EV Charging Cables — AC, DC Fast and Megawatt Charging Infrastructure

01Why the cable problem in EV charging is not the cable in the driver's hand

Ask what an EV charging cable is and most people picture the flexible coiled lead between the dispenser and the car. That lead is the last metre of a chain that starts at a grid transformer. Three of the four layers sit underground or inside a building, are specified once, and are expected to outlive the charger they feed. Getting those three wrong is what makes a charging site expensive, because the cost of correcting them later is measured in excavation and reinstatement rather than in metres of cable.

This page is part of cable solutions by industry, the cross-industry overview that compares the application families Sorivo supplies and the standards governing each.

The four layers behave differently enough that treating them as one specification is the origin of most site-level errors. A buried feeder is selected against ampacity and voltage drop over decades of thermal cycling. The charger supply cable is selected against mechanical protection and fire performance because it terminates at a pedestal a vehicle can hit. The charging lead is selected against flex fatigue and abrasion because a driver coils it ten to twenty times a day. These are not variations on one requirement; they are three separate qualification problems.

The asymmetry between layer 4 and layers 1–3 is the single most useful fact for a procurement team. The charging lead is a consumable that will be replaced several times over the life of the site, and its failure is visible and cheap. The feeder under the car park is not replaceable without digging, and its failure is an outage. Budget attention accordingly: the cable nobody will ever look at again deserves more of the specification effort than the one the driver handles every day.

02AC charging: the installation rulebook that sizes the cable

AC charging leaves the AC-to-DC conversion inside the vehicle. That single architectural fact sets where the thermal stress goes: into the car's on-board charger rather than into the cable. It also means the cable on an AC circuit is largely an installation problem, governed by whichever wiring regulation the jurisdiction has adopted, and the two dominant frameworks differ in how they express the same requirement.

02.1 Continuous load, and why diversity does not apply

In the United Kingdom the governing document is BS 7671 Section 722, Supplies for Electric Vehicles. The 2026 amendment (BS 7671:2018+A4:2026) retains the provisions that matter most to cable selection and withdraws the previous amendment from October 2026. Electric vehicle charging is classified as a continuous load, because sessions routinely exceed thirty minutes. The cable must therefore be rated for 100 per cent of the charger's maximum demand, and no diversity factor may be applied. That single sentence is the most frequently violated rule on charging sites, and applying a diversity factor copied from a domestic installation load schedule is how a feeder ends up undersized before it is even buried.

Two further provisions follow the same logic. Each charging point requires its own dedicated circuit, with no sharing against other loads, and Mode 3 AC charging requires RCD protection at a minimum of 30 mA Type A. Where the supply is TN-C-S, the earthing arrangement needs separate assessment, because the continuous charging current in the neutral can raise the potential at the charger.

Europe works to IEC 60364-7-722, harmonised through CENELEC HD 60364, which reaches the same place by a different route. The material difference is on the fire side: many European jurisdictions require CPR-compliant cable to EN 50575 with a minimum of Class Dca for charging installations in public-access buildings, and Class Cca may be required for runs between buildings.

02.2 North America expresses it as 125 per cent

NEC 2026 Article 625 governs EV charging infrastructure in the United States, and it arrives at a comparable conductor size by requiring branch circuits to be sized at 125 per cent of the continuous load, with GFCI protection for charging outlets. The two approaches look different on paper and converge in practice: in neither framework is an EV charging circuit sized at its nominal rating alone.

02.3 Where the size actually comes from

Once the continuous-load factor is applied, voltage drop rather than ampacity usually decides the conductor on AC charging work, because chargers cluster at the far end of car parks. A 32 A single-phase charger on a 6 mm² conductor reaches roughly 5.6 per cent voltage drop at 60 m — well past the three per cent design target that is widely adopted for charger compatibility. The table below states the sizes for a 30 m run with XLPE-insulated armoured cable clipped direct; longer runs move up a size.

Two corrections have to be layered on top of the table before it becomes a specification. The first is grouping: when several charger circuits share a trench or a tray the cables heat one another, and a group of six circuits can attract a derating factor in the region of 0.60 to 0.70, which raises the required cross-section by forty to sixty per cent. The second is earthing practice. Because the armour of an armoured cable has to provide a continuous parallel earth path and be earthed at both ends, a two-core construction is the wrong answer for charging work: specify three-core for a three-phase charger and three-core for a single-phase charger as well, carrying line, neutral and earth. Our earthing and bonding guide to IEC 60364 sets out the sizing method behind that choice.

03DC fast charging: where air cooling runs out

DC fast charging moves the AC-to-DC conversion into the charger cabinet, which is what shifts the thermal burden onto the cable and connector. Two different cables are involved, and confusing them is a common source of over-specification. The cable from the site switchboard to the charger cabinet is an ordinary AC feeder, sized by the input current the cabinet draws, not by the DC power it advertises. The cable between the cabinet and the vehicle is a DC cable with its own standard framework, and it is the one that has to survive repeated handling at 500 A.

The cabinet input is larger than the nameplate DC output, because conversion is not lossless. A 150 kW DC charger at 94 per cent efficiency draws approximately 160 kW from the grid, which is roughly 230 A per phase at 400 V three-phase at unity power factor — and closer to 242 A where the power factor is 0.95. Sizing the feeder from the DC nameplate alone is the error this calculation is meant to prevent.

03.1 The vehicle-side cable has its own sizing logic

On the vehicle side the conductor cross-section is set not by ampacity tables alone but by the temperature rise that remains acceptable under sustained full-power charging. Air-cooled constructions follow a straightforward relationship until they stop being practical.

The turning point is worth stating plainly because it is a physical limit rather than a commercial one. Above roughly 375 A sustained, doubling the copper from 95 mm² to 185 mm² buys about forty per cent more ampacity while roughly doubling cable weight and stiffness. Past that point the cable stops being something a person can reasonably handle, and a heavier cable is not a solution. Liquid cooling breaks the trade-off differently: a closed circuit circulating a water-glycol mixture or dielectric fluid through channels alongside the DC conductors removes heat to a pump and heat exchanger in the cabinet, and allows conductor cross-sections to fall by sixty to eighty per cent for the same current. A 500 A liquid-cooled cable can end up lighter and more flexible than a 300 A air-cooled one.

The published range answers the DC duty at both ends of that transition: the liquid-cooled ultra-fast charging cable rated 600 A at 1,000 V DC for the highway-hub end, and the flexible DC conductors in the EV charging cable category for the air-cooled 50–350 kW band.

04Megawatt charging: why liquid cooling stops being optional

Megawatt Charging System work began as a CharIN task force answering one demand from the truck and bus industry: charge a long-haul vehicle inside a driver's mandatory rest break rather than over several hours. The standard side settled in March 2025 when SAE published J3271 as a Technical Information Report covering DC charging on the order of 1 MW and above, with a ceiling of 1,250 V and 3,000 A — approximately 3.75 MW at the top end.

J3271 is not one document but a family, and the split matters to anyone buying cable. Part J3271/1 covers the coupler, J3271/2 the communication and control, J3271/3 the cable, cable handling, cooling and automated connection, J3271/4 the use cases including grid interconnection and bidirectional transfer, and J3271/5 the interoperability test procedures. A claim of being ready for megawatt charging should resolve to a specific part — cable and cooling live in J3271/3 — rather than to a marketing page.

04.1 The arithmetic that forces coolant

Conductor losses scale with the square of current. Moving from today's high-power CCS ceiling of roughly 500 A to 3,000 A is a sixfold current increase, which is a factor of about thirty-six in ohmic heat for the same conductor size. Cooling that with air, or with more copper, runs into the limits of cable weight, stiffness and the connector face a person can still plug in. This is why liquid cooling is the design basis rather than an option in J3271/3, and why the standard gives cable handling and cooling a part of its own. A liquid-cooled charging cable is better understood as a thermal system with electrical terminations than as a heavier version of an air-cooled cable, which is why coolant flow rate, pressure drop, connector thermal design and pump redundancy belong on the datasheet next to ampacity — and are the items most often missing from it.

The first-generation figures matter more than the ceiling. Orderable megawatt-equipped trucks in 2026 charge at up to 1,000 A and 750 kW — already around twice what CCS2 delivers — with the 3,000 A ceiling as headroom the ecosystem grows into over several years. For a specifier that argues for leaving conduit and civil works able to accept the full rating later without paying for the 3,000 A cable on day one.

The binding constraints in 2026 are not on the cable side at all. Grid interconnection and transformer availability dominate project timelines, and full-rate refuelling sessions remain rare globally. Budget the site electrical scope before budgeting the cable, then match the cable generation to the vehicle generation actually being delivered.

05The standard map: which framework governs which part

EV charging cable sits at the intersection of several standard families, and which one applies depends on the charging mode, the region, and whether the cable is attached to the charging station or to the vehicle. The table below is the reference to work from when a specification cites a standard by number and the number needs checking against the duty.

For a European charger manufacturer the primary cable standard is EN 50620. For DC fast charging assemblies sold globally, cross-certification — EN 50620 alongside UL 2251 and TÜV 2PfG 1908 — is common, and the practical caution is to confirm which standard the destination market's building regulation actually references before assuming a certificate travels. State the destination market at the start of an enquiry, because it changes the construction and sometimes the connector interface as well.

06Sheath and insulation: why TPU is a specification, not a preference

A charging lead lives outdoors, gets dragged across concrete and asphalt, is exposed to oil and fuel spillage, UV, and repeated coiling at temperatures that are freezing for half the year in northern markets. The sheath is the first line of defence against all of it, and the material choice is the difference between a lead that lasts its designed life and one that cracks through its first winter.

The table explains a field observation that surprises operators: charging leads with PVC jackets stiffen and crack within the first cold winter, while a TPU lead on the adjacent dispenser keeps working. The material cost difference is small against the cost of a warranty replacement and the downtime of a dispenser, which is the argument for treating the EN 50620 sheath designation as a specification line rather than an optional upgrade.

06.1 Conductor class matters as much as the sheath

EN 50620 requires Class 5 or Class 6 extra-fine stranded conductor. A Class 2 rigid stranded conductor will pass a bench inspection and fail the flex-life test within weeks, because the strands do not share bending strain evenly. The same logic runs through the SORIVO EV range: the 22 kW Type 2 assembly and the portable unit are both built on Class 5 flexible stranding, which is the construction the standard expects for a cable a person handles daily.

06.2 Where data cable fits

Most smart chargers need a data connection for load management, OCPP communication or current-transformer monitoring, and that circuit has its own requirements. Historically it meant a separate data cable in the same trench, and composite power-plus-data armoured constructions now exist that reduce trench space. On the data side the relevant published item is the Cat6A and Cat8 S/FTP shielded data cable with an LSZH jacket and CPR Eca/Dca rating, which is worth one line in the bill of materials and not more — the electrical and mechanical duty on a charging site belongs to the power cables.

07What to verify before a charging cable is accepted on site

The charging cable market carries a high proportion of products described as compliant without a certificate behind the description. Five checks separate a genuine EN 50620 assembly from a lookalike, and all five can be done from a sample and a document pack.

1. Read the jacket printing. A genuine cable carries the standard number, the conductor count and cross-section, the voltage rating and the manufacturer mark, printed along its length. A marking in the form “EN 50620 3×6 mm² + 2×0.5 mm² 450/750 V” is what to look for. A jacket that says only “EV charging cable” is not making a verifiable claim.
2. Confirm the sheath designation. EN 50620 requires the EVM-1 Z5 material designation. A datasheet that lists a PVC sheath, or that declines to name the compound at all, is not describing an EN 50620 construction whatever the certificate headline says.
3. Check the conductor class. Class 5 or Class 6 is what the standard mandates. Class 2 rigid stranding is a bench-passing, field-failing substitution.
4. Ask for the abrasion test report. The standard requires 2,000 scraping cycles without exposing the conductor, which a genuine TPU construction passes without difficulty.
5. Do the cold bend test. Hold a 50 cm sample at −25 °C for four hours and bend it 180 degrees around a mandrel at six times the diameter. A genuine TPU cable returns without cracking; a poor substitute shows stress whitening or splits.

Two further checks apply to the broader delivery rather than to the cable alone: verify the certificate number with the issuing body instead of accepting a scan, and confirm that the certificate covers the model code actually being shipped rather than a similar construction in the same family. Our global cable certification guide and the cable supplier qualification checklist set out the same process from the buyer's side.

08Selection toolset for charging infrastructure engineers

08.1 Routing the application to the right construction

08.2 The specification errors that recur on charging sites

• Applying diversity to an EV circuit. Charging is a continuous load and the installation rules prohibit diversity. The conductor is rated for the full charger demand, and a load schedule copied from a domestic installation will not reflect that.
• Sizing on ampacity when voltage drop governs. Long trench runs to the far end of a car park decide the conductor. Calculate the actual run, including the trench route rather than the straight-line distance, before defaulting to a standard size.
• Specifying a PVC sheath in an enclosed car park. In an underground structure with hundreds of vehicles, dense smoke and hydrogen chloride are a life-safety problem rather than a materials preference. Where the route is enclosed, an LSZH construction such as CU/XLPE/LSZH/SWA/LSZH armoured power cable is the appropriate answer, and the PVC-sheathed equivalent belongs on open-ground runs.
• Ignoring grouping derating. Multiple charger circuits in one trench heat each other. The correction factor can be severe enough to change the conductor size, and it is easy to omit when each circuit is calculated in isolation.
• Sizing the grid connection for day one. The feeder under a car park is the hardest cable on the site to replace. Sizing it for the target fleet rather than the initial installation costs very little at the trenching stage and avoids a second excavation later.
• Using a two-core construction on a buried run. The armour has to be earthed at both ends as a parallel earth path, and retrofit installations on TT systems may need a dedicated earth core instead. Three-core is the practical minimum.

08.3 What a working bill of materials looks like

A ten-bay AC hub at 22 kW per bay with a switchboard 80 m from the transformer and a distribution board 50 m away lands on three quite different constructions, which is a useful illustration of how little the layers have in common. The transformer-to-switchboard feeder, sized for a final target around 330 kW rather than the day-one 220 kW, runs to parallel 4-core 185 mm² armoured cable or an equivalent busway. The switchboard-to-distribution-board run carries 320 A with no diversity and needs 185 mm² to hold inside the three per cent voltage drop target over 50 m, before the grouping factor is applied. The final connection to each 32 A charger is a 4-core 10 mm² armoured cable at a 25 m average run. Three layers, three constructions, and a single diversity assumption applied anywhere in that chain is what breaks it.

09Engineering questions we are asked most often

These are the questions that come up in specification reviews, answered from the same standard basis used above. Where a figure is a design threshold rather than a company measurement, it is described as such.

Can one cable cover AC charging and DC fast charging?

No, and the reason is structural rather than commercial. An AC charging assembly carries three phase conductors for three-phase supply plus signalling cores at a small cross-section, while a DC fast charging assembly carries two large DC conductors plus earth, signalling and sometimes coolant channels. The connector interfaces differ as well, and the current levels are an order of magnitude apart. They are two product families sharing a supply chain, not one product with two uses.

Do I need liquid cooling for a 180 kW DC charger?

Not at that level. Air-cooled constructions remain practical up to roughly 350 kW, which is where the conductor becomes too heavy to handle rather than where it stops conducting. The 350–600 kW band is where liquid cooling earns its complexity, and the reason is user handling as much as thermal capacity: a liquid-cooled cable can be lighter and more flexible than a much lower-rated air-cooled one.

Is LSZH the same thing as fire-resistant?

No, and the two are commonly conflated. LSZH describes the chemistry of the jacket when exposed to flame — low smoke emission and no halogen acid gases, measured by IEC 61034-2 and IEC 60754-2. Fire resistance describes how long the cable continues to conduct under direct flame and mechanical shock, measured by BS 6387 or EN 50200. A cable can be halogen-free and fail in minutes, and a fire-resistant construction with a PVC jacket can still produce dense toxic smoke. Charging infrastructure in enclosed spaces needs the low-smoke property; circuit-integrity requirements are a separate specification question.

Why does the cabinet feeder look oversized against the charger's DC rating?

Because the feeder is sized on the cabinet's AC input current, not on its DC output power. Conversion losses mean a 150 kW charger draws approximately 160 kW from the grid once efficiency is accounted for, and the power factor affects the current further. A feeder calculated from the nameplate DC rating alone will land one or two sizes below what the cabinet actually draws at full output.

What is the difference between EN 50620 and TÜV 2PfG 1908?

EN 50620 is the harmonised European standard and sets the baseline requirements. TÜV 2PfG 1908 is a supplemental certification that adds a tighter bend radius of at least 8×OD against the harmonised figure of 5×OD for fixed installation, extended mechanical durability testing and additional environmental tests. Many European charging station manufacturers now write both into their procurement specifications, which is a reasonable practice provided the certificate for each is checked against the exact model.

Does the same charging cable work on both sides of the Atlantic?

Not automatically. European assemblies are specified against EN 50620 or IEC 62893 with the Type 2 interface, while North American work references UL 62 and UL 2251 with SAE J1772 or SAE J3400 connectors, and installation rules differ between BS 7671 and the NEC. Cross-certified assemblies exist and are common for export programmes, but the destination market's regulation, not the assembly's certificate portfolio, decides whether a given construction can be installed. State the destination at the start of an enquiry.

10Why Sorivo, and how to check our claims

10.1 Team experience, stated precisely

Sorivo's senior team brings more than 15 years of combined experience in cable design, production control and quality management. We describe it as team industry experience rather than company age, because that is what it is — the brand began exporting industrial cable in 2023 on the back of a long-standing team and a manufacturing partnership, and the figures are published that way on our about page. Projects and buyers currently span 300+ clients across 30+ countries.

10.2 Application engineering before the quotation

Charging infrastructure specifications are calculation-heavy in four places: conductor sizing against derated ampacity in a grouped trench, voltage drop over the actual route, sheath chemistry matched to the enclosure and to UV exposure, and the earthing arrangement around a TN-C-S supply. Support covers those four before a quotation is issued rather than after a site problem appears. For repetitive work the cable engineering tools section carries free utilities, including the ampacity calculator and a derating guide covering temperature, altitude and grouping correction factors.

10.3 The qualifications that come with the cable

For any published model we can provide the certificate reference covering the exact construction, the relevant test report and batch traceability from the delivered drum back to production. On the charging side the published range covers the AC assemblies from 16 A to 32 A three-phase, the GB/T variant for dual-standard sites, the portable Mode 2 unit and the 600 A liquid-cooled DC cable. Two checks are worth doing regardless of supplier: verify the certificate number with the issuing body rather than accepting a scan, and confirm that it covers the model code actually shipping.

Reviewed by Luo Qiang — Senior Cable Application Engineer, Sorivo
15+ years in industrial and renewable energy cable specification. Experienced in cable specification aligned with IEC standards.

How this was written: every standard number and product claim on this page is taken from the corresponding published Sorivo product or technical article, and the standard identifiers were checked against the issuing body's own catalogue before publication. Where a figure is a general design threshold rather than a company measurement, it is described as such.

Standards referenced on this page

10Frequently asked questions