EV charging station installation — from an empty
slab to a charging
vehicle.

Objective: know what can kill you on an EV charging site, and the sequence that stops it. This module comes first because a charger has three independent energy sources and two of them stay dangerous after you switch off.

The three energies in a charging station

Lock-out / tag-out, in the order you actually do it

1. Identify every source. Walk the line: charger → feeder breaker → LT panel → DT. Note anything that can back-feed (solar inverter, DG set, BESS, a second incomer, a bypass link).
2. Notify. Site owner, security, CMS/control room. Remote-disable the charger in the backend so nobody starts a session on you.
3. Isolate. Open the feeder breaker, rack it out or open the isolator. For DC work, open the internal DC isolator too.
4. Lock and tag. Your own padlock, your own tag, your name and number, your date. Multi-lock hasp if more than one person is working.
5. Dissipate. Wait the OEM-specified capacitor discharge time. Nameplate or manual — do not guess.
6. Verify. Three-point test. Then apply earthing/short device if the OEM specifies one.
7. Only now, open the enclosure.

PPE, matched to the task

Objective: understand the one distinction that explains the entire product line — where the rectifier lives. Once you have this, AC vs DC, slow vs fast, cheap vs expensive all fall out automatically.

Every lithium battery in every EV can only be charged with DC. The grid only supplies AC. So somewhere between the grid and the cells, something has to convert AC to DC. That converter is called a rectifier, and the single question that defines a charger is: is the rectifier inside the car, or inside the box on the wall?

The consequences you will meet in the field

Objective: the working subset of electrical theory you need on a charging site — not a diploma course. Three-phase, kW vs kVA, earthing systems, and the arithmetic you will do standing in front of a customer.

Single phase vs three phase

Indian LT supply is 415 V between any two phases and 240 V between a phase and neutral, at 50 Hz. Three phases are three AC voltages offset by 120°. Because their peaks are staggered, three-phase power arrives smoothly and continuously — which is why every charger above about 7.4 kW is three-phase.

kW, kVA, and why the DISCOM cares

• kW (real power) is what does work and what the customer is billed for in energy terms.
• kVA (apparent power) is what the cable, breaker and transformer must actually carry. kVA = kW ÷ power factor.
• Sanctioned load from the DISCOM is normally expressed in kVA or kW as per the connection agreement, and exceeding it triggers penalties and, on many industrial tariffs, maximum-demand charges.

Modern DC chargers run active PFC and hold power factor above ~0.98, so kVA ≈ kW. Older or badly loaded units drift lower. Efficiency also matters: to deliver 60 kW to a car at 95% efficiency, the charger draws roughly 60 ÷ 0.95 ≈ 63 kW from the grid. Size the supply on input, not on nameplate output.

Current calculator — the arithmetic you do on site

Three-phase current: I = kW × 1000 ÷ (√3 × 415 × PF). Single-phase: I = kW × 1000 ÷ (240 × PF). Then apply efficiency to get input, then apply the breaker sizing rule.

Earthing systems — know which one you are standing on

The earthing arrangement decides how a fault current returns to source, which decides which protective device operates, which decides whether your customer gets a shock or a trip. In India you will mostly meet TN-C-S (combined neutral-earth from the DISCOM, split at your panel) and TT (your own local electrode, independent of the supply earth).

Terms you must be fluent in

Objective: identify any gun on sight, know what each pin does, and know which regulatory body governs which part of your job.

The connector family, as you meet it in India

Who governs what

Objective: open the lid of a 7.4/22 kW AC unit and name every part, know what each does, and know which ones you are allowed to replace in the field.

Part by part

Objective: understand the power train of a 60–360 kW cabinet so that a fault code points you at a subsystem instead of at a mystery.

Six subsystems, six kinds of trouble

1. Power modules. Parallel rectifier bricks. Lose one of four in a 120 kW unit and you still deliver 90 kW — the site keeps running with reduced output, which is exactly why nobody notices for a month. Watch the CMS for silent capacity loss.
2. DC/DC and isolation. Provides the galvanic barrier between grid and vehicle. Its failure is not gradual — the unit refuses to start.
3. Output stage. Precharge resistor gently brings the output up to the battery's voltage before the main contactors close. Without it you get an enormous inrush and welded contacts. If a charger fails right at "starting", suspect precharge.
4. Insulation monitoring. Before any real current flows, the charger measures resistance from DC+ and DC− to earth. Below the threshold it aborts. Water in a gun, a nicked cable, or a wet vehicle inlet all show up here.
5. Thermal. Fans, filters, and on big units a liquid loop with a pump, radiator and coolant level sensor. In Indian summer, a clogged filter is the most common cause of "charger is slow" — it is derating, not faulting.
6. Control and comms. Sequencing, limits, logging, HMI, OCPP uplink, PLC downlink to the vehicle.

Objective: master the one wire that decides whether a car charges. If you learn nothing else in this course, learn this module — it resolves the majority of "it just won't start" callouts.

Before a single ampere flows, the charger and the car have a conversation on the control pilot (CP) wire. It is a ±12 V, 1 kHz square wave. The charger sends it out; the car pulls the positive peak down with resistors to answer. The height of the positive peak is the car's answer. The duty cycle is the charger's offer.

Duty cycle is the current limit

The charger tells the vehicle how much current it may draw by varying the width of the pulse. The conversion is fixed by the standard:

Proximity pilot — the cable's own rating

PP is simpler: a resistor moulded into the plug tells the car what the cable can carry, so a thin cable cannot be pushed to a fat cable's current. On a Type 2 assembly, the resistance between PP and PE encodes it. Indicative values you will measure:

A damaged PP resistor is a classic "charges at 13 A on every charger" complaint — and it is the customer's cable, not your station. Measuring PP-to-PE on their cable settles the argument in thirty seconds.

The full session, message by message

The three protocol layers, kept straight

Objective: walk a site once and come back with everything design, procurement and the DISCOM will ask for. A weak survey is the single biggest cause of cost overrun and schedule slip on charging projects.

Six questions the survey must answer

1. Is there enough power, and whose is it? Existing sanctioned load, current maximum demand, spare headroom, transformer rating and its loading, whether the connection is dedicated or shared with the host's building.
2. Where does the cable run? From LT panel to charger, metre by metre. Route length, obstacles, road crossings, existing ducts, who owns each section of ground.
3. Where does the charger physically go? Bay geometry, cable reach to the vehicle's inlet, turning and reversing space, canopy, drainage, flood level.
4. Will it earth? Soil type, water table, space for earth pits, existing earthing to test and connect to.
5. Will it connect? Mobile signal strength on all carriers at the exact charger location, not at the gate.
6. Who says yes? Landowner, building society, DISCOM, fire, municipal body, mall management. Each one is a date on your schedule.

Survey kit

Measuring tape + laser distance meter · clamp meter · multimeter · earth resistance tester · insulation tester · phase sequence meter · signal strength check on 2+ carriers · camera · torch · chalk/marker · this checklist · previous electrical drawings if any.

Power availability worksheet

Layout geometry — how the bay actually gets used

Replace these slots with your own site photographs. Every survey should return with a photo set in this order — it becomes the evidence pack for design review and the DISCOM file.

Objective: turn the survey into a design that is safe, compliant and not gold-plated. Cable and protection are where most money and most risk sit.

The protection chain, source to gun

Cable sizing — the four checks, in order

1. Current: the cable's rated capacity must exceed the design current.
2. Derating: multiply capacity down for ambient temperature, grouping with other cables, and installation method (buried, tray, conduit, in-air). India's 45 °C+ summer ambient bites hard here.
3. Voltage drop: keep within limits over the real route length. Long runs, not high current, are what usually forces the next size up.
4. Fault withstand: the cable must survive the prospective fault current for the time the protective device takes to clear it.

Sizing helper

Panel design notes that save you later

• Leave spare ways. Every charging site expands. Two spare outgoing ways and busbar headroom cost almost nothing now and save a panel replacement later.
• Separate the EV feeder from the host's building load with its own metering wherever the commercial model needs energy attributed to charging.
• Label everything with engraved traffolyte, not marker pen. Circuit name, charger ID, cable size, source. The next technician is not you.
• Space for the cable bending radius at the gland plate. Large-section armoured cable will not turn as tightly as your drawing assumes.
• Ingress and vermin. Seal every gland and every unused knockout. Rodents in a panel are a leading cause of outdoor site failures in India.
• Panel heat. An outdoor LT panel in direct sun with no louvres or fan will run 15–20 °C above ambient and shorten everything inside it. Sunshade, ventilation, correct IP.

Objective: build a base and a route that will still be sound in ten monsoons. Civil defects are the failures you cannot fix without digging everything up again.

The foundation

A DC cabinet can weigh several hundred kilograms and must not settle, tilt or flood. The plinth does four jobs at once: carries the weight, lifts the cabinet above standing water, presents the cable entry at the right place, and holds the foundation bolts in the exact pattern the OEM template specifies.

Trenching and cable route

• Depth and protection. Buried LT cable goes in a sand bed, with protective tiles or bricks above and a coloured warning tape above that, at the depth your standard and the local authority specify for the location — deeper under roads and vehicle crossings.
• Never share a trench carelessly. Keep separation from water, gas, telecom and existing power. Where crossings are unavoidable, cross at right angles with mechanical protection.
• Sleeve every road and hard-surface crossing in GI or HDPE, sized so the cable can be pulled and, one day, replaced.
• Draw pits at every long run and every direction change so pulling tension stays within limits. Exceeding pulling tension damages insulation invisibly.
• Record the as-laid route with dimensions from permanent features before you backfill. Take photographs of the open trench. This drawing is worth more than the cable itself the day someone else digs.

Mounting, enclosure and environment

Objective: make joints that will not loosen, heat, or corrode, and build an earth that will actually carry a fault. This is where craft separates a technician from a wire-puller.

Termination discipline

1. Strip to the right length. No copper showing beyond the lug barrel, no insulation trapped inside it. Both are failures.
2. Correct lug for the conductor and the stud. Match cross-section and hole size. A lug that is one size up "so it fits" makes a loose, hot joint.
3. Crimp with the matched die, full cycle, hydraulic or ratchet tool that does not release until complete. Pliers are not a crimping tool.
4. Torque to the manufacturer's figure with a calibrated torque wrench or screwdriver. Under-torque runs hot; over-torque cracks the terminal or crushes the conductor. Take the value from the device label or manual — never from memory or from another brand's chart.
5. Mark the torqued joint with a paint pen line across nut and terminal. Later, a shifted line tells you instantly the joint has moved.
6. Ferrule and label both ends of every conductor, matching the drawing.
7. Dress the cables. Bending radius respected, no strain on terminals, weight supported by cleats and not by the lug, doors close without pinching anything.
8. Re-check after first load. Thermal cycling settles joints; a torque check and a thermal scan after the first weeks of operation catches the ones that moved.

Earthing — build it, then prove it

• Separate pits, interconnected, with accessible inspection chambers and removable test links so each electrode can be isolated and measured individually.
• Bond everything: charger body, panel, canopy, cable tray, gantry, bollards, gate frames if within reach of the installation.
• Earth conductor sizing follows the phase conductor size per the standard — it must carry fault current for the clearing time without melting.
• Joints below ground should be exothermic or purpose-made clamps rated for burial. A bolted joint in soil corrodes and quietly goes open-circuit.
• Record the readings on a handover sheet with date, instrument serial and technician name. This is the document that protects you if there is ever an incident.

Objective: energise in a controlled sequence, prove the installation with measurements, and hand over a documented, working asset. An installed charger is not a commissioned charger.

Stage 1 — dead tests, before any power

Every one of these is done with the installation isolated and proven dead. Record the number, not "OK".

Stage 2 — first energisation

1. Confirm the DISCOM connection is complete, metered and approved. Do not back-feed from a temporary supply for a "quick test".
2. Everything downstream off. Close the incomer only.
3. Measure at the panel: phase-to-phase, phase-to-neutral, phase-to-earth, neutral-to-earth. Neutral-to-earth should be near zero; a significant reading means a neutral problem you must resolve now.
4. Check phase sequence R-Y-B with a rotation meter.
5. Close the feeder breaker. Charger powers up. Watch the boot sequence and the HMI.
6. Test the RCD with a proper RCD tester — trip time and trip current, both recorded. Press the test button too, but the button only proves the mechanism, not the sensitivity.
7. Verify the emergency stop actually removes power, and that recovery from E-stop requires a deliberate reset.

Stage 3 — functional proving with a real vehicle

1. Charger appears online in the CMS. Boot notification and heartbeat visible.
2. Authorise by every method the site offers: RFID card, mobile app, remote start from the CMS.
3. Plug in. Confirm the gun locks and the pilot moves through its states.
4. Deliver a real session — at least long enough to reach steady power. Compare the charger's displayed power against a clamp meter reading on the input.
5. Stop from each source: the vehicle, the charger's stop button, the app, and the CMS remote stop. All four must work.
6. Compare energy delivered on the charger, in the CMS transaction, and on the site meter. They should agree within the meters' accuracy class. A large mismatch is a billing problem you fix now, not after a month of invoices.
7. Emergency stop during an active session — power must drop and the session must close cleanly.
8. If the site has multiple guns, test simultaneous sessions to confirm real shared behaviour matches what was sold.

Commissioning checklist — tick as you go

• Isolation proven, LOTO in place during dead tests
• Continuity of all protective conductors recorded
• Insulation resistance recorded, electronics disconnected
• Polarity confirmed at every termination
• Earth electrode resistance measured and within target
• All power terminations torqued and paint-marked
• Panel clean, glands sealed, entries plugged, labels fitted
• Supply voltages and neutral-earth voltage measured
• Phase sequence R-Y-B verified
• RCD trip time and current tested with an instrument
• Emergency stop verified, including mid-session
• Charger online in CMS, heartbeat confirmed
• All authorisation methods tested
• Real vehicle session completed at rated power
• Stop tested from vehicle, charger, app and CMS
• Energy reconciled: charger vs CMS vs site meter
• Simultaneous session behaviour verified (multi-gun sites)
• Signage, tariff display, emergency contact fitted
• Handover pack complete and issued

Objective: get the charger talking to the backend and keep it talking. A charger that is offline is invisible, unbillable and unmanageable, even while it charges cars perfectly.

What OCPP actually does for you

OCPP is the language between the charge point and your central management system. In 1.6J it runs as JSON over a WebSocket connection. The messages you will care about in the field:

Connectivity build, in order

1. Signal survey at the charger location. Inside a metal cabinet, at the antenna position — not on your phone at the gate. Check every carrier; pick the strongest, not the cheapest.
2. Antenna placement. External antenna where signal is marginal. A metal enclosure is a Faraday cage and will cost you 15–20 dB.
3. SIM and APN. M2M SIM with a static IP or a VPN if your CMS requires it. Record the SIM number and its billing account against the charge point ID.
4. Backend configuration. Charge point ID, OCPP endpoint URL, protocol version, security profile and credentials. A single character wrong in the URL produces exactly the same symptom as no signal at all.
5. Firmware. Update to the version your fleet is standardised on, before handover, and record it. Never leave a site with an unknown firmware version.
6. Tariff, connector configuration and display. Confirm what a customer sees matches what the CMS charges.
7. Redundancy where the site justifies it. Wired broadband with a 4G fallback for high-value sites; consider a local fallback mode so the charger still serves cards or free sessions when offline.

Objective: convert a vague complaint into a located fault, fast, without shotgun part-swapping. Method first, then the fault library.

The method

1. Get the real symptom. "It's not working" is not a symptom. Ask: at what point in the sequence does it stop? Which vehicles? All guns or one? Every time or intermittently? Since when? What changed?
2. Read the logs before you read the charger. CMS status history, error codes and the last few sessions tell you more than the front panel. Half of all calls are solvable before you arrive.
3. Split the system. Charger-side or vehicle-side? Local or network? Power path or control path? Every test should eliminate half the remaining possibilities.
4. Test with a known-good reference. Your own test vehicle, your own cable, another gun on the same site. Substitution is the fastest proof there is.
5. Measure; do not assume. Voltage, current, resistance, temperature. A number ends an argument that an opinion cannot.
6. Fix the cause, not the symptom. Resetting a charger that faults every Tuesday is not a repair.
7. Record it. Fault, evidence, action, result, parts. Your logs are the only route to finding the pattern behind repeat failures.

Fault library — open the one that matches your symptom

Instruments you should have in the van

CAT III/IV multimeter · clamp meter with DC capability · insulation tester · earth resistance tester · RCD tester · phase rotation meter · oscilloscope or dedicated EV pilot tester · EV charging simulator / adapter that can emulate states B and C · thermal camera · torque wrench and screwdriver · laptop with the OEM service tool.

Objective: keep the asset earning. Uptime is not luck; it is a schedule, a spares kit, and a discipline about root causes.

Preventive maintenance schedule

The numbers that run the operation

Objective: recognise the archetype on arrival. Each one has a different power problem, a different customer, and a different way of failing.

Objective: the topics that separate a technician who installs chargers from one who engineers charging sites.

Dynamic load management

Instead of sizing the supply for every charger at full power simultaneously, you install a controller and a current transformer at the incomer. It measures the whole site's draw in real time and reduces the current offered to chargers so that the total never exceeds the sanctioned limit.

Solar and battery storage on a charging site

• Solar reduces energy cost during the day and reduces grid draw, but it does not reduce your sanctioned load requirement — the grid must still be able to serve the peak when a cloud passes.
• Battery storage does something solar cannot: it shaves the peak. A 240 kW charger fed by a 100 kW grid connection plus a buffer battery is a real architecture for constrained sites, and often beats waiting eighteen months for a transformer.
• Technician implications: two more energy sources that can back-feed. Your isolation procedure must include the inverter and the battery, and the site's single line diagram must show every source and every changeover. Never assume the LOTO you learned on a grid-only site is sufficient here.

Billing integrity

You are selling a measured commodity. That makes metering a legal matter, not an engineering preference.

• Use an approved, verified meter of the required accuracy class for any energy you bill.
• Reconcile three numbers regularly: the charger's meter, the CMS transaction record, and the site's utility meter. Understand and document the gap — auxiliary loads, cable losses and charger efficiency all live in it.
• Charger clock drift silently misprices sessions on time-of-day tariffs. Confirm time sync at commissioning and check it at every service visit.
• Keep session records with the meter start and stop values, not just totals. A customer dispute you cannot evidence is a customer you refund.

V2G and bidirectional charging

Bidirectional units can take energy back out of the vehicle to a building or the grid. The technician-relevant points: the equipment is a generator as well as a load, so grid protection, anti-islanding and interconnection approval enter your scope; the vehicle must support it; and your isolation procedure now has a source that can energise the installation from the parking bay. Treat any bidirectional site as a generation site.

This course covers method and craft. Three things turn it into a training programme that holds up in your organisation:

1. Bind it to your actual hardware. Add an OEM annex per charger model you operate — fault code tables, torque values, capacitor discharge times, service procedures, spare part numbers. That is the layer this course deliberately does not guess at.
2. Bind it to your state. Application forms, DISCOM process, tariff category, inspection requirements and approval timelines differ by state and change often. Keep them as a living annex with the source document and the date it was pulled.
3. Bind it to your incidents. Every repeat fault should end as a new line in module 13 and, where the cause was upstream, a new check in modules 07 to 11. A training document that does not absorb field experience goes stale in a year.