a. The tests described in this section have been developed to check the skill of the welding operator as well as the quality of the weld metal and the strength of the welded joint for each type of metal used in ordnance materiel.

b. Some of these tests, such as tensile and bending tests, are destructive, in that the test Specimens are loaded until they fail, so the desired information can be gained. Other testing methods, such as the X-ray and hydrostatic tests, are not destructive.


a. This test is used to determine the soundness of a weld. The acid attacks or reacts with the edges of cracks in the base or weld metal and discloses weld defects, if present. It also accentuates the boundary between the base and weld metal and, in this manner, shows the size of the weld which may otherwise be indistinct. This test is usually performed on a cross section of the joint.

b. Solutions of hydrochloric acid, nitric acid, ammonium per sulfate, or iodine and potassium iodide are commonly used for etching carbon and low alloy steels.


The quality of the weld metal at the face and root of the welded joint, as well as the degree of penetration and fusion to the base metal, are determined by means of guided bend tests. These tests are made in a jig (fig. 13-1). These test specimens are machined from welded plates, the thickness of which must be within the capacity of the bending jig. The test specimen is placed across the supports of the die which is the lower portion of the jig. The plunger, operated from above by a hydraulic jack or other device, causes the specimen to be forced into and to assure the shape of the die. To fulfill the requirements of this test, the specimens must bend 180 degrees and, to be accepted as passable, no cracks greater than 1/8 in. (3.2 mm) in any dimension should appear on the surface. The face bend tests are made in the jig with the face of the weld in tension (i.e., on the outside of the bend) (A, fig. 13-2). The root bend tests are made with the root of the weld in tension (i. e., on outside of the bend) (B, fig. 13-2). Guided bend test specimens are also shown the in figure 13-3.


a. The free bend test has been devised to measure the ductility of the weld metal deposited in a weld joint. A test specimen is machined from the welded plate with the weld located as shown at A, figure 13-4. Each corner lengthwise of the specimen shall be rounded in a radius not exceeding one-tenth of the thickness of the specimen. Tool marks, if any, shall be lengthwise of the specimen. Two scribed lines are placed on the face 1/16 in. (1.6 mm) in from the edge of the weld. The distance between these lines is measured in inches and recorded as the initial distance X (B, fig. 13-4). The ends of the test specimen are then bent through angles of about 30 degrees, these bends being approximately one-third of the length in from each end. The weld is thus located centrally to ensure that all of the bending occurs in the weld. The specimen bent initially is then placed in a machine capable of exerting a large compressive force (C, fig. 13-4) and bent until a crack greater than 1/16 in. (1.6 mm) in any dimension appears on the face of the weld. If no cracks appear, bending is continued until the specimens 1/4 in. (6.4 mm) thick or under can be tested in vise. Heavier plate is usually tested in a press or bending jig. Whether a vise or other type of compression device is used when making the free bend test, it is advisable to machine the upper and lower contact plates of the bending equipment to present surfaces parallel to the ends of the specimen (E, fig. 13-4). This will prevent the specimen from slipping and snapping out of the testing machine as it is bent.

b. After bending the specimen to the point where the test bend is concluded, the distance between the scribed lines on the specimen is again measured and recorded as the distance Y. To find the percentage of elongation, subtract the initial from the final distance, divide by the initial distance, and multiply by 100 (fig. 13-4). The usual requirements for passing this test are that the minimum elongation be 15 percent and that no cracks greater than 1/16 in. (1.6 mm) in any dimension exist on the face of the weld.

c. The free bend test is being largely replaced by the guided bend test where the required testing equipment is available.


The back bend test is used to determine the quality of the weld metal and the degree of penetration into the root of the Y of the welded butt joint. The specimens used are similar to those required for the free bend test (para 13-15) except they are bent with the root of the weld on the tension side, or outside. The specimens tested are required to bend 90 degrees without breaking apart. This test is being largely replaced by the guided bend test (para 13-14).


a. The nick break test has been devised to determine if the weld metal of a welded butt joint has any internal defects, such as slag inclusions, gas pockets, poor fusion, and/or oxidized or burnt metal. The specimen is obtained from a welded butt joint either by machining or by cutting with an oxyacetylene torch. Each edge of the weld at the joint is slotted by means of a saw cut through the center (fig. 13-5). The piece thus prepared is bridged across two steel blocks (fig. 13-5) and stuck with a heavy hammer until the section of the weld between the slots fractures. The metal thus exposed should be completely fused and free from slag inclusions. The size of any gas pocket must not be greater than 1/16 in. (1.6 mm) across the greater dimension and the number of gas pockets or pores per square inch (64.5 sq mm) should not exceed 6.

b. Another break test method is used to determine the soundness of fillet welds. This is the fillet weld break test. A force, by means of a press, a testing machine, or blows of a hammer, is applied to the apex of the V shaped specimen until the fillet weld ruptures. The surfaces of the fracture will then be examined for soundness.


a. This test is used to measure the strength of a welded joint. A portion of a to locate the welded plate is locate the weld midway between the jaws of the testing machine (fig. 13-6). The width thickness of the test specimen are measured before testing, and the area in square inches is calculated by multiplying these before testing , and the area in square inches is calculated by multiplying these two figures (see formula, fig. 13-6). The tensile test specimen is then mounted in a machine that will exert enough pull on the piece to break the specimen. The testing machining may be either a stationary or a portable type. A machine of the portable type, operating on the hydraulic principle and capable of pulling as well as bending test specimens, is shown in figure 13-7. As the specimen is being tested in this machine, the load in pounds is registered on the gauge. In the stationary types, the load applied may be registered on a balancing beam. In either case, the load at the point of breaking is recorded. Test specimens broken by the tensile strength test are shown in figure 13-3.

b. The tensile strength, which is defined as stress in pounds per square inch, is calculated by dividing the breaking load of the test piece by the original cross section area of the specimen. The usual requirements for the tensile strength of welds is that the specimen shall pull not less than 90 percent of the base metal tensile strength.

c. The shearing strength of transverse and longitudinal fillet welds is determined by tensile stress on the test specimens. The width of the specimen is measured in inches. The specimen is ruptured under tensile load, and the maximum load in pounds is determined. The shearing strength of the weld in pounds per linear inch is determined by dividing the maximum load by the length of fillet weld that ruptured. The shearing strength in pounds per square inch is obtained by dividing the shearing strength in pounds per linear inch by the average throat dimension of the weld in inches. The test specimens are made wider than required and machined down to size.