Never Underestimate the Importance of Weld Bead Shape

Many weld failures occur due to lack of fusion when proper attention is not given to this characteristic. Here is a remedy that may help fabricators avoid such problems.

When welding multiple-pass, complete-joint-penetration (CJP) butt joints, weld bead shape is of the utmost importance. As such, numerous “lack of fusion” weld failures have occurred when proper attention is not given to this characteristic. And because this is such a common problem, this month is an opportunity to describe one of my most recent experiences where lack of fusion was a primary concern. I’ll also recommend a remedy that may help fabricators avoid such problems.

Not long ago, Frank Dragolich Jr. and I were tasked with developing gas metal arc welding (GMAW) procedures for long vertical sections to be used in the construction of a containment vessel. The CJP joint design used in this application (shown in Figure 1) consisted of a single-V butt joint with a backing bar. The included angle was 20 deg with a root opening of ¼ in. The plate thickness was ¼ in and had a feather-edge (i.e., no land). The welding process to be used was Pulsed-GMAW, and the welding position was 3G vertical-up. And oh, by the way, we had to make sure that our welding heat input stayed below 35 kilojoules per inch to keep distortion to a minimum.

Because the joint was more than 60 ft long, there was really no other alternative than to weld the joint with some type of automated welding package, as this would be the case at the customer’s jobsite. But for the purposes of procedure development for this project, we used a test plate that was only a few ft long. As one of our colleagues once said, “If you can weld an inch, you can weld a mile.”

Readers of this column who are well versed in the vertical-up GMAW process may be cringing after reading “¼ in root opening”, as this narrow groove presents a potential problem in achieving root pass penetration through the joint (lack of penetration is not to be confused with lack of fusion). This root opening of ¼ in is what I like to call a “tweener” (or in-between). I’ll explain the difficulties associated with this ¼ in root opening shortly.

But first, one might ask what advantages could be realized by simply changing the gap dimension? Well, if the gap was narrower – say 1/8 in – we might be able to weld straight down the middle of the joint using a stringer pass with minimal fear of lack of penetration. If, on the other hand, the gap was wider, we could use two stringer passes, with each pass focused on the point of contact of the feather edge and the backing bar.

Narrowing the gap, however, was not without latent problems. Under normal conditions where the weld joint is only a few feet long or the joint is tacked in place securely, narrowing the gap could work. But because in this situation the vertical weld joint was more than 60 ft long, and one side of the joint was barely tacked into place at all, the weldment was going to undergo its “fair share” of distortion.

Although the customer’s main reason for limiting the heat input was to prevent the plate from taking on a “banana” shape, distortion also would force a 1/8 in gap between the plates to close during welding to a distance that for all intents and purposes is equal to zero. Once this happens, complete penetration through the joint will become almost impossible with the low heat input constraint mentioned above.

Now let’s think about the opposite scenario. Widening the root opening to greater than ¼ in would have been ideal for two reasons. First, opening up access to the joint with a larger gap (or perhaps an included angle greater than 20 deg) allows the welder to focus the arc at the intersection of the backing bar and the feather edge of the sidewall. By “focusing the arc,” I mean that he or she can angle the torch and bisect the 110 deg angle (shown in Figure 2).

A second benefit to opening up the gap wider than ¼ in was that it would give us the ability to use two stringer passes, as opposed to one pass down the middle of the joint. In so doing, we would rely less on the papilla (or penetration profile) of the single stringer pass to sweep all the way to the side of the joint and catch the feather edge.

Nevertheless, the customer balked at the idea of widening the joint, as this would essentially double the number of passes required to fill the joint. And their concern was quite valid when factoring in the not-so-minor detail that the sections of steel to be welded were several yards in length. A 100 percent increase in arc time and the resulting loss in productivity were simply out of the question.