Product and process review leads to key welding quality recommendations
When a Yorkshire engineering company contemplated sub-contracting welding fabrication work to a distant European supplier it wanted absolute assurance that the finished job would be up to scratch. It needed independent advice from a UK expert.
The job carried the column and arm of an aerial work platform and was unquestionably deemed safety critical. The client was taking no chances. It called in TWI to inspect the prototype and comment upon the quality of the welding.
The work started out as a Product and Process Review under TWI's Joining Forces Yorkshire and Humber programme. But as an analysis of the engineering plot thickened over the next couple of months the scope was broadened to a Feasibility Study and a Near Market R&D project.
TWI's specialist welding engineers decided to divide the work into two broad sections; first a visual inspection, followed by assessment of the quality control and quality assurance aspects of the job.
It was this first part of the work that revealed some very questionable welding.
For instance most of the vertical external fillet welds, such as those on the pedestal corner joints, and the outrigger to the oil tanks, were deemed to be of an acceptable profile.
Other welds however were very poor. For example, the internal fillets attaching the slewing ring plate to the pedestal had been ground, presumably in an attempt to improve the contour.
The drawings specified 6mm fillet welds but the inspection revealed leg lengths of 15 to 16mm.
Numerous similar examples exposed that this quality was thematic of the fabrication's general weld quality.
A number of issues arose from the inspection, perhaps the most significant being that the drawings gave insufficient or ambiguous detail.
One by one TWI identified these deficiencies and advised both that they be addressed urgently, and that welding requirements should be specified using BS EN 22553 and BS EN 25817. The latter goes into some depth on surface preparation, cleanliness, consumable control, joint design and crack avoidance.
But the secondary outcome of the inspection led TWI to recommend that the welding staff should be trained to a level of competence in line with BS EN 287 Pt 1 or BS EN 4872 Part 1.
It also suggested that the welds be categorised with respect to criticality and that a proportion of them should be inspected using magnetic particle inspection.
The conclusion emanating from this consultation was that; when welding work is sub contracted a clear statement of the quality requirements should be specified by the client.
Clearly there was more to the problem than appeared on the surface and which would be revealed by a visual inspection. So the job was moved on to take up the status of a feasibility study, also under the Joining Forces Yorkshire and Humber banner.
This involved macro examination of eight weld samples taken from the prototype. They were sectioned by TWI and polished for inspection.
The butt welds were deemed acceptable but the fillet welds all failed having exhibited extensive lack of sidewall fusion defects and porosity. And in one case a weld had been made in a component by inserting a round bar to fill a gap in the mis-matched components with the exposed weld ground flush to disguise the fact.
By this stage TWI recommended that the fabricator be asked to supply approved welding procedure and welder approval certificates, and also ensure that only approved and competent welders be used.
Stage three took the investigation to what Joining Forces Yorkshire and Humber call a Near Market R & D stage. It demanded a full metallurgical report.
This detailed appraisal of the structure confirmed the initial findings, reported defect sizes and types and sentenced the structure against a relevant inspection standard. The main conclusion was that there were such serious flaws that the integrity of the whole structure was in doubt and that to accept such flaws could result in premature failure of the component.
In its final recommendations TWI's specialist welding engineer returned to the fundamental issue of welder skill. It strongly advised that welders involved in these fabrications receive re-training before being qualified to the requirements of either BS EN 287 Part 1or BS 4872 Part 1.
And once qualified they should be required to ensure that the welders maintain the required quality by producing satisfactory magnetic particle inspected welds and also regular welder re-testing at say, six monthly intervals.
For information about TWI's capabilities please contact us.