Welding’s vital part in major American historical events
By Bob Irving
The finest hours for U.S. shipbuilding were during World War II when 2710 Liberty ships, 531 Victory ships and 525 T-2 tankers were built for the war effort. Through 1945, some 5171 vessels of all types were constructed to American Bureau of Shipping (ABS) class during the Maritime Commission wartime shipbuilding program. At this time in shipbuilding history, welding was replacing riveting as the main method of assembly.
The importance of welding was emphasized early in the war when President Roosevelt sent a letter to Prime Minister Winston Churchill, who is said to have read it aloud to the members of Britain’s House of Commons. The letter read in part, “Here there had been developed a welding technique which enables us to construct standard merchant ships with a speed unequaled in the history of merchant shipping.”
The technique the President was referring to was undoubtedly submerged arc welding, which was capable of joining steel plate as much as 20 times faster than any other welding process at that time.
As the war progressed, new shipyards opened. In 1943, no fewer than 17 shipyards in the United States were building Liberty ships for the war effort. In June of 1943, California Shipbuilding Corp. broke the U.S. record that month by delivering 20 Liberty ships. Its work staff included 6000 welders and 160 operators of submerged arc welding equipment. Each ship was consuming 135,000 lb (60,750 kg) of welding electrodes. The shipyard used a carload of 65,000 lb (29,250 kg) of welding electrodes every three shifts. The Bethlehem Shipyard in Baltimore was in second place, delivering 14 Liberty ships that same month.
At Southeastern Shipbuilding Corp., Savannah, Ga., more than 2000 welders delivered at least three Liberty ships per month. Chicago Bridge & Iron Co. even got into the act and was weld fabricating ships for the U.S. Navy.
During this period of assimilation, eight Liberty ships were lost due to a problem called brittle fracture. At first, many blamed welding, but history would soon prove that the real cause of brittle fracture was steels that were notch sensitive at operating temperatures. The steel was found to have high sulfur and phosphorus contents. Another cause was design-related discontinuities, such as hatch openings, vents and other interruptions in the structure. By far the highest incidence of fracture occurred under a combination of low air temperature and heavy seas.
On more than 1400 ships, crack arrestors were used to prevent crack propagation. No crack was known to grow past an arrestor. This safeguard helped reduce casualties from 140 to 20 per month.
In 1939, the American Bureau of Shipping had 92 surveyors on staff. By World War II, those numbers increased dramatically, reaching 479 in 1944.
The ASME Code
In the late 1920s and early 1930s, the welding of pressure vessels came on the scene. Welding made possible a quantum jump in pressure attainable because the process eliminated the low structural efficiency of the riveted joint. Welding was widely utilized by industry as it strove to increase operating efficiencies by the use of higher pressures and temperatures, all of which meant thick-walled vessels. But before this occurred, a code for fabrication was born from the aftermath of catastrophe.
On April 27, 1865, the steamboat Sultana blew up while transporting 2200 passengers on the Mississippi River. The cause of the catastrophe was the sudden explosion of three of the steamboat’s four boilers, and up to 1500 people were killed as a result. Most of the passengers were Union soldiers homeward bound after surviving Confederate prison camps. In another disaster on March 10, 1905, a fire tube boiler in a shoe factory in Brockton, Mass., exploded, killing 58, injuring 117 and causing damages valued at $250,000. These two incidents, and the many others between them, proved there was a need to bring safety to boiler operation. So, a voluntary code of construction went into effect in 1915 – the ASME Boiler Code.
As welding began to be used, a need for nondestructively examining those welds emerged. In the 1920s, inspectors tested welds by tapping them with hammers, then listening to the sound through stethoscopes. A dead sound indicated a defective weld. By 1931, the revised Boiler Code accepted welded vessels judged safe by radiographic testing. By this time, magnetic particle testing was used to detect surface cracks that had been missed by radiographic inspection. In his history of the ASME Code, A. M. Greene, Jr., referred to the late 1920s and early 1930s as “the great years.” It was during this period that fusion welding received widespread acceptance. Nowadays, thousands of individuals who make their living in welding live and breathe the ASME Code every minute of the working day.
As far as welding interests are concerned, probably the most important part of the ASME Code is “Section IX – Welding and Brazing Qualifications.” This section relates to the qualification of welders and welding operators, and the procedures they must follow to comply with the code. Under procedure qualifications, each process is listed, and the essential and nonessential variables of each are spelled out. Welding performance qualifications are also included.
In the early years of the ASME Code, fabricators were known to spend their own research dollars to develop a process so it could be used in Code construction work. Eventually, however, a code case would have to be submitted to the appropriate ASME Code committee, but first a procedure qualification had to be developed. The code authorities expected the quality of the weld metal and the heat-affected zone to be equal to that of the base metal.
For years, considerable research has been carried out in support of the ASME Code, now called the Boiler and Pressure Vessel Code. Much support of the Code’s welding content came from the Pressure Vessel Research Committee, or PVRC, of the Welding Research Council, New York, N.Y. PVRC was launched shortly after World War II.
In 1977, Leonard Zick, chairman of the main committee of the ASME Code, said, “It’s more than a code; the related groups make up a safety system. Our main objective is to provide requirements for new construction of pressure-related items that, when followed, will provide safety to those who use them and those who might be affected by their use. “And, since the use of the code item might be for any type of process or for any discipline involving energy, the committee’s activities do not play one against the other. We want all code items to be safe, period.”
A triumph of the code was the huge aluminum spheres built by General Dynamics in Charleston, S.C. They were built to criteria established by the U.S. Coast Guard and were based on Section VIII, Division 1, of the ASME Code.
At about 2 a.m. on October 2, 1976, the first welded aluminum sphere for a liquefied natural gas tanker was rolled out of a building in Charleston, then moved over to a special stand for final hydropneumatic testing. It soon passed the test with flying colors. The sphere itself weighed 850 tons and measured 120 ft (36 m) in diameter. Each sphere consisted of more than 100 precisely machined plates, “orange peel” in shape. The plates were gas metal arc welded together using 7036 lb (3166 kg) of filler metal. Total length of the welds on each sphere was 48.6 miles. Completed spheres were barged along the coast and delivered onto steel tankers under construction at General Dynamics’ shipyard in Quincy, Mass. This type of LNG tanker was based on the Moss-Rosenberg design from Norway.
Meanwhile, Newport News Shipbuilding and Dry Dock Co. was building LNG tankers in Virginia. Based on the Technigaz design, the tankers featured a waffled membrane of stainless steel for containing the gas. Avondale Shipyards, Inc., New Orleans, La., was building still other LNG tankers based on the Conch design, which featured prismatic tanks of aluminum.
At General Dynamics’ facility in Charleston, 80% of the metalworking manhours were spent welding. Much of the filler metal deposited in Charleston was 5183 aluminum. The vertical joints were welded using special equipment from Switzerland in which the operator rode in a custom-designed chair alongside the welding arc. At this distance, he was able to monitor the weld and observe the oscillation of the 1Z16-in. (1.5-mm) diameter filler metal. Actual welding was controlled remotely. About 30 weld passes were required for each joint.
The thicker 11Z2-in. (3.8-cm) joints were welded by Big MIG equipment, operating with a 1Z8-in. (3-mm) diameter wire at 500 A. The equipment was capable of completing the joint in four passes.
The massive equatorial ring was welded outdoors. In this setup, nine heavily machined, curved aluminum extrusions had to be welded together. To do it, 88 GMA weld passes were made from the outside and 60 more from the inside.
One engineer on site at the time had just transferred from one of the company’s aerospace divisions, a division known for extremely high weld quality on sensitive material. “As far as quality is concerned,” he noted, “there’s really not that much difference (between Charleston and the aerospace division). I do think that the weld quality achievement here in Charleston is as high, if not higher, than it is in aerospace, but then 5083-0 is a very forgiving aluminum alloy.”
The Alaska Pipeline
Perhaps no single welding event in history ever received so much attention as did the Alaska Pipeline. Crews of seasoned welders braved Alaska’s frigid terrain to weld this large-diameter pipeline, from start to finish. At one point, 17,000 people were working on the pipeline – 6% of the total population of Alaska. The entire pipeline only disturbed about 12 square miles of the 586,000 square miles of the state of Alaska.
Welders were called upon to handle and weld a new steel pipe thicker and larger than most of them had ever encountered before, using electrodes also new to most. And, the requirements were the stiffest they had ever seen.
The requirements for the welds in double jointing were even more severe. For double-jointed pipe, welds were expected to meet an average of 20 ft-lb (27 J) and at least 15 ft-lb (20 J) in both the welds and the heat-affected zones at -50°F (-45.5°C). About 38,000 of these joints were welded for the Alaska pipeline. The joints were made using submerged arc welding and a wire that contained 3% nickel. About 80,000 lb (36,000 kg) of that wire were used in the project. Field construction of this 798-mile-long pipeline began in March 1975 and it was no picnic. The U.S. Department of the Interior and a new pipeline coordinating group representing the state of Alaska instituted some changes.
So, the original specifications for field welding were tossed, replaced by much stiffer requirements for weld toughness. Instead of the conventional pipeline welding electrode planned originally for the bulk of field welding, the new requirements required higher quality. The only electrode the engineers could find that met the new requirements was an E8010-G filler metal from Germany, so it was soon flown over by the planeload. Some in Local 798 of the Pipeline Welders Union out of Tulsa, OkIa., then welding in Alaska, had used this electrode while working on lines in the North Sea, but most welders were seeing it for the first time.
One of the requirements was 100% X-ray inspection of all welds. The films were processed automatically in vans that traveled alongside the welding crews.
Welders worked inside protective aluminum enclosures intended to protect the weld joint from the wind. Lighting inside the enclosures enabled welders to see what they were doing during Alaska’s dark winter.
On the main pipeline, preheat and the heat between weld passes was applied at first by spider-ring burners. Induction heating was used later during construction.
All of the 48-in. (122-cm) diameter main line pipe came from Japan. The total order called for about 500,000 tons of pipe. Two wall thicknesses were required, 0.462 and 0.562 in. (12 and 14 mm). About 407 miles of the pipeline are buried underground. Due to concern about the melting of permafrost from the heat of the pipeline, the rest of the line was installed above ground and rested on vertical support members weld fabricated from 18-in. (45.7-cm) diameter 5LX steel piping. About 120,000 tons of this type of piping were used on the spread. The welds were made using E8018-C3 low-hydrogen electrodes.
About 30 years ago, steel construction went into orbit. The 100-story John Hancock Center in Chicago and the 110-story twin towers of New York’s World Trade Center were under construction. Above ground, the World Trade Center required some 176,000 tons of fabricated structural steel. The Sears Tower came later. Bethlehem Steel Corp. had received orders for 200,000 tons of rolled steel products for the South Mall complex in Albany, N.Y. Allied Structural Steel Co. was reported to have used multiple-electrode gas metal arc welding in the fabrication of the First National Bank of Chicago Building.
In a progress report on the erection of the critical corner pieces for the first 22 floors of the 1107-ft (332-m) high John Hancock Center, an Allied Structural Steel spokesman said various welding processes were being used in that portion of the high-rise building. More than 12,000 tons of structural steel were used in that section. Webs and flanges for each interior H column were made up of A36 steel plate with thicknesses up to 61Z2 in. (16.5 cm). The long fillet welds at the web-to-flange contact faces were made using the submerged arc process, while the box sections were being welded by CO2-shielded E70T-1 flux cored electrodes. On the box sections, welding operators were averaging deposition rates of 90 lb (40.5 kg) per day. A total of 310,000 lb (139,500 kg) of weld metal was consumed in shop fabrication for this building, while 165,000 lb (74,250 kg) of weld metal was consumed during field erection.
Weld metal consumption in shop fabrication for the U.S. Steel Building in Pittsburgh, Pa., reached 609,000 lb (274,050 kg).
During this same period, Kaiser Steel Corp. had used the consumable guide version of electroslag welding to deposit 24,000 welds in the Bank of America world headquarters building in San Francisco. At the time, this building was regarded as the tallest earthquake-proof structure ever erected on the West Coast.
In terms of welding, one of the most intensive structures built during this period was NASA’s Vertical Assembly Building on Merritt Island, Fla. Shop-welded sections for this giant structure consumed 830,000 lb (373,500 kg) of weld metal.
For the World Trade Center, Leslie E. Robertson, a partner in charge of the New York office of Skilling, Helle, Christiansen, Robertson, said a computer was used to produce the drawing lists, beam schedules, column details and all schedules for exterior wall panels. Millions of IBM cards were then sent to every fabricator. These cards gave fabricators the width, length, thickness and grade of steel of every plate and section in all of the columns and panels. “In addition,” he said, “the fabricators are given all of the requirements of every weld needed to make up the columns and panels. Many of these cards are used as equatable to the production of drawings. They are sent directly from the designer to the fabricators. Draftsmen never become involved.”
Elsewhere in New York City, Leo Plofker, a partner in one of the city’s leading design engineering firms, extensively used welding in design. Among Plofker’s achievements are the Pan-Am Building and the all-welded, 52-story office building known as 140 Broadway.
“Our decision to make extensive use of welding is strictly based on economics,” he said. “Welded design results in savings in steel. Field welding can cause some problems, but they are not too serious as long as you maintain control over the welders and you insist that qualified personnel be employed to perform nondestructive testing of the welds.”
1. Iron Age.
2. The History of American Bureau of Shipping.
3. The Code – An Authorized History of the ASME Boiler and Pressure Vessel Code.
4. Welding Journal.