FABRICATING RECOMMENDATIONS IN THE WELDING OF TWO-SIDED GROOVE WELDS

Here are some real-world considerations and practices to improve quality and productivity during the welding of two-sided butt joints that are frequently used in pressure vessels and wind tower construction.

As mentioned in last month’s column (“Choosing A Butt Weld Joint Preparation”, Welding Tips, May 2011), fabricators have traditionally tended to prefer the use of two-sided butt joints to achieve complete penetration in groove welds when the plates to be welded are greater than one inch in thickness. There are a few distinct advantages to two sided groove welds that makes them a very attractive joint design for use in industries such as pressure vessel fabrication and wind tower construction.

The first advantage to two-sided butt joints over their single-sided complements is that joint economy improves above as the plate thickness increases due to the joint geometry. This improvement in joint economy by moving from a single-V weld to a double-V weld is displayed graphically in Figure 1. In addition to the joint geometry creating a 50 percent reduction, an additional component of the superior joint economy of two-sided groove welds is attributed to the smaller opening of the joint at the surface of the plate. Because this distance is smaller, there is less weld “reinforcement” in two-sided joints compared to their single-sided counterparts. This concept of reinforcement is displayed in Figure 2.

The second advantage is that the residual stresses can be forced to be more balanced in a two-sided groove weld, provided that the appropriate pass sequence is utilized. By proper pass sequence, I am referring to an alternating pass deposition progression, where a weld layer is deposited on one side followed by a weld layer on the second side of the joint.

However, two-sided groove welds are not a panacea. It was also mentioned last month that two-sided complete joint penetration groove welds require that two root passes be deposited. The inherent fear regarding root passes is that incomplete penetration and trapped slag can easily occur and these conditions may not be detected until after the joint has been fully welded. Without a doubt, the fabricator will want to minimize the risk that this will occur in order to prevent a costly weld repair.

To pre-empt the possibility of these weld defects as a potential problem, most fabricators preferentially choose to perform the air-carbon arc gouging process on the un-welded second side of the joint. By “gouging to sound weld metal” on the back-side of the root pass, chances for a clean X-ray are virtually assured. A schematic of a back-gouged two sided butt joint is shown in Figure 3. However, back-gouging is quite often unnecessary, as other less-costly fabrication techniques can be employed that will ensure complete penetration through the root without weld defects.

Consider an example where a pressure vessel is being fabricated using 1.5 in thick sections. These segments are being butted together and welded in a double-V joint configuration. Both plates have a 3/16 in root face (land) in the middle of the through-thickness, a symmetrical 30 deg bevel on the first side and a 30 deg bevel on the second side (in other words, a 60 deg included angle for both Vs). This 3/16 in root face is actually quite common, as this typically ranges from 3/16 in to ½ in.

The next step would naturally be to place a “quick” root pass on the first side. This root pass can be deposited with any of a variety of welding processes. For example, either the gas-shielded flux core arc welding (FCAW-G), gas metal arc welding (GMAW) or submerged arc welding (SAW) processes can all be used to deposit this pass.  When the fit-up is not perfect in some places – say greater than 0.100 in – it is required that the welder be wary that the weld could burn through and leave a hole in the root of the joint. That is why, in these situations, it is very common for the FCAW-G or GMAW processes to be used for the quick root pass, because these two processes operate at lower output power and can be more forgiving when large gaps are encountered.

But if the fabricator’s manufacturing process provides a tight fit-up along the entire length of the joint then the SAW process, accompanied by its generally higher output amperage and voltage, can be employed. With the joint configuration dimensions listed above, a tried-and-true root pass procedure that has been developed using the single-arc AC Square Wave SAW process can be used. The wire diameter that is (or should) most often be selected is 5/32 in diameter, as the top end of the output range of this diameter electrode aligns with the top amperage rating of 1000 amp welding power sources. A proven procedure in this situation is 625 amps, 28 volts, a contact-tip-to-work distance (CTWD) of 1.5 in, and a travel speed of 30 ipm.

Note that we have chosen to use AC Square Wave polarity over DC+ polarity for this pass since we will achieve a higher melt-off rate “per amp” of output, which will provide us with the same deposit layer thickness at a faster travel speed. This, in turn, will reduce the level of penetration on this root pass. This higher melt-off rate with AC polarity (compared to DC+ polarity) is a consequence of a higher proportion of the welding “heat” being transferred to the electrode in AC polarity due to the back-and-forth flow of electrons.

However, if there is more heat in the electrode with AC polarity, then there is less heat in the plate. This may necessitate the use of DC+ polarity (and hence a larger root face) for the first several passes to get better bead wetting on extremely thick plate (say 4 in or greater). After approximately ¼ in of the joint has been welded, the AC Square Wave polarity can be implemented to achieve higher melt-off rates.

The root pass on the first side should be followed immediately by a second or “hot” pass on the same side. In staying with the same train of logic – that is, the maximization of productivity – we will continue our use of the SAW process. A good starting procedure with AC Square Wave polarity and a 5/32 in diameter electrode would be 750 amps, 32 volts, 1.25 in CTWD and a 20 ipm travel speed.

At this point, you may wonder why we would want to continue welding on the same side of the joint. After all, it was stated above that we would want to minimize the build-up of residual stresses by alternating the deposition of weld layers between sides. This would be true for all of the weld layers – except for this one. What we are trying to accomplish here is a combined thickness of the root and hot pass that will provide enough depth to prevent burn-through for the first pass deposited on the SECOND side.

This leads us to the next step in the process of constructing of our pressure vessel, which may be new to some fabricators. Our stated desire is to avoid the common practice of “back-gouging” the second side before depositing weld metal. As mentioned earlier, this quality measure is performed due to fears of weld defects in the root. However, with the SAW process, a properly designed weld joint and second-side, root-pass welding procedure will allow the arc to “punch” into the weld bead(s) that have deposited on the first side. Note that no additional preparations are required to be performed on the second side before this pass is to be deposited. Rather, all that is needed is simply experience and belief in the axiom that amperage is proportional to the depth of penetration.

Exactly what amperage is necessary to achieve this complete joint penetration? This, of course, depends upon the thickness of the land and the bevel angle. But successful welding of the second root pass on our two-sided joint example has been conducted with the following procedures: AC Square Wave polarity and a 5/32 in diameter electrode, 750 amps, 32 volts, 1.25 in CTWD and a 20 ipm travel speed. A similar procedure for the remaining passes can be used, resulting in a weld joint profile similar to that shown in the macro photograph in Figure 4.

It is noteworthy that this welding pass procedure is identical to what was used for the hot pass on the opposite side of the plate. While we do want the root pass welds to “meet in the middle”, the weld is not actually any “stronger” with 115 percent or greater penetration. There is a common misconception that the deeper the penetration of the weld bead, the better. For this reason welders often say “let’s crank up the amperage”. But the only logical rationale for greater than 115 percent penetration would be insurance against the possibility of the wire “tracking away” from the center of the joint. So if we could ensure that the weld beads meet in the middle, then all that is really required is 105 percent penetration. Therefore, in order to accommodate either wire or joint wander we can increase the width of the weld bead by increasing the output voltage. This will have the same effect of insuring against joint tracking issues.

Not coincidentally, there are actually two distinct benefits to having just over 100 percent penetration versus having 115 percent penetration. The first is that width-to-depth cracking issues are mitigated, because the weld bead will be wider due to a lower output amperage and higher output voltage. This relatively wider bead creates a weld microstructure in which the “grains” will be less likely to be aligned perpendicular to the centerline of the weld. This reduces the chances of centerline cracking in the weld. Furthermore, when the microstructure aligns itself in this fashion, it can have the effect of improving weld metal toughness.

An additional, incremental improvement to weld metal toughness can be attained due to the fact that when there is 105 percent penetration, there is more re-heated weld metal present in a Charpy V-notch sample taken from the center of the weld joint. With 115 percent penetration, the majority of the Charpy V-notch sample is as-cast weld metal, which tends to have lower impact toughness.

The above discussion is just a small sample of some real-world considerations and practices that can be implemented to improve quality and productivity during the welding of two-sided butt joints. This type of joint design is frequently used in applications such as the fabrication of pressure vessels and wind tower construction.