Ft.JNDAMENTAL STUDY OF JOINING TITANIUM ALLOYS USING ELECTROSIAG TECHNOLOGY Sin-Jang Chen, ph.D. Oregon Graduate Institute, 1991 SupervisingProfessor:Jack H. Dev1etian At present, titanium alloys can not be welded in thick sections (>25mm) with commercial welding processes such as submerged arc, gas metal arc and flux cored arc welding because of titanium's excessive high reactivity and resistivity at high temperatures. As a result, the only methods to weld titanium alloys in thick section are electron beam welding which is not practical for large structures and gas tungsten arc welding which is extremely labor intensive. Thus, a substantial need for cost-effective welding of titanium alloys exists. In adapting the electroslag technique for joining titanium alloys, many fundamental problems had to be overcome. As a result, the mechanisms of ohmic heat generation in both the slag and electrode were studied. Furthermore, the electrochemdcal effects between the slag and titanium filler metal were analyzed in order the model the heat generation mechanism in electroslag processing for application to both consumable guide and non-consumable guide ESW. Frequent experimental comparisons between ESWof titanium and steel (for reference) were drawn to better verify the models and predictive equations. xv

Fundamental study of joining titanium alloys using electroslag technology

This report summarizes the results of a preliminary examination of four Pennsylvania bridge structures thought to contain electroslag welds. They included the 179 multiple girder bridges at Meadville in Crawford County, the Blue Route Bridges over the Schuylkill River in Montgomery County, the 195 Girard Point approach pier caps in Philadelphia and the Spring Garden Street Bridge over the Schuylkill River in Philadelphia. Nondestructive tests were carried out on the Crawford County and Blue Route Bridges. These tests were supplemented by visual examination and surface grinding and polishing of the electroslag weldments in the field. Core samples were removed from several Crawford County, Blue Route and Girard Point electroslag weldments in order to permit metallographic studies and to evaluate the Charpy V-Notch absorbed energy of the welds. The cores removed f·rom the Crawford County Bridges contained cracks, and this prevented Charpy V-Notch tests from being carried out on these core samples. None of the welds examined in the Spr~ng Garden Street Bridge were found to contain electroslag weldments. As a result, no further work was carried out on this structure. The studies revealed significant cracking existed in the electroslag weldments of the Crawford County Bridges at Meadville. Most of these cracks were observed to follow the grain boundaries of the electroslag weldment near the fusion lines. Forty-eight of eighty-four electroslag

Evaluation of electroslag welds in three bridges

A Fitness-for-Purpose Evaluation of Fracture Critical Electro-Slag Welds

Live load stresses, thermal stresses., and the temperature gradients were measured on two of the main longitudinal girders of the Blue Route Bridge on I~476 over the Schuylkill River. Core samples were removed from several electroslag welds that cQlntained discontinuities and were examined in order to assess the defects and correlate the defects with the nondestructive tests. The fracture toughness characteristics of the weldment were also determined. The measured live load stresses indicated that full composite action existed between the concrete deck and the main girders. The live load stresses measured in the top flange electroslag welds were always less than 0.5 ksi under all loading conditions. The thermal gradient was recorded at several sections. The temperature gradient was used to compute stresses ,in the-·-t'bp flange, which were found to be about 1.5 k.si at the most severe thermal conditions. Destructive examination of the weld core samples revealed that the defects were not critical. The fracture toughness of the embedded top flange electroslag weldments was found to be adequate, and no retrofitting of the welds is considered necessary.

An assessment of the blue route bridge electroslag weldments, December 1981

The electroslag welding process is a rapid deposition, high heat input process in which the weld is made vertically. The metal is deposited under a blanket of molten slag which is maintained throughout the time of welding. The intimate contact of the weld deposit with the slag pool for a relatively long period of time affords an opportunity to obtain clean, well refined weld metal. The Electroslag Refining process conducted under similar conditions is used in the steel industry for the sole purpose of obtaining clean metal, ingots in that case rather than weld metal. Despite the potential for excellent weld deposits, problems have been encountered in electroslag welds. One such problem has been the occurrence of microcracks up to 4 mm (0.160") long in the weld metal. The-cracks are usually located in the proeutectoid ferrite following the prior austenite grain boundaries. The present research was directed toward finding out what is the cause of the microcracking and how it might be controlled. The research has indicated that the microcracking is a low temperature hydrogen induced phenomenon which begins at w-125 C (257 F) during postweld cooling. Sources of hydrogen were found to include moist flux, atmospheric moisture and the filler wire. A saturated 53 C (128 F) atmosphere consistently produced microcracking. Welding wire containing ;r20 ppm hydrogen caused cracking while wire containing <-15 ppm hydrogen did not cause cracking when welded under similar conditions. The amount of external restraint on the weldment was a determining factor in how much cracking was obtained at a given hydrogen potential. High basicity flux was found to produce larger amounts of weld metal hydrogen than a neutral flux. The hydrogen sometimes manifested itself as weld metal porosity (wormholes.) The possibility that more highly alloyed and/or finer austenite grained weld metal would exhibit less detrimental cracking has been raised by this research. Mechanical tests made on electroslag weld metal indicate that the microcracks do not cause a loss in Charpy V-notch or tensile ultimate and yield strength test values. The microcracks were found to cause a drop in tensile test elongation from 28% to 22%. Although there are many questions still left unanswered regarding the cracking phenomenon,• the results of this research suggest possible solutions to the microcracking in a production environment. The most obvious remedy is more careful control of the hydrogen supplied to the weld metal by the filler wire, the flux and the atmospehre. Also, the desirability of using retarded cooling, possibly with preheat or postheat, is readily apparent. Comparisons have been made with the apparently similar microcracking obtained in shielded metal arc welding.

Grain boundary separations in electroslag welds.

Electroslag welding is a very attractive process for making welds in building and bridge construction. There are a number of variations of the electroslag process and all variations have common characteristics. The process is initiated in a sump at the bottom of a joint. The sump is later removed and discarded and never considered a part of the finished joint for structural purposes. The process is ideally suited for welding heavy plates that can be positioned in the vertical or near-vertical position and the process becomes more economical as the plate thickness increases. The most common usage of electroslag welding appears to be for plate thicknesses that range from 1 to 8 inches and equipment is commercially available for butt-welding plates as thin as inch and as thick as 36 inches. The advantages of this process are: (1) once the process is started it is continuous and the weld can be completed in one pass and therefore there is only one setup and no need for interpass cleaning or repositioning of materials or parts; (2) extremely high depostion rates of 35 to 45 pounds per hour per electrode are achievable; (3) a savings of welding material because of no stub or spatter losses and flux consumption is only about 5 percent of deposited-meal weight; (4) joint preparation is minimized; (5) smooth, predetermined contours are produced on the weld enforcement; (6) high quality, sound weld deposits can be produced; (7) weld chemical composition can be controlled by adjusting the welding consumables and parameters; (8) no angular distortion in the horizontal plane because of the symmetry of most vertical welds; (9) preheating is generally not required; (10) electoslag welding is the fastest welding process for large thick joints; and (11) ease of operator training and good operator comfort. The disadvantages are: (1) welds must be made in a vertical or near-vertical position; (2) centerline cracking can occur because of unfavorable dendrite orientation unless welding parameters are properly controlled; and (3) if the welding process is discontinued for any reason, restaring will usually result in a major discontinuity that requires repair.

W P Benter

Papers

Acceptance criteria for electroslag weldments in bridges