively fast cooling rate," explained Ravi Menon, vice president of technology at Stoody Co., a division of Thermadyne in Bowling Green, Ky. He added that steels with high carbon content or hardenability are particularly susceptible, because their HAZs may exceed 300 to 350 Vickers hardness (HV) (approximately 35 HRC), at which point they become sensitive to hydrogen cracking. This is where measuring hardenability comes into play, typically evaluated through a carbon equivalent (CE) calculation, such as the one in the annex of the American Welding Society (AWS) D1.1 code, which takes into account the weighted effects of the different alloying elements in the steel.
Because the cracks may not transmit through to the surface, "the best way to look at it may be a magnetic particle inspection or ultrasonic inspection," said Menon. "The dye penetrant would just reveal surface cracks."
Hydrogen can come from the moisture in the electrodes, electrode coating constituents, and ambient humidity, along with moisture, oil, and grease in the joint area. As hydrogen diffuses into the weld, various metals react differently. In the 300 series (austenitic) stainless steels, hydrogen stays dissolved in the microstructure. Thinner mild steels usually don't have much of an issue with hydrogen, because hydrogen readily dissipates from the welds, and less stress builds up during the welding process. But in metals like 400 series (martensitic) stainless and high-tensile-strength steels, hydrogen accumulates to a certain threshold level, and after hours or even days, a crack may develop.
Hydrogen lingers. As Roger Bushey, product compliance manager at ESAB Welding & Cutting Products, Florence, S.C., explained, a welded subassembly may be allowed to shrink and shift to relieve stresses. But what if that subassembly is welded to a larger component during final assembly? More stresses may exist because of shrinkage forces, depending on the part design, and those stresses may contribute to hydrogen cracking in joints that have been welded days earlier.
The cracks usually develop in areas of high stress—in an overlap or undercut area on the edges of a fillet weld, for instance. This is where the weld cools first, and if cooling is too abrupt, stresses build and cracking occurs. Other stress-risers include incomplete fusion at the weld root, elongated slag inclusions, or a sharp change in weld contour because of poor joint fit-up.
As Menon explained, "In general, the more hardenable a steel, as measured by the carbon equivalent, and the greater the thickness, the lower the threshold level of hydrogen where cracks may be generated."
Typical mild steel welding usually doesn't produce hydrogen cracks unless the mild steel is extremely thick—not an uncommon situation. High-strength steels such as A514, HY80, HY100, and high-strength, low-alloy (HSLA) steels are another story. "Special precautions should be used when welding these materials," Bushey said.
When carbon contents are low but the strength is high, "cracking tends to be finer and harder to detect," Menon explained. "With high-carbon steel, you can have a HAZ with a hardness level well exceeding 35 HRC. When a weld inspector sees the cracking, it probably will be at or near the surface and may be visible to the naked eye. But newer steels have lower carbon content, and the strength is provided by the alloying elements in the steel, so the cracks tend to be finer and may require sophisticated detection, such as magnetic particle or ultrasonic inspection."
Menon added that for many steels, the optimal preheats, interpass temperatures, postweld heat treatments, and electrode handling procedures have been developed. "The problem happens when you're welding a steel of unknown composition, or when you weld with higher-strength steels; there is less experience in handling those steels."
Hydrogen cracking also occurs when welding certain dissimilar metals, even if one usually doesn't have issues with hydrogen. Say you're welding HSLA steel to 300 series stainless. The hydrogen dissolved in the stainless steel during the welding operation could contribute to the diffusible hydrogen level found in the interface of the high-strength steel.
The heat input of a welding process can be used to control the cooling rate in the HAZ. Higher heat inputs, as well as a higher preheat temperature, result in slower cooling rates that allow hydrogen to diffuse out of the weldment. As Menon explained, "Good fabricators will go with the more conservative approach, with higher preheat temperatures and longer postweld heat-treat times, and they will ensure they know the hydrogen levels on their electrodes. Obviously, this has to be balanced against the economics of the welding operation." He added, though, that "the costs resulting from unexpected hydrogen cracking will far exceed any costs associated with incremental preheat or postheat."
Purchasing low-hydrogen electrodes may not prevent hydrogen cracking on its own, but it is a necessary part of the equation. These electrodes use designations such as H4, H8, or H16 at the end of their AWS classifications; the numbers indicate the maximum amount of milliliters of diffusible hydrogen the electrode will evolve per 100 grams of weld deposit. AWS A5.5 covers classifications for low-alloy shielded metal arc welding (SMAW) electrodes, while A5.29 covers low-alloy, flux-cored wires.
In low-hydrogen SMAW electrodes, absorbed moisture usually can be baked out at temperatures between 400 and 600 degrees F, though electrode manufacturers give more specific ranges for their products. Unused electrodes removed from hermetically sealed packaging should be stored in holding ovens at temperatures of 250 degrees F or above to prevent moisture pickup.
But in conventional electrodes there is chemically combined moisture (hydrogen), borne in the consumable's makeup, that can build up in the weld metal and the HAZ during welding. "Low-hydrogen electrodes are designed to limit this problem by using the proper materials to [limit] the contribution of diffusible hydrogen found in a weld deposit," Bushey explained.
Arc-stabilizing chemicals in the electrode, as the name implies, stabilize the arc as the molten electrode droplets transfer to the weld joint, making the droplets smaller and creating an even metal transfer. Various compounds of chemical elements are effective arc stabilizers, but they tend to pick up moisture. Some electrodes, such as the E6010 pipe electrode, have moisture (hydrogen)-containing coatings that help them attain good penetration characteristics. Without moisture, these cellulosic electrodes wouldn't weld properly—which is why they aren't used when hydrogen is a concern.
Low-hydrogen electrodes are made differently. A low-hydrogen, flux-cored electrode's interior may consist of potassium and/or sodium compounds agglomerated together with materials like silicon dioxide (sand) and titanium dioxide. The agglomerate is baked at high temperature to drive off the absorbed and chemically combined moisture. What's left is a specialized potassium/sodium-compound arc stabilizer in the form of a "frit," or ashlike material, which helps reduce diffusible hydrogen content in the weld deposits.
Many situations call for certification of diffusible hydrogen levels in selected batches of electrodes. The test weld is made by running a continuous bead on three separate coupons placed next to each other. After the run-on and run-off tabs (the start and end of the weld) are removed, "the remaining coupon is placed in a specialized container to capture the hydrogen that escapes from the weld deposit," Bushey said. "This level then is reported as a certain number of milliliters per 100 grams of weld deposit."
Detailed procedures for this are spelled out in AWS A4.3.
Low-hydrogen electrodes don't prevent hydrogen cracking alone, sources said, and they may not be the right choice for every application. For one thing, low-hydrogen electrodes don't totally eliminate the chance of absorbed moisture making its way into the wire or electrode coatings, particularly in humid workplaces, so good electrode handling practices are essential. Such practices are written into the welding codes for a reason.
Paul Cameron, a certified welding inspector at PWC Inspection Service, Rochester, Minn., added that inspectors can take note of such electrode handling practices. "A broad storage program is required for low-hydrogen electrodes," he said. "Does the shop or construction site have the tools needed for this? [As a welding inspector,] I would certainly note this in a report."
For instance, if not stored in a heated holding oven, low-hydrogen electrodes could be susceptible to moisture pickup. A 25-pound spool of low-hydrogen, flux-cored wire, used over many shifts, won't have its specified low-hydrogen content for very long if left on a humid shop floor. In these cases, Menon said, wire can be supplied on wire spools (instead of plastic), which can be placed in a holding oven at the end of a shift to minimize the risk of moisture pickup.
Flux-cored wire, simply because of the drawing process involved in making it, can cause issues with hydrogen levels in a weld deposit. The drawing lubricants typically are comprised of animal fats, and if any remain on the wire, they could contribute to the diffusible hydrogen level in the weld deposit. So historically, hydrogen-sensitive work was either left to solid-wire welding (tough to weld out-of-position) or SMAW.
"The control of these drawing lubricants is extremely important in governing the amount of hydrogen in the weld," Menon said. Also, flux-cored filler typically is manufactured with a seam, which can let in moisture if left exposed to the elements, although flux-cored wires with welded seams (so-called "seamless" flux-cored wires) are now available for extremely crack-sensitive applications.
Still, low-hydrogen, flux-cored electrodes have matured enough to take on high-strength steel work previously welded with solid wire. The flux-cored process (FCAW) offers higher deposition rates when compared to SMAW or solid wire because of the current density used, and FCAW can be run out-of-position without needing to use a pulse or other specialized equipment. Today you can be reasonably confident these electrodes won't spur hydrogen issues. And according to sources, because of advancing electrode production techniques, that confidence is likely to grow in the future.
The FABRICATOR® is North America's leading magazine for the metal forming and fabricating industry. The magazine delivers the news, technical articles, and case histories that enable fabricators to do their jobs more efficiently. The FABRICATOR has served the industry since 1971. Print subscriptions are free to qualified persons in North America involved in metal forming and fabricating.