Can Copper Tungsten Be Electron Beam Welded? When EB Welding Is Specified and What the Alternatives Are
Short answer: yes. Copper tungsten can be electron beam welded to copper and to chromium copper, and several European HV breaker designs specify exactly that for arcing contact rods. The longer answer is that EB welding of CuW is a demanding process with a narrow window, most contact suppliers do not run it in-house, and a drawing that says "EB welded" deserves a conversation about whether the process itself is the requirement or the joint properties are.
What Actually Happens When You EB Weld CuW
Copper tungsten is a two-phase composite: a sintered tungsten skeleton with copper filling the pores. Under an electron beam in vacuum, the copper phase melts and the tungsten does not. The weld pool is copper, with tungsten particles from the CuW side dispersed into it near the interface. On the other side of the joint, the copper or CuCr shank melts normally.
That gives EB welding three characteristics that matter on an arcing contact:
- Very narrow heat-affected zone. The beam is focused to well under a millimetre and travel speed is high. A CuCr1Zr shank a few millimetres from the joint stays at its age-hardened strength. This is the main reason OEMs specify EB over brazing when the shank is chromium copper.
- No filler. The joint is parent copper to parent copper, so conductivity across it is close to that of the shank. No braze layer to creep under arc heating.
- Deep, narrow penetration. A single pass can weld a 30 to 40 mm diameter rod joint. Multi-pass or double-sided welds are used for larger sections.
Why It Is Difficult on CuW Specifically
The same physics that make EB attractive make it hard to control on copper tungsten.
Thermal expansion mismatch. Tungsten at roughly 4.5 µm/m·K against copper at 17. The narrow, fast-cooling EB weld locks that mismatch into a small volume, and the result is cracking along the tungsten-copper interface unless the joint is preheated and the beam parameters are tuned to the specific CuW grade. Higher tungsten content, CuW80 and above, is more crack-prone than CuW70.
Porosity. Copper has high thermal conductivity and absorbs gas readily when molten. Infiltrated CuW can carry residual porosity from the infiltration step. Both feed gas into the weld pool. Porosity in an EB weld shows up as scattered voids on ultrasonic inspection, and a drawing with a tight unbonded-area limit will reject it.
Tungsten in the pool. Tungsten particles carried into the copper weld pool do not melt. They change the pool's viscosity and can leave a band of poor fusion at the CuW side. Beam offset toward the copper side is the usual answer, but that shifts the fusion line and has to be validated by metallography for each part geometry.
Equipment. An EB welder is a vacuum chamber, a high-voltage gun and CNC motion, and the chamber size limits the part length. This is specialised capital equipment. It exists at HV breaker OEMs, at a handful of contract welding houses, and at very few powder metallurgy contact producers. A supplier that infiltrates CuW well does not automatically weld it well, and the reverse is also true.
When OEMs Specify EB Welding
From the drawings we see, EB welding is called out when at least one of these applies:
- The shank is CuCr or CuCrZr and must stay hard. Brazing at 700 to 800 °C over-ages chromium copper. EB keeps the shank properties.
- The joint is in the high-current path and a filler layer is unacceptable. Some 245 kV and 420 kV arcing contact rods carry the full short-circuit current through the joint.
- The part is a long round rod, and the OEM's own process for that part number has always been EB. Once a part is inside a type-tested assembly, the OEM will not change the joining method without requalification, so the spec persists.
- Legacy design from a specific European OEM lineage. Certain breaker families were developed around EB-welded contact rods and the drawings still say so decades later.
In practice, the second and third reasons overlap almost completely with what friction welding also delivers.
The Alternatives
Friction welding
Rotary friction welding is the direct competitor to EB for round CuW-to-CuCr joints. Solid-state, no filler, narrow heat-affected zone, CuCr keeps most of its hardness. Joint strength close to the copper parent. The process is far more common at contact producers than EB, because a friction welder is a fraction of the cost of an EB chamber and needs no vacuum.
What friction welding cannot do: non-round joints, and very thin-walled tubes where forge pressure would collapse the section. For a solid rod arcing contact, it does what EB does. A fair number of drawings that once said EB now say "EB or friction welded, joint per [acceptance criteria]" for exactly this reason.
Cast-on
If the shank behind the tip can be pure copper rather than CuCr, integral infiltration (cast-on) gives the best joint conductivity of any method with no welding step at all. It is not an alternative when the drawing needs a hardened shank. See Cast-On vs Brazing vs Friction Welding for the full comparison.
Brazing with a low-temperature cycle
For MV parts and some 145 kV designs, a short induction brazing cycle with a lower-melting silver filler limits the time the CuCr spends above its ageing temperature. The shank loses some hardness, not all. Whether that is acceptable is the OEM's call and depends on the spring load on the contact.
How to Handle a Drawing That Says "EB Welded"
The useful question is: is EB the requirement, or is the joint performance the requirement and EB the way the OEM happened to achieve it?
If the drawing states joint acceptance criteria (unbonded area by ultrasonic test, tensile or shear minimum, shank hardness after welding, metallographic requirements), then a friction-welded joint that meets all of them is technically equivalent, and many OEMs will accept it after a first-article qualification with sectioned samples. Some will not, because the part sits inside a type-tested assembly and their internal process says requalification. That is their decision to make.
If the drawing states only "EB welded" with no joint criteria, the specification is incomplete for any supplier, EB-capable or not. Ask for the criteria.
Our own practice: when a drawing specifies EB welding, we state at the quotation stage whether the part can be produced on an EB route for that size and geometry, or whether we are proposing friction welding as an equivalent, with the supporting sample plan. We do not substitute a joining process without written agreement. The OEM's drawing is the contract, and a joint made by a different process than specified is a nonconformance even if it tests well.
Summary
Copper tungsten can be EB welded, and for CuCr-shanked HV arcing contact rods it is a legitimate specification. It is also a niche process with a small supplier base, sensitive to CuW grade, porosity and preheat. Friction welding covers most of the same ground for round parts and is far more widely available. Cast-on covers the pure-copper-shank case with no welding at all.
If you are sourcing to a drawing that says EB, get the joint acceptance criteria and ask suppliers to quote against those. If you are writing the drawing, consider stating the joint requirements and listing the acceptable processes rather than naming one.
Related Reading
- Cast-On vs Brazing vs Friction Welding — Joining methods compared
- Brazing Copper Tungsten to Copper — Filler and process notes
- How to Specify Copper Tungsten Contacts on a Drawing — Joint callouts, grade, density, hardness
- CuW Static Arc Contact — Made-to-drawing fixed arcing contacts