Weld Distortion: The Basics

It’s a basic fact of metal fabricator’s lives that all welds shrink as the molten metal cools. This weld shrinkage introduces stresses into the fabrication and results in local distortions that affect the final shape of the part.

Welds shrink in all directions. Usually, the effect is greater in one direction than another. Since the parent metal away from the weld joint does not shrink, the uneven stress distribution causes the joined parts to distort.

This weld induced distortion is not easy to predict. This is especially so in more complex structures where multiple parts are welded together, with each weld adding its own distortion to the whole.

Some simple cases illustrate the basic distortions.

If two flat plates are welded side by side, the finished assembly will assume a shallow vee cross section and it will be slightly curled over its length. The vee is sharper if all the welding is all from one side.

If those two flat plates are butted together and welded to make an “L” or “T” right angle cross section, the angle between the legs will typically be less than 90° on the welded sides because the weld cross section shrinks.

If the flat plates are formed into cylinders and welded, the distortions are different. The longitudinal welds will flatten the curve of the formed shape at the weld, producing a reduction in the local radius at the seam. Typically, the cylinders stay straight because the formed cylinders are relatively stiff along their length. If two cylinders are then butted together and welded, the circumferential weld will shrink inward, producing a locally smaller diameter at the weld.

The degree of each of these distortions is influenced by the configuration of the weld joint, by the size of the weld along with the relative stiffness of the joined sections and even by the heat input of the chosen weld process.

It is possible to predict the weld shrinkage and the resulting distortions for simple cases like this, and there are more sophisticated methods to predict distortions in more complex cases. However, it often comes down to experience to anticipate the degree of shrinkage, where distortions will occur and to what extent the distortions will affect the final part.

Once the potential shrinkages and distortions are identified, designers, manufacturing engineers and welders can mitigate the effects to produce a finished part that meets all the customer’s expectations.

Planning for and Mitigating Weld Shrinkage and Distortion

Mitigating the effects of weld shrinkage and distortion is a function of good design, sound planning and careful welding practices.

Designers can start the mitigation process by minimizing the size and number of welds and by selective placement of welds. Placing welds at thinner sections, for example, can minimize distortion. Specifying double sided welds where practical gives welders an opportunity to balance out the distortions as they proceed. In some cases, designers can specify the degree of welding required for strength without specifying a particular weld configuration. A typical callout that allows this freedom is “Complete Joint Penetration” (CJP). This allows the manufacturer to use their expertise to select a weld joint design that will minimize weld distortions. Another way is to allow fabricators to reduce the number of welds by combining and machining several parts from a single piece of metal.

Manufacturers have several techniques available to them to control, anticipate and minimize the effects of weld distortion.

Manufacturing engineers can specify the sequence of assembly welds to take advantage of the accumulation of stiffness as parts are added to the assembly. If the customer can allow some leeway in the weld configuration, manufacturing engineers can minimize distortions by selecting weld designs that produce a more balanced joint with less distortion. Manufacturing engineers may also be able to reduce the number of weld joints and orient them in a way that reduces the accumulation of distortion. An important consideration for manufacturing engineers is to leave extra material on weld joint interfaces between subassemblies so surfaces can be trued up, including an allowance for shrinkage, in a planned machining step. Similarly, extra material should be allowed on critical surfaces for machining to print after all the fabrication is complete.

Welders can reduce weld distortions by keeping weld deposit sizes within the limits of the appropriate welding standards and by sequencing their weld deposits minimize heat input and to balance out the distortions as they occur. Bracing is another technique that is used to good effect, especially if the assembly and welding can be sequenced to be self-bracing.

In a production environment, where multiples of a welded structure are made, fabricators can often afford to make practice pieces to iron out the fabrication process to minimize difficulties at the production level. Specialty fabricators don’t often have this luxury, so experience becomes an important factor in anticipating the degree that welds will shrink and the direction and extent of the resulting distortion. Using the techniques and practices described above, the effects of weld shrinkage and distortions can be minimized and accounted for in subsequent processing.

Ultimately, distortion in a weldment comes down to the accumulation of weld shrinkages. Planners, therefore, have to examine each weldment individually and decide how it is going to be built and welders need to follow best practices to minimize the distortions caused by each and every weld.

The complexity of the welded fabrication, the number of welds, their placement and geometry, and weld the weld processes and practices all have an impact on distortion levels. It is critical to work with manufacturers who understand weld distortion and have the experience and skills to predict it, plan for it and mitigate it.