Technical Insight: Electron Beam Welding

Electron beam welding (EBW) is a fusion welding process that joins metals through a focused, high-energy beam of electrons. The kinetic energy of the electrons creates heat upon impact to melt the materials and create a deep, narrow weld with a small heat affected zone (HAZ) and minimal distortion. Although the process does not typically require the use of filler wire, it does need to be performed in a vacuum to prevent the electrons from be scattered by air molecules as well as to create a clean final weld.

TWI has decades of experience in electron beam welding, developing the process and technology as well as creating innovations that have progressed the technique. We also offer electron beam welding training and bespoke solutions for our Industrial Members.

CORE RESEARCH

Formative Research

Some of our earliest work in electron beam welding dates back to the later 1970s, when we created several core research programme (CRP) projects to solve electron beam welding challenges for the benefit of our Industrial Members and wider industry.

This included an investigation into high voltage breakdown in electron beam welding equipment. This early electron beam CRP project tackled what was an inherent weakness in the process caused by particle streams generated in the weld capillary that led to high voltage breakdown of electron beam guns. TWI experts developed double and triple magnetic trap devices that minimised beam spot movement without introducing excessive beam aberration. The triple deflection system eliminated gun discharging for gassy and volatile materials up to high power levels.

At the same time as addressing high voltage breakdown in electron beam (EB) welding equipment, our experts developed a test to investigate the factors affecting solidification cracking during EB welding, creating a regression analysis using crack lengths measured in the tests to make a formula to predict crack length from compositional factors of the form.

As TWI addressed the existing challenges of electron beam welding, we were also working on the computer design of compact electron beam guns, to continue advancing the process for industry, as well as investigating the development of a high power (60kW) non-vacuum electron beam (NVEB) welder, assessing the effects of process parameters (welding speed, beam power and working distance) and joint preparation on weld penetration depth.

By the 1980s, our core research was investigating EB welding for different materials in a variety of configurations, such as the 1982 CRP project to examine the mechanical properties of welds made with non-arc joining processes. The 1982 fatigue strength of EB transverse butt weld joints in steel project specifically examined the fatigue properties of fully penetrating single-pass electron beam welds in a 13mm thick steel plate to BS 4360: 1972 Grade 50D. The effect of top bead undercutting on the fatigue performance of the welds was investigated along with the effect of dressing the weld beads by grinding and applying an electron beam cosmetic pass. It was shown that transverse butt welded joints, made by the electron beam process, possessed similar properties to joints made by arc-welding processes, e.g. submerged-arc or manual metal arc, and that they could be classified using the existing fatigue design rules for welded steel joints. This research was continued and extended through further CRP projects in 1984, 1985 and 1989. By 1995 our experts were performing tests were to determine the fatigue strength of EB butt welds in seven different conditions: as-welded (with or without backing bars); welds (without backing bars) having received a second smoothing pass; welds with backing bars with the bar removed and the root side ground flush; welds (with or without backing bars) having received uncontrolled flush grinding on both sides; and welds without backing bars having received full grinding on both sides.

As work continued to assess the fatigue strength of EB welds, TWI’s electron beam experts were also conducting research into the origin and effect of magnetic fields in EB welding and examining the effect of process parameters on electron beam welds in steel.

However, electron beam welding gun discharge was still an issue that needed researching, so our we created a 1985 CRP project to compare EB gun discharge tendencies in a range of materials at different thicknesses and also under beam fade-in conditions and at low current level. From this it was possible to classify groupings in order of difficulty. For materials and thicknesses where excessive gun discharging occurs, with a conventional electron column, a magnetic trap device was shown to eliminate discharge events almost entirely. TWI also compared fusion zone width, without and with the magnetic trap device, which generally indicated no undue beam aberration effects. This project continued into a second phase, to re-design the magnetic trap and control unit to solve the problem of equipment malfunction from electron gun discharges caused by metal vapour generated at the weld pool entering the electron gun. An improved design was devised and manufactured, leading to a reduction in beam aberration. EB weld discharges were also the subject of a 1994 CRP project to improve their control through an assessment and improvement of the control electronics for 5kHz switchmode power supplies for electron beam welding.

Defect control was also a subject of formative EB welding research at TWI, with a 1986 CRP project investigating defect control of thick section electron beam welds. This project assessed the factors affecting the creation of defects in the fade-out region where beam current is reduced to zero, particularly in thick section welds. The results of an experimental programme was also presented with particular reference to the effects of beam focus position on partial penetration weld shape. This work continued with a 1988 CRP project that conducted constant current partial penetration (CCPP) welds and fade-out tests in 70mm thick C-Mn steel, isolating the effects of various process variables on weld quality. Leading to the creation of methods by which satisfactory fade-out integrity could be achieved for this material and thickness.

Process Improvements

As we reached the end of the 1980s, the work of TWI’s core research into EB welding began to focus on process improvements. This began with a project to improve the microstructure and properties of electron beam welds, which investigated three techniques; producing a highly alloyed weld, with microstructure and phases present that were completely different from those normally encountered, producing a weld with a lower carbon equivalent than the parent plate, and producing a weld with a fine, tough 'acicular ferrite' (AF) microstructure. Various filler materials were added to the weld metal in order to promote such changes, with their effects assessed by metallography, hardness determination, and Charpy testing.

Magnetism remained a problem for thick section EB welding of ferritic materials, leading to a 1990 CRP project the examined residual magnetism and magnetism generated by dissimilar metal joints. In both cases, experimental techniques were evolved and preliminary results reported. Although filler wire is not typically required for EB welding, a 1992 project examined the use of filler wire as a potential technique for improving the tolerance of the process to poor fit-up and for reducing dependence on parent metal composition to achieve good toughness when welding thick section steel. EB welding was also studied as a solution for encapsulating a medium carbon (0.4%) steel created by hot isostatic pressing (HIP). EB sealing was compared to TIG canning in terms of room temperature Charpy fracture energy and joint microstructure.

As well as seeking to progress the technique, TWI’s work in this period also sought to improve monitoring of EB welds, including research into the use of backscattered electrons to obtain a video image of the workpiece surface during welding. This work continued into 1995 with further improvements to imaging for in-vacuum electron beam welding, allowing systems to operate continuously at all beam powers up to 75kW without any image deterioration due to metal vapour deposits. This later work also removed the need to pulse the electron gun during imaging while making the vision system mechanically and electrically rugged and easily installed in new equipment or retrofitted later. By 1996, our EB experts continued to improve real-time seam tracking using backscattered electrons for the detection of the joint position. These improvements focused on the signal processing components of the tracker to enable more consistent operation at up to 30kW beam power.

While the work continued to advance the process, our experts were still investigating solutions for EB welding defects. This included a 1996 CRP project to improve the response to electron beam weld flashovers. These occur when the accelerating potential of an electron beam welding machine breaks down, creating weld defects. Breakdowns were shown to typically occur when the vacuum is degraded by gases from the weld pool, causing the beam current to surge or get interrupted, causing defects in the weld. TWI’s experts advanced previous research into this problem through the design and construction of a device for better detection of gun discharges, improving the power supply response time as well as determining the best characteristics of the beam feedback circuit in order to allow rapid control of the high voltage during discharges. Demagnetisation for EB welding was also examined further in a 1997 report addressing missed joints in electron beam welding.

Local and Non-Vacuum Research

The need to contain the entire structure to be welded enclosed in a vacuum envelope had limited the adoption of the process by industry, so we sought a solution from the mid-2000s to develop a local vacuum system for reduced pressure electron beam welds. This preliminary project work involved an examination and critical review of methods for reliable generation of local vacuum or reduced pressure conditions in which EB welding could be carried out. This was followed by the design and construction of an experimental local sealing and pumping test facility to permit evaluation of the various seal/pumping concepts. This work continued through into 2008 with research to establish the optimum local sealing method in the pressure regime consistent with reliable welding performance (ie 0.1-10mbar) and to create a demonstration facility at TWI with sufficient flexibility to permit the practicality of EB welding using local vacuum on a variety of component types. By 2010, we were undertaking reduced pressure local vacuum electron beam welding trials with the creation of an experimental mobile seal system at TWI which illustrated that, with local sealing and pumping, the pressure levels required for successful high power EBW at reduced pressure (1mbar) can be sustained whilst moving at a typical welding speed. Further work, in 2012, examined out-of-chamber thick C-Mn steel electron beam welding, with the objective of developing a local vacuum mobile sliding seal established in previous work to a position of repeatable and robust laboratory application before establishing a prototype small scale welding facility to assess welding performance and aligning welding performance to identified large scale fabrication needs. At the same time TWI was researching non-vacuum electron beam welding solutions with a comparison of the advantages and disadvantages of the process with laser welding.

EB Quality and Characterisation

The 2000s also gave TWI an opportunity to advance EB welding quality and characterisation, building upon work begun in the mid-1990s to produce high speed beam analysis equipment to allow for the recording and display of beam intensity distribution. By 2011, we were ready to review the available beam quality assessment methods and decide which was our preferred method for measuring the quality of electron beams and by 2013 TW had developed a Mk3 beam probing and analysis system, BeamAssure^TM, to determine laboratory and industrial beam qualities and the relationship of those qualities to weld integrity. Laboratory tests were made at TWI with industrial deployment within a highly controlled aerospace production facility to validate electron beam probing equipment.

Electron beam gun design was the focus of a 2015 CRP project, which characterised an RF excited plasma cathode EB gun that offered the potential to provide reliable and consistent beam processing. The use of RF simplifies the power supply and enables rapid beam pulsing.

Combining our work with local chamber EB welding with efforts to monitor the process in real time, we created a 2020 CRP project aimed at using ultrasonic testing (UT) to monitor the reduced pressure electron beam welding (RPEBW) process, including the design of a phased array ultrasonic testing (PAUT) prototype system for real-time monitoring of RPEBW in process conditions (Figures 1-3). This core research continued into 2025, with work to correlate between data collected during welding and flaws that are likely to appear in the joints.

EBOBend

Our experts have also been available to test the capabilities of new technological innovations in EB welding, including an assessment of novel inside out electron beam welding using EBOBend. The EBOBend apparatus allows the electron beam to access the inside of tubes or cavities for processing, making welding and processing possible at the full power of the machine, facilitating welds up to and beyond 20mm in titanium. The process was deemed suitable for electron beam welding of components such as fuel tanks for satellite systems, vacuum vessels, and aerospace landing gear applications, highlighting the possibilities available with the EBOBend equipment (Figures 4-13).

INDUSTRY-SPECIFIC PROJECTS

Of course, not all of our work in EB welding has been conducted as part of our core research, as a number of projects have been undertaken over the years for the direct benefit of specific industry sectors.

Wind Power

We joined the collaborative HiWeld project to assist the wind power sector with the challenge of producing thick section steel structures. At the time of the project, the fabrication of structures was limited by welding times and costs, with a typical 40m long monopile (nominally 60mm thick) taking approximately 6,000hrs to fabricate. Electron beam (EB) welding systems were shown to be able to reduce this welding time to less than 200hrs, equivalent to a reduction in cost of over 85%.

Defence

Moving to the defence sector, and TWI provided support for the joining of aluminium alloys with EB for military vehicles, conducting extensive weld development studies.

Aerospace

The aerospace industry has adopted EB welding for its ability to create, precise, low distortion welds under clean vacuum conditions. Our experts conducted studies on behalf of the industry to use advanced electron beam deflection techniques to split the heat source and allow a portion of the beam’s energy to complete other tasks such as heating the workpiece or making a second, cosmetic weld pass at the same time as the main welding operation.

Formula One

Formula One racing relies on the structural performance of components to succeed, making materials validation, inspection and joining integral to race teams. We were approached by a team who wished to fabricate small pressure vessels in a thin wall aluminium alloy, so we applied EB technology to weld two hemispherical caps to each assembly as the thin walls of the structure and predicted in-service conditions required the highest degree of quality and joint integrity (Figure 14).

Nuclear

TWI provided expertise and process guidance for nuclear waste management programmes across the world, determining that electron beam welding was a reliable process for the manufacture of spent nuclear fuel containers. This work began in the late 1980s and early 1990s, with later R&D establishing best practice for materials selection as well as setting down a series of manufacturing and high integrity welding methods and parameters for fabricating and sealing containers (Figure 15).

Aerothermal

Cambridge Aerothermal contacted TWI for assistance in EB welding of a novel temperature sensor probe for use with industrial gas turbines. The probe’s complex design used numerous small components made of various materials, which required the careful application of EB welding with minimal heat input and distortion. This included the EB welding of precipitation hardening copper alloy and the creation of a prototype component using a set welding plan. The prototype probe was subjected to extensive testing ahead of further work on additional prototypes (Figures 16-19).

JOINT INDUSTRY PROJECTS

Whereas our core research is designed to progress capabilities and knowledge for the wider benefit of our Industrial Members and other projects are conducted on behalf of specific Members, our joint industry programme projects allow interested parties the opportunity to pool their resources to gain access to research on specific areas of interest.

This included a 2007 project to exploit power beam welding of thick section steel for structural applications using ytterbium doped fibre laser and non-vacuum electron beam processes. By 2012, there had been a renewed interest in welding thick section steels for the renewable energy, nuclear a fossil-fuel industries. Although EB and narrow gap (NG) arc welding processes provide a cost effective and high integrity joining solution, the presence of residual magnetism in the materials impeded the effective application of these processes. To address this, TWI launched a JIP project related to the demagnetisation of thick-section ferritic steel components for electron beam fabrication. The project examined the origins of residual magnetic fields in large structures, and developed techniques for demagnetising large components using numerical and experimental methods, providing a framework for a European standard on assessment of residual magnetism and guidelines for demagnetisation for welding requirements.

A 2022 JIP project continued our research into EB welding solutions for the offshore wind sector, with the validation of out-of-chamber electron beam welding for the fabrication of offshore wind turbine support structures. Building on recent developments, the project sought to establish out-of-chamber EB welding capabilities for a range of thicknesses and steel grades suitable for the fabrication of offshore wind support structures, generate fatigue data to demonstrate the performance of the welds and compare with the qualification requirements of relevant certification bodies, before producing a recommended practice guide.

This is just a glimpse at some of the EB welding R&D conducted at TWI over the decades in support of industry and our Industrial Members. As well as progressing the technology and delivering innovation, TWI also provides bespoke training in electron beam welding, focussing on machine operation, CNC programming for EB welding and beam deflection optimisation.

You can find out more about electron beam welding services at TWI, here:

https://www.twi-global.com/what-we-do/research-and-technology/technologies/welding-joining-and-cutting/electron-beam-technology