TOWARDS A BETTER WELD

By L. R. Parkes (British Welding Research Association)

Reproduced from 'Engineering News' (No. 196 - 29th April 1965) with kind permission of the author end the publishers – The Certificated Engineer December 1965.

It is pointless today to produce a new structural alloy unless it is weldable. The almost casual adoption of welding in the industry is responsible for a large proportion of the problems faced by fabricators. 

These problems arise from the wrong choice being made of material or process, faulty operator techniques and fundamental faults in the design. 

If the wrong process is used possibly the worst result will be uneconomic production. But if the material selection or design is at fault then failure will almost inevitably result. 

Poor design can lead to a rejection, following failure, of material which would normally be satisfactory in the mistaken belief that it was that, rather than the design itself, which caused the failure. 

Again one must have regard for how the welded structure has come about. In the majority of traditional structures, welding has replaced riveting, bolting and casting. 

Heavy Sections 

These last three give rise in common to a high weight/ strength ratio variously accounted for by heavy sections, gussets, cover plates, stiffeners, webs, etc., incorporated into designs where the tendency has been to allow for an excessive safety margin. 

"Design for welding" has in large measure been based on the development of these designs. 

But it is not just a question of replacing one form of jointing by another. Good welding design call for considerable re-thinking and restraint, particularly concerning details. 

On the drawing board, a design correctly interpreted f r welding looks naked. Thus there is a temptation, which must be resisted, to adorn it with some of the details more appropriate to mechanical joints and castings. 

Worse than adding nothing to the strength of the resulting structure these embellishments can, under fatigue loading conditions, seriously impair strength. 

Essentially, whether a joint is riveted, bolted or welded, it is always the weakest part of a structure. But unlike riveting or bolting, welding does not rely solely on mechanical factors. 

Microstructural changes occur in the welded material due to thermal cycling and, if these are associated with faulty design, service performance will suffer. 

On the other hand, the monolithic nature of a welded structure, well designed, affords material economics, integrity and aesthetic quality that could never be obtained by other means. 

Where, then, does one go from here? Much depends on whether or not the final structure is to be subjected to fatigue loading. 

Static Loading 

For static loading conditions, for example, a complete penetration butt weld in mild steel can be regarded in design as being of the strength of the parent material, the cross-section for cross-section. 

Under fatigue loading conditions, however, the strength of a similar joint may be halved. Nor is the use of a stronger material alone the answer to this. High strength materials have been shown to afford little or no increase in fatigue life if simply substituted in an unchanged design here mild steel had previously failed. 

First and foremost it is necessary to appreciate that the only virtue of a fixed joint is that it holds two or more pieces of material together, allowing for changes in the shape or the creation of structures of a size that could not be otherwise achieved. 

Welded joints should therefore be reduced to the minimum consistent with structural requirements. AU joints are in themselves stress raisers, irrespective of whether or not they are free from defects, and it is thus clear that they should not be positioned at points of high stress particularly if fluctuating or impact loading is anticipated. 

Analysis of a structure from stress concentrations has been likened to imagining a liquid flowing through the structural members at high speed. Any point where eddies or turbulence would occur is a point of stress concentration. 

As in a liquid system, all sharp changes should be avoided, for example, abrupt changes in material thickness in one plane. 

Finally, joints should be as clean as possible so that overlapping welds are avoided, and should be so arranged as to give the operator maximum access for making the weld. 

Growing Practice 

It is a growing practice at present, particularly in machine frame and bedplate production, to use fabrication method rather than casting to take advantage of the higher stiffness/weight and stiffness/volume ratios of steel. 

It should be remembered, however, that casting enables certain things to be done that could not be bettered by total welded fabrication. 

A typical example would be the detail where several stiffened sections radiate from one point in, say, the base or the table. To attempt to effect this with welded sections would be to end up with a complicated array of joints, difficult to reach, full of overlapping welds and notched and at a point where loading would be high. 

Here is an excellent case for marrying together the advantages of casting and welding, by employing steel castings for the intersection and making up the stiffener assembly from the cut plate or bar stock welded to the casting clear of the area of directional change and thus of maximum stress. 

How Rife? 

How rife is fatigue failure? From time to time metal fatigue makes news and in the main, it is in connection with aircraft. 

One way or another aircraft failures are newsworthy and the most recent example was the grounding of the Valiant bomber force from this cause. 

Metal fatigue is, however, a universal problem costing the individual firm and the nation dearly in continuous repair work alone. The majority of fatigue failures are non-catastrophic, occurring as a crack in structures during live loading service. They are frequently accepted by the user as an inevitable consequence of wear and tear and a such are generally repaired during routine maintenance. 

As a result, the supplier may even be unaware that failures are taking place. Thus the user is involved in meeting a continuous repair bill and the maker is left unintentionally to incorporate the same failure propensity into other structures.

Many cases of fatigue failure have been seen in objects ranging from brackets or ceiling mounted fans to cranes and earthmoving equipment, which could have been avoided by simple modifications to design. 

It is a natural consequence of all failures, if self-developed remedies are unsuccessful, for manufacturers to seek advice and a solution to the problem. This is the wrong time, since the material will have been bought and processed, extensive tooling up and jigging may have been made, and some goodwill may already have been lost. 

Correct Moment 

The correct moment for seeking advice for those who have doubts or want to verify the correctness of a design is at the design stage. Machinery exists for this and happily, the number of fabric tors following this line is increasing. 

If this article has struck a depressing note it should be remembered that problems have come about because welding L basically too easy and too versatile and its industrial application has been rapid. 

Rather it is felt, the note should be one of cautious optimism caution because of the fundamental metallurgical and design influences that exist and optimism because welding, correctly applied, vastly extends the scope and competitiveness of the fabricator.