EV Charger Installation Testing: Kit, O-PEN and RCD Rules

Quick answer

For EV charger work you need three things sorted before you quote: a PEN fault protection strategy (most decent modern chargers have open-PEN protection built in, so you usually don't need an earth rod), the right RCD arrangement (Type A plus 6 mA DC detection, or Type B), and a way to test at the vehicle connector, which in practice means an EVSE adaptor such as the Metrel A 1532 or Megger EVCA210 alongside your normal MFT. Everything else is standard initial verification to BS 7671 plus the manufacturer's commissioning procedure.

The regulation that shapes the whole job: PEN fault protection

If you only read one part of Section 722 properly, make it Regulation 722.411.4.1. It exists because of one ugly failure mode: a broken PEN conductor on a PME (TN-C-S) supply.

On a healthy PME supply, the neutral and earth are combined back to the transformer and everything is fine. Lose that PEN conductor somewhere in the street and the "earth" at the property floats up towards line potential through the connected loads. Inside a house that's largely masked by equipotential bonding. Outside, a car is a big lump of exposed metal sat on tyres, connected to your earthing system, being touched by someone standing on wet ground in trainers. That is the scenario 722 is written around.

So BS 7671 restricts using the PME earth for EV charging unless you apply one of the permitted measures. In practical terms you have three realistic routes on a domestic or small commercial job:

Route	What it means	When it makes sense
Charger with built-in open-PEN (O-PEN) protection	The charger monitors the supply and disconnects line, neutral and earth if it detects conditions consistent with a PEN fault	The default for most domestic installs in 2026
Separate O-PEN device	A standalone unit (Matt:e and Garo are the names you'll see) doing the same job upstream of a charger that lacks it	Commercial units, budget chargers, retrofits
Earth electrode / TT	Either the electrode option within 722.411.4.1 limiting touch voltage to 70 V, or converting the charging circuit to TT	Rural supplies, outbuildings, sites where you don't trust the alternatives

Why most chargers now have O-PEN built in

The voltage-monitoring method permitted by 722.411.4.1 is what made rod-free installs mainstream. Instead of needing a reference electrode, the device watches the line-to-neutral voltage and disconnects all conductors, including the CPC, if the voltage moves outside the normal window (the commonly quoted band is 207 V to 253 V, i.e. 230 V plus or minus 10 per cent). A shifting L-N voltage is the classic signature of a deteriorating or broken PEN, because the loads on the network pull the floating neutral around.

Nearly every mainstream UK domestic charger now ships with this built in, and it's the main reason the earth rod outside every driveway has quietly disappeared. Two things to do on every job though:

1. Check the paperwork, not the brochure. You want the manufacturer's declaration that the device meets the requirements of 722.411.4.1, and you keep a copy with the cert. "PEN fault protection" on a datasheet is not automatically the same thing.
2. Understand what it protects. The O-PEN device protects the vehicle and charging circuit. It does nothing for other exposed metalwork outside on the PME earth. If the job includes outside sockets, hot tubs or garden buildings, think about those separately.

When an earth electrode still makes sense

Don't write off the TT route. You'll still reach for it when:

• The charger (often commercial three-phase kit or the cheap end of the market) has no compliant O-PEN function and adding a Matt:e-type unit makes the job cost silly money.
• The installation is already TT, or it's a detached garage or barn a long way from the intake where TT-ing the charging circuit is clean and simple.
• You simply don't fancy hanging the safety case on electronics, which is a legitimate engineering position, particularly on long-life commercial installs.

Two honest warnings from doing this in the real world. First, the "70 V touch voltage" electrode option under PME needs a genuinely low electrode resistance once you run the numbers with realistic PEN currents; the worked examples in the IET Code of Practice for EV charging equipment installation are sobering, and in most soil you won't get there with one rod. That's why in practice the electrode route usually means converting the charging circuit to TT, not bolting a token rod onto PME. Second, a TT island only works if it's actually an island. In a terraced street with metal service pipes, bonded metalwork and the neighbour's PME earth a metre away, you often cannot achieve separation from PME-earthed metalwork, and that pushes you straight back to an O-PEN solution. Survey first, quote second.

If you do go TT: get the electrode resistance measured properly, not guessed. Under 200 Ω is the usual stability ceiling quoted, but I want to see well under 100 Ω before I'm happy signing it, and I want it measured with a stake test or a dedicated electrode measurement, not inferred from a wobbly loop reading on a wet day.

RCDs: Type A plus 6 mA DC detection, or Type B

Section 722 (Regulation 722.531.3.101 in the current numbering) gives you the RCD rules, and they trip up a lot of people moving into EV work:

• Each charging point gets its own RCD, rated not more than 30 mA. You can't hang it off the shared RCD covering half the house.
• The RCD must be at least Type A. Type AC is dead for EV work, full stop.
• The RCD must disconnect all live conductors, including the neutral. This is the one that catches people out: plenty of compact single-module RCBOs protect the neutral but don't actually switch it. For an EV circuit you need a device that breaks the neutral too, so check the RCBO spec before you order, not after.
• Where the charging equipment can push smooth DC fault current above 6 mA into the installation (which is the assumption with a vehicle on charge), you need either a Type B RCD, or a Type A RCD plus a 6 mA DC detecting device (RDC-DD to IEC 62955).

Here's the practical shortcut: almost every serious charger on the UK market has the RDC-DD built into the charger itself. The manufacturer does this precisely so you can protect the circuit with an ordinary Type A RCBO instead of a Type B RCD, because a Type B device costs several times as much and eats board space. So on a typical domestic job the arrangement is: Type A double-pole RCBO (or Type A RCD plus MCB) at the origin, charger's internal 6 mA DC detection doing the rest. Read the charger's installation manual and it will tell you exactly what it expects upstream; that document is part of your design evidence.

Where you do end up with a Type B RCD (some commercial chargers, some DC-capable kit), remember the blinding problem, which is the whole reason these rules exist: smooth DC can saturate the core of an upstream Type A or AC device and stop it tripping. Don't put a Type A ahead of a Type B on the same circuit and assume everything stacks up.

If your consumer unit needs surgery to make room for any of this, that's a job in itself; see our consumer unit upgrade guide.

The EVSE adaptor question: do you actually need one?

Short answer: if EV installs are going to be a regular line of work, yes, and it's not really optional.

The problem an EVSE adaptor solves is simple. A Mode 3 charge point won't energise its output until it has negotiated with a vehicle over the control pilot (CP) line. No vehicle, no closed contactor, no supply at the Type 2 connector, and therefore no way to measure Zs or test the RCD at the point that actually matters. The adaptor plugs into the vehicle connector and pretends to be a car: it signals the CP states (vehicle connected, vehicle ready to charge), which persuades the charger to close its contactor, and it brings the supply out so you can connect your tester. The better ones also simulate faults, like a CP diode fault or a PE fault, so you can prove the charger responds correctly and shuts down.

Two units I'm happy to put my name to:

• Metrel A 1532 - Metrel's EVSE adaptor, the natural partner for a Metrel installation tester. State simulation plus fault simulation, and it turns commissioning from guesswork into a repeatable sequence.
• Megger EVCA210 - Megger's equivalent, and the obvious pick if your MFT is already blue. Same core job: simulate the vehicle, get the contactor closed, test through the connector.

In practice either adaptor will work with whatever MFT you carry, so buy the one that matches the rest of your kit and the accessories you already own. For a one-off install, hiring one or leaning on a colleague is defensible. For anyone doing an EV install a week, the adaptor pays for itself in the first month just in time saved not chasing a customer's car keys around.

On the instrument side, any competent MFT covers the basics; our multifunction tester guide covers the field properly. The Megger MFT1741+ is a sound workhorse choice that also handles earth electrode stake testing for TT jobs. If you want one instrument that goes furthest on EV work, look at the Megger MFT-X1, which covers the wider range of RCD types including Type B, or the Metrel MI 3152 EurotestXC, which likewise tests beyond Type A and pairs naturally with the A 1532 adaptor.

Commissioning: the test sequence that actually works

1. Dead tests as normal. Continuity of the CPC, insulation resistance on the new circuit. Test the cable before final connection to the charger, or drop to 250 V with the charger connected; most manufacturers do not want 500 V across their electronics, and the manual will say so.
2. Polarity, Ze and prospective fault current at the origin, as any initial verification.
3. Zs at the vehicle connector, through the adaptor with the charger in the "charging" state so the contactor is closed. This is the reading that proves the whole protective loop, charger included. Compare against the limits properly; see maximum Zs values explained.
4. RCD tests. Test the upstream Type A device conventionally at the board, then through the adaptor at the connector. For the 6 mA DC detection inside the charger, follow the manufacturer's functional test procedure; many chargers have a documented self-test. Some instruments can perform a DC test on RDC-DDs, but do not fail a charger on an instrument test the manufacturer doesn't recognise; the manual's procedure is the reference.
5. O-PEN functional check per the manufacturer, where a test mode is provided. What you must never do is try to create a genuine PEN fault or start lifting neutrals on a live PME supply to "prove" it. You verify the function the way the manufacturer specifies, and you file their conformity documentation.
6. Electrode measurement if the job is TT, by stake test or dedicated electrode method, recorded on the schedule.
7. A real charge. Adaptors prove states and protection, but the only true load test is a vehicle actually charging. Watch the current, check the CT clamp for load curtailment is on the right conductor and reading the right way (a backwards CT is probably the most common snag on smart chargers), and confirm the app/connectivity side is commissioned, which matters for compliance with the Smart Charge Points Regulations on domestic installs.
8. Paperwork. EIC and schedule of test results, the manufacturer's commissioning record, DNO notification through the ENA process (most single domestic chargers are connect-and-notify, but check maximum demand and talk to the DNO first on looped or 60 A supplies), and labelling, especially where you've created a TT arrangement.

The kit list, summarised

• MFT: Megger MFT1741+ as the dependable all-rounder, or step up to the Megger MFT-X1 / Metrel MI 3152 EurotestXC for the widest RCD coverage.
• EVSE adaptor: Metrel A 1532 or Megger EVCA210, matched to your tester brand. Mid-range money, earns its keep fast.
• Earth stake kit for TT work.
• The basics you should already carry: a proper two-pole tester and proving unit (our guide), locks and lock-off kit.

FAQ

Do I need an earth rod for every EV charger? No. If the charger has built-in open-PEN protection meeting 722.411.4.1 (most current UK domestic chargers do), you can install on a PME supply without an electrode. You still need the manufacturer's evidence on file, and rods remain the right answer on some sites.

Can I run the charger off the existing RCD in the consumer unit? No. Each charging point needs its own RCD, at least Type A, 30 mA, disconnecting all live conductors including the neutral. A shared front-end RCD covering other circuits doesn't satisfy that, and many compact RCBOs don't switch the neutral, so check the spec.

Type A or Type B RCD? If the charger contains a 6 mA RDC-DD (check the installation manual; nearly all good ones do), a Type A device upstream is compliant and is the normal domestic arrangement. Type B is for chargers without that internal detection, and never sit a Type A or AC upstream of a Type B on the same circuit.

How do I test the 6 mA DC detection inside the charger? Follow the manufacturer's commissioning procedure; most provide a self-test or functional check. Instrument-based DC tests exist on some MFTs, but the manufacturer's documented procedure is what you record.

Do I have to notify the DNO? Yes, every charge point installation goes through the ENA notification process. Most single domestic chargers are connect-and-notify; apply first where maximum demand is marginal, the cut-out is 60 A, or the supply is looped.