A lug 4-in. long, 5-in. high, and 1-in. thick attaches to the beam flange. Through a hole in the lug, the hydraulic cylinder applies a force of 22 ton, which is transferred to the rolled beam through a pair of fillet welds. The designer determined the size of the fillet welds by taking the 44,000-lb force, dividing it by the two 4-in.-long welds, and obtaining a unit force per length of 5,500 lb/in. Using E70 weld metal and a weld metal allowable of 30%, the designer calculated a required fillet weld size of 0.370 in. and specified a 3 8-in. fillet weld.

In the second version, the designer lengthened the weld to reduce the leg size proportionally. The designer added a 1-in.-wide, 11-in.-long bar to the assembly to lengthen the weld.

Company workers built the second log splitter, and after a few loading cycles, the lug tore away from the beam. Once the horizontal weld ripped, the 1-in. bar bent, twisted, and also tore away.

The problem was solved when the larger piece was welded to the beam with a true L-shaped weld group, consisting of a 7 16-in. PJP groove welds to the web. Adequate support enabled the whole weld group to function as intended, permitting successful operation of the log splitter.

Providing proper support for welds is critical as illustrated in the design of a log splitter's main member, which uses a rolled-beam section.

When the first log splitter was fabricated, the weld immediately broke, prompting the chagrined designer to reconsider his initial calculations. Since the hole was 4.5 in. above the beam, the fillet welds were subject to a bending moment of 44,000 lb 3 4.5 in., or nearly 200,000 in.-lb. To resist this bending moment, the welds would require leg sizes of 2.5 in., more than 6 3 larger than those originally supplied.

Understandably reluctant to place such huge welds on this assembly, the designer sought another solution. The first design had restricted the available length for making the weld to a total of 8 in. (4 in. on each side).

In his next attempt, the designer specified a groove weld to attach the original lug to the bar, and this assembly was welded to the beam. A pair of L-shaped "weld groups" was used; in theory, the whole weld group would resist the load. With a total of 20 linear inches of joint on which the weld could be placed, and with a configuration that better resisted the bending loads, the required weld leg size dropped to only 0.579 in. Therefore, the designer specified a ⁵ /8-in. fillet. The connection to the web was best suited as a partial-penetration (PJP) groove weld with a throat of ⁷ /16 in. on each side.

The designer's reasoning for adding the 1-in. bar was correct, but he had failed to properly support the additional two 6-in.-long PJP groove welds to the beam web. As a result, those welds became ineffective, and participated in load transfer only after the horizontal weld tore.

The flexible 1 3 1-in. bar could not transfer the applied load into the vertical portion of the weld group, and the horizontal portion took virtually all the load until it failed. Then, the vertical portion simply peeled off. Because the vertical welds were not supported, the welds did not act as an L-shaped group, but rather as a couple of unconnected linear welds.

The designer realized that the ultimate solution was to make the lug longer to supply ample support to the weld group. He made a lug from a single piece of 11-in.-high, 6-in.-long, 1-in.-thick material. Then he specified that a section 4-in. long by 6-in. high be cut from the piece to fit over the beam.

Omer W. Blodgett, Sc.D., P.E., senior design consultant with The Lincoln Electric Co., struck his first arc on his grandfather's welder at the age of ten. He is the author of Design of Welded Structures and Design of Weldments and an internationally recognized expert in the field of weld design. In 1999, Blodgett was named one of the "Top 125 People of the Past 125 Years" by Engineering News Record. Blodgett may be reached at (216) 383-2225.