Electron beam welding & radiation protection. Do you need a consent?

Do you need a consent?

Before looking at the potential radiation risks and exposure potential from operating an electron beam welder, and comparing with other radiation generators (e.g. an airport x-ray cabinet security system), we need to tackle the issue of registrations and consents - which are required in particular circumstances due to the Ionising Radiations Regulations 2017 (IRR17) in the UK. 

We have already provided some guidance such as the following (will all open in a new tab). 

• IRR17 (5) - Notification of certain work
• IRR17 (6) – Registration of certain practices
• IRR17 (7) – Consent to carry out specified practices
• X-ray cabinet sterilisation (irradiation) – UK registration or consent (IRR17)? [blog article]
• New UK Consent process for users of Ionising Radiation (as of February 2023) (Updated January 2025) [blog article]

Notwithstanding the above guidance and comment, we can summarise the requirements as follows:

• Notification of certain work - lowest level of radiological risk (e.g. for limited quantities of radioactive materials, occupational exposure to radon).
• Registration of a certain practice  - medium level radiological  risk (e.g. using a radiation generator such as an airport security x-ray machine).
• Consent to carry out specified practices - highest level of radiological risk (e.g. operation of an accelerator, industrial irradiation, use of high activity sealed sources etc).

The above is intended to represent the graded approach to radiation protection as outlined in IRR17, and also the likely degree of regulation required from HSE. 

[Ionactive comment: There will hopefully be a 10 year review of IRR17 during 2027 (it is desperately needed) and we think the graded approach will need some significant improvement. For example, notification (lowest risk level) is required for occupational exposure to radon gas (where radon concentrations are > 300 Bq/m³). Yet radon gas probably yields the highest occupational exposures to any group of workers (or other persons) in the UK, whilst also being the most undeclared exposure of all - partly due to ignorance. Radon exposure can be serious  - take a look at the following Ionactive blog article (will open in a new tab): Radon gas in schools (and other workplaces). The exposures derived in this blog article are way above those received by most workers who knowingly work with ionising radiation in the UK. ]

With the consent  - what matters is that a particular specified practice is taking place, the type of ionising radiation source matters not. So industrial irradiation could use a radiation generator, an accelerator or radioactive sources - the relevant point is only that a specified practice is taking place. 

The definition of industrial irradiation is found in IRR17 (2) and states specifically 'the use of ionising radiation to sterilise, process or alter the structure of products or materials'. It matters not what the equipment is, the physical aspects of the work, the shielding employed, the surface dose rates in normal use, whether a radiation accident is reasonably foreseeable, the output of a radiation risk assessment required by IRR17 (8). All that matters is that a specified practice is taking place. 

It is true that some forms of industrial irradiation (such as large scale industrial sterilisation plants) carry significant radiation hazards and have massive radiation protection infrastructure. The relevant plants will have walk-in irradiation cells (e.g. for maintenance access) and require multiplicative and diverse safety systems, since they are capable of delivering say 50 kGy to items. But as already outlined, industrial irradiation cares not what the source of ionising radiation is  - there is simply an assumption (?!) that in the worse reasonably foreseeable case, it could yield high unplanned radiation exposures. 

So now we come to electron beam welding. A radiation generator (electron beam  / secondary x-rays from bremsstrahlung) is used to alter the structure of a material (there can be little doubt about that). Therefore, under IRR17, electron beam welding is a form of industrial irradiation (a specified practiced) and requires a consent.  The type of equipment, the kV, the mA, the surface dose rate, the output of the risk assessment is not relevant when making this declaration.

So - do you need a consent? If you do not have a consent you should not be operating an electron beam welder in the UK.

This is not just an Ionactive interpretation, the HSE state this clearly here:  Work with ionising radiation that requires consent (will open in a new tab). Under examples for industrial irradiation is clearly states: 'This applies where you: work with a particle accelerator to perform ion implantation or electron beam welding'. 

Now we will leave you to debate this with the HSE as their wording is not particularly helpful here. For example. 

• Accelerator IS defined in IRR17 (2) as 'apparatus or installation in which particles are accelerated and which emits ionising radiation with an energy higher than 1 MeV'. However, particle accelerator (which the regulator uses above) is NOT defined in IRR17. Is there a difference?

Please don't get your hope up though - further up the paragraph of the above link, the following is stated:

You must apply for consent if you perform industrial irradiation using:

• radioactive sources (for example HASS)

• accelerators

• radiation generators

There is no doubt that an electron beam welder is a radiation generator. IRR17 (2) defines a radiation generator as: 'a device capable of generating ionising radiation such as x-rays, neutrons, electrons or other charged particles'. 

In case the above linked text is changed for whatever reason, the graphic below shows the website wording as of October 2025.

Ionising radiation generated when operating an electron beam welder

Before considering the implications of needing a consent, and in support of the above risk comparison, we will delve a little deeper into the ionising radiation generated within a typical electron beam welder. This is not a wasted exercise either, since a safety assessment (SA) required to apply for a consent, would certainly be enhanced using this type of analysis. 

We are going to assume the following parameters/ settings:

• 150 kV acceleration (and also 60 kV).
• 40mA current  (and also 60 mA at 60 kV).
• 6 kW power / 3. 6kW.
• Vacuum chamber (parts for welding are held in place within a vacuum chamber).

The above figures are for a pretty significant electron beam welding system, but the lower power systems (e.g. 60 kV at 60mA = 3.6 kW) will also be considered. 

Being exposed to a 150 keV electron beam of this power would not be advisable (after all, it is being used to weld metal!). Such exposures are not reasonably foreseeable - for the vast majority of cases it takes place in a vacuum vessel.  In addition, electron beams are poorly penetrating (i.e. easily shielded) so are not the dominant ionising radiation hazard in such equipment.  We will therefore concentrate on the bremsstrahlung x-ray component, as this radiation will be far more penetrating and will need to be shielded. 

We will borrow some resource from an Ionactive article (written for us by Dr Chris Robbins): Calculate an estimate of x-ray dose rate from an x-ray tube given kV and mA (opens in a new tab), but note we are not suggesting that the electron beam welder uses an x-ray tube (it does not). 

The electron beam welder is not a particularly efficient producer of x-rays (it is not optimised for this purpose). The bremsstrahlung x-ray fraction of the total energy input can be approximated by as \[ F_e = 1 \times 10^{-6} Z E \] where:

• E is the energy in keV.
• Z is the atomic number.

The table below shows the atomic number (Z) for a few materials that may undergo electron beam welding, as compared to tungsten (W) which is the most common anode material used in industrial x-ray machines. The value F_e (bremsstrahlung x-rays) is also given. The table clearly shows that whilst x-ray generation from an electron beam (for the energies considered) is not that efficient for an x-ray tube, it is much less efficient for electron beam welding.

The above table does not take into account characteristic x-rays which are less important for this analysis. 

It can be shown (gross approximation), that the dose rate in Gy/h can be estimated if we know the current, voltage and target material (being welded - not the anode) using the following expression.  \[ D_{Gy hr^{-1}} = \frac{9 \times 10^{-7} \times I V^2 Z}{\pi r^2} \left( { \frac {\mu}{\rho}} \right) \]

The mass energy attenuation coefficient \( ({\mu}/{\rho}) \) is in units of \( m^2 kg^{-1}\), I is in amps, V is in volts and r is in meters (from the welding object). The expression is derived in the article linked above (the derivation will not be considered further here). For simplicity, in this assessment we assume average energy occurs at approximately 1/3 peak energy and is 57.3 keV for 150 kVp. If we calculate the dose rate beyond the confines of the vacuum (ignoring the shielding provide by the chamber for the moment) then we can estimate a dose rate in air using an air absorption coefficient of \(3.29 \times 10^{-3} \space m^{2}kg^{-1} \). 

The above expression is then used to construct the following table which provides an estimated dose rate in Gy/h at 1m for the different target materials. [Ionactive comment: each target material will significantly alter the x-ray spectrum which should be considered if this analysis was a formal assessment.  However, all we are looking for is some reasonable estimation which provides an order of magnitude likely dose rate which we can then consider further by adding some shielding. This simple analysis is also used to demonstrate that the electron beam welder is a radiation generator - even though this is not the intended purpose. 

Indicative absorbed dose rates (in air) are given below for each material (150 kV / 40mA). 

For comparison, indicative dose rates are given for the same materials at 60kV and 60 mA as shown in the table below. 

[Ionactive comment: It may surprise you that the dose rates in air at 60 kV and 150 kV appear similar?! This is true, but rather academic once shielding is applied as noted shortly below. At 60 kV the average energy of the bremsstrahlung x-rays will be about 30 keV and here the mass energy attenuation coefficient in air rises considerably  - not surprising if you think about.  The data above will shortly be combined with shielding and the headline dose rates will be mitigated swiftly, especially those from the 60 kV scenario. If we were trying to measure these dose rates (if that were even possible), we might want to look at tissue equivalent doses which we would modify to yield effective whole body dose rate. For the purposes of this article we are just comparing values, so absorbed dose in air is fine.]