Quality Inspection and Fatigue Assessment of Welded Structures

Abstract: Recently, Volvo Construction Equipment AB has developed a weld class system forimperfections in welded joints, which contains demands for the toe radii, cold laps, undercutsetc. and where root defects are treated as requirements on the drawing. In this master thesis, thetoe radius has been studied more carefully along with the selection of reliable measurementsystems which are able to measure the toe radius along the weld. A computerized vision systemhas been evaluated by performing a measurement system analysis. FE-simulations anddestructive fatigue testing has also been carried out to determine which radial geometry beingcritical to the fatigue life.The results show that the currently used methods and gauges do not provide the requiredaccuracy when measuring the toe radius. The gauges are handled differently by differentoperators – even when using the vision system – which makes the methods subjective andtherefore unreliable. There are measuring systems that can gather surface data along the weldwith high accuracy, but there is no reliable method to assess the data. Therefore, the authors havedeveloped an algorithm – named STELIN – that assess the gathered surface data andautomatically identifies and calculates the toe radius and the toe angle along the weld. Using thatinformation an opportunity to improve the process control when welding is possible.The performed FE-calculations show that the surface roughness in the weld toe probably has aninfluence at the fatigue life of the joint. A more precise separate study should be made todetermine the impact of the surface roughness on the fatigue life. Those results should serve as abase when reviewing the theory used when predicting the fatigue life. Currently, stress averagingapproach is used in the notches of the root and the weld toe. In the future though, there might beanother stress condition to be taken into account, if the goal of reducing weight of the finishedproduct shall be achieved. Regarding measuring the surface roughness in the weld toe, theevaluated vision system has enough accuracy to deliver reliable data.More work remains with the STELIN-algorithm. The method used when assessing the calculatedtoe radii should be based on the conclusions from the performed FE-calculations. Integrating theSTELIN-algorithm in a fast feedback measurement system – for instance, on a laser – willprobably provide good opportunities for a better process control in order to achieve higherfatigue life of the welded joint.