Welding Inspection



In a sense, everyone connected with the job, as well as the formal inspector, participates in visual inspection. A conscientious worker does not knowingly pass on work in which he recognizes defects of his making. Nevertheless, it is usually desirable that someone be assigned responsibility for quality checking each operation. The tools for visual inspection are simple-a pocket rule, a weld-size gage, a magnifying glass, and sometimes a straight edge and square for determining straightness, alignment and perpendicularity.
Visual inspection should begin before the first arc is struck. The materials should be examined to see if they meet specifications for quality, type, size, cleanliness and freedom from defects. Foreign matter-grease, paint, oil, oxide film, heavy scale-that could be detrimental to the weld should be removed. The pieces to be joined should be checked for straightness, flatness and dimensions. Warped, bent, improperly cut, or damaged pieces should be ordered for repair or rejected. Alignment and fit up of parts and the fixturing should be scrutinized. Joint preparation should be checked. Often, little more than a passing glance is required in this preliminary inspection, but, despite, its almost casual nature, such inspection can be a significant factor in weld quality.
Inspection prior to welding also includes verification that the correct process and procedures are to be employed-that the electrode type and size and the equipment settings for voltage and amperage are as specified-and that provisions are made for the required preheat or postheat.
Assuming the preliminary requirements are in good order, the most productive inspection will take place while the weldments is being fabricated. Examination of a weld bead and the end crater may reveal quality deficiencies such as crack, inadequate penetration, and gas and slag inclusions to a competent inspector.
On simple welds, inspection of a specimen at the beginning of the operation and periodically as the work progresses may be adequate. When more than one layer of filler meal is deposited, however, it may be desirable to inspect each layer before a subsequent layer is placed.
The root pass in multipass weld is the most critical one from the standpoint of weld soundness. It is especially susceptible to cracking and because it tends to solidify quickly, is prone to trap gas and slag. Subsequent passes are subject to a variety of weld defect-creating conditions that result from the shape of the weld bead or change in the configuration of the joint. These can be visually detected by the welder and repair cost minimized if the problem is corrected before welding progresses.
A workmanship standard, constructed for the specific purpose, can be helpful both to the welder and the inspector in visually appraising the production weld during the stages of its formation.
Visual inspection at an early stage of the fabrication will also detect underwelding and overwelding. Underwelding is in violation of specifications and cannot be tolerated. Overwelding should be of as much concerns to the purchaser’s inspector as to those members of the shop responsible for monitoring costs, since it is a major cause of distortion. Usually the designer has specified a weld size approaching the limit possible in good practice. The welder-perhaps wanting to make certain that the joint is strong enough, or having been criticized for making undersize welds-takes it upon himself to add 1/16 in. to a ¼-in. fillet. Since the weld metal deposited increases as the square of the size, the 1/16-in. increase in leg size increases the amount of weld metal deposited 56%, and has thesame effect on shrinkage stress and cost.


Visual inspection after the weldment has been completed is also useful in evaluating quality, even if ultrasonic, radiographic, or other methods are to be employed. Here, as with visual inspection as welding progresses, surface flaws such as cracks, porosity, and unfilled craters can be detected, and may be of such consequence that repairs are required or the work is rejected without use of subsequent inspection procedures. Therefore is no point in submitting and obviously bad weld to sophisticated inspection methods.
Dimensional variations from tolerances, warpage, and faults in appearance are detected visually at this stage. The extent and continuity of the weld, its size, and the length of segments in intermittent weld can be readily measured or noted.
Welds must be cleaned of slag to make inspection for surface flaws possible. A glass with a magnification of up to 10 diameters is helpful in detecting fine cracks and other defects. Shotblasting should not be used in preparing the weld for examination, since the peening action may seal fine cracks and make them invisible.
The objective of visual inspection at this stage is not only to seek defects not permissible under the quality standard, but also to give clues to what may be amiss in the entire fabrication process. If the inspector has a sound knowledge of the welding, he can read much from what he sees. Thus, the presence of excessive porosity and slag inclusions may be a tip-off to the fact that current is not adequate, no matter what the dial readings may be. Subsequent tests will also give clues to faults in equipment or procedures, but the information acquired through visual examination permits corrections to be made before the results from complicated tests are available.
Only the surface defects in welds are visible to the eye, and the specifications or applicable codes may require that the internal portion of the weld and the adjacent metal zones also be examined. Additionally, the application may require assurance that the chemical composition of the weld metal has not change beyond specified limits. Thus, other inspection methods capable of gathering such information may be necessary.
These methods may be destructive or nondestructive. Destructive methods, obviously, cannot apply to production fabrication, other than for the testing to destruction of prototypes, “first unit,” or sparsely selected samples. In large weldments-building framing, for example-their use would be out of the question, but in an airframe component a periodic test to destruction may be regarded as essential to assurance of quality.
Nondestructive methods of testing welds include radiographic, ultrasonic, magnetic-particle and penetrant techniques. Proof testing is a mechanical method of determining whether the weld and other parts of the fabrication will withstand certain stresses encountered in service.
Chemical and metallographic methods of inspection may be completely nondestructive or destructive in a very minor way and to a reparable extent. If a sample for chemical or metallographic analysis is taken from a run-out portion of the weld bead, there is no damage; if a core drill is used to remove the sample from the weld proper, a hole results that must be repaired.
It should be noted that nondestructive proof of the existence of a flaw does not measure its influence on the serviceability of the product. Only destructive tests have the potential for giving such information. Presumably, the specification or code requiring the nondestructive test is based on a correlation between the flaw and some characteristic affecting service. If such correlations have not been established, the nondestructive test may well give nothing more than someone’s opinion as to what is good or bad.
The tendency in all matters pertaining to welded structures is to overdo factors affecting strength and safety, which means that flaws unlikely to influence serviceability may be the cause for rejection of the weld. In highly critical engineering structures, this overly cautious approach is justifiable, but in less critical applications some relaxation in the demands for flawless welds may be in order. Both laboratory destructive tests of segments of fabricated structures and tests to destruction of whole units where size and cost permit are useful in determining how much significance should be attributed to nondestructively revealed flaws. Also the judgments of skilled and experienced interpreters of nondestructive data may be invaluable when a correlation between the flaw and serviceability has not been established.
If, however, the code or specification demands that a weld be rejected or repaired because of a discovered flaw, that code or specification must be followed.


No comments yet.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: