CE marking extended for UK, UKCA redundant for electronics products

The government have introduced legislation [1] that indefinitely recognises CE marking for UK market access. This means that applying the UKCA mark for electronics products is no longer required.

This extension comes into force on 1st October 2024, before the (most recently) proposed UKCA introduction date of 31 December 2024

Specific Regulations

This has been done for a number of regulations. The ones that we are interested in (along with their EU counterpart laws) are:

EMC The Electromagnetic Compatibility Regulations 2016 EMC Directive 2014/30/EU
Electrical Safety The Electrical Equipment (Safety) Regulations 2016 Low Voltage Directive 3014/35/EU
RoHS The Restriction of the Use of Certain Hazardous Substances in Electrical and Electronic Equipment Regulations 2012 RoHS Directive 2011/65/EU
Radio Equipment The Radio Equipment Regulations 2017 Radio Equipment Directive 2014/53/EU
Machinery The Supply of Machinery (Safety) Regulations 2008 Machinery Directive 2006/42/EC
ATEX The Equipment and Protective Systems Intended for Use in Potentially Explosive Atmospheres Regulations 2016 ATEX Directive 2014/34/EU


What does it not apply to?

UKCA marking will still be required for the other equipment including Medical Devices and Marine Equipment.  See the page on the UK government website for more details.

Essentially the government departments in charge of these regulations have not yet decided whether or not.


Amendment Wording

What does the amendment do? The Explanatory Note at the bottom of the regulation makes it clearer

The amendments made to each of these pieces of legislation revoke provision that sets out the expiry of provisions which themselves allow obligations in the legislation as it applies in Great Britain to be met by complying with requirements of the corresponding EU law…

Meaning: the expiry date of CE marking is withdrawn

…they replace this provision with provision allowing relevant economic operators to meet any conformity assessment or testing requirements in the legislation as it applies in Great Britain by complying with the corresponding EU conformity assessment or testing requirements.

Meaning: CE marking is acceptable for placing products on the market in the UK


Option for continuing to apply UKCA

If you want to keep applying the UKCA mark then you can do this. One reason the UK government website [2] gives for continuing with UKCA marking is:

This is designed to provide longer-term certainty and flexibility for businesses in case the UK mandates UKCA for certain regulations in the future.

If you wanted to try and bet on the future, given how much of a success UKCA has been so far, then you are more than welcome to do so.

Their is a “fast track” available for equipment compliant with CE marking requirements whereby the manufacturer:

  1. Ensure conformity with CE marking Essential Requirements
  2. Affix the UKCA marking
  3. Draw up UK Declaration of Conformity and list compliance with EU regulations

There are future plans for the UKCA mark to have the option to be applied separately to the main label or in the accompanying documentation (either by the manufacturer or the importer), or to use digital labelling. These are supposed to be coming later in 2024 but with a general election looming, let’s all pretend to be surprised when this gets delayed.


Effect on manufacturers?

All the hard work you have spent updating documentation, labelling, technical Documentation, producing UK Declarations of Conformity to support the application of UKCA marking in parallel to CE marking has been regrettably rendered rather redundant (wasted).

On the plus side, your paperwork burden just significantly reduced for the future!


What are we doing at Unit 3 Compliance?

From now on, our Quotations for testing for the UK market will only make reference to CE marking, EU Directives (the right hand column above), and Harmonised Standards.

We are working on the assumption that most of our customers want to minimise their costs, and minimise the amount of time spent creating paperwork and labelling. Instead, you’d rather be designing your next product, managing the production, answering emails, etc!

Any questions, please get in touch – hello@unit3compliance.co.uk



[1] SI 2024 No. 696 “The Product Safety and Metrology etc. (Amendment) Regulations 2024”

[2] Using the UKCA marking – gov.uk

Problems Observed During EFT Testing of DC Power Port Equipment

(Special guest blog from Lawrence)

Just in case achieving a pass during EFT testing wasn’t tricky enough, how about throwing in an additional hurdle…the added inductance and capacitance the transient generator CDN applies to the DC power lines.

A customer provided us with the opportunity to debug an EFT issue seen during the original round of testing at another laboratory. The failure mode was pretty obvious and a real showstopper, the EUT will crash/ lock up/ stop functioning and a full power off and on cycle is required to establish a normal operating condition.

Sure enough, on arrival at Unit 3 Compliance the failure mode was repeated, however, the failure immediately stood out as being a little odd because it occurred before the actual EFT test had begun. This was certainly a new phenomenon for me and something I haven’t encountered before. The failure certainly didn’t feel like a proper EFT failure and something else was afoot. To figure out what is changing during the generators initialisation stage EN61000-4-4:2012, section 6.3.1 provides some valuable information and defines the internal make up of the EFT generator:

The 33nF capacitors are switched into the circuit as soon as the “start test” button is pressed on the transient generator; this is the very first thing the generator does during its initialisation stage, before any transients are applied. Charging of the added capacitance requires a gulp of energy that must be sourced from either the DC power supply or the EUT. The capacitors don’t care where the energy comes from, they just want charging right now.

So, what’s the best source of energy? The DC power supply would be great, however, energy is going to be supplied relatively slowly because of the >100uH inductors in the decoupling section, so that leaves the EUT to supply the energy. If the EUT only has a small input capacitance then this will cause a drop in voltage across the input of the EUT. If the EUT has some kind of Under Voltage Lock Out (UVLO) circuit then there’s a very good chance the equipment will fall over in some fashion!

At first glance this failure will present itself as an EFT issue, but this failure wasn’t caused by a transient voltage.

Adding sufficient de-coupling capacitance at the input of the EUT provided a good enough reservoir of charge that can be consumed in this test situation. More importantly it isn’t beyond the realms of possibility that the EUT will face a real world installation that is similar to that of the EFT test set up (long, inductive power supply lines), so having sufficient de-coupling will go some way to help with the smooth installation and compatibility of the EUT.

With the larger capacitance in place we tested this EUT up to 2kV (against a specification of 1kV) and it performed fine.



  • Make sure the input to your power supply has enough decoupling capacitance to account for inductive supply lines
  • Make sure you correctly classify your input power supply port – is it a DC Power Port or not?



Pocket Near Field Probe Boards (R5) – Now with 100% more layers!

Hello and welcome to Unit 3 Compliance.

We test your products for EMC and Electrical Safety, fix them if they don’t pass, and help you with all that boring standards and documentation stuff.

We also provide customised design for EMC training via Think EMC.

Hot off the press are the revision 5 of our Pocket Near Field Probe kits. Better than revision 4? “How is that possible?” I hear you ask…

New Features

  • Now on a 4 layer PCB (100% more layers!)
  • All sensing and signal traces are enclosed within ground plane sections on the outside of the PCB – preventing pickup from stray fields and reduces effects of fingers
  • Removed snap off sections
  • Better silkscreen markings
  • More formulae! (ok, just 3)

Download our Guide

But first things first – if you are new to Near Field Probing then you simply must check out our free eBook that has everything you needed to know (and probably a lot that you didn’t) about the art of near field probing to help solve EMC problems.

* Click here to download *

Assembly Instructions

Refer to the instructions in our Revision 4 guide. It’s exactly the same, only we’ve removed the snap off sections.

We hope you enjoy using it!


Schaffner NSG 2070 Programming Manual

This workhorse has been a part of our lab for years. Used for generating conducted RF immunity test signals for EN 61000-4-3 and ISO 11452-4

Whilst this is one from the archives, I wanted to upload this to the internet in case there was another, just like me, who was searching for such a thing.

Yes, it is written in German but still usable.

Download Schaffner NSG2070 Programming Manual

EMC radiated emissions problems from Riverdi LCD panel

Lawrence has been working on an Industrial (emissions Class A) pump product and has fixed an interesting EMC radiated emissions problem associated with the display panel.

Narrowband emissions caused by harmonics of a digital clock are nothing new when it comes to LCD panels and the signals that drive them. Inadequate filtering of these digital interfaces means that there is plenty of high frequency energy available to excite any parasitic antenna structure that we have created with these interconnected elements.

Normally the source of the emissions are the digital video signals from the CPU to the display. In this case, near field probing showed that the source of the noise was strongest around the display controller itself.

The first radiated emissions scan showed that there were emissions of a 60MHz clock signal present at around 300MHz and 660MHz (green = chamber background, red = measurement)

Here’s the display in question, a Riverdi RVT43 series touch screen for the Human Machine Interface (HMI) of the pump.

Lawrence took a closer look at the PCB with a contact probe and a spectrum analyser to find there was a Winbond QSPI flash running at around 60MHz.

Measuring each of the pins of the display cable (on the quad SMT resistor packs) showed significant levels of 60MHz noise on each trace, with the bottom most ones being the worst.

Looking at the circuit board we can clearly see where the problem lies:

We have a 60MHz clock trace (in red) running immediately adjacent to one of the signal traces leaving the PCB on an unshielded ribbon cable.

This is poor design practice on the part of the display manufacturer.

There is enough coupling between these traces (inductive and capacitive) for noise on the clock trace (source) to appear on the cable signal trace (victim). If we consider the fields around the traces in cross section through the PCB…

There is also the possibility of coupling inside the IC given that the output pins are immediately adjacent.

We can address the problem a number of ways

  1. Remove the noise at source by changing the PCB = not possible, this is a third party component. Changing the value of the series resistor packs could have worked in this case.
  2. Detune the antenna structure by changing the relationship of the two PCBs relative to each other = difficult to achieve, not guaranteed to work
  3. Adding common mode suppression on the cable to reduce the energy of the signal driving the parasitic antenna

Point #3 above was the only practical course of action.

A small ferrite core was placed over the cable and the results speak for themselves (green = before, red = after)

There is no compliance / EMC information / CE marking present on the display. But surely, as a piece of equipment or sub-assembly capable of causing radio interference, this falls under the scope of the EMC Directive and should be CE marked and assessed for EMC?

Yes, it should.

In this case, the host equipment manufacturer is having to deal with the poor hardware implementation by the display manufacturer.

Good work by Lawrence for getting to the bottom of this EMC problem.

Pocket Near Field Probe Boards (R4)

Hello and welcome to Unit 3 Compliance.

We test your products for EMC and Electrical Safety, and fix them if they don’t pass.

We also provide customised EMC training via Think EMC.

If you have picked up one of our new Pocket Near Field Probe kits from us then, oh my goodness, we hope you like it as it is one of our best yet. We’ve added cool new features and some useful EMC reference formulae as well as the near field tools from the original.

Download our Guide

But first things first – if you are new to Near Field Probing then you simply must check out our free eBook that has everything you needed to know (and probably a lot that you didn’t) about the art of near field probing to help solve EMC problems.

* Click here to download it *


What’s In This Probe?

Get started assembling by snapping off the two parts shown.

The first part is our contact details and, on the back, a quick reference of useful EMC formulae.

The bit from the top is a new addition: an SMA spanner (careful with that torque Eugene) and a key for opening Wurth Elektronik snap on ferrite cores.

You can then assemble the rest as described below.


Assembly Guide

Components required:

0805 resistors (10:1 voltage probe)

R1 = 470R and R2 = 10k, gives a 450 ohm parallel combination. This is required for the 10:1 divider into a 50 ohm input.

0805 capacitor (10:1 voltage probe)

C1 = 220pF to 2.2nF, C0G/NPO dielectric, 50V

This AC couples the signal from the probe into the measuring instrument. Many spectrum analysers have a 0VDC requirement on their RF input.

The actual value isn’t too important at the frequencies we are interested in.

Metal pin for contact probe

For the contact probe pin we recommend using a male pin header from a 2.54mm pitch connector strip and filing the end to sharp point with a metal file.

These parts are available from all major component distributors and from eBay again.

We also recommend securing this in place with a generous blob of 2-part epoxy adhesive to provide some additional mechanical strength.

Edge mount RF connectors

The 1.6mm thick PCB is designed to take SMA edge mount female connectors to connect to an SMA-SMA RF cable.

These connectors are available from a wide range of sources but most readily and cheaply from eBay

You can solder a coaxial cable onto the pads at the expense of ease of swapping the connection between probes.

Ferrite cores (current transformer adaptor)

The current transformer adaptor was designed to use the excellent Wurth Elektronik 742715x series of ferrite cores which are available in different sizes.

If you do get one, tie the two-pronged plastic unlocking key to the probe with a bit of string so that you don’t lose it!

An M3 screw and nut will secure the core to the board, but again a blob of two-part epoxy adhesive or some hot melt glue will do the trick.

Experiment with the number of turns but we generally use between 5 and 10 as a starting point.


Finished Product

Here are our probes that we use in the lab for all kinds of EMC problem solving:


And if you really know what you are doing with these probe boards, there’s a secret power user Easter egg mode available.

Tea up!


Mug is from Mike’s Electric Stuff




Do I need to EMC test a pre-approved power supply? – EMC Explained

One of the most common questions we get asked when we send an EMC Test Plan / quotation to our customers is along the lines of:


“Our equipment is powered from a pre-approved CE marked power supply so we don’t need to do any AC mains EMC testing… right?”


If a power supply has already been EMC tested (if it has a CE or UKCA mark you would hope that this was the case) then it is a fair question – why should we retest it?

Adding AC mains specific tests into the EMC Test Plan adds time and therefore cost, something that some of our customers would like to avoid. For smaller businesses, the cost of assessment for EMC might be one of the largest external costs incurred on a project.

The main assumption driving this question is that EMC emissions – the noise that is coming out of the power supply and either back onto the AC mains or radiated from the power supply – is the only EMC problem we have to worry about. It’s the main one, but not the only one.

The pre-approved power supply will have been tested for immunity, but only the immunity performance of the power supply itself, not the equipment that it is powering.

Some noise will get through the power supply and into the equipment being powered. How does your product respond to this noise?

Also, how low are the AC mains conducted emissions from the power supply? Have you seen a test report? How reputable is the vendor?

Testing is the most reliable way to find out.


Our Recommendations

We generally recommend to our customers that they perform all of the applicable tests to the product.

(What, a test lab recommending testing? I’m shocked!).

Firstly, the tests are called up in the EMC standards, and for CE/UKCA marking, testing to a Harmonised Standard gets you a “Presumption of Conformity” to the requirements of the Directives – a pass without any further Risk Assessment or justification on your part.

Deciding not to perform the testing puts the responsibility on you to assess the remaining EMC risks. If you needed us to do this assessment for you or advise on it, the cost of a few hours of consultancy time would be equivalent to just doing the tests in the first place.

Secondly, EMC performance is often dictated by parasitic capacitances and inductances, component values that are not on the datasheet or intentionally designed into the product. Even knowing their magnitude does not give a good understanding of how they will interact. Testing allows us to measure their interaction under standardized conditions.



Risk Assessment Factors

As discussed above, our recommendation is always to perform testing on applicable ports, the AC mains port included.

If you are worried about costs or time taken for testing, then you might decide to omit some of the specific tests. The below table outlines some of the factors you may wish to consider when making this decision.

The more items that apply from the Risk Increasing Factors column, the less strong your argument becomes for not carrying out testing.


Risk Reducing Factors Risk Increasing Factors
Class II power supply (un-earthed)


Class I power supply (earthed)

Especially if the DC negative of the power supply output is connected to Protective Earth in the system.

Power supply comes from reputable vendor (e.g. Meanwell, XP Power, Recom, Traco, TDK Lambda, Puls, etc) Power supply comes from cheap or from far east supplier
Power supply external to product Power supply internal to product
No analogue or sensitive circuitry Analogue circuitry e.g. audio, 0-10V I/O, 4-20mA I/O

Sensitive, low level signals e.g. thermocouple, RTD

No other long (>3m) cables connected to equipment One or more long (>3m) cables connected to equipment
Main use in Basic (residential, commercial) EM environment Flexible use, could be used in Light Industrial or Industrial EM environments


If you are at all unsure then you should test the AC mains port with your intended production power supply.

For the ultimate in performance, or if the equipment is for flexible use (could be powered from an AC/DC supply or from a distributed DC power supply) then we would recommend treating the DC power input to your product as a signal port with a length greater than 3m.

This would then call up Conducted RF Immunity (EN 61000-4-6) and Electrical Fast Transient (EFT, EN 61000-4-4) testing to the power port at the appropriate levels for the end EM environment (e.g. Basic or Industrial)

One step further would be to apply line-to-line and line-to-earth surges to the DC input, assuming that the design already contains a transient surge voltage suppressing element like a TVS diode or an MOV.

Let’s take a look at some of the technical justification behind the selection of these items.


AC Mains Port vs DC Power Port

If you typically derive your equipment power from an AC mains power supply, then it is unlikely that you will fall under the DC Power Port classification.

The term DC Power Port in EMC terms means a very specific classification of port. We discuss this in some length in this article.


Power supplies do not always meet the regulations

A scenario that we have experienced on several occasions: the power supplies that end up with our customers or in our test lab are not the same as the ones in the manufacturer supplied EMC test report.


Another customer had similar problems on  power supply that they had received samples of in that the EMC performance varied wildly. In this case the clue was that the weight of the two samples was significantly different.



These power supplies were almost identical on the outside but significantly different on the inside. Same manufacturer and model number, different components. Imagine the conversation:

“I’d like to order some HM-A132 power supplies please”

“Certainly sir, which ones?”



This is mostly related to cheaper power supplies sourced from China. We often see significantly different results to those shown in the manufacturer test report.

The worrying thing is if changes like this are being made on the basis of EMC, what changes are being made that affect Electrical Safety that are going unchecked? We can check that for you as well.


Cable Routing

If your power supply is integrated into your equipment then there is the possibility of noise on the AC mains cable coupling onto other nearby cables.

It is also possible for noise to couple (both to and from) components connected to the AC mains and internal system components. This could be an emissions (noise getting out) or an immunity (noise getting in) risk.


This is particularly likely if you are using slotted trunking and mixing AC mains cabling in with other cables.




This is less important for an external power supply like a laptop type charger or a plug top power supply as the AC mains cable remains outside of the equipment enclosure.



Power Supply Common Mode Impedance

Electrical noise inevitably gets coupled onto the AC mains bus. Normally this noise is coupled onto the AC mains Common Mode. This means all the lines together in relation to a high frequency “ground” reference plane.

The noise current through the power supply and equipment will flow something like this:

The noise reaching the equipment will have been attenuated by the Common Mode impedance of the power supply and the currents diverted through the parasitic capacitance of the power supply relative to the HF ground reference plane used in the tests.

Crucially, some noise still gets through to the power supply and will flow through the product. The magnitude of this current can be estimated or measured but relies on electrical parameters that are not on the power supply datasheet.

It is this noise current that we are interested in. How does it affect your product? The only way to find out is to perform testing.


Class I vs Class II Power Supply CM Impedance

The construction of a typical switch mode AC/DC power supply is broadly similar across a wide range of topologies. One of the main EMC variations results from if the power supply is Class I (earthed) or Class II (unearthed).


Class II

A Class II power supply relies on Double or Reinforced insulation between Live parts and user accessible secondary low voltage parts for Electrical Safety. There is no connection to Protective Earth. This kind of power supply is usually identifiable by:

  • the square-in-a-square double insulation symbol (IEC 60417 symbol # 5172)
  • a plastic earth pin on a UK mains plug (technical name is an ISOD or Insulated Shutter Opening Device)
  • An IEC C8 “figure-8” AC mains inlet socket with just two pins


Looking at the typical internal structure of a Class II AC/DC SMPS we can see that the components providing Common Mode noise attenuation are

  • the inductive common mode filter (Lcm)
  • the components across the safety isolation barrier, transformer Tx and class Y capacitor Cy

The value of parasitic parallel capacitance of the choke or transformer (or wanted series capacitance of Cy) will reduce the impedance ( Xc = 1 / [ 2 * pi * f * C ] ) and allow more noise current to flow at higher frequencies.

This capacitance is usually a low value to prevent too high a touch / leakage current to flow which would compromise Electrical Safety.

However, at EMC frequencies of MHz and higher this presents a much lower impedance allowing noise currents to flow through the cable.


block diagram of a class II power supply showing EMC immunity noise current through the power supply


Because current always flows in a loop, and because current always returns to the source, to close this common mode current loop we need to have return currents flowing. We usually think of these coupling capacitively onto a nearby metallic element like a nearby metal structure.

In the test lab we simulate this with a nearby metal plate but in real life this could take a number of forms (building steelwork, conductive cable trays, other wiring).



Class I (Or Class II with Functional Earth) Power Supplies

With a Class I power supply, the Protective Earth is connected to accessible metalwork for Electrical Safety reasons (prevention of electric shock). Basic insulation (or higher) is required between the live parts and user accessible secondary parts.

Possibly the protective Earth is also connected to DC negative somewhere in the system as well.

A Class II with Functional Earth power supply is similar from an EMC point of view but very different from an Electrical Safety point of view. In this case, the Earth is connected for functional reasons (reducing noise or EMC emissions) but the power supply still relies on Double or Reinforced insulation for safety.

This isn’t a very common power supply topology choice, so I was surprised to see it marked on my laptop charger power supply.



In both cases, when we apply common mode noise to the AC mains input (L+N+E) then the Protective Earth conductor allows the noise to bypass the common mode impedance of the power supply. It is for this reason that we view the use of a Class I earthed power supply as a higher EMC risk for immunity reasons.


block diagram of a class I power supply showing EMC immunity noise current through the power supply



How this noise couples into the rest of the equipment, its magnitude, and how it affects it depends massively on the construction of the equipment. Again, testing is the best way to determine this.



Power supplies and the equipment they power are not perfect and can have varying EMC performance depending on how you connect them and how the equipment is designed.

It isn’t always easy to estimate how likely EMC issues are, even for experienced engineers and problem like us at Unit 3 Compliance. It is for this reason that we would always recommend testing to characterize the unknown EMC performance.

If you do decide to omit some testing, then the Risk Reducing or Increasing Factors above should help with that decision.

Again, we hope that this guide was useful to you in some way. Get in touch with us if you have any thoughts, questions, observations, or (obviously) a need for EMC or Electrical Safety testing.

All the best!




sketch showing dc power distributed around a building on busbars to a vriety of loads, and with a battery bank. There is an AC/DC charger for the batteries.

What is a DC Power Port? – EMC Explained

Everyone knows what a DC power port is, right? It’s this…

sketch showing an ac/dc adaptor and a piece of equipment with a dc power input - this is classified as a signal port for emc purposes

It’s got DC power on it, and it is a port on the equipment. DC. Power. Port.

Not in the context of EMC I’m afraid. Despite the similar name, the EMC definition for a DC Power Port (from the IEC / EN standards) is very different.

The DC Power Port is unfortunately mis-named. A better term would be “DC Mains Port” to indicate how similar it is in construction and EMC requirements to its counterpart “AC Mains Port”.

In this guide we will refer to it in this guide as a DC power/mains port and look at:

  • The EMC definition of a “DC Power Port”
  • The EMC implications of classifying a port as a “DC Power Port”
  • Examples of a DC Power/Mains Port
  • Examples of NOT a DC Power/Mains Port

Any port that doesn’t meet ALL of the definitions of a DC Power Port is just classed as a Signal Port, albeit one that happens to carry DC power.

Those key parameters are:

Criteria Met?
Local supply in a site / building / infrastructure? ???
Flexible use by different types of equipment? ???
Supply independent from AC mains? ???



The definitions in the Generic EMC standards of EN 61000-6-1 (immunity) and EN 61000-6-3 (emissions) lays out what a DC Power/Mains Port is:


EN 61000-6-3:2007+A1:2011, Clause 3.8

“d.c. power network

local electricity supply network in the infrastructure of a certain site or building intended for flexible use by one or more different types of equipment and guaranteeing continuous power supply independently from the conditions of the public mains network

NOTE Connection to a remote local battery is not regarded as a DC power network, if such a link comprises only power supply for a single piece of equipment.”


Let’s break out the key terms to understand the definition:


“…local electricity supply network in the infrastructure of a certain site or building…”


This suggests something wiring that is built into or spreads around a large area. A good example is the way that AC mains wiring is distributed around a building. Imagine this carrying DC instead of AC.

Typical cable lengths are probably around 10m or longer. Longer cables means they can act as antennae for low frequencies (longer wavelength). So we need to be concerned with power supply noise from our equipment on these cables that could radiated from them.

Longer cables will also pick up lower frequency common mode disturbances (conducted RF and surge) and present a larger surface for capacitive coupling of fast transients (EFT).


“…for flexible use by one or more different types of equipment…”


Use of the word flexible implies ease of use and simple connection to this power distribution system. Perhaps a common power connector (similar in nature to an AC mains plug) is used, or an agreed connector standard.

A DC Power/Mains bus that requires tools and time to connect to (example a fire alarm wired with Mineral Insulated Copper Clad (MICC) or “pyro” cable) might not meet the definition of “flexible” in terms of “ease of connection”. Nevertheless it would be flexible in terms of connection of different types of equipment (sounders, detectors, etc.)


“…guaranteeing continuous power supply independently from the conditions of the public mains network…”


The likely scenarios here are:

  • A “DC UPS” system where a bank of batteries are kept topped up by an AC mains charger
  • A DC micro-grid system where power is generated from sources like solar power



1) Any port that doesn’t meet ALL of these definitions is just classed as a Signal Port, albeit one that happens to carry DC power.

2) Any piece of equipment connecting to this DC power supply is classified as a “DC Power Port” regardless of whether it supplies or consumes the power


EMC Tests Required for a DC Mains/Power Port

The classification of a port as a DC Power/Mains Port invites extra EMC testing to be applied.


Port Length Conducted


EN 61000-4-4


EN 61000-4-6

Conducted RF

EN 61000-4-5


DC mains/power Any YES YES YES YES
Signal (with DC) <3m NO NO NO NO
Signal (with DC) >3m and <30m NO YES YES NO
Signal (with DC) >30m NO YES YES YES


Almost inevitably, unless the equipment has been explicitly designed as a DC Mains/Power port, there will likely be EMC test failures.

Conducted emissions invariably fails the limits. Usually the first system component after the input power connector are a series of DC/DC buck converters to change the input voltage down to levels that are needed in the system.

Buck converters suffer from noisy input nodes because of the high dI/dt requirements of the switching transistors. This needs to be mitigated through good quality high frequency decoupling and can cause noise at 20MHz upwards. Common mode chokes in the DC input may be required to mitigate this noise.

At lower frequencies, there will be current draw from the supply at the switching frequency of the DC/DC and at it’s harmonics. Unless low impedance electrolytics and a differential mode filter (usually an inductor in the 2.2uH to 10uH range forming a pi-filter) are used, the emissions from the port will fail the average limits in the 150kHz to 1MHz range.

DC Mains/Power also requires the addition of the surge test in both line-to-line (DC+ to DC-) and line-to-earth (DC+ and DC- together relative to Earth) coupling modes.

The line-to-line surge of 500V (commercial/light industrial EM environments) or 1kV (industrial EM environments) with a 2 ohm source impedance is capable of damaging the first switching transistor it comes across on the DC line unless a Transient Voltage Suppressor (TVS) is employed between DC+ and DC-.

The line-to-earth test with a series impedance of 42 ohms (not the 12 ohms as used for the AC mains port test) tests the insulation of any isolated power supply and depends heavily on how (or indeed if) a Protective Earth connection is made within the system.


Examples of A DC Power/Mains Port

The sketch below tries to capture a typical DC Mains/Power port application

sketch showing dc power distributed around a building on busbars to a vriety of loads, and with a battery bank. There is an AC/DC charger for the batteries.


Criteria Met?
Local supply in site / building / infrastructure? Yes
Flexible use by different types of equipment? Yes
Supply independent from AC mains Yes


Specific examples include:



48V distribution around telecoms switching / data centers to power the equipment and to provide low levels of power to handsets in a Plain Ordinary Telephone Service (POTS)


Computing Data Centres

Large data centre and cloud computing providers like Facebook, Microsoft, Google, and Amazon are moving away from traditional DC>AC UPS systems and towards DC power distribution (380V, 200V, 48V depending on standards) to servers and other electrical loads.

The efficiency savings from not having to convert from AC power to DC in every load, multiplied by the number of loads makes for significant energy efficiency savings and heat reduction – some of the biggest costs for such facilities.

In addition, the DC to AC conversion loss in the UPS from battery DC voltage to AC voltage is removed. Instead there are just the batteries connected to the DC power bus.


Electricity Substations

Battery Tripping Units (BTU) are used to power monitoring and control equipment in electricity substations. The LV AC mains supply to the substation equipment (derived from the HV or MV feed) is considered to be an “auxiliary” supply. Control of the equipment is a requirement even if this power is not present. Common DC voltages are 220V, 110V, 48V, 36V, 24V.


DC Micro-Grid

Local power generation from renewable sources like Solar PV might be distributed around a power generating plant or a local area.


Emergency Lighting Central Battery Units

There is a requirement in Building Regulations to have fire exit emergency lighting powered separately so that in the event of a power cut the building occupants can find their way out of the building safely.

In smaller buildings this is usually achieved using emergency lighting with independent battery backup. However in larger buildings, a Central Battery Unit is used to provide power (and often control / monitoring functionality) to emergency lights spread throughout the structure.

The combination of data and DC power blurs the lines between a DC mains/power port and a Wired Network port. Both call up conducted emissions tests and similar levels of immunity.


Fire Alarm System

DC power is passed to different critical components of the fire alarm system (e.g. smoke / fire detectors, displays, alarm sounders) in a loop system from a central control panel.

sketch showing the connection of fire alarm components to a central panel - emc dc power port example 2

Criteria Met?
Local supply in site / building / infrastructure? Yes
Flexible use by different types of equipment? Yes [1]
Supply independent from AC mains Yes


[1] May be difficult to connect to and reconfigure but certainly flexible in terms of variety of equipment that could be connected

Interestingly, the EMC product family standard that deals with fire, security, and social alarms (EN 50130-4) only focuses on emissions from the AC mains port with no mention of DC power outputs. Since other standards address EMC requirements for DC Power Ports, including the Generic EN 61000-6-x series mentioned above, we have a path to bring in these requirements to the EMC Test Plan as part of the EMC Risk Assessment.

If using MICC / pyro cable, whilst the joints are required to be fireproof, there is no requirement for quality of termination for EMC purposes. Reliance on the shielding formed by the outside of the cable is contingent on a low impedance electrical termination which is not necessarily guaranteed.




Examples of NOT DC Power/Mains Ports

AC/DC Power Adaptor

sketch showing an ac/dc adaptor and a piece of equipment with a dc power input - this is classified as a signal port for emc purposes

Criteria Met?
Local supply in site / building / infrastructure? No
Flexible use by different types of equipment? No
Supply independent from AC mains No


In this event, the power bus with long cables is the AC mains interface that our AC/DC power supply plugs into (for non-UK readers: that is a UK AC mains plug).

The AC mains has all the EMC characteristics discussed above: long cables that can radiate noise (emissions) or have noise coupled onto them.

One question we get a lot is along the lines of:

“My product is powered from a pre-approved / CE marked power supply, so we don’t need to do any EMC testing on it… right?”

We’ve written a separate article to cover this interesting question.


DC power distribution around a typical DIN rail electrical cabinet

sketch showing typical dc power distribution around a DIN rail equipped electrical cabinet - again this would be classed as a signal port

Criteria Met?
Local supply in site / building / infrastructure? No
Flexible use by different types of equipment? Yes
Supply independent from AC mains No


In this example, the Load represents the equipment we are interested in. There is the probability of noise coupling onto the DC power cable from other equipment inside this cabinet. For example a large industrial machine would typically have contactors and large Variable Frequency Drives running close by.

If we think this could be the case then we would recommend testing Conducted RF immunity (61000-4-6) and EFT (61000-4-4) regardless of the anticipated maximum length of power supply cable.

This would form part of the EMC Risk Assessment for the equipment, an important part of the decision-making process for what EMC tests to apply. If you’ve not considered EMC Risk Assessments before then get in touch with us and we can help!


Power over Ethernet (PoE)

sketch showing an example power over ethernet distribution - these are classed as Wired Network Ports under EN 55032


Criteria Met?
Local supply in site / building / infrastructure? Yes
Flexible use by different types of equipment? Yes
Supply independent from AC mains No [1]


[1] Depends on the power source for the switch, it could come from a UPS for no-interruption requirements like security or network infrastructure.

Supplying DC power over an Ethernet cable is a thoroughly good idea. High speed data, enough power to run a simple device, all over cables approaching 100m in length? Sounds great!

Each port in a PoE switch will have power provided from a dedicated isolated power supply. This provides isolation (both in terms of EMC emissions and immunity) between different segments of the PoE network.

Despite the potentially long cables, it still doesn’t quite meet our criteria for a DC power port. However similar EMC requirements for a DC power port are called up by other standards:

  • EN 55032 (emissions of multimedia equipment) calls up a requirement for conducted emissions on wired network ports
  • IEEE 802.3 specifies a voltage isolation between Ethernet cabling and the circuit at each end of 1500Vac. This will often help (but not completely resolve) with the surge requirements
  • The surge test of EN 61000-4-5 is not applied line-to-line as the Ethernet lines are considered to be “symmetrical” in the language of this Basic standard. The tight coupling between the pairs in the cable and floating / isolated nature of the signaling means that coupling onto these cables generating line-to-line surges is considered unlikely. Only line-to-earth surges are applied.


Daisy chain of DC powered devices all running from the same bus

sketch showing a daisy chained series of DC powered loads - classified as a signal port

Criteria Met?
Local supply in site / building / infrastructure? No
Flexible use by different types of equipment? Yes
Supply independent from AC mains No




Hopefully this guide has cleared up some of the confusion about DC power ports in the context of EMC.

If you are unsure about whether your equipment falls into this classification then you can always contact us if you need help.

We generally advise that if you aren’t sure if your equipment could be used in this fashion then you should design and test your product as if they do apply. It is easier to “not-fit” or link out unwanted components than to try and add them in later.


Translating Product Documentation into Another Language

From one of our customers

We have a question as to whether there is a legal requirement to supply documentation in a local language, such as French, if supplying equipment into France.

This is not a question specifically for France, but supplying equipment to any other country.

If the answer is yes, what documentation is required to be translated?

I hope you can help.

The New Legislative Framework EU Directives (e.g. Radio Equipment, EMC, Low Voltage) have similar wording regarding translation of instructions and information into the language of the EU Member State

EMC Directive

Obligations of manufacturers

7. Manufacturers shall ensure that the apparatus is accompanied by instructions and the information referred to in Article 18 in a language which can be easily understood by consumers and other end-users, as determined by the Member State concerned. Such instructions and information, as well as any labelling, shall be clear, understandable and intelligible.

Blue Guide page 36 (pdf page 38)

4. accompany the product with instructions and safety information (110) (111) as required by the applicable Union harmonisation legislation (
112), in a language easily understood by consumers and other end-users, as determined by the Member State concerned (113). Unless otherwise specified in specific legislation, instructions and safety information need to be provided (114), whether the product is intended for consumers or other end-users.

This should include all the necessary information for the safe use of the product, to enable the consumer to assemble, install, operate, store, maintain, and dispose of the product. Instructions for assembly or installation should include the inventory parts and special skills or tools. Instructions on operation should include information for restriction of use, need for personal protective equipment, maintenance and cleaning or repair. It is for the manufacturer to determine the relevant information which should be included in the instructions and safety information for a particular product.

Manufacturers have to look beyond what they consider the intended use of a product and place themselves in the position of the average user of a particular product and envisage in what way they would reasonably consider to use the product. Furthermore, a tool designed and intended to be used by professionals only might also be used by nonprofessionals, the design and instructions accompanied must take this possibility into account. Instructions and safety information must be clear, understandable and intelligible;


So yes, information should be translated into the local language, especially where it conveys information critical to compliance with Directives or Laws. I would list this as at least:

  • EU Declaration of Conformity
  • Product manual, instructions, quick start guide
  • Any text on safety specific labels e.g. warning, hot. This is where IEC defined symbols such as warning triangles come in handy
  • Product label

The Regulatory Technical Documentation (Technical File) would be exempt from this requirement. This document is not meant for users / customers of the product. Rather it is for market enforcement authorities only.

EMC Directive

9.   Manufacturers shall, further to a reasoned request from a competent national authority, provide it with all the information and documentation in paper or electronic form necessary to demonstrate the conformity of the apparatus with this Directive, in a language which can be easily understood by that authority. They shall cooperate with that authority, at its request, on any action taken to eliminate the risks posed by apparatus which they have placed on the market.

My understanding of this requirement would be that the Technical Documentation should be available in one of the EU member state languages.

Official languages of the EU are listed here.

In the context of medical devices, but broadly applicable to any electronics product, this table from Mastermind Translations is a good starting point. Belgium, Finland, Ireland, Luxembourg, and Malta have multiple options for languages.

Image is “Tower of Babel” by M. C. Escher.





Safety voltages: ELV, SELV, PELV, and FELV

As an ex rock climber, I always liken the system of safety voltages (ELV, SELV, and PELV) to the slightly opaque British Trad grading system for climbing (which I think should have been extended to include Tricky, Jolly Tricky, Bit Spicy That One, Whoops There Goes My Lunch, and Guaranteed Hospital Visit)

One effect of this is non intuitive nature is that people often say “SELV” when they mean “voltages too low to exceed the 50Vac / 75Vdc threshold of the Low Voltage Directive” or “a low voltage supply”.

This isn’t accurate, as one could meet the ELV/SELV voltage limits for a DC supply of 100VDC, which would fall under the LVD.

Importantly, these terms capture not only the voltage levels but the nature of the supply from mains.

  • ELV defines safe(ish) voltage levels for user accessible parts
  • ELV limits correspond to ES1 levels for AC and ES2 for DC from EN 62368-1
  • SELV, PELV and FELV levels do not exceed ELV limits, even under no-load conditions (a useful consideration when using an unregulated power source like a 50/60Hz mains transformer)
  • SELV indicates a double insulated mains supply (Class II mains)
  • PELV indicates a single insulated with earth (Class I mains)

One thing I do like about EN 62368-1 is that it skirts around this non intuitive terminology and makes voltages and insulation much easier to understand.

A breakdown of the terms can be found below:

Intialisation ELV SELV PELV FELV
Meaning Extra Low Voltage Safety Extra Low Voltage Protective Extra Low Voltage Functional Extra Low Voltage
Electropedia Definition 195-05-24 195-06-28 195-06-29 n/a, defined in BS 7671 wiring regulations
Voltage Limit
(Conductor-to-Conductor or Conductor-to-Earth)
50Vac/120Vdc 42Vac
50Vac/120Vdc with no load
50Vac/120Vdc with no load
50Vac/120Vdc with no load
Insulation to Hazardous Mains n/a Double
Basic + earthed screen
Less than Basic
Earthed n/a No Yes Yes
Can be Accessible? Yes Yes Yes No
Can exceed voltage limits under conditions:
Normal No No No No
Single Fault No No No No
Earth Fault No No No No
Example ELV just defines acceptable voltage levels Class II a.c. mains with un-earthed output e.g. typical USB phone charger Class I earthed a.c. mains suply with low voltage outputs e.g. desktop PC power supply Low voltage that is not separated from a.c. mains e.g. voltage for SMPS controller primary side not accessible to user